Calibration...are we just fooling ourselves? [Text View]

stanger89

02-23-09, 12:46 PM

I think we use the term "Calibration" a bit too loosely around here.

When you calibrate an instrument it generally means you will change it to a specified standard (or be able to quantify the difference).

Lets see,
We have standards for proportion of colors to produce white
We have standards for location of primaries and secondaries
We have standards for the peak brightness
We have standards for the relationship of input stim to output brightness
We have "standards" for how inputs should be interpreted
We have measurements quantifying how far off from those standards we are (delta E)
Heck, I think if you dig you can even find standards for contrast ratio (I think DCI spec calls that out)

When someone does a full "calibration" the end result is something that is (most likely) closer to those standards than before, and with a quantifiable error to those standards. That seems to meet your definition of what calibration is.

In other words it will do exactly what the other instrument will do.

With projectors this is simply impossible.

The problem with projectors is not that you can't calibrate them, it's that many projectors don't have the adjustments required to get them within some accepted delta of the specified standards. But that doesn't mean they can't be calibrated.

Can you make two projectors of different technology look identical? NO
Can you make two projectors of the same technology look identical? NO
Can you make two of the same projector look identical? NO

Can you make two measurement instruments identical? NO The best you can do is calibrate them to within an accepted delta of the specified standard.

Take a projector with a full CMS and you truely can properly "calibrate" it to the standards. You can get greyscale to track within x dE across the luminance range, you can get primaries located within y dE, you can get peak brightness to be 16 ftL +/- z ftL.

At best about all you can do is "tune" the projector. Maybe that is even too much to claim.

Your thoughts?

I think calibration is absolutely, 100% the right term. It's a process that attempts to get the projector to match given standards for reproduction. If people want to "tune" to their tastes afterwards that's their choice and their perogative. If people want to not do any calibration at all, that's also their choice, and if people want to do only a "basic" calibration (getting the projector setup interpretting inputs correctly, ie brightness/contrast/color/tint calibration) that's fine as well.

Just because our standards, adjustments, and techniques may not achieve "the same level" of precision that some may expect from calibrating scientific measurement instrument, does not mean it's not "calibration". Even in the scientific area, calibration precisions differ based on use.

calibrate

1: to ascertain the caliber of (as a thermometer tube)
2: to determine, rectify, or mark the graduations of (as a thermometer tube)
3: to standardize (as a measuring instrument) by determining the deviation from a standard so as to ascertain the proper correction factors
4: to adjust precisely for a particular function
5: to measure precisely ; especially : to measure against a standard

3-5 all seem to apply directly to what we do when we "calibrate" a projector.

Art Sonneborn

02-23-09, 12:51 PM

Art Sonneborn

02-23-09, 01:01 PM

At best about all you can do is "tune" the projector. Maybe that is even too much to claim.

Your thoughts?

My thoughts... your conclusion is false.

Art

anirbana

02-23-09, 01:11 PM

Good thread.

Sometimes I think ignorance is bliss as it does not take away the enjoyment of the actual content. Like when I watch Shawshank, I get immersed into the movie but if i look for too much on the presentation I may never enjoy the movie.

But Calibration is fun and helps to achieve close to a standard.

HogPilot

02-23-09, 02:06 PM

Some time spent in the calibration forum would have probably prevented the creation of this thread in the first place. No one claims calibration makes displays look identical; the goal is to remove one's display as a variable (as much as possible) that can cause the material to deviate from what was seen during the mastering process.

As the end user, anyone can choose to have their display adhere to those standards or to paint the screen as they see fit. If your goal is minimal display interference and accurate (as possible) reproduction, calibrating one's display is a must. But everyone is different.

Varrius

02-23-09, 02:17 PM

Some time spent in the calibration forum would have probably prevented the creation of this thread in the first place.

Uh, Tryg has been a member for 9 years with almost 10k posts. I'm fairly sure he's ventured into the calibration forum a time or two. If I do find one thing to disagree with in this thread, this is definately it. Of course, if I were to disagree with any of the rest of it, it would be just about as relevant as enimas of wolverines are. Personally, I just think Tryg was bored and doing his best George Carlin impression, poor guy.

Josh Z

02-23-09, 02:20 PM

You cant imagine how much better a wolverine can be once it has a good enema :)

Now that you mention it, he does look kind of constipated. :D

http://www.etonline.com/media/photo/2008/02/41716/400_wolverine_080214_michaelmuller.jpg

HogPilot

02-23-09, 02:24 PM

Uh, Tryg has been a member for 9 years with almost 10k posts. I'm fairly sure he's ventured into the calibration forum a time or two. If I do find one thing to disagree with in this thread, this is definately it. Of course, if I were to disagree with any of the rest of it, it would be just about as relevant as enimas of wolverines are. Personally, I just think Tryg was bored and doing his best George Carlin impression, poor guy.

I'm well aware of who he is and how long he's been a member - as I'm sure you'll find as you spend some more time here, longevity of membership and post count are completely inconsequential to whether someone knows what they're talking about or not.

Gibbah

02-23-09, 02:49 PM

The way I see it, calibration is not done to get the PERFECT picture. It is done in order to be able to see the picture the way the director wanted it. Without calibration you never really know how the picture was intended to look like.

To get the picture the way a director makes it you both must have the same reference and the only way to achive that is to calibrate to follow a standard. Only when you speak the same language can you REALLY understand what he is trying to tell you =)

Varrius

02-23-09, 03:02 PM

I'm well aware of who he is and how long he's been a member - as I'm sure you'll find as you spend some more time here, longevity of membership and post count are completely inconsequential to whether someone knows what they're talking about or not.

Your starting to make me want to discuss enima's... and wolverines.

Jason Turk

02-23-09, 03:18 PM

"Calibration" doesn't equal perfection. It merely means that it is more dialed in to known standards. Of course one cannot make a DLP look like a DILA, etc... but that isn't what calibration is anyways. If it were, most people would be able to make a digital look like a film projector. :)

jedi.master.dre

02-23-09, 03:22 PM

HogPilot

02-23-09, 03:43 PM

Your starting to make me want to discuss enima's... and wolverines.

Your post count defense and misuse of the apostrophe were more than enough.

Varrius

02-23-09, 04:14 PM

Your post count defense and misuse of the apostrophe were more than enough.

Haha. I've been trying to be nice to Tryg, for starting such a discussion, in the wrong section no less (of course I noticed, before you pointed it out).

I think it's quite obvious what the purpose of the thread was, but then I don't really know Tryg, so perhaps I'm wrong. I was simply making light of your serious response to what I would consider a not-so-serious thread (at least I don't consider his approach all that serious).

And then you had to go and correct my grammar, in my not-so-serious response, and all is explained. I really do enjoy these forums, and the wealth of knowledge here is unbelievable, but does that mean we can't have some fun once in a while? I hope not.

R Harkness

02-23-09, 04:17 PM

The thing is calibration is so often conflated with a "better" image. I often see claims that after calibration the image looked so much better in every respect, including "so much more 3D etc." But I have yet to experience a calibrated image that looks significantly "better" to my eyes than what I can typically get out of tweaking the image myself. By that I mean an over all impression of richness, clarity, dimensionality etc.

I had a calibration done on my plasma (years ago by a very well respected calibrator) and I had expected, or at least hoped, from all the hype about ISF calibration to have the experience that so many reported "Wow...I never knew my display could look this good until I had it calibrated!" But instead I actually found the image less impressive and less believable over my own tweaked settings. I realised then that calibration was about "accuracy to a standard" and not in of itself "a better or more believable image."

Definitely when talking about displays that are out of whack color/contrast-wise a calibration to industry standards is also going to have the result of a "better" more believable image. But it's not precisely the goal. As people have pointed out, being more accurate to the source material does not always mean a more pleasing or a more believable image. Now, I did later come to appreciate the ISF settings on my plasma, once I realised that pure image realism wasn't going to be my ultimate goal; I started to appreciate the more film like image. But if I wanted to impress someone with a sheer looking-through-a-window realism, my settings were definitely the ones I'd choose.

Now in terms of accuracy-to-the-source (e.g. ISF standards) there's no way my eyes can compete with a calibrationist and his equipment. Presuming the projector has a good CMS or using an outboard VP like the Lumagen Radiance, the main thing I believe I'd get out of a professional calibration, in terms of what I'd notice that I couldn't get myself, is more accurate colors. The OOTB settings of projectors like the JVC are clearly skewed such that some colors just strike me as "wrong," e.g. the greens of vegetation. When a color strikes me as wrong-enough, it can distract me. So getting the colors in line would be a nice thing to have.

But just as an example when I viewed the THX mode on the RS20 it did look somewhat more accurate, but also a bit bland and slightly flat compared to other settings. After seeing the THX setting for a while I'd put it into an OOTB setting and frankly in many cases I preferred the OOTB setting. Why? Because, while it gave up ultimate color accuracy to the THX mode, it excelled in other areas like image dynamics, apparent contrast, punch, and in that way it had an added "wow" and believability factor over the THX mode. What I'd like is to have both: a dynamic image with accurate colors that doesn't look bland.

I don't mind possibly throwing the money-dice one more time on a calibration to see if I end up happier this time.

Jason Turk

02-23-09, 04:44 PM

"Better" is indeed in the eye of the beholder. I think again the main purpose is that many don't know how an image technically should look, thus preferring to have it dialed in to a certain set of standards. Of course as you said, it doesn't mean one is going to like that look. Case in point the vast majority of us are used to poorly setup displays at various stores (Best Buy, Walmart, etc...)...almost all on display are just taken out and put on display. When one then sees a "calibrated" unit, it can go either way as to whether it is better or worse. What I can say is that usually when one is used to a display setup as intended, switching back can be hard to do.

HogPilot

02-23-09, 04:53 PM

Haha. I've been trying to be nice to Tryg, for starting such a discussion, in the wrong section no less (of course I noticed, before you pointed it out).

I think it's quite obvious what the purpose of the thread was, but then I don't really know Tryg, so perhaps I'm wrong. I was simply making light of your serious response to what I would consider a not-so-serious thread (at least I don't consider his approach all that serious).

And then you had to go and correct my grammar, in my not-so-serious response, and all is explained. I really do enjoy these forums, and the wealth of knowledge here is unbelievable, but does that mean we can't have some fun once in a while? I hope not.

The way I see it, if he meant to be humorous he did a horrible job of communicating that - this is the internet after all.

If he was serious, I gave him a serious reply. Regardless of the intent, this thread is flame bait at best, so I wouldn't mind seeing it disappear.

R Harkness

02-23-09, 05:01 PM

Just to be clear: the typical OOTB setting of most displays, like those you see in the stores, make me want to throw up. I always tweak a display to look much smoother and more believable, with the deepest black levels, most apparent contrast and biggest helping of shadow detail I can get. And I don't like an over-saturated image. But most of that I can do by eye. What I can't get is the most accurate colors by eye.

WOLVERNOLE

02-23-09, 06:23 PM

Hey !!! I resemble that remark !;)

QUOTE=R Harkness;15891903]I'm of two minds about it. I'd like the option of a well calibrated image. At the same time I think I'd sooner spend my time trying to give enemas to wolverines rather than spend as much time in the rats nest of techno-hell, as some folks do on calibrating their display.
[/QUOTE]

mdputnam

02-23-09, 07:50 PM

The way I see it, calibration is not done to get the PERFECT picture. It is done in order to be able to see the picture the way the director wanted it.

Most movie directors, cinematographers, and directors of photography work in a film medium. Their focus is how a finished piece will look on film, not what it looks like once the thing is telecined and crammed into a limited 8 bit color gaumet. I doubt any DP has told a director "we better not film it that way it won't look good when it gets released on DVD." :D

Art Sonneborn

02-23-09, 08:15 PM

R Harkness

02-23-09, 08:18 PM

Most movie directors, cinematographers, and directors of photography work in a film medium. Their focus is how a finished piece will look on film, not what it looks like once the thing is telecined and crammed into a limited 8 bit color gaumet. I doubt any DP has told a director "we better not film it that way it won't look good when it gets released on DVD." :D

Well, DPs and directors have decided to film things certain ways knowing the movie will eventually end up on TV (especially before DVD and widescreen TVs became ubiquitous).

I can tell you in the sound end of things we are cognizant of how the content will tend to be viewed on average. For instance we'll often play-back 5.1 mixes on the tiny monitors ("cans") that more closely mimic the quality most people will hear on their TV, and then adjust the mix accordingly.

(To my dismay, sometimes).

HoustonHoyaFan

02-23-09, 09:06 PM

Just so I know,you are saying they don't care about what the home media presentation looks like ?Increasingly Directors do care about home presentation. FF Coppola (Dracula,...) and Peter Jackon (LOTR, ...) have had contractural oversight ofer the home presentation process. There is growing sense that the DVD/Blu ray market is a different medium and needs creative vision oversight to properly assure Director/DOP intent. Coppola even gave his HD Colorist top level credit on FFC Dracula!

Glimmie

02-23-09, 10:43 PM

Most movie directors, cinematographers, and directors of photography work in a film medium. Their focus is how a finished piece will look on film, not what it looks like once the thing is telecined and crammed into a limited 8 bit color gaumet. I doubt any DP has told a director "we better not film it that way it won't look good when it gets released on DVD." :D

Yes what you say is true, they don't compromise the film image for the benefit of home video.

However the home video versioning is also done to eaxcting standards and critical subjective evaluation often by the DP. Yes, you can only do so much with HD video but that doesn't mean is just thrown to the wolves.

Glimmie

02-23-09, 10:48 PM

Colorimetry can be calibrated using a spectroradiometer like a PR650 that itself is periodically calibrated to known standards. Some displays can be worked to those standards which go back to the wellhead. So, at least in that regard ,it is calibration.

If one wants this ,cares about it or even perhaps doesn't even like it, is a different subject.

Art

Absolutly. But how many of the home theater calibration techs has one of those? At $35,000 I doubt any HT only installation company has one. Plus like any sophisicated test instrument, the more accurate and professional it is, the more you need to know to operate it and understand it's reporting.

As someone who works around the highest end video monitoring and projection equipment I find this whole calibration craze of late out of control. Many of these consumer services leave the unit in worse condition than the factory.

Yeah right, I'm going to trust some hack from Geek Squad to know anything about monitor calibration.

CT_Wiebe

02-24-09, 01:30 AM

Glimmie -- I don't think anyone on these forums thinks that a Geek Squad "hack" qualifies as a "real certified calibrator" - except for the complete novices. On the other hand there does seem to be a lot of AVS members who appear to get carried away with the calibration theme (as Tryg is trying to point out, I think). I agree that it seems to be "out of control", as you put it.

I tend to agree with R Harkness, that spending a lot of time and effort tweaking a display (seemingly on a daily basis) is a total waste of time. I, too, would rather spend my time enjoying my PJ (currently a 1080p at 11' from a 106" screen).

Ever since I got my first projector, some 20+ years ago, I have bought and used calibration discs - Joe Kane's "A Video Standard" on laserdisc, and subsequent DVD iterations, etc. (I have 7 "calibration discs total). I have done my own "rough adjustments" to improve on the OOTB settings for a more pleasing display (to my eyes, with my screen and my equipment, in my room environment). I might be able to get a better picture, if I bought and used measurement instrumentation (or hired a professional ISF certified calibrator). However, I'd rather watch my PJ than spending a lot of time, and/or money, for another 5%-10% improvement (maybe).

Is that calibration? In Tryg's terms, NO (even though that's the term we loosely associate with the process). Is it an improvement in picture quality? Definitely!

DaGamePimp

02-24-09, 03:56 AM

Without set standards where would we be , would we all be better off if we owned projectors incapable of D65/6500K ? I think the only way I would be fooling myself is if I tried to convince my eyes that I would rather watch a last gen JVC for the sake of purchase justification (then I'd be pulling the wool over my own eyes) .

Bottom line here is it's all preference anyway , some of us are neat tidy perfectionists that floss daily while others could harvest dust farms on their av equipment ... neither is right or wrong , it's a personal choice and one that I am thankful to have :) .

Disclaimer : Yes sometimes those inaccurate images can look stunning but I find it's like Vegas with all the pretty lights , I luv to visit but I wouldn't want to live there ;) .

- Jason

[ Just my opinion , to each their own ... that's the beauty of it ]

CaspianM

02-24-09, 06:50 AM

Art Sonneborn

02-24-09, 09:37 AM

Absolutly. But how many of the home theater calibration techs has one of those? At $35,000 I doubt any HT only installation company has one. Plus like any sophisicated test instrument, the more accurate and professional it is, the more you need to know to operate it and understand it's reporting.

As someone who works around the highest end video monitoring and projection equipment I find this whole calibration craze of late out of control. Many of these consumer services leave the unit in worse condition than the factory.

Yeah right, I'm going to trust some hack from Geek Squad to know anything about monitor calibration.

Maybe back at you but the OP said that what is done can't be called calibration which is false as a sweeping statement. Yes ,if all I do is buy Joe Kane's disc and set a few controls but that isn't what the OP said. He came right out and called the whole process into question. Of course as usual the motives are obvious for the thread.

Art

Art Sonneborn

02-24-09, 09:51 AM

Calibration of a pj to a given standard i.e., D65 & proper color ought to be a "matter of fact" not a "matter of opinion" whether or not it can be done correctly with a given PJ.
Let's not give up on that! That is the only common ground we all should care about imo.

Great point ! Just because a given device can't achieve standards doesn't mean it isn't worthwhile to reach for it.

Some projectors are capable of essentially perfect color,for example,so go for it. Others might be short some place but can get better by using standards of calibration and setting as close as possible.

Art

Jason Turk

02-24-09, 09:56 AM

Jason Turk

02-24-09, 09:57 AM

I personally am an amateur filmmaker...I prefer to shoot on VHS-C. :D

Art Sonneborn

02-24-09, 10:14 AM

I personally am an amateur filmmaker...I prefer to shoot on VHS-C. :D

We respect you for coming out.;):D

Art

Varrius

02-24-09, 10:27 AM

Regardless of the intent, this thread is flame bait at best...

That's more or less what I was getting at.

I believe the merits of calibration are a worthwhile discussion, I just felt his approach was intending a flame war. Fortunately, the discussion has turned more productive, so I will leave it at that.

Jason Turk

02-24-09, 10:40 AM

we respect you for coming out.;):d

art

:d

GeorgeAB

02-24-09, 11:01 AM

Regarding the validity of consumer display calibration:
http://www.avsforum.com/avs-vb/showthread.php?t=1021933

An award winning cinematographer's perspective:
http://www.jkpi.net/allen_daviau.php

Regarding reliance on your eyes when adjusting for image fidelity:
http://www.avsforum.com/avs-vb/showthread.php?t=849430

Best regards and beautiful pictures,
G. Alan Brown, President
CinemaQuest, Inc.
A Lion AV Consultants Affiliate

"Advancing the art and science of electronic imaging"

Kelvin1965S

02-24-09, 11:23 AM

I believe the merits of calibration are a worthwhile discussion, I just felt his approach was intending a flame war.

My thoughts exactly, especially as the OP hasn't posted a single reply, I thought it more akin to trolling. :rolleyes: I choose not to pay much attention to the OP's comments generally, hence why I prefer to watch a calibrated display(s) and also why I don't have a HP screen.

High post count, it seems, is no guarentee to providing useful indication. ;)

Art Sonneborn

02-24-09, 11:23 AM

GeorgeAB

02-24-09, 11:38 AM

I was atcually surprised that no calibrators have chimed in till now on this. The OP essentially negated the validity of your whole profession.

Art
I don't come to any section of the forum on a frequent basis, other than the display calibration section. That seems to be the one section I have a consistent interest in, and I've found it to be the least encumbered by erroneous information. I don't have much in common with viewers who don't value image fidelity above individual preference.

emdawgz1

02-24-09, 11:42 AM

Calibration is about image FIDELITY.

From Merriam Webster.
Fidelity.
2: the degree to which an electronic device (as a record player, radio, or television) accurately reproduces its effect (as sound or picture)

So while a picture may look good to you, is it reproducing an image ACCURATELY??? Calibration helps this matter. Is it the end of civilization if your picture is inaccurate? No but you might like it to be accurate.If you do, calibrate.

Lawguy

02-24-09, 11:51 AM

It is naive to think that a calibrated display captures film makers' intent.

There is value in calibrating but calibration has been sold as accomplishing something that it doesn't. I wish that calibrators were more open about this. I think that many actually believe it or strongly want to believe it.

GeorgeAB

02-24-09, 11:57 AM

It is naive to think that a calibrated display captures film makers' intent.
This is what is known as setting up a straw man. Should I prefer your perspective on this issue, or the award winning cinematographer referenced above (post 48)?

lcaillo

02-24-09, 11:58 AM

How have calibrators not been open? Who is selling it as something that it isn't? The best way to capture the film maker's intent for how an image should appear is to follow the industry accepted standards throughout the process of distribution and display. How would one better accomplish the goal of revealing the intent of the maker of programming with regard to presentation than adhering to standards?

The fact is that the best calibration professionals in the world have been extremely open about what they do, what can be done, and educating the AV community regarding the process of display calibration. There are some who propose nonsense, but that is true in any profession. As a whole, I have found the most serious calibration specialists to be very open and helpful. The mission has largely been one of teaching people what the standards are and how to better realize them.

If you have issue with something that someone has said in promoting calibration, please deal with that person by name and the offending statements directly. Please do not generalize unfairly. Certainly there are some calibrators who are secretive about what they do, and some who get a bit condescending at times, but in general they are among the most helpful people on AVS.

mdputnam

02-24-09, 12:03 PM

Just so I know,you are saying they don't care about what the home media presentation looks like ?
Art

I'm not involved in the telecine business but, my very limited observation is that few directors are directly involved in the telecine process and the final product often does not require approval by the director. Now, that is not to say they don't care or that it doesn't happen, it appears to be mostly an economic issue. For example, here is what Lou Levinson said in his 2005 interview with Suzanne Hodges in WSR:

To what extent do cinematographers and directors
have an interest in the telecine process?
LL: There is a lot of interest. Some of them will
make the time to be involved. One factor that gets
into it is the economic factor. The studios are well
aware that if you get a cinematographer or director
involved, it’s going to take longer and cost more.
There it depends on the stature of the project as to
what kind of involvement the studios would like to
see. There are cinematographers out there who
are more than just seriously interested and make a
point to be there because it is what they’re being
judged by.

For anyone interested in film and digital production techniques I would recommend a subscription to the ACS magazine "American Cinematographer" it's only $25/yr. I believe you can still receive a complementary digital copy of the November issue for free, check it out (http://www.theasc.com/magazine_dynamic/February2009/current.php).

HoustonHoyaFan

02-24-09, 12:04 PM

It is naive to think that a calibrated display captures film makers' intent...Why?

Kevin 3000

02-24-09, 12:17 PM

QUOTE"Video displays used by program professionals are designed, manufactured and calibrated to tightly adhere to industry standards. They include in their design certain features that allow technicians to adjust them periodically (sometimes several times each day), using reference test signals, to insure picture accuracy."

If the pros have to do this on their £££££ displays to keep within specs unless you can calibrate yourself daily/weekly your own display is going to be off, AFTER BEING CALIBRATED, as soon as you have watched a couple of films?

lcaillo

02-24-09, 12:23 PM

I know of no professionals that calibrate their displays several times a day.

No display is perfect, even after being calibrated. The point is that once calibrated, variances will occur within a range closer to where the display should be to begin with. An uncalibrated display will likely not even be close to where it should be, and variances can occur that make it even worse.

If you adjust anything, it will drift or vary. Would you like the alignment of your car to vary around the correct alignment or around an alignment that causes it to pull to one side severely or wear your tires unevenly?

Lawguy

02-24-09, 12:24 PM

lcaillo

02-24-09, 12:27 PM

Lawguy

02-24-09, 12:35 PM

What hype? Once again, if you want to reproduce what was intended, how better to do so than to adhere to accepted standards? You admit yourself that this is the case. Who is hyping anything? There simply is no better way to see what was intended than to calibrate to accepted standards. What is hype about that?

I can't describe this any better than I have.

I suggest you reread what I wrote because the answers to your questions are all addressed in my last post.

A few years ago we had (I think) a cinematographer of some very respected movies post on these forums. He recounted how much effort he put into in making sure the screening of a premiere of a particular movie looked good and that, as far as he could tell, it would never look as good again.

These guys have no illusions about what happens to their work after they are done with it.

lcaillo

02-24-09, 12:38 PM

We do not agree on the degree to which your concern is "hyped" in the marketing of calibration.

Lawguy

02-24-09, 12:46 PM

We do not agree on the degree to which your concern is "hyped" in the marketing of calibration.

My only point is that it is hype to talk about calibration as capturing filmmaker's intent. Even though this is not true, it is commonly said. Again, I think many people believe it because some well-respected people in the industry speak in this kind of language.

Again, calibration has value and is a good thing but it should not be sold as doing something that it cannot.

R Harkness

02-24-09, 12:48 PM

I've also found the calibrationists who visit AVS to generally be quite reasonable, helpful and open in their views as to what they are trying to achieve.

As I said, once I grew less interested in pure "realism" for it's own sake, I grew to like my ISF'd settings on my plasma for it's more cinematic look.

But the expectations for what an ISF calibration will do for an image gets a bit misty, mixed up with legitimate claims of improved image accuracy to the source, and legitimate claims that the image will likely look more believable than any OOTB settings, with, in my view, more dubious claims that it takes an ISF calibration to get an image that looks "the best" or "more 3D" or more detail, or more realistic etc.

I compared the ISF'd settings against my own on my plasma (and in some other displays that had been ISF'd vs my own tweaking) and there were no gains I could spot whatsoever in clarity, image detail, color detail, shadow detail, black levels, dimensionality etc. In fact, on those issues my settings looked "better" to my eyes.

But the ISF settings I'm sure were more accurate to the source, at least to the source colors and color temperature. And it looked more film-like.

I'm not sure how to square this with other people who have different experiences, like "I tried calibrating myself but the image after ISF calibration looks waaay better."

I firmly support the ISF guys in how they offer consumers a more accurate image (and often a better image).

Art Sonneborn

02-24-09, 12:53 PM

lcaillo

02-24-09, 12:54 PM

In what way is calibrating to industry standards not going to be more likely to allow a display to render the intent of the creator of the material? Your point is very clear but it makes no sense. If one wants to see what a director, editor, or cinematographer intended, there is no alternative other than to properly calibrate the display. An uncalibrated display will produce an image that is farther from the intent, unless one just happens to have a display with settings that under some strange set of conditions just happens to compensate for any anomolies in the production or distribution of the material. While there are many variables involved in these processes, there is simply no reasonable way to try to reproduce what was intended other than to calibrate.

You can argue that any aspect of marketing something is hype. Please cite some examples of how calibrators use this aspect of calibration in an unfair manner? Would you also state your suggested alternative?

Lawguy

02-24-09, 12:58 PM

You act like if we can't achieve perfection that we should give up getting as close as possible. I certainly don't subscribe to this.The alternative to calibration is further inaccuracy.

Art

That is not my opinion and I have tried to make that clear by saying repeatedly that calibration is a good thing. I will qualify this by stating that you also need a good calibrator, who isn't mechanically taking measurements and adjusting controls.

theelviscerator

02-24-09, 01:06 PM

Can a colorblind person appreciate calibration?

I would say most people have some defect in their own ability to see color correctly.

I myself have been through numerous color blindness tests as I work in the electrical/electronic fields....(used to anyway).....

So in other words, its probably best to calibrate to a standard...then tweak it towards a "house eye curve" as a matter of practicality........

One day a color analyzer tool will use the output from your optic nerve to calibrate.....then that really IS a good fleshtone...red/green balance!

Lawguy

02-24-09, 01:07 PM

In what way is calibrating to industry standards not going to be more likely to allow a display to render the intent of the creator of the material?

We can't calibrate to the right industry standards because neither our equipment or our sources were designed to do it. It is as simple as that and I am repeating myself now.

I am no expert in how calibrators market themselves. There are several examples in this thread alone of people expressly stating or implying that watching material on a calibrated display reveals the filmmakers' intent. Are those people accurate when they say these things?

The truth is much more complicated than what people appear to want to believe. If it were me and I walked around all day with an overly simplistic view of the something, I would welcome being corrected. A lot of people don't want to be.

Drexler

02-24-09, 01:31 PM

LG,

Isn't this just semantics? Yes, we could never exactly reproduce the film from the premiere night. However, we could get as close as is darn possible on the display in question.

Do you think there is any better way at getting as close as possible to the directors intent than calibrating? If not, why not claim that calibration is the best way to go if you want to achive this?

Best regards

HoustonHoyaFan

02-24-09, 01:41 PM

We can't calibrate to the right industry standards because neither our equipment or our sources were designed to do it...You are not likely to find 1 industry expert witness who would support your case counseller. :D

Try telling Coppola or Jackson that the time, effort, passion, and money that they have spent in creating home video (Rec 601/Rec 709) interpretations of their work is not the "right industry standards".

I think most calibrators are selling "conforming to industry standards" which is a necessary but not sufficient step towards achieving creator intent.

lcaillo

02-24-09, 01:46 PM

We can't calibrate to the right industry standards because neither our equipment or our sources were designed to do it.

Cite some references that support this statement, please.

GeorgeAB

02-24-09, 01:47 PM

Lawguy,

Should I prefer your perspective on this issue, or the award winning cinematographer referenced above (post 48)? You are trolling here.

Lawguy

02-24-09, 03:19 PM

Lawguy,

Should I prefer your perspective on this issue, or the award winning cinematographer referenced above (post 48)? You are trolling here.

I don't care. My reasoning speaks for itself. If someone agrees with it. Fine. If they don't agree, fine. No one has suggested that any fact that I cite to is incorrect. If I misstated anything, please point it out.

I never meant to suggest that some filmmakers are at times involved in the DVD or bluray creation process. Of course there are and some disks are obviously quite good. This fact does not detract from anything that I have written, in my opinion.

I don't know how I am trolling in a thread that I did not create that is captioned " Calibration...are we just fooling ourselves?"

lcaillo

02-24-09, 03:34 PM

No one has suggested that any fact that I cite to is incorrect. If I misstated anything, please point it out.

What facts have you cited? Sorry, but I missed them. I have seen nothing but your opinions and misconceptions. Those have been pointed out. There are clear standards from a number of industry organizations that are recognized as being appropriate to follow in the production and display of film and images. Those of us who calibrate displays do our best to adhere to those standards, as is the case in most production.

Lawguy

02-24-09, 03:40 PM

What facts have you cited? Sorry, but I missed them. I have seen nothing but your opinions and misconceptions. Those have been pointed out. There are clear standards from a number of industry organizations that are recognized as being appropriate to follow in the production and display of film and images. Those of us who calibrate displays do our best to adhere to those standards, as is the case in most production.

Either you don't read very carefully or you are pretending not to because you completely miss my point.

mlang46

02-24-09, 03:48 PM

There are standards which are not perfect that the company making a video disk expects the projector to respond to. I don't even think it is very subtle. I think a calibrated projector which has a good CMS looks better and the colors look more natural than an uncalibrated projector.

Once you have had an uncalibrated projector calibrated and seen the difference it makes to the image than you will always have your projectors calibrated.

Unlike spending 8 times the money for a 3 chip dlp as opposed to a single chip Dlp, it really does make a difference in the quality of the image you see.

Lawguy

02-24-09, 03:53 PM

There are standards which are not perfect that the company making a video disk expects the projector to respond to. I don't even think it is very subtle. I think a calibrated projector which has a good CMS looks better and the colors look more natural than an uncalibrated projector.

Once you have had an uncalibrated projector calibrated and seen the difference it makes to the image than you will always have your projectors calibrated.

I agree with this almost completely and this is the most compelling argument for calibration. There is no need to oversell it as capturing "filmmakers' intent" or other such nonsense.

Art Sonneborn

02-24-09, 03:57 PM

Unlike spending 8 times the money for a 3 chip dlp as opposed to a single chip Dlp, it really does make a difference in the quality of the image you see.

You were doing so well...then.:( But of course this point uncovered some of your intent in other threads.

Art

Art Sonneborn

02-24-09, 04:01 PM

It is naive to think that a calibrated display captures film makers' intent.

I think there are enough film makers either quoted here or some of us know who would say that a calibrated display is the best way we have of getting there. It doesn't represent perfection since film and digital displays and storage systems aren't the same but no doubt better than just seasoning to taste.

Art

lcaillo

02-24-09, 04:02 PM

I still have not heard from you Lawguy, regarding how you would propose to best display an image to reveal what the maker intended us to see. Should we calibrate to the 196M standard for film theaters? If you want to make the point that there should be an option for that color temp on a calibrated display, I can buy that. My question is, what do you suggest IS the best practice for display of film originated content? SMPTE would seem to be pretty clear that D65 is the preferred standard for monitors.

coldmachine

02-24-09, 04:08 PM

I think something thats being lost here is that, in Post, a video master is created, that adjusts the color of the film to accommodate the rec709 gamut. This is what a well calibrated machine will help us approach.

This isn't just a simple colorimetric conversion, as that can result in "out of gamut" clipping.

An LUT can be applied to preserve as much of the look as possible. The transfer function is adjusted for color appearance and "soft clipping" is applied to "out of gamut" colors to preserve hue.

Lawguy

02-24-09, 04:12 PM

SOWK

02-24-09, 04:16 PM

Ok then... Now bring in animated films into the mix... :)

They are much closer to being perfect, and would need a "standard" to look just like the developer intended.

Sometimes I absolutely Love my PC and PC Gaming background. ;)

coldmachine

02-24-09, 04:20 PM

Unlike spending 8 times the money for a 3 chip dlp as opposed to a single chip Dlp, it really does make a difference in the quality of the image you see.

Thats a rather ill informed statement, but very much in keeping with your current agenda.

Lawguy

02-24-09, 04:22 PM

I think something thats being lost here is that, in post, a video master is created, that adjusts the color of the film to accommodate the rec709 gamut. This is what a well calibrated machine will help us approach.

This isn't just a simple colorimetric conversion, as that can result in "out of gamut" clipping.

An LUT can be applied to preserve as much of the look as possible. The transfer function is adjusted for color appearance and "soft clipping" is applied to "out of gamut" colors to preserve hue.

Everything you write is true. Calibration is a good thing. But, for those in a profession that concerns itself with accurracy, they should be careful to describe what they can accomplish, well . . . accurately.

Lawguy

02-24-09, 04:24 PM

coldmachine

02-24-09, 04:30 PM

Everything you write is true. Calibration is a good thing. But, for those in a profession that concerns itself with accurracy, they should be careful to describe what they can accomplish, well . . . accurately.

Thanks.

I think there is also a certain amount of misunderstanding on the part of many domestic viewers with regard to what they are actually viewing and how it is derived. This is even more prevalent when we consider that most film based images are a result of a D.I. process anyway.

SOWK

02-24-09, 04:30 PM

Thats a rather ill informed statement, but very much in keeping with your current agenda.

Besides possible deeper color saturation, what are the "Significant" improvements that 3 chip DLP can provide over a very good 1 chip DLP?

Art Sonneborn

02-24-09, 04:32 PM

For those who want to read a little more about actual filmmakers' intent, read this (http://www.avsforum.com/avs-vb/showthread.php?p=13659472&highlight=#post13659472) post, this (http://www.avsforum.com/avs-vb/showthread.php?p=13658710&highlight=#post13658710) post, this (http://www.avsforum.com/avs-vb/showthread.php?p=13659550&highlight=#post13659550) post, this (http://www.avsforum.com/avs-vb/showthread.php?p=13652486&highlight=#post13652486)post and this (http://www.avsforum.com/avs-vb/showthread.php?p=12495508&highlight=#post12495508)post.

These are all by Rob Hahn (http://www.imdb.com/name/nm0353963/),

I know Rob,I doubt that that series of posts ,despite their grim tone,he would say anything but an uncalibrated display would be anything more than adding insult to injury.

I think you are off base and you can't backpeddle enough except to simply say that in fact a calibrated display would do a better job at capturing the intent than would a seasoned to taste display.

Art

coldmachine

02-24-09, 04:36 PM

Besides possible deeper color saturation, what are the "Significant" improvements that 3 chip DLP can provide over a very good 1 chip DLP?

Start another thread, preferably in the forum more native to 3 chip DLP, and I'll explain it to you. I have no desire to derail this thread, or encourage the douche fest that would ensue. Im not referring to your good self there.

SOWK

02-24-09, 04:37 PM

I will, thanks.

Lawguy

02-24-09, 04:45 PM

I know Rob,I doubt that that series of posts ,despite there grim tone,he would say anything but an uncalibrated display would be anything more than adding insult to injury.

I think you are off base and you can't backpeddle enough except to simply say that in fact a calibrated display would do a better job at capturing the intent than would a seasoned to taste display.

Art

Art

It is not my intention to backpeddle.

It is my intention to illustrate how what the concept of "intent" is, is in many ways illusory and it is certainly not something that can be assuredly captured merely through calibration.

For example, in the post where Rob speaks of his temptation to compensate for the fact that most displays out there have their white point set incorrectly, let's assume that he gave in to his temptation and mastered a particular disk so that it looked best on that uncalibrated display. In that case, what is the intent?

Nobody seems to care about these kinds of angel dancing on the head of a pin issues but me, so I'll just join the crowd and chalk up all that edge enhancement I see everywhere to filmmakers' intent and call it a day.

Art Sonneborn

02-24-09, 04:52 PM

Nobody seems to care about these kinds of angel dancing on the head of a pin issues but me, so I'll just join the crowd and chalk up all that edge enhancement I see everywhere to filmmakers' intent and call it a day.

You actually know better than this and the hyperbole does not do you justice.You are out on the end of a limb and you are sawing it off yourself.

Art

joeycalda

02-24-09, 04:55 PM

To some extent calibration is totally subjective. Case in point, the new Samsung Led based LCD. The picture when using dynamic color and motion enhancement on medium to high looks almost too good and sort of like HD video and a more staged looking movie set . I prefer the settings on low enhancement and standard mode, so it seems more like the movies that I have grown up with. In twenty years what a movie should look like might be totally different than yesterday.

The standards, cameras and displays are constantly changing so what's correct today is not necessarily correct tommorow. If that makes any sense.

JC
JC

Lawguy

02-24-09, 04:56 PM

You actually know better than this and the hyperbole does not do you justice.You are out on the end of a limb and you are sawing it off yourself.

Art

I really don't see how. Edge enhancement, DNR, the whole cast of evils are put on our disks in the same process involved in setting color and other things.

The process always does a great job with things like color but always screws up with DNR and EE. Makes sense to me. :rolleyes:

Art Sonneborn

02-24-09, 04:59 PM

I really don't see how. Edge enhancement, DNR, the whole cast of evils are put on our disks in the same process involved in setting color and other things.

The process always does a great job with things like color but always screws up with DNR and EE. Makes sense to me. :rolleyes:

Calibration of a display(the thread topic) has essentially nothing to do with DNR or EE (evil by any estimation IMO).Bringing them into the conversation is a red herring.

Art

Lawguy

02-24-09, 05:09 PM

Calibration of a display(the thread topic) has essentially nothing to do with DNR or EE (evil by any estimation IMO).Bringing them into the conversation is a red herring.

Art

Not so. People are claiming that what we see on our disks captures filmmakers' intent. If that is so than EE and DNR are part of that intent.

The truth is that it is patently obvious that what is on our disks often times has nothing at all to do with filmmakers' intent. I don't understand why otherwise smart people don't want to appreciate this fact.

Lawguy

02-24-09, 05:11 PM

Calibration of a display(the thread topic) has essentially nothing to do with DNR or EE (evil by any estimation IMO).Bringing them into the conversation is a red herring.

Art

Not so. People are claiming that what we see on our disks captures filmmakers' intent and this fact justifies strict adherence to standards. If it is so that disks embody intent, than EE and DNR are also part of that intent.

The truth is that it is patently obvious that what is on our disks often times has nothing at all to do with filmmakers' intent. I don't understand why otherwise smart people don't want to appreciate this fact.

Josh Z

02-24-09, 05:21 PM

Lawguy, what is the purpose of your argument? All you've done is set up straw men. No one here but you has argued that Calibration = Director's Intent.

The purpose of calibration is to bring the display as close as possible to the standard used when mastering the video content. Many (most?) displays are not capable of 100% matching that standard. Does that mean you shouldn't even attempt to calibrate them? What would be the alternative? Rather than get as close as possible, don't even bother to try at all?

Basically, the argument you're trying to create here doesn't make any sense at all. I suspect that even you realize as much, but are continuing to defend it out of stubbornness.

None of the rest of us here care. Please move on. This was an interesting thread until you derailed it.

CaspianM

02-24-09, 05:25 PM

We run into poor trasfer all the time but that should not distract us as we also get nice trasfers as well.
All I want is my display be as transparent as it could be.

Lawguy

02-24-09, 05:46 PM

None of the rest of us here care. Please move on. This was an interesting thread until you derailed it.

I am accused of derailing a thread called "Calibration...are we just fooling ourselves?" and I am the one who has the problem here? :rolleyes:

You are wrong to say that people in this thread did not equate calibration with director's intent or state that it provides image fidelity. You can presumably read as well as I can so that fact that you missed these things says something. It was those posts that I was addressing and I think they are actually on topic.

lcaillo

02-24-09, 05:48 PM

Calibration of a display(the thread topic) has essentially nothing to do with DNR or EE (evil by any estimation IMO).Bringing them into the conversation is a red herring.

Art

The whole argument that he makes is a red herring. While some may advertise that calibration will allow one to see the intent of the filmmaker, this is a relatively minor basis for arguing that calibration is useful. That "hype" that LG proposes to exist is minimal. It is little different than claims that calibration reduces energy use. Some will exaggerate calims such as these, but most professionals will qualify such statements and assure that their clients understand how they apply to their context and application.

Rather than discuss the matter intelligently and the standards that may apply, it is easier to claim hype and generalize.

SOWK

02-24-09, 05:49 PM

Lawguy, you skipped over my post about animated movies / video games, they usually are near "perfect" and a calibrated display would allow you to get what the developers wanted you to see. :)

coldmachine

02-24-09, 06:09 PM

Lawguy, you skipped over my post about animated movies / video games, they usually are near "perfect" and a calibrated display would allow you to get what the developers wanted you to see. :)

Thats not the case for animated movies, for the reasons outlined in post #83.

Your post attributes an equivalence to animated movies and video games that doesn't actually exist.

A calibrated display will allow you to see an image closer the the video master.

Art Sonneborn

02-24-09, 06:25 PM

Not so. People are claiming that what we see on our disks captures filmmakers' intent. If that is so than EE and DNR are part of that intent.

The truth is that it is patently obvious that what is on our disks often times has nothing at all to do with filmmakers' intent. I don't understand why otherwise smart people don't want to appreciate this fact.

You are still doing it ,we can't calibrate out EE. The thread is about if calibration can be done and it can.

I think you need to drop the crusade against intent as a reason for calibration. It is a good , reasonable and valid goal even if fully getting there may not always be possible.

Art

DaGamePimp

02-24-09, 07:12 PM

I have to wonder if this thread would even exist if it were not for the last gen JVC's ? Last gen JVC owners are actually convincing themselves that seeking D65 is a pointless endeavor for , 'grabbing at straws' , technical reasons simply because the units cannot do it natively without costly add-ons . I don't think we would have seen this same discussion from people 'in the know' before the release of those JVC units . You have to give JVC credit for having such an impact , kudos ;) .

What we are seeing is what we are intended to see from the disc and striving for those same standards that the disc was mastered in seems a pretty logical path to me ( regardless of directors intent which we may or may not get on the discs that we purchase ) .

- Jason

lcaillo

02-24-09, 10:42 PM

Which JVCs cannot achieve D65? I think you are referring to the location of primaries, not white point. This is a different matter.

Varrius

02-25-09, 01:03 AM

I am accused of derailing a thread called "Calibration...are we just fooling ourselves?" and I am the one who has the problem here? :rolleyes:

You are wrong to say that people in this thread did not equate calibration with director's intent or state that it provides image fidelity. You can presumably read as well as I can so that fact that you missed these things says something. It was those posts that I was addressing and I think they are actually on topic.

I'm going to agree with you here, slightly. I also believed that some arguments are made in this thread that calibration allows us to see the directors intent. Certainly that is on topic. I also agree with your argument that we don't really know what the directors intent was, perhaps he (and in most cases he isn't even involved in the transfer) adjusted the picture for the 'average' tv set, in which case the standards are out the window. Of course, additional processing like EE has nothing to do with any of this discussion (I definately agree with Art on that).

Simple example to illustrate why I sort of agree with lawguy: Director (or telecine transfer dude) wants us to see a greenish-blue murky lake. IF he assumes our sets are set too high in color temp, he will adjust it such that the lake is more green, so that we actually see greenish-blue. However, on a calibrated set, we will see green. This isn't exactly lawguys argument, but it's a theory on how calibrating may or may not give us the directors intent.

Now, I will also agree with the others here. If we dismiss calibration as useless, then we stand little to no chance of obtaining the directors intent. It is the best we can do, and we as consumers have the power to determine what the industry will adhere to, if we utilize that power.

Glimmie

02-25-09, 01:25 AM

Simple example to illustrate why I sort of agree with lawguy: Director (or telecine transfer dude) wants us to see a greenish-blue murky lake. IF he assumes our sets are set too high in color temp, he will adjust it such that the lake is more green, so that we actually see greenish-blue. However, on a calibrated set, we will see green. This isn't exactly lawguys argument, but it's a theory on how calibrating may or may not give us the directors intent.

Wrong, that is not how it's done. Mastering is done on CRT monitors adjusted to SMPTE specifications. Today Plasma monitors are starting to replace the CRT but the Plasmas are pulled into calibration with external LUT boxes - at least where I work. Consumer displays are assumed to be totally uncalibrated in all directions. So we stick to published standards. Therefore if you calibrate your monitor / projector to the same standards you will see the colorists "intent" to the best of your display device capabilities.

The transfers are paid for and approved by the studio. On an A title the DP or director is often involved to evaluate or in some cases supervise the color correction.

DaGamePimp

02-25-09, 01:57 AM

Which JVCs cannot achieve D65? I think you are referring to the location of primaries, not white point. This is a different matter.

My mistake using D65 as the example but my point still stands regarding their inability to be tuned accurately (without costly add-ons) ;) . I think it is a very related matter and the main motivation behind the original post ( do Tryg and Lawguy not still own these last gen JVC models ? ) .

- Jason

TomHuffman

02-25-09, 04:30 AM

Calibration is sold as capturing capturing the filmmaker's intent. If, by filmmakers' intent, one is referring to how a movie was intended to look in a movie theater, someone will have to explain how this can be possible.

In fact, it is not possible to capture a filmmaker's intent with today's displays. This is so for many reasons. Among them are that the color gamut used either in film or in digital cinema is larger than those used in a consumer display that is calibrated either to SMTPE-C or REC 709. Second, all available consumer media (DVD or Bluray) is mastered in these smaller color gamuts (often times it is not apparent which is even used) and, even given the best display and the best available source material, we, as consumers, start out with a severe handicap.

I have even been told that DCI cinema projectors are not calibrated to 6500k, but rather 5500k. How many of us calibrate to 5500k?

At best we can hope to emulate the work that was done when mastering a film for release. At times, this can reflect what a filmmaker chose to release to the public on DVD or Bluray but it most certainly is not the film in all of its original glory.

The other thing to consider is that most of our displays are incapable of being properly calibrated to any standard. I have been told that many displays have non-linear color gamuts and that it is foolish to think that calibrating them to a color standard produces results that are accurate in any measure. These things take us even farther away from any "filmmaker's intent."

There are a plethora of other issues. Check out the DVD or Bluray forums and see the problems that are commonly identified with transfers including bad/wrong colors, edge enhancement, overuse of DNR and others. These issues take us still farther away from "filmmaker's intent."

Please don't get me wrong, a well calibrated display generally produces what I think is the best picture. But, that is my subjective assessment and I don't have any illusions that what I am seeing on a calibrated display has anything but an accidental relationship to the filmmaker's intent.

There will hopefully come a time when our sources and equipment does not limit our viewing experience. We are not there yet.

Talking about filmmaker's intent when that is not possible is both inaccurate and deceptive and takes on some kind of absolutist view that is not justified.

In short, calibration will make your picture better, just don't believe the hype.The DCI white point is closer to 6300K. However CCT is not a very useful metric. DCI white is greenish compared to D65, which CCT doesn't capture. You wouldn't calibrate to this white point unless the medium was mastered in DCI, which Blu-ray is not.

http://home.comcast.net/%7Etlhuffman/dig.jpg

"Filmmaker's intent" does not refer to what he/she intended to be seen in the theater. It is what was intended during the transfer of the source elements to a home theater medium. Directors are often involved in the process, especially with the advent of Blu-ray where the greatly increased image fidelity often demands a remaster. See this article on William Friedkin's participation in the Blu-ray version of "The French Connection."

http://www.iht.com/articles/2009/02/24/arts/friedkin.1-432290.php

Money quote: "The Blu-ray release offers a new transfer from the original negative, for which Friedkin has devised a different color palette, inspired by experimental work of the cinematographer Oswald Morris."

When the filmmaker is not involved in the process, then the intent refers to the telecine artist supervising the master. If he/she adds edge enhancement, then a well-calibrated display will just reveal it clearly in all its glory. There's nothing to be done when the offending image property resides in the source material. The point of calibration is to make the display as transparent as possible, revealing what is contained on the source material as accurately as possible. Good masters will look good. Bad masters will look bad.

coldmachine

02-25-09, 05:25 AM

Lawguy

02-25-09, 08:00 AM

My mistake using D65 as the example but my point still stands regarding their inability to be tuned accurately (without costly add-ons) ;) . I think it is a very related matter and the main motivation behind the original post ( do Tryg and Lawguy not still own these last gen JVC models ? ) .

- Jason

Not. When I did, I color calibrated with a Lumagen.

My point is basically this and I don't understand how anyone would object to it.

Personally, I keep my display calibrated to standards because things tend to look best to me on most material.

However, there are good transfers and there are bad transfers. Sometimes, a disk will look much better when viewed with an expanded gamut, with some DNR or with the sharpness control adjusted. In other words, I have no issue trying to correct things that don't look right but can be at least partially remediated by adjusting available controls. Some people would argue that the disk represents filmmakers' intent and we should suffer suffer through bad transfers without using the controls at our disposal because that is the way we were intended to see things. It is this view that I consider to be naive.

Someone earlier raised the point Peter Jackson supervised the transfers of some of his movies. Well, read this link (http://www.videophile.info/Review/FOTR/FOTR_01.htm)and look at the screenshots and tell me if you think they reflect Peter Jackson's intent? Assuming you agree that Peter Jackson would not have approved of all of that filtering, whose intent is reflected there and should we really be bound to the decisions that those people make? Is George Lucas's intent reflected in this (http://www.videophile.info/Review/TPM/TPM_01.htm)edge enhancement?

I have said all I have to say about this topic and I don't appear to be winning anyone over but you haven't won me over either. :)

jmanthey

02-25-09, 09:37 AM

Lawguy

02-25-09, 09:40 AM

coldmachine

02-25-09, 09:44 AM

Calibration should be done for the value it brings (which it certainly does), not for some debatable philosophical reason.

Its certainly not a debatable philosophical reason. Its done to best approach the video master, this has been explained a number of times already.

I've often wondered why the third or fourth version of a DVD is called the "Director's Cut". Who made the first one -- the grip?

The editor edits as the film progresses and produces a rough cut. This progresses with the shoot.

In conjunction with the director,normally after shooting is complete, they produce a Directors Cut. That addresses any plot holes, chronology issues etc.

There is then a Final Cut, where the studio and/or production company supervise the final cut. Top ranking directors will have contacted a "Final Cut" privilege, giving him final say.

The "Directors Cut" allows us to see the film cut the way the director edited it.

Hope this helps.

Bob Sorel

02-25-09, 10:14 AM

I haven't bothered to read this thread, so I am probably just repeating what has already been said, but I'll add my 2 cents anyway:

I want all equipment to be as neutral as possible within the audio/video chain simply because I don't want to add more variables to an already complex and uncalibrated chain. The idea is that I want to see and hear everything as close to possible to what the artist/director intended, not to create something that is "lifelike". Sure that end is not attainable, but by buying and calibrating my equipment I can get things to be "reasonably close", and that is good enough for me.

With my display device, I don't require perfection, but I don't want my colors grossly misrepresented either. I spend about 4 to 6 hours or so calibrating the projector when it has ~100 hours on the lamp, and I find that in most cases it is time well spent. Once calibrated, I don't touch it again until I get a sense that the color balance is drifting (maybe around 1000 hours?), and touching it up at that point is a relatively quick and painless process (usually under 1 hour). Is my display perfect? No, but it is VERY VERY good, at least by my standards, and I have not wasted countless hours trying to achieve absolute perfection.

A couple of points I wanted to make just from quickly scanning some of the posts:

1. Not all "accomplished professional calibrators" are as good as you might think, so I can fully understand why there are a great many people who are not happy with so called "calibrated" displays. I have personally seen calibration jobs by well respected pros that were just horrible. I know of one calibrator who adjusted color saturation by turning down the "color" control (which is actually a luminance control) in order to bring the x,y values to their correct points, totally disregarding the effect on Y (luminance). The result was, as you could well imagine, a lot less than spectacular, but many people would have simply accepted that the "pro" knew what he was doing and that that was what a properly calibrated display should look like...NOT!!

2. Which brings me to my second point: A lot of people seem to be under the mistaken belief that a calibrated display lacks pop and color vibrancy, and this is simply not true. If you have seen a display that seemed dull and lifeless, then something is wrong. Either the calibrator doesn't know what he is doing OR the display has other problems (like very low contrast, dim lamp, wrong screen material, ambient light, etc.) that are causing the inferior image. For any given projector, the colors produced by a properly calibrated unit are every bit as vibrant and full of life as uncalibrated colors, the main difference being that the colors track correctly (by film industry standards) on the calibrated display and incorrectly on the uncalibrated one. I repeat, proper calibration in no way creates a lesser, duller, more lifeless image than one that is not calibrated.

I think the myth about dull colors on calibrated units largely comes from people who own and have lived with projectors that have oversaturated colors. In a lot of cases people have attempted to correct the colors by turning down the color (luminance) control in order to fix saturation, and that simply does not work. The result was a picture with lower luminance, and THAT produces a duller, more lifeless image. If the saturation is fixed through a properly designed CMS, then ONLY saturation is affected, with luminance (and thus the "life") remaining intact.

Ok...enough from the peanut gallery...;)

CaspianM

02-25-09, 10:24 AM

It is the chroma not the luminance BTW.

Lawguy

02-25-09, 10:34 AM

I agree with Bob that a properly calibrated display will not create a duller, more lifeless image.

I look at it this way.

I like steak. If I have a really good steak, my philosophy is to do as little to it as possible. A few minutes on the grill, perhaps a little salt and pepper, but that's it. If I have a bad steak, I would never treat it that way because it would just taste bad. Whatever intervention I take (seasoning it, pounding it with a mallet, etc.) will only improve it, assuming I know what I am doing.

This is exactly my point with source material. If it is a really good source, just stay out of its way. If it is not, well, try to make it look better.

(I regret in advance using the phrase "pounding it with a mallet" because I suspect that coldmachine will do some damage to me for having used it.)

Bob Sorel

02-25-09, 10:45 AM

It is the chroma not the luminance BTW.
I think that luminance is the correct term, or at least that is how I learned it. I think that "chroma" refers to 2 different attributes of color - "hue" and "saturation", but I could be wrong. The standard "color" control should have minimal effect on hue and saturation, but in real life are not as clean as they should be.

At any rate, my reference was to the coordinates x,y,Y which represent saturation, hue, and luminance respectively. If the term "chroma" applies here, I don't know how to use it and will thus avoid confusing myself...:)

lcaillo

02-25-09, 10:45 AM

For what little it's worth, LG, I completely agree with you. Calibration should be done for the value it brings (which it certainly does), not for some debatable philosophical reason. When "director's intent" is invoked as the primary reason for calibration, we might as well switch immediately to a discussion of religion since both are an issue of faith without proof. I've often wondered why the third or fourth version of a DVD is called the "Director's Cut". Who made the first one -- the grip? But like religion, this debate will not be resolved.

The value that calibration brings is that it moves the system closer to the ability to resolve inputs using the selected standards. One result of this is that one is more likely to see the material as was intended. There is no assurance of this, and I know of no true professionals that would mislead clients into thinking that there are not many more variables involved that affect the ability to view what the artist intended. LGs argument is that this is a faulty selling point and that it is "hyped" to sell calibration. I simply don't see the hype. Certainly you can see it used in Sencore advertising, and there is some discussion of it, but the vast majority of professional calibrators do not make this the primary argument for calibration. Most professionals are very careful to assure that their clients understand what is possible and what is not, and to provide a context in which such statements can be understood for what they are worth. What I take issue with is the claim that there is excessive hype. This thread has done more to create hype for the matter than ever really existed.

Varrius

02-25-09, 11:02 AM

Wrong, that is not how it's done. Mastering is done on CRT monitors adjusted to SMPTE specifications. Today Plasma monitors are starting to replace the CRT but the Plasmas are pulled into calibration with external LUT boxes - at least where I work. Consumer displays are assumed to be totally uncalibrated in all directions. So we stick to published standards. Therefore if you calibrate your monitor / projector to the same standards you will see the colorists "intent" to the best of your display device capabilities.

The transfers are paid for and approved by the studio. On an A title the DP or director is often involved to evaluate or in some cases supervise the color correction.

Fair enough. I was merely attempting to inject a theory that would allow us to conclude that calibration = directors intent is false. It sounds as if that theory has no merit and does not need to be explored, so be it.

When the filmmaker is not involved in the process, then the intent refers to the telecine artist supervising the master. If he/she adds edge enhancement, then a well-calibrated display will just reveal it clearly in all its glory. There's nothing to be done when the offending image property resides in the source material. The point of calibration is to make the display as transparent as possible, revealing what is contained on the source material as accurately as possible. Good masters will look good. Bad masters will look bad.

I like and agree with this summary of the issue, and it explains it better than I could.

SOWK

02-25-09, 11:02 AM

The point of calibration is to make the display as transparent as possible, revealing what is contained on the source material as accurately as possible. Good masters will look good. Bad masters will look bad.

This is the best statement yet on calibration.

It is the same in high end 2 channel audio. (including some 5.1 and 7.1 channel audio setups)

You want everything in the chain as transparent as possible.

HoustonHoyaFan

02-25-09, 11:07 AM

...Someone earlier raised the point Peter Jackson supervised the transfers of some of his movies. Well, read this link (http://www.videophile.info/Review/FOTR/FOTR_01.htm)and look at the screenshots and tell me if you think they reflect Peter Jackson's intent?...Jackson had 100% creative control of the LOTR home video production including shooting new footage! Whatever is on those masters he actively participated in creating.

Most Directors/DOPs are more interested in the overall look/feel/mood of the scene than the incidental technical issues pointed out in that link. Would they add EE/NR/color changes if they felt it created the correct look they want in the scene, absolutely!

Snippet of an exchange I witnessed:

DOP: Manny, that shot does not right it is too soft. Can you tighten it up?
Colorist: How is that.
DOP: A little more
Colorist: How is this, I don't think we want to go much sharper.
DOP : OK, now the background is too in my face, can you smooth it out.
...

Varrius

02-25-09, 11:08 AM

.. Consumer displays are assumed to be totally uncalibrated in all directions. So we stick to published standards. Therefore if you calibrate your monitor / projector to the same standards you will see the colorists "intent" to the best of your display device capabilities.

If this is the case most of the time (or hopefully, all the time), that statement pretty much puts and end to the discussion, at least for me.

SOWK

02-25-09, 11:11 AM

I think that luminance is the correct term, or at least that is how I learned it. I think that "chroma" refers to 2 different attributes of color - "hue" and "saturation", but I could be wrong. The standard "color" control should have minimal effect on hue and saturation, but in real life are not as clean as they should be.

At any rate, my reference was to the coordinates x,y,Y which represent saturation, hue, and luminance respectively. If the term "chroma" applies here, I don't know how to use it and will thus avoid confusing myself...:)

Blink blink.... Am I worng here...

The color control, controls saturation & luminance, not Tint.

If you bring color control to the point were you have no color you have a reading that should be on the d65 white point...

luminance is the one you can not see on the CIE Chart.

lcaillo

02-25-09, 11:12 AM

Fair enough. I was merely attempting to inject a theory that would allow us to conclude that calibration = directors intent is false. It sounds as if that theory has no merit and does not need to be explored, so be it.

Like most questions and debates, the real answer is "it depends." The point that is relevant is that there is no better way than calibration to accepted standards to be most likely to see what was the intent of the artists involved. We are far more likely to see material the way it was intended when a display is calibrated than when it is not, just like we are more likely to see realistic depiciton of color in a live broadcast. No one is claiming that there are not deviations and lots of variables involved that mitigate the process.

CaspianM

02-25-09, 11:17 AM

Blink blink.... Am I worng here...

The color control, controls saturation, not luminance / nor Tint.

If you bring color control to the point were you have no color you have a reading that should be on the d65 white point...

luminance is the one you can not see on the CIE Chart.

Right. Chroma function has two primery control i.e., amplitude (color gain) and phase response which is the hue control. Luminance level is the amplitude of black and white portion of signal simply put.

Art Sonneborn

02-25-09, 11:46 AM

I certainly was never one to use directors intent as the primary reason but cetainly without intent (whether it be director or otherwise) being used as some sort of grounding ,the whole quality of presentation concept gets distorted.

When we get a BD review if we get things like color evaluation ,shadow detail, holding black etc etc it is meaningless without common standards to which displays are set .As I've said, some displays are better than others at adhering to those standards but that is not a good reason to say oh what the **** we should just give up,as the OP recommended.

The is not the PC comment to make but the elephant is in the room.... I think the ones who feel that the concept of calibration is not valid are the ones who own displays that are not able to achieve the standards or aren't able to without add ons(the OP for one).

Art

R Harkness

02-25-09, 11:57 AM

This is the best statement yet on calibration.

It is the same in high end 2 channel audio. (including some 5.1 and 7.1 channel audio setups)

You want everything in the chain as transparent as possible.

I want the option for a transparent chain but I also have no qualms about adjusting an image to my satisfaction.

For instance, the DVD transfer for North By NorthWest is one of my favorites with the exception that the black levels seem unusually high and washed out compared to other movies. An adjustment of the brightness control, lowering the black levels, makes a big difference in the richness, dimensionality and the viewing experience as a whole. Why I would want to suffer with a poorer image and higher black levels just to be true to how it was transferred is to me a strange sort of priority.

As a long time audiophile I certainly am aware of the rationals for trying to get as transparent a chain as possible; one benefit is that great recordings will be shown to even greater effect if they aren't dumbed down in some sugar coated way that makes everything sound good. On the other hand, the obsession for pure transparency insofar as it focuses so heavily on the recording quality over the musical enjoyment can go off the rails, at least from my perspective.

It's like the audiophile abhorrence of equalizers and having the chain anything less than "transparent."...The audiophile intones: "If it's a good sounding recording I want to hear it...if it's a bad sounding recording I want it to sound like a bad sounding recording."

Which, unless you are working at a mastering suite, is an odd priority. It shows the obsessiveness of the audiophile who listens to "recordings" rather than to the music. I'd prefer the music to sound enjoyable, because that's what I'm listening to and wish to be able to enjoy: the music.

The same can be said for video calibration. I want the option of as transparent a chain as possible, so calibration = good. But I certainly have
no desire to be slave to poor source quality and transfers: If tweaking the image means I enjoy the viewing experience more, that's the most important
thing and enjoying the viewing/sound experience is why I'm doing home theater.

mdputnam

02-25-09, 11:58 AM

"Filmmaker's intent" does not refer to what he/she intended to be seen in the theater. It is what was intended during the transfer of the source elements to a home theater medium. Directors are often involved in the process, especially with the advent of Blu-ray where the greatly increased image fidelity often demands a remaster.

Although I think we're splitting hairs here and making broad generalities is always a dangerous thing. I think it would be safe to say that the majority of films that make it to DVD/Blu-ray are not telecined with the director's, cinematographer's, or director of photography's direct involvement through the telecine process, and that is not due to lack of intrest on their part. High stature projects will and do receive more attention. Again in generalities, the telecine artist is contracted by the studios, he has a very limited schedule and an even tighter budget to produce a product. Most are good at interpreting the film makers intent and remain true to it within the limitations of the medium and the budget. However, for example, studio execs know that edge enhancement and elimination of film grain "improves" the product in the eyes of the public so we get it put in. I know it's a matter of semantics but to me "Film Makers" are the folks out on the front lines during planning, production and post production of the project for theatrical release. Based on this limited definition I would say in many cases we are getting the telecine colorist's interpretation of the film makers intent plus the studio executive's requirements on disk. I do agree whole heartedly that without a calibrated display we wouldn't be able to tell what anyone involved wanted us to see.

TomHuffman

02-25-09, 12:05 PM

I think that luminance is the correct term, or at least that is how I learned it. I think that "chroma" refers to 2 different attributes of color - "hue" and "saturation", but I could be wrong. The standard "color" control should have minimal effect on hue and saturation, but in real life are not as clean as they should be.

At any rate, my reference was to the coordinates x,y,Y which represent saturation, hue, and luminance respectively. If the term "chroma" applies here, I don't know how to use it and will thus avoid confusing myself...:)This is mostly a semantic distinction. The main color control adjusts chroma, not luminance. However, the primary EFFECT of modest changes to the chroma control is to change the luminance of the color. Larger changes will also substantially effect saturation.

millerwill

02-25-09, 12:21 PM

This is mostly a semantic distinction.

Agreed, wrt to just about this whole thread. I think we're mainly just bored because there's not much interesting to talk about now in the way of projectors, so Tryg just stirred the pot to keep us active.

HogPilot

02-25-09, 01:04 PM

i certainly was never one to use directors intent as the primary reason but cetainly without intent (whether it be director or otherwise) being used as some sort of grounding ,the whole quality of presentation concept gets distorted.

When we get a bd review if we get things like color evaluation ,shadow detail, holding black etc etc it is meaningless without common standards to which displays are set .as i've said, some displays are better than others at adhering to those standards but that is not a good reason to say oh what the **** we should just give up,as the op recommended.

The is not the pc comment to make but the elephant is in the room.... I think the ones who feel that the concept of calibration is not valid are the ones who own displays that are not able to achieve the standards or aren't able to without add ons(the op for one).

Art

+1

Bob Sorel

02-25-09, 01:13 PM

This is mostly a semantic distinction. The main color control adjusts chroma, not luminance. However, the primary EFFECT of modest changes to the chroma control is to change the luminance of the color. Larger changes will also substantially effect saturation.
Thanks for the clarification, Tom!
I want the option for a transparent chain but I also have no qualms about adjusting an image to my satisfaction.

I agree with both you and Lawguy here. Having a properly calibrated display and "fixing" bad transfers are not mutually exclusive events. I keep both my audio and video systems properly calibrated, yet I use additional gear to "fix" bad recordings when and if I feel the need to do so. But without a calibrated system you really have no point of reference in order to make an informed decision as to whether the recording requires "fixing" or not.

Dammit, I just can't get rid of that nasty green tint in The Matrix no matter what I do! :D

Josh Z

02-25-09, 01:15 PM

Wrong, that is not how it's done. Mastering is done on CRT monitors adjusted to SMPTE specifications. Today Plasma monitors are starting to replace the CRT but the Plasmas are pulled into calibration with external LUT boxes - at least where I work. Consumer displays are assumed to be totally uncalibrated in all directions. So we stick to published standards. Therefore if you calibrate your monitor / projector to the same standards you will see the colorists "intent" to the best of your display device capabilities.

I don't think his analogy is really that far off. It has indeed happen on at least one occasion to my knowledge. The Japanese DVD edition of Spirited Away was mastered with the color saturation cranked up too far because the post-house had just installed new plasma monitors, and felt that the movie's colors didn't "pop" enough. So they tried to adjust for their mointors' inaccuracies by tweaking the master. The result was that the animated characters looked deeply sunburned when viewed on displays that closer conformed to the NTSC standard.

Due to complaints about that disc, Disney had the movie retransferred with accurate colors by the time they released it in Region 1.

Glimmie

02-25-09, 02:20 PM

I don't think his analogy is really that far off. It has indeed happen on at least one occasion to my knowledge. The Japanese DVD edition of Spirited Away was mastered with the color saturation cranked up too far because the post-house had just installed new plasma monitors, and felt that the movie's colors didn't "pop" enough. So they tried to adjust for their mointors' inaccuracies by tweaking the master. The result was that the animated characters looked deeply sunburned when viewed on displays that closer conformed to the NTSC standard.

Due to complaints about that disc, Disney had the movie retransferred with accurate colors by the time they released it in Region 1.

There are always bad transfers. And there are facilities with qualified engineering staff and then there are those without!

DrewL

02-25-09, 02:55 PM

I'm always shocked by how long AVS Forum threads are. Now it's my turn to make one even longer! :)

One of our clients is a production company with an overseas subsidiary where most production is done. They needed a modest screening room to view footage so they could make judgments and issue comments to the directors and animators.
Since we can't do a side-by-side comparison, the best approach is to calibrate the display to a common standard, such as Rec.709. Calibration is needed, end of story.

When we dissect calibration with such scrutiny, I'd shy away from anyone who uses the words "good," "better," or "best" to describe or judge picture. Give me my delta from spec, acknowledge that personal human preference is a separate adjustment, and close this thread already :D

Now, if you talk about the benefit of calibration for consumers, or its marketability for calibrators... that is a totally different topic. Like math vs. economics.

Josh Z

02-25-09, 05:03 PM

There are always bad transfers. And there are facilities with qualified engineering staff and then there are those without!

Absolutely agreed. Which is why I think this whole "director's intent" argument is derailing the thread. There will always be some good transfers and some bad transfers. That can't be helped. The point of calibration isn't to adjust the display for each and every individual disc played. It's to adjust the display to the known standard by which those discs should have been authored.

Glimmie

02-25-09, 06:19 PM

Absolutely agreed. Which is why I think this whole "director's intent" argument is derailing the thread. There will always be some good transfers and some bad transfers. That can't be helped. The point of calibration isn't to adjust the display for each and every individual disc played. It's to adjust the display to the known standard by which those discs should have been authored.

That's how I see it too.

orion456

02-26-09, 12:10 AM

I, personally, would rather use my own eye (ie. DVD Essentials, etc) than a calibrated instrument to setup my system. Everyone interprets what they see different and it should be pleasing to your own eyes. If a piece of equipment is used to calibrate, then you think it looks a little 'off' - then what is the point? Just my opinion.

Your eye and DVE can only do a limited amount on the gray scale. The eye is easily fooled when comparing colors and shades of gray. Once you have seen an instrument calibrated projector, I for one could NEVER go back to the old way. Of course you might like the image with a bit more contrast, darker, or a different gamma and you can adjust that to your liking.

It is calibration within the available parameters. However, there are many variables that need to be juggled so calibrations is not just a science but also requires some judgement and experience to attain a high level result. It is possible to make two good projectors appear very similar within a reasonable level of uncertainty. Calibration always has some uncertainty, in this case with many variables, the uncertainty is simply larger.

Tryg

02-26-09, 10:03 AM

If you cant get a projector to a designated standard can you call that calibration?

in my book that's just tuning

Josh Z

02-26-09, 11:03 AM

If you cant get a projector to a designated standard can you call that calibration?

in my book that's just tuning

To me, "tuning" implies that you're adjusting the picture to your own personal preferences, based only on your own whims. Calibration implies at least the attempt to adjust the picture to a defined standard. Maybe the hardware isn't capable of exactly matching that standard, but we try to get it as close as possible.

GeorgeAB

02-26-09, 11:13 AM

Tryg

02-26-09, 11:14 AM

A lot of the talk in this thread clearly demonstrates you cannot definitively achieve those standards. Also, If you could, each projectors image would look very similar. They simply do not.

When you take your car in to get "optimized" for the standard it's components allow it to do. Is this Calibration? or Tuning?

GeorgeAB

02-26-09, 11:30 AM

If you cant get a projector to a designated standard can you call that calibration?

in my book that's just tuning
You've been around long enough to know better and have heard all this before (as have most of the rest of the posters in this thread). I have become convinced that you just enjoy being an agent provocateur for your own amusement. Be happy with your "book" and live according to your own ideas.:)

Best regards and beautiful pictures,
G. Alan Brown, President
CinemaQuest, Inc.
A Lion AV Consultants Affiliate

"Advancing the art and science of electronic imaging"

Lawguy

02-26-09, 11:36 AM

We obviously need and rely on the fact that people in the production chain attempt to adhere to standards. If they typically did not, the quality of our sources would vary even more than it does now. It has often been stated that there is great reluctance for people in this production chain to move away from using the Sony CRTs that they have been using for years, even though Sony no longer makes them.

I would bet that part of the reluctance in using new kinds of studio monitors (LCD, Plasma, etc.) is that they look different from CRTs and thus, using them will have some kind of effect on the finished product. This would be true even if everyone is sticking to standards.

It is another "angels dancing on the head of a pin" kind of question but I agree that what we on the consumer side of things do is more akin to tuning than to calibration, but I don't feel strongly about it.

Dr. Spankenstein

02-26-09, 11:49 AM

If you cant get a projector to a designated standard can you call that calibration?

in my book that's just tuning

So, by extension, you would say that if you didn't WIN a foot-race you were not "running" a race, you were "losing" a race.

Semantics...

Reminds me of my conservatory days when the first chair player's mantra was: "I'd rather be sharp than out-of-tune!" :rolleyes:

SOWK

02-26-09, 11:52 AM

If Calibration doesn't matter at all then throw all your video displays to dynamic mode, and be happy you "tuned" your displays to look like poo! :)

BTW.

I'm pretty sure you set your displays to the most accurate mode, that is closest to d65...

Yeah, I thought so! :)

SOWK

02-26-09, 05:58 PM

Good, maybe I shut down this thread... :cool:

ChrisWiggles

02-26-09, 10:22 PM

R Harkness

02-26-09, 10:34 PM

There is no way to discuss image quality in a way other than discussing its accuracy. The goal is the most accurate possible reproduction of an original image source.

Any discussion of image quality other than a discussion of accuracy is an ignorant and pointless discussion since it meaninglessly subjective and completely ignores the purpose of having video displays at all.

Wow. I couldn't disagree more!

Image Accuracy and Image Quality can be (and are) separate and both valid and important issues in of themselves. This is recognized by...well...pretty much everyone, professional calibrationists included.

There is the question "Is this display re-creating the source image accurately" and there is ALSO the valid, separate question of "Is this a good image?"

If the issue of image quality and accuracy weren't separate, valid issues on their own then we could not talk of the much acknowledged fact that you can have an image that is accurate (to the source) but which is a poor image (because the source is a poor image).
So your conflation of "accuracy" with "image quality" is both wrong and needlessly restrictive.

And there are plenty of factors by which image quality can be discussed separately from the strict question of accuracy to whatever source; e.g. the richness of contrast, shadow detail, clarity, realism or lack of realism in the color, dimness, brightness...you name it. Not once do we have to mention "accuracy" to any particular video source in order to talk about whether an image possesses the attributes of a "good image."
If one were to believe what you just wrote, we could stand in front of a display watching an image that totally wowed us, amazing contrast, resolution, detail, dimensionality, realism etc. Yet we could never talk about how good the image quality was...unless we knew how the source material was transferred or the specifics of the display calibration. Which is silly - of course we could talk about how impressive the image looked...then we could also ask the separate question of "Ok, impressive image quality...but is this image accurate to the source material?" And if it weren't accurate it wouldn't suddenly make what is an impressive image "bad" in terms of image quality...it would just make it inaccurate to the source.

But I'm sure you realise this...maybe it's just the way it came out in your post...

Cheers,

HoustonHoyaFan

02-26-09, 10:54 PM

..."Is this a good image?"...The answer to that question so subjective as to be useless as a discussion point. Take any single image. One could crank as part of post processing up color and contrast to the point where it looks almost like a cartoon ( Speed Racer), one could emphasize green to the point where people look slightly sick (Matrix), one could have a soft dream-like 20s style feel (FFC Dracula). Which one is a good image. As Chris said the goal is to reproduce the image exactly as produced bt the source.

R Harkness

02-26-09, 11:40 PM

The answer to that question so subjective as to be useless as a discussion point.

No it's not. And in fact the very way we discuss displays and images around here would make little sense if image quality could not be discussed
separately from the issue of strict calibration-type accuracy.

Think about it: Why is a display capable of higher contrast and ever lower black levels considered desirable? Why do we on these forums get excited by progress in these areas? Because higher contrast has an effect on picture quality. That is a fact that can be pointed out without having to talk about "accurately calibrated images." We can talk about how higher contrast will increase the perceived image clarity, richness, depth, punch, realism etc. And it's not merely one guy's opinion - it's based on objective facts (which have been studied) about the nature of human perception. You can talk about the impact of black levels and contrast to image quality, the impact of resolution, the impact of color resolution, the impact of brightness etc, in terms of how we perceive their effects independently of the question of whether "X image is accurate to Y source material."

Research into how people perceive image quality has been undertaken by the film/broadcast/display industries since the beginning, limited in no way by the question "But is the image accurate to X source material?"

This is just a basic fact and it underlies much of the discourse, "scientific" and unscientific, of image quality. This does not presume that every person in the world will agree on which display they like. But it is far from accurate to say that therefore questions of image quality are useless
unless conjoined with accuracy in the calibration sense.

It's why we could stand in front of two displays, "A" having much better contrast, clarity, more realistic color etc than "B" and we can easily agree that A has the better picture quality. Whether the particular image we are watching is accurate to whatever source is another question.

I really think people are getting a bit mixed up on this issue, conflating "image quality" with "accuracy" when, if you even examine how we discuss issues around here you'd know they weren't the same and don't have to be. All you have to remember to realise this is that you can
have an accurate display showing you a very poor source image. You can simultaneously express "The image is accurate" and "The image is poor" without contradiction...because image accuracy and image quality are not necessarily the same thing.

Calibrationists themselves have often acknowledged this. Why some folks seem to be confusing the issue is puzzling.

Lawguy

02-27-09, 07:39 AM

The answer to that question so subjective as to be useless as a discussion point.

What is wrong with it being somewhat subjective? Not everything has a clear black and white answer and the absense of them does not make them useless as discussion points.

In the law, we have "rules" and we have "standards." A rule is like a speed limit. Go over 55 mph and you are guilty of speeding. Standards are more nebulous. An example is that in matrimonial proceedings, the custody of a child is often determined by the judge deciding what is in "the best interests of the child."

Rules are easy to apply. If it is proven that you were driving over 55 mph, you are guilty of speeding. Standards involve the exercise of some discretion. Both parents firmly believe it is in their child's best interest to have custody awarded to them. There is often no clear answer to who is right, but that does not hinder these disputes from being resolved.

Rich alludes to different factors by which we can judge image quality. Among them are contrast, color, image clarity, etc. I think that it is inappropriate to use a hard "rule" on this issue stating that a display fails at its primary purpose solely because someone adjusts a control that makes an image demonstrably less accurate to its source Rather, I would argue that it is more appropriate that we apply "standards" to an image and thus judge an image against how well it achieves the things that we know comprise a good image.

Most times, it may be that the best image quality is achieved by viewing a source on a display calibrated according to specifications and there is thus no conflict been image accuracy and image quality. Sometimes, however, the most accurate image does not have the best image quality and the judicious use of the controls at our disposal can improve image quality.

lcaillo

02-27-09, 08:37 AM

Bottom line is that televisions and other home displays are fundamentally image reproduction devices. Their primary and ONLY function is the reproduction of an image created to strict professional standards. As such, the ONLY way we can discuss the quality of such a reproduced image is to judge it by its accuracy to the referent.

There is no way to discuss image quality in a way other than discussing its accuracy. The goal is the most accurate possible reproduction of an original image source.

Any discussion of image quality other than a discussion of accuracy is an ignorant and pointless discussion since it meaninglessly subjective and completely ignores the purpose of having video displays at all.

Subjective image quality enters the picture if you are an artist, a director, or whatnot and are creating your own content with your own artistic intent.

But in the cinema, the home theater, or watching TV, the task is fundamentally one of reproduction, a task which fundamentally eschews subjectivity. And the only judgement of the quality of that reproduction is one of accuracy. All others are finger-in-the-wind opinion and subjective nonsense that destroys the artistic intent.

I'd say this goes a bit far, Chris. The fact is that the funtion of home displays is very much NOT limited to "reproduction of an image created to strict professional standards." The purpose is to statify the user in reproducing a wide range of material that may or may not adhere to standards. In addition, the user can add his/her subjective interpretation and preferences as they desire. Users can choose to watch their sets any way they choose, just like people can buy art and place it wherever they choose under whatever lighting they choose. The artist's intent is not a matter of concern for many. If it is, then using a calibrated dislpay is certainly a good starting point.

Perhaps the arguement that there is a lot of hype about calibration enabling one to see artist intent is not so far off, considering your post...because it certainly fits into that category, IMO. I still do not see such hard line opinions from any but a handful of the most exremely condescending calibration specialists. Most are much more grounded in the reality of variance in program quality and limitations in most displays. There are subjective aspects to calibration and viewing that we deal with every day because of those, and the differences in priorities of the user.

Lawguy

02-27-09, 08:50 AM

I'd say this goes a bit far, Chris. The fact is that the funtion of home displays is very much NOT limited to "reproduction of an image created to strict professional standards." The purpose is to statify the user in reproducing a wide range of material that may or may not adhere to standards. In addition, the user can add his/her subjective interpretation and preferences as they desire. Users can choose to watch their sets any way they choose, just like people can buy art and place it wherever they choose under whatever lighting they choose. The artist's intent is not a matter of concern for many. If it is, then using a calibrated dislpay is certainly a good starting point.

Perhaps the arguement that there is a lot of hype about calibration enabling one to see artist intent is not so far off, considering your post...because it certainly fits into that category, IMO. I still do not see such hard line opinions from any but a handful of the most exremely condescending calibration specialists. Most are much more grounded in the reality of variance in program quality and limitations in most displays. There are subjective aspects to calibration and viewing that we deal with every day because of those, and the differences in priorities of the user.

After all this, we have found some common ground. :)

anirbana

02-27-09, 10:30 AM

Really interesting to see that this thread is going so strong. Initially I thought it was going bit on a kind of war type but later some wonderful posts and lots of information.

if I may add my opinion here. I am not very technical I see things the way I understand to my capability.
Some points I like about Calibration

1. It is the only way (subjected to use of right equipments) to go closer to a standard defined for proper viewing. And I think it is a wonderful effort for people who have spent their lives in making these standards. Like all other standards a reference for anything is necessary to make a informed decision or choice by others.

2. It helps you to get rid of image anomalies that normally comes OOB with different devices and unfortunately devices are hardly ever close to that standard. I have no idea why they shouldn't be at the first place to start with. But I guess that is why quality is such a difficult thing to manage considering many other factors.

3. It generates an awareness that is important to know if not essential (since it is subjected to taste of different individuals). Spreading the awareness is more important.

Why I like the idea of tuning with respect to calibration

1. You can't calibrate unless you have the tools and techniques for calibration that are standards by themselves not forget the eye factor of the person who is doing this on his own. This is with most of us who cannot afford or don't want to hire a professional.

2. It wil be wrong to say to someone that if you have not calibrated to standards so you are not getting what you should get as a standard or as the director or the technician authoring discs the way they intended specially to someone who has spent lot of time to calibrate (I would say tune) his device based on his best satisfaction and is satisfied with the image.

We need to remember few things

a. People tend to like different tones and ranges of colour. Some like bolder, softer, highlighted colours. Different cultures like different shades and strengths of colours. It doesn't matter what the director intended. They look for those elements in things they see. Now in theatres they don't have a choice to change settings. But at home they have a choice to 'Tune' either away or towards a set calibrated standard and I think it is their right and there is no right or wrong. It is just plain and simple preference... This is also the reason why many people find calibrated displays to be dull and flat. Please note they are not wrong from their perspective. My wife likes bolder colours and I like natural colours. I can't change that. So when they try to change settings (which they may understand as calibration though it is not it is actually tuning)

b. If a viewer tunes or calibrates to his heart's content (considering he is aware of calibration and has seen calibrated displays), it is unfair to say you are not getting what the director wanted. It is own choice of seeing things the way he intends...what’s wrong with that. I keep two settings on my PJ.

c. Again back to OP's point - even if you spend loads and calibrate to standards, 10 different make and spec display may show similar colour balance but what about greyscale? That where the difference will be like day and night? Point is - it is very much variable to certain factors otherwise all display would look same after calibration and that does not happen.

d. Movie disc qualities vary between most titles. If you read disc reviews, you will come across many such issues that vary between films. Now if I have a 100% guaranteed calibrated display with a 100% reference disc, I can get exactly a 100% correct picture if the other title I am watching is 100% reference quality. In real world it does not happen. Now think of this. PJs have stored settings. The best one can do is have a calibrated settings ISF D/N or whatever and when a movie is played, first apply that. In many cases you will see that there is still something missing and that is where the so called ‘Tuning’ comes into play. Now some may not do it stating that it won’t be a good idea to deviate from my calibrated standard (director intends or I don’t want to do or …) but most will do….remotes are always handy…
Now here is a interesting part
If you are tuning at this stage you may start moving towards a settings that you may start liking even though the Calibrated preset is there. Or you may notice it is way too off, so revert back to your calibrated settings. But again the fact is it is your choice and preference. If you like something bit deviated from the standard better accept it and enjoy instead of going back to standard even if that will need to digest pebbles.

To conclude, I like the idea of calibration very much and know what value it can add to my viewing pleasure but I cannot stop myself from fiddling around with some settings from disc to disc due the differences between them.

And at the end it is a preference of the way you like to see things. The director is not sharing the sofa with you at your home. So spoil yourself at your own den but do give calibration a try to know what standards are all about. To you it could be a setting that you may start liking and you will never know unless you have seen it.

The satisfaction is within you and not necessarility if you have calibrated....though Calibration have a high chance of making you satisfied...

Ani

ChrisWiggles

02-27-09, 11:51 AM

Wow. I couldn't disagree more!

Image Accuracy and Image Quality can be (and are) separate and both valid and important issues in of themselves. This is recognized by...well...pretty much everyone, professional calibrationists included.

There is the question "Is this display re-creating the source image accurately" and there is ALSO the valid, separate question of "Is this a good image?"

No, I reject this. I reject this vehemently. Now, we are all advanced enough in this forum to understand that in approaching an idealized and perfectly accurate image there are compromises, and subjective choices do come into play because no display is perfect. I fully understand and accept that, but it is important to point out for novices that the starting point of a quality image has nothing subjective involved. I absolutely reject the idea that there is a difference between a "good/quality" image and an accurate image at least in terms of the reproduction chain that we're dealing with here. There simply is no meaningful way to discuss what a good image looks like other than in terms of accuracy. Doing so destroys the entire concept of accuracy and the goal of home displays which is the reproduction of an original image.

Now, if you are a photographer, or a film-maker and are creating your own images the discussion is an entirely different one because your intent may have nothing to do with accuracy to the original scene, and indeed the discussion is a subjective one since you are in control of the image yourself and what it intends to convey. But in order to maintain that intent in any robust way via a delivery system, we must adhere as strictly as possible to an accurate reproduction. Once again, most of us here understand that approaching that accurate image reproduction is complex and inevitably involves certain degrees of compromise which inevitably involve subjective preferences (which departure from perfect accuracy is least objectionable to me personally in my display or particular calibration choices, etc).

If the issue of image quality and accuracy weren't separate, valid issues on their own then we could not talk of the much acknowledged fact that you can have an image that is accurate (to the source) but which is a poor image (because the source is a poor image).
So your conflation of "accuracy" with "image quality" is both wrong and needlessly restrictive.

No. I reject this strongly. How would you define a good image then, if it differs from accuracy? Do you define it as most realistic? Do you define it as most colorful? Do you define it as brightest? Do you define it as the sharpest? If so, how do you approach a film like Finding Nemo in terms of realism, or Saving Private Ryan? How do you approach a film like Ballast or Fargo in terms of colorfulness? How do you manipulate Dark City in terms of brightness? How do you manipulate Sky Captain and the World of Tomorrow in terms of sharpness?

If you choose your own metric of what defines a "good" image in any fashion other than accurate, it requires seriously manipulating and destroying the artistic intent of various films that are chosen for very particular emotional reasons.

And there are plenty of factors by which image quality can be discussed separately from the strict question of accuracy to whatever source; e.g. the richness of contrast, shadow detail, clarity, realism or lack of realism in the color, dimness, brightness...you name it. Not once do we have to mention "accuracy" to any particular video source in order to talk about whether an image possesses the attributes of a "good image."
If one were to believe what you just wrote, we could stand in front of a display watching an image that totally wowed us, amazing contrast, resolution, detail, dimensionality, realism etc. Yet we could never talk about how good the image quality was...unless we knew how the source material was transferred or the specifics of the display calibration. Which is silly - of course we could talk about how impressive the image looked...then we could also ask the separate question of "Ok, impressive image quality...but is this image accurate to the source material?" And if it weren't accurate it wouldn't suddenly make what is an impressive image "bad" in terms of image quality...it would just make it inaccurate to the source.

That can be discussed outside the realm of image reproduction. But the goal in the home is one of reproduction and nothing more.

The only exception that I make for this at a fundamental level is where the reproduction chain completely breaks down, and that really only happens in the example you gave above of a seriously poorly mastered title that is obviously made in error. In that instance, the content does not match (as it should) the director's intent. But once again, how do we know that things are amiss? Because we have a referent, either somebody familiar with the film presentation to say: hey the colors are all wrong, or hey the black level is artificially elevated wrongly, or whatever, or we make intelligent and educated presumptions for instance a scene that goes fully to black that is significantly above 16, we can assume that the director intended that to be black rather than how it was mastered which I think we can intelligently conclude was in error. And we go about fixing this by returning the image back to one that closely matches what it should be, once again approaching an accurate image. In this case the purposeful 'miscalibration' of the playback chain or the manipulation of the content by the user is intended to restore an accurate, hence good image.

Otherwise, if we inject subjectivity into a fundamental judgement of image quality of a reproduction chain, the entire concept of image reproduction falls apart and there is no point to having image standards at all. There is no point to calibrating displays, and we should just set things however we please. In other words, put your TV in vivid and call it a day.

ChrisWiggles

02-27-09, 12:03 PM

What is wrong with it being somewhat subjective? Not everything has a clear black and white answer and the absense of them does not make them useless as discussion points.

In the law, we have "rules" and we have "standards." A rule is like a speed limit. Go over 55 mph and you are guilty of speeding. Standards are more nebulous. An example is that in matrimonial proceedings, the custody of a child is often determined by the judge deciding what is in "the best interests of the child."

Rules are easy to apply. If it is proven that you were driving over 55 mph, you are guilty of speeding. Standards involve the exercise of some discretion. Both parents firmly believe it is in their child's best interest to have custody awarded to them. There is often no clear answer to who is right, but that does not hinder these disputes from being resolved.

Rich alludes to different factors by which we can judge image quality. Among them are contrast, color, image clarity, etc. I think that it is inappropriate to use a hard "rule" on this issue stating that a display fails at its primary purpose solely because someone adjusts a control that makes an image demonstrably less accurate to its source Rather, I would argue that it is more appropriate that we apply "standards" to an image and thus judge an image against how well it achieves the things that we know comprise a good image.

Most times, it may be that the best image quality is achieved by viewing a source on a display calibrated according to specifications and there is thus no conflict been image accuracy and image quality. Sometimes, however, the most accurate image does not have the best image quality and the judicious use of the controls at our disposal can improve image quality.

But you completely ignore the fundamental problem of how one defines "quality" in any meaningful fashion other than by accuracy.

The home reproduction chain is exactly that. It is akin to a photocopier or a fax machine. When you make a photocopy of something, the entire purpose and intent of the device is to make an exact copy of an original as cloesely as possible. We judge the quality of the photocopy by how closely it matches the original.

So in what objective way can we define quality as something other than accuracy when the task is fundamentally one of reproduction? Your argument fails.

Rich alludes to different factors by which we can judge image quality. Among them are contrast, color, image clarity, etc.

Again, what do these things mean? Who decides the attributes that are important. For instance color: What makes a "good image" with regards to color? The most colorful and oversaturated? The most accurate to the real world? What is the choice? Should we then make every film look the same way depending on what we decide? If so, we should colorize all B&W film, we should totally change the color tones of Saving Private Ryan, we should make Dark City bright and colorful and cheery. If not, then what is the purpose of the decision? It seems to be whimsical and meaningless.

In other words, I absolutely do not believe that there is any meaningful way whatsoever to divorce image quality from accuracy in an image reproduction chain. It simply is not possible, or it ceases to be a reproduction chain.

R Harkness

02-27-09, 02:10 PM

I absolutely reject the idea that there is a difference between a "good/quality" image and an accurate image at least in terms of the reproduction chain that we're dealing with here. There simply is no meaningful way to discuss what a good image looks like other than in terms of accuracy. Doing so destroys the entire concept of accuracy and the goal of home displays which is the reproduction of an original image.

Chris that's not true.

You can say an image is "accurate" or not in terms of meeting calibration standards. We can likely agree.

You can make an argument that displays "ought to reproduce the source content as accurately as possible" and perhaps find many agreeing with you.

That's fine.

What I take issue with is conjoining your argument with erroneous statements - an indefensible claim that "image quality" must therefore only reference "image accuracy." Which is nonsense.

I mean, you may have your own, constrained idea of what constitutes image quality, but it simply doesn't account for our ability to discuss image quality in of itself. (And keep in mind I am not arguing against the concept or desirability of our displays being able to accurately reproduce the source as intended. I'm arguing against questionable claims you are using to argue your position).

Now what may be happening here to some degree is that the word "good" is being used, which is a value judgement. (Nothing wrong with that). I'm using it as a short hand for the fact there are parameters in image quality that, generally, when pushed in one direction most people would tend to call "good." But we can for a moment drop the term "good" and keep talking about "Image Quality" (in a descriptive sense).

In other words: The qualities of an image can obviously be described independently of it's accuracy. For instance: whether an image is low or high in contrast, has very high or low black levels, rich color saturation
or washed out, high or low resolution, is comparatively bright or dim, has 3 dimensionality or appears flat and on and on.

You know this. This is the way we describe images to one another - the quality of a source material, the characteristic abilities of a display etc. And such descriptions are useful observations independent of mentioning accuracy to any particular source material. It's why we can say, for instance, that the big advances JVC made in their projectors that make for deeper black levels, while not compressing bright areas (no DI), are significant and in large part are advances toward a desirable goal.

The goal can be expressed purely in terms of image quality: Deeper black levels presented simultaneously with brighter bright areas - expanded contrast - contribute to a greater sense of vividness, realism of contrast, image depth etc.

That is saying something valid and important, and comprehensible...and uncontroversial...without having to mention "accuracy to X source material."

But if we mean "image quality" where "quality" equates to "desirable traits" which essentially equates to "good," then in those terms image quality becomes a value judgment. In which case you are trying to make your own value judgement "what is good" apply for everyone else, which doesn't work. Why should Chris Wiggles' "good" determine what everyone else will see as "good?"

Chris, you can say an image is "accurate" (in terms of reproducing source material) or "inaccurate" and, really, that says all you need to say. Trying to attach the value judgement "good" ("image quality") is unnecessary and
will fail.

If it's understood that value judgements are separate from fact statements (always fun to weave in moral philosophy :) ) then we make more sense of how people discuss these issues and allow important nuance into our approach.

So my claim is:

1. There is the issue of "accuracy," in terms of "Am I seeing the source material reproduced accurately?" The attempt to set up objective parameters can help us come to agreement on whether an image is being
reproduced more or less accurately.

2. There is the issue of "image quality" in it's descriptive sense. We can talk about how an image "looks" - it's qualities - and it's general effect on
how we perceive images. A lot of which can be discussed in fairly objective terms as to how pushing an image parameter in X direction will tend to be perceived by people as having Y effect. (E.g. increasing contrast increases apparent sharpness and image dimensionality).

3. The issue of value judgement: Is X image "good?" This is subjective,
which is why we'll see variance in preference for different types of images.

Conflating all of those into "We can only talk of Image Quality in terms of Accuracy" is really, really getting things wrong, both logically and in it's inability to account for how we actually discuss and inform one another
about images.

(Just to add about #3...while there is going to be variation in the value judgement "Is X image good?" to a degree, the facts underlying #2 will mean that there will be meaningful trends to observe in what people will
tend to call "good." And it has been studied: how changing certain image parameters produce trends in what subjects will tend to evaluate as being preferable).

I think all this is pretty obvious and in fact we would have little disagreement
when talking about images in "real life." To the extent you want to say a display device "ought only reproduce accurately the source" - that's a value judgment. And it's fine to argue for that value judgment, you may get many on your side. I'm simply being pedantic and pointing out you've mingled in some indefensible propositions ("it only makes sense to talk of image quality in the sense of image accuracy") while trying to argue your position.

Cheers,

CaspianM

02-27-09, 03:12 PM

High image quality and accuracy are not two seperate entities. You are confusing yourself.
That is the nirvana in image quality that be able to reproduce the source signal faithfully.
What is the point of having a PJ that can only do color or gamma and so forth based on whatever?

Lawguy

02-27-09, 03:31 PM

Chris Wiggles is conflating Image Quality with Display Quality and they are definitely not the same. We all want a display that can produce an image that is as accurate to the source as possible. But, to me, it doesn't end there. I really want a display with controls that that permit me to have full control over the source so that I can also tweak the image if we need to.

A perfect example is my current display, the RS20. It is nearly perfectly accurate (both in color and greyscale) but it has a broken CMS. Many would say, "your display is accurate, what more do you need?" The truth is that with good sources, I don't need anything. However, every once in a while something comes a long that just looks better when the image is tweaked. The controls that I have are (currently) imperfect (a fix is said to be on the way) for the purpose of tweaking the picture the way I want.

I suspect that these issues are mostly philosophical and in practice there is really no disagreement among us.

I understand why people would be against making random changes to a picture because odds are they will be doing more harm than good. However, this is not what I am advocating.

Tryg

02-27-09, 03:33 PM

High image quality and accuracy are not two seperate entities.

I would beg to differ...obviously

Just because she may have all the parts that make her a woman, it doesn't mean she's beautiful. :)

R Harkness

02-27-09, 03:42 PM

High image quality and accuracy are not two seperate entities. You are confusing yourself.

So...every accurate image will be of high quality?

That is what you are saying if we take your statement seriously. Yet can you not see how obviously wrong that is?

Hint: Not all sources are of good quality. A display that is recreating a poor source image accurately will be showing you a poor image.

That's why one can simultaneously and correctly say "This image is accurately reproduced" and "This is a poor image."

Why? Because "image accuracy" and "image quality" are not precisely the same thing.

You've never heard of poor source material or bad transfers? I'm not the confused one here....this is all (or should be) grossly obvious stuff.

CaspianM

02-27-09, 03:52 PM

Can you not see how obviously wrong that is?

....

wrong!! No you see it wrong.:)
Your entire argument of seperating quality and accuracy is silly imo. The bottom line whatever the feed if your device cannot do it trnaspantly or close enough you have have low quality device.

I believe Chris set up a clear argument of the case and I'll leave it at that. This is a forum of resistance and argument and eventually right and wrong.:confused:

Lawguy

02-27-09, 04:09 PM

The bottom line whatever the feed if your device cannot do it trnaspantly or close enough you have have low quality device.

Exactly! This, however, has nothing to do with image quality.

HoustonHoyaFan

02-27-09, 04:36 PM

...In other words: The qualities of an image can obviously be described independently of it's accuracy. For instance: whether an image is low or high in contrast, has very high or low black levels, rich color saturation
or washed out, high or low resolution, is comparatively bright or dim, has 3 dimensionality or appears flat and on and on....

The goal can be expressed purely in terms of image quality: Deeper black levels presented simultaneously with brighter bright areas - expanded contrast - contribute to a greater sense of vividness, realism of contrast, image depth etc....Those examples do not define the quality of an image they are just parameters. one director may choose an ultra contrasty cartoon like approach ie. Hitman, another may choose a more realistic muted approach like say Eyes Wide Shut. Are you suggesting that Hitman has better image quality than Eyes Wide Shut because the Hitman has a more colorful, contrasty, saturated, brighter, 3D appearance? Are you saying that directors like Spielberg and Coppola do not produce "good quality image" because they frown on that "gimmicky" look?

TomHuffman

02-27-09, 04:48 PM

I don't want to put words into anyone's mouth, but it seems to me that this image quality vs. accuracy is more of a verbal dispute than a real one.

Image quality is a function of the display, the source material, and the environment in which it is viewed.

Watch a great transfer on a perfectly calibrated, high-quality projector in a room flooded with ambient light and you won't have high image quality. Use a perfectly calibrated, high-quality projector installed in a bat cave to watch a terrible transfer and you won't have high image quality. Watch a great transfer in a bat cave on a POS projector or one that is poorly adjusted, it won't be good image quality.

All the above can be true and Chris is still correct about what constitutes display calibration, which is adherence to accepted standards which, when achieved, makes the display a device that reveals the source material as transparently its underlying performance allows.

On the other hand, when people start intentionally changing the gamut or the white point because they just prefer the look, then that's not calibrating a display, it's tweaking it to serve one's personal preference. There's nothing wrong with that, but this activity should not be confused with calibration. That may sound like a purely verbal distinction as well, but I think it is necessary to distinguish what individuals may do to serve a subjective preference and what calibration does to bring a display as close to specs as possible. They are inherently different activities.

dovercat

02-27-09, 05:06 PM

Image quality maybe subjective. But we are all humans with similiar eyes - brains. We tend to like the same images. Lots of academic papers have been written detailing ways to objectively measure image quality without reference to it being accurate to its source. Objective methods of no-reference perceptual quality assesment of images exist and they get over a 90% correlation with peoples subjective opinions of which images look best. It is not all down to personal taste with different people preferring different images. We all tend to prefer the same images and the reasons why can be objectively measured.

As for calibration making the image more true to source. I agree up to a point. As I understand it a standard ISF calibration is to D65 color temperature. This makes the greyscale more neutral, enabling subtle shades of color to be seen as they no longer have to overcome the bias of a non-grey greyscale.

However in most displays the gamma curve will also effect color hue and saturation. Many displays do not have controls for adjusting - fine tuning the gamma curve. With a ISF standard calibration gamma will not be fine tuned even if possible. A preset gamma will be picked with anywhere between 2.2 and 2.6 being acceptable and ignoring dE - they do not look the same. The higher the gamma the lower the average picture brightness and the higher the perceived contrast, but also the higher the color saturation and the more biased the color hue towards the color most present in a mix.

Accurate color primaries are also needed for accurate color representation particularly of highly saturated colors. Standard ISF calibration does not cover color managment systems even if present in the display.

Color saturation - gain is set by eye in a standard ISF calibration using a filter and test screen, not the most accurate method. Also white, black level and sharpness are set up by eye using test screens again this can be inaccurate.

As I understand it many - most displays are incappable of being calibrated to meet the standards, displays that can meet the standards cost a small fortune and are used as reference displays. What display you choose, what it is capable of, and how you choose to set it up or which calibrator you choose to do it for you, comes down to personal preference.

If displays were capable of meeting the standards and the calibration was strictly to the standard, it would not matter which display you bought or who you got to calibrate it. I believe it does matter and this means display calibration can not In my opinion be viewed as a proper calibration to a standard. The displays are not capable and the calibration is too dependent on the calibrators skill, experience, personal preferences.

R Harkness

02-27-09, 05:52 PM

Those examples do not define the quality of an image they are just parameters. one director may choose an ultra contrasty cartoon like approach ie. Hitman, another may choose a more realistic muted approach like say Eyes Wide Shut. Are you suggesting that Hitman has better image quality than Eyes Wide Shut because the Hitman has a more colorful, contrasty, saturated, brighter, 3D appearance? Are you saying that directors like Spielberg and Coppola do not produce "good quality image" because they frown on that "gimmicky" look?

I already gave a detailed response to that issue, where I separated the three issues of Image Accuracy, Image Qualities in terms of describing the qualities of an image and it's perceptual effects, and then there is the issue of Preference, where we rate certain images as preferable.

Your reply is another example underlying my point. One can describe an image in terms of it's qualities, e.g. "Contrasty," "Saturated," "Brighter," "3D Appearance etc. And those description impart relevant information about how an image looks, independently of whether it is accurate to the source because describing the quality of an image is not logically the same as describing whether it is accurate to X source.
We can talk about how the image LOOKS, it's effect, whether we like it. AND/OR we can talk about whether the image is an accurate reproduction from the source material.

People who research our perceptual system and what people perceive as preferable or realistic image quality routinely do so without reference to whether something is accurately reproducing a source or not. In fact they routinely manipulate images however they wish to test how people report the differences impact the image quality, or what people tend to prefer. This would not be possible or make sense if the arguments you are using, making "image quality" inseparable from "accuracy" were valid.

How much more obvious and clear can this be? I don't know how anyone can possibly deny it.

You have to answer the question:

Is there such thing as "poor" source image quality?

If you answer "yes," then you have understood that "Accurate" and "Good Image Quality" are not one and the same. Because an accurately reproduced image can nonetheless be acknowledged as poor image quality, due to the source image being of poor quality. The logic is simple

If you answer "no," there is no such thing as a poor quality source image, then you've just made nonsense of film restoration, film transfer processes, reviews of image transfer quality...and most everything else we discuss.

So, which is it?

R Harkness

02-27-09, 05:53 PM

TomHuffman gets it!

Cheers.

Lawguy

02-27-09, 07:37 PM

i don't want to put words into anyone's mouth, but it seems to me that this image quality vs. Accuracy is more of a verbal dispute than a real one.

Image quality is a function of the display, the source material, and the environment in which it is viewed.

Watch a great transfer on a perfectly calibrated, high-quality projector in a room flooded with ambient light and you won't have high image quality. Use a perfectly calibrated, high-quality projector installed in a bat cave to watch a terrible transfer and you won't have high image quality. Watch a great transfer in a bat cave on a pos projector or one that is poorly adjusted, it won't be good image quality.

All the above can be true and chris is still correct about what constitutes display calibration, which is adherence to accepted standards which, when achieved, makes the display a device that reveals the source material as transparently its underlying performance allows.

On the other hand, when people start intentionally changing the gamut or the white point because they just prefer the look, then that's not calibrating a display, it's tweaking it to serve one's personal preference. There's nothing wrong with that, but this activity should not be confused with calibration. That may sound like a purely verbal distinction as well, but i think it is necessary to distinguish what individuals may do to serve a subjective preference and what calibration does to bring a display as close to specs as possible. They are inherently different activities.

+1

ChrisWiggles

02-27-09, 10:47 PM

Chris that's not true.

You can say an image is "accurate" or not in terms of meeting calibration standards. We can likely agree.

You can make an argument that displays "ought to reproduce the source content as accurately as possible" and perhaps find many agreeing with you.

That's fine.

What I take issue with is conjoining your argument with erroneous statements - an indefensible claim that "image quality" must therefore only reference "image accuracy." Which is nonsense.

I mean, you may have your own, constrained idea of what constitutes image quality, but it simply doesn't account for our ability to discuss image quality in of itself. (And keep in mind I am not arguing against the concept or desirability of our displays being able to accurately reproduce the source as intended. I'm arguing against questionable claims you are using to argue your position).

Now what may be happening here to some degree is that the word "good" is being used, which is a value judgement. (Nothing wrong with that). I'm using it as a short hand for the fact there are parameters in image quality that, generally, when pushed in one direction most people would tend to call "good." But we can for a moment drop the term "good" and keep talking about "Image Quality" (in a descriptive sense).

In other words: The qualities of an image can obviously be described independently of it's accuracy. For instance: whether an image is low or high in contrast, has very high or low black levels, rich color saturation
or washed out, high or low resolution, is comparatively bright or dim, has 3 dimensionality or appears flat and on and on.

You know this. This is the way we describe images to one another - the quality of a source material, the characteristic abilities of a display etc. And such descriptions are useful observations independent of mentioning accuracy to any particular source material. It's why we can say, for instance, that the big advances JVC made in their projectors that make for deeper black levels, while not compressing bright areas (no DI), are significant and in large part are advances toward a desirable goal.

The goal can be expressed purely in terms of image quality: Deeper black levels presented simultaneously with brighter bright areas - expanded contrast - contribute to a greater sense of vividness, realism of contrast, image depth etc.

That is saying something valid and important, and comprehensible...and uncontroversial...without having to mention "accuracy to X source material."

Okay. I think we are both advanced enough, and many others in this conversation are advanced enough to see this point, and I would agree with you. Perhaps we could call this image attributes or the qualities of the image or something, but this is wholly different from a judgement about whether it is a "good" image or not. And I would argue further that discussions about particular attributes of displays still are discussions that in the end revolve around accuracy. For instance greater contrast ratio performance, greater MTF allows greater image accuracy, hence we deem that as qualities to a display which are preferable to another display.

Now, simply because we frequently discuss these attributes on their own and we don't say "this display is sharper hence more accurate" doesn't mean that the reason we prefer that is in the end something other (even if not explicit) than accuracy. We often just say "this display has better contrast ratio" or "this display has better MTF performance" or "this display has less banding" or whatnot, but in the end I think the underlying point of those judgements still gets at accuracy.

But if we mean "image quality" where "quality" equates to "desirable traits" which essentially equates to "good," then in those terms image quality becomes a value judgment. In which case you are trying to make your own value judgement "what is good" apply for everyone else, which doesn't work. Why should Chris Wiggles' "good" determine what everyone else will see as "good?"

Exactly. And that's my point. Nobody should care what my "good" is, or what your "good" is, at least insofar as an image reproduction chain is concerned.

So if anything, perhaps I am pushing an absolutist position in a simplistic way because I don't want to cede territory to the people who want a sort of "personal-preference" calibration and that's what they mean by "image quality vs. accuracy." And in that sense I absolutely reject any difference.

Chris, you can say an image is "accurate" (in terms of reproducing source material) or "inaccurate" and, really, that says all you need to say. Trying to attach the value judgement "good" ("image quality") is unnecessary and
will fail.

I absolutely agree, and that is my point.

If it's understood that value judgements are separate from fact statements (always fun to weave in moral philosophy :) ) then we make more sense of how people discuss these issues and allow important nuance into our approach.

So my claim is:

1. There is the issue of "accuracy," in terms of "Am I seeing the source material reproduced accurately?" The attempt to set up objective parameters can help us come to agreement on whether an image is being
reproduced more or less accurately.

Yep!

2. There is the issue of "image quality" in it's descriptive sense. We can talk about how an image "looks" - it's qualities - and it's general effect on
how we perceive images. A lot of which can be discussed in fairly objective terms as to how pushing an image parameter in X direction will tend to be perceived by people as having Y effect. (E.g. increasing contrast increases apparent sharpness and image dimensionality).

3. The issue of value judgement: Is X image "good?" This is subjective,
which is why we'll see variance in preference for different types of images.

Conflating all of those into "We can only talk of Image Quality in terms of Accuracy" is really, really getting things wrong, both logically and in it's inability to account for how we actually discuss and inform one another
about images. [/quote]

But I don't think it does at all. I spend as much time as anyone discussion the 'image qualities' or 'image attributes' that you talk about here. But I think as I said above in this post, this really still revolves around the goal of greater accuracy which is gained by greater performance for all these attributes.

I understand your point here, but I don't think that you need to separate the concept of accuracy from qualities. Clearly I don't believe that discussing these attributes is bad since I do so all the time! But in the end, isn't this all part of a discussion of achieving a more accurate image? Even if the term accuracy never comes up in these many threads, that really is the background motivation for all of us, I would argue.

What I fear is the novice wandering into threads like this and coming away with the idea that there is some fundamental difference between an accurate image and a 'good quality image' (as defined purely subjectively by preference).

(Just to add about #3...while there is going to be variation in the value judgement "Is X image good?" to a degree, the facts underlying #2 will mean that there will be meaningful trends to observe in what people will
tend to call "good." And it has been studied: how changing certain image parameters produce trends in what subjects will tend to evaluate as being preferable).

I think all this is pretty obvious and in fact we would have little disagreement
when talking about images in "real life." To the extent you want to say a display device "ought only reproduce accurately the source" - that's a value judgment. And it's fine to argue for that value judgment, you may get many on your side. I'm simply being pedantic and pointing out you've mingled in some indefensible propositions ("it only makes sense to talk of image quality in the sense of image accuracy") while trying to argue your position.

Cheers,

I think we do agree, and we're probably both being pedantic. But you surely can understand my point about not ceding ground to the subjectivist willy-nilly knob turners. But if we are discussing the attributes of an image, when we discuss whether it is good or bad, in the end isn't that always concerned with its accuracy?

If we discuss as we all have, the behaviors and attributes of a DI system, don't we judge the strength and weaknesses of that in terms of how it improves certain things and perhaps worsens others? And isn't that always in some way or another concerned with how that matches the ideal image? I think it's certainly clear that I'm not claiming that those discussions are somehow invalid, clearly they are very valid. What I am saying is that they still are fundamentally involved with accuracy. Now, again, as I said before there are subjective grey zones that we all understand as more advanced users, and DI is a great example. In approaching the ideal, one person may choose to utilize a DI and view certain improvements in accuracy as more important, while another may choose not to use a DI because certain departures from accuracy may be too much of a concern.

I think we agree overall, but are getting a little bit lost in two different senses of "image quality" versus "image qualities (attributes)." And I guess my point is that the former is fundamentally the same thing as accuracy, and your point is that the latter is something we can discuss as well. I see that point, but I still think that the latter is concerned with accuracy, but isn't always discussed with accuracy explicitly in mind.

dovercat

02-28-09, 09:39 AM

HoustonHoyaFan

02-28-09, 02:26 PM

...The film Matrix may give the matrix-world a green hue or another film may give night scenes a blue hue. If you tune the display to remove the hue you improve perceived image quality...:eek::eek::eek:

Art Sonneborn

02-28-09, 07:35 PM

Is it just me or has this thread gone so far off topic as to no longer even be applicable to the OP ? Calibration results in images as close to being transparent to the source.Whether we end up liking that (just like the post above about removing the green hue of the matrix) isn't even germain.

We could go on forever arguing over things like that which are, by definition , subjective.

Art

WOLVERNOLE

02-28-09, 08:28 PM

GeorgeAB

03-01-09, 01:15 AM

Losing sight of, ignoring, or denying the fundamentals always results in confusion, misunderstanding, and chaos:
http://www.avsforum.com/avs-vb/showthread.php?t=1021933

dovercat

03-01-09, 08:26 AM

"Is it just me or has this thread gone so far off topic as to no longer even be applicable to the OP ? Calibration results in images as close to being transparent to the source.Whether we end up liking that (just like the post above about removing the green hue of the matrix) isn't even germain."

"We could go on forever arguing over things like that which are, by definition , subjective."

Art

My point was relevent to previous posts arguing over if calibration automatically equals good picture quality. I do not believe it does, a point you obviously agree with but think is irrelevent. As I pointed out in my previous post perception of image quality is not purely subjective, it can be objectively measured.

The fact that films are mastered using the standards and the makers want the films to have wide appeal so give them a picture quality they hope most people will like, shows that film makers think having a good picture quality is relevent. Calibration to standards enables you to best appreciate their efforts. But not everyone prefers a green tinted matrix world or on some films set mostly at night using a muted color palette and blue tint. I am not always impressed with or in awe off the film makers efforts.

The thread title is "Calibration... are we just fooling ourselves?"
and makes the point "When you calibrate an instrument it generally means you will change it to a specified standard (or be able to quantify the difference). In other words it will do exactly what the other instrument will do. With projectors this is simply impossible."

In my last post the last paragraph states that overall I agree with the OP. Earlier I state that I think the standards them selves are less than ideal. But go on to say I think calibration as far as possible is a good idea because otherwise you will end up adjusting the image on a per film basis, rather than sitting back and watching the movie.

In my post prior to that one I point out that some calibration is not even trying to meet all the existing standards, just D65 color temperature and get in the ball park for gamma, black and white level, and color saturation. That post goes on to agree with the OP that most displays do not meet and are incapable of being adjusted to meet all the standards, and ends by clearly agreeing with the OP that it is not calibration to a standards in the true sense.

Are posts agreeing with the OP not relevent to the thread, only posts defending calibration?

R Harkness

03-01-09, 08:43 AM

Losing sight of, ignoring, or denying the fundamentals always results in confusion, misunderstanding, and chaos:
http://www.avsforum.com/avs-vb/showthread.php?t=1021933

That's a nice link. That said, I don't see that it says anything that people on this thread don't already know and that hasn't already been acknowledged.

Art Sonneborn

03-01-09, 08:55 AM

The thread title is "Calibration... are we just fooling ourselves?"
and makes the point "When you calibrate an instrument it generally means you will change it to a specified standard (or be able to quantify the difference). In other words it will do exactly what the other instrument will do. With projectors this is simply impossible."

Are posts agreeing with the OP not relevent to the thread, only posts defending calibration?

First of all we can have things like extremely high quality time pieces calibrated to an atomic clock but they aren't ultimately as precise or accurate as that AC. So on that standard yea one can calibrate a projector. It seems the problem in this thread is the less accurate projectors have owners who feel that it isn't worthwhile then hammer the whole idea but not from the calibration perpective itself but from the perepective of they don't like the picture after, which is irrelevant to the thread IMO.

The whole pretty picture seasoned to taste ,like the torch mode crushed blacks used at Best Buy on the floor for example is subjective.

If I said calibration often increases the life of the projector compared to torch mode which one personally may like (as a defense of calibration) that would be totally irrelelvant to the OP.

Art

ddingle

03-01-09, 09:28 AM

I always think back to my first ISF training with Joe Kane. He chaired the SMPTE commitee to set standards for white balance.
6500 was decided on for a variety of reasons. Joes demonstrations always looked fabulous. It convinced me to get a test instrument and start calibrating my own set and others.
Currently Joe is promoting his Samsung DLP set to exactly the SMPTE standard. If anyone here wants to know the value of calibrating to a standard they should audition this set.
Not to say that single chip DLPs are the best display device,but Joe's set might be the best example of what "properly calibrated" means.

dovercat

03-01-09, 09:29 AM

"First of all we can have things like extremely high quality time pieces calibrated to an atomic clock but they aren't ultimately as precise or accurate as that AC. So on that standard yea one can calibrate a projector"

At least at the point when it is calibrated it is accurate to the atomic clock. It just goes out of calibration overtime.

"It seems the problem in this thread is the less accurate projectors have owners who feel that it isn't worthwhile then hammer the whole idea but not from the calibration perpective itself but from the perepective of they don't like the picture after"

I think calibration is worthwhile. But the OP point was the use of the word calibration is inaccurate. I agree with the OP that tune-up to best - most accurate picture possible is a better description.

If my projector was calibrated in the normal sense of the word the report would have a big red failed at the top. It is incapable of meeting all the display standards and is incapable of simultaneously meeting some of the display standards. It is not a reference quality display calling it calibrated will not make it so.

Tune-up is a better description because at least with my projector I do not aim to strictly adhere to the standards. My settings are guided by the standards as an ideal. I can not for example adhere strickly to D65 color temperature across the top end of the greyscale because doing so cripples contrast ratio.

The methods used for calibration are not extremely accurate, sensors are inaccurate at the bottom end of the greyscale and using things like DVE disc and filter to set color saturation is not precise. Setting the display up to what you believe is the standard and not making fine adjustments by eye for best picture quality strikes me as odd, after all I believe if you get several calibrators to calibrate the same display you do not get identical results. My ultimate guide will always be if the image looks good on lots of different films. The only time the film should look odd - wrong is if there is a known issue with the film like matrix green world.

Also calibration seems to be equated only to setting color temperature to D65. D65 color temperature is important for a good picture, but it is not the only thing that is. Other things are also crucial to a good image.

Tryg

03-01-09, 09:42 AM

dovercat gets it!

the rest are just fooling themselves :)

Gary Lightfoot

03-01-09, 09:54 AM

Tune-up is a better description because at least with my projector I do not aim to strictly adhere to the standards. My settings are guided by the standards as an ideal. I can not for example adhere strickly to D65 color temperature across the top end of the greyscale because doing so cripples contrast ratio - gamma.

D65 is a standard but contrast ratio isn't, and different gamma will achieve different results with different contrast capable displays so that isn't an ideal aim either.

I agree that if you want more lumens and contrast then you can get that by aiming for a higher colour temp at higher IREs, but then your sacrificing accuracy for a personal preference, so you're making a choice between accuracy and something else.

The methods used for calibration are not extremely accurate, sensors are inaccurate at the bottom end of the greyscale and using things like DVE disc and filter to set color saturation is not precise. Setting the display up to what you believe is the standard and not making fine adjustments by eye for best picture quality strikes me as odd, after all I believe if you get several calibrators to calibrate the same display you do not get identical results. My ultimate guide will always be if the image looks good on lots of different films. The only time the film should look odd - wrong is if there is a known issue with the film like matrix green world.

I feel that even with a display that can't be completely accurate when calibrated, it will still be more accurate than a display that is left as delivered, so no programs will be anywhere near close to accurate with that display, but a calibrated display (as close as it will allow) will be better than one that isn't, if accuracy is your aim. Different calibrators may end up with different results, but they would all be in the same ball park and still more accurate than the out of the box settings. Professional calibrators would have spent more money on their calibration kit than those who buy a Spyder or Eye One colorimiter, so you would expect better results from them than from a consumer probe.

I do agree with you that if your eye notices something is wrong (pink tint in the greyscale even though the calibration software says D65 for example), then it probably is wrong. Otherwise I'm in the calibration camp and not the tuning camp. :)

I also agree that there's more to an image than D65, but if you don't have the other things but you do have a greyscale that's at or close to D65, you're still more accurate than a display that doesn't have the other things or a greyscale that's at or close to D65.

Gary

Art Sonneborn

03-01-09, 10:04 AM

I think calibration is worthwhile. But the OP point was the use of the word calibration is inaccurate. I agree with the OP that tune-up to best - most accurate picture possible is a better description.

If my projector was calibrated in the normal sense of the word the report would have a big red failed at the top. It is incapable of meeting all the display standards and is incapable of simultaneously meeting some of the display standards. It is not a reference quality display calling it calibrated will not make it so.

And in fact that may well be the case with your projector. As I've said some projectors can't accurately be set to the standards.

The methods used for calibration are not extremely accurate, sensors are inaccurate at the bottom end of the greyscale and using things like DVE disc and filter to set color saturation is not precise. Setting the display up to what you believe is the standard and not making fine adjustments by eye for best picture quality strikes me as odd, after all I believe if you get several calibrators to calibrate the same display you do not get identical results. My ultimate guide will always be if the image looks good on lots of different films. The only time the film should look odd - wrong is if there is a known issue with the film like matrix green world.

This goes back again to my season to taste comment. Many calibration devices (PR650 for example) are extremely accurate. Just because you set up your display with inaccurate techniques does not negate the concept of calibration.

Also calibration seems to be equated only to setting color temperature to D65. D65 color temperature is important for a good picture, but it is not the only thing that is. Other things are also crucial to a good image.

I've not seen anyone here say that color was the only calibrated parameter where did you get that ?:confused:

dovercat

03-01-09, 10:42 AM

"dovercat gets it! the rest are just fooling themselves" Tryg

At least the OP seems to think I am being relevent, even if I do ramble and make my points poorly.

"D65 is a standard but contrast ratio isn't, and different gamma will achieve different results with different contrast capable displays so that isn't an ideal aim either"
"I agree that if you want more lumens and contrast then you can get that by aiming for a higher colour temp at higher IREs, but then your sacrificing accuracy for a personal preference, so you're making a choice between accuracy and something else."
Gary Lightfoot

Specifications exsit for contrast, white and black level and gamma. I am in the UK, PAL rather than NTSC land so EBU-Tech 3320 User requirements for video monitors in Television, for a grade 1 monitor used for high-grade technical quality evaluation of picture capturing, postproduction, transmission and storage, these are: Black Level below 0.1 cd/m2. White Level 70 to at least 100 cd/m2 ( ITU-R BT.500-11 requires monitor brightness up to 200 cd/m2 for tests simulating domestic viewing conditions.)
Full screen (1% patch) contrast ratio above 1000 to 1. Simultaneous contrast ratio (with EBU box pattern) above 200 to 1.
Gamma. The luminance gamma characteristic (electro-optical transfer function) of the screen should be equivalent to those of a reference CRT with the rendering intent (dim-surround) expected of a TV system. It is believed that a nominal value of 2.35 is appropriate. The ideal electro-optical transfer function shall be followed such that it remains within a tolerance of +/- 0.15 of the ideal gamma value from 10% to 90% of input signal level.

In EBU-TECH 3321 Guide lines for consumer flat pannel displays.
White level for display upto 50inch diagonal small area peak white at least 200 cd/m2, for larger displays less.
Black level less than 1 cd/m2
Contrast ratio simultaneous at least 200:1
Display Gamma 2.35

So a gamma of 2.2 to 2.5 is desirable to meet the specification. I think desirable for good image depth and color accuracy. I aim for 2.2, the gamma dictates the contrast ratio 10% to 100% white. I achieve the desired gamma by increasing my white level at the expense of D65 color temperature accuracy (my display does not have gamma correction adjustable at multiple points in the greyscale.) This is personal preference in that I am sacrificing strict adherence to one specification D65 color temperature to meet another specification Gamma. The only other way to meet the gamma specification with my display is to lower the brightness - black level, this crushes/clips shadow detail and I find it less desirable than a off D65 90% white. I also find strict adherence to D65 color temperature and letting gamma fall where it may, gives a less pleasing image.

"that may well be the case with your projector. As I've said some projectors can't accurately be set to the standards" Art Sonneborn

How many projectors meet all the display standards to be reference quality?

"Many calibration devices (PR650 for example) are extremely accurate. Just because you set up your display with inaccurate techniques does not negate the concept of calibration" Art Sonneborn

I agree thats why people pay for professional calibrators. Less hassle and more accurate, at least in theory. But how accurate - finely adjustable are most displays. You may know more precisely how far off target you are, but does it do you any good. Are most displays capable of that amount of precise adjustment to exactly meet standards.

"I've not seen anyone here say that color was the only calibrated parameter where did you get that ?" Art Sonneborn

True, but in general the likes of ISF seem to emphasize it.

Bob Sorel

03-02-09, 12:53 PM

So a gamma of 2.2 to 2.5 is desirable to meet the specification. With my display I aim for 2.2. A gamma of 2.2 is I think desirable for good image depth and color accuracy. The gamma dictates the contrast ratio 10% to 100% white. I achieve the desired gamma by increasing my white level at the expense of D65 color temperature accuracy (my display does not have gamma correction adjustable at multiple points in the greyscale.) This is personal preference in that I am sacrificing strict adherence to one specification D65 color temperature to meet another specification Gamma.
Apparently you know this because you have spent some time "tuning" your unit with equipment. Since you can never achieve a totally calibrated picture anyway, why not just eyeball it until it looks "good" to you?

Lawguy

03-02-09, 01:03 PM

dovercat

03-02-09, 04:02 PM

"Apparently you know this because you have spent some time "tuning" your unit with equipment. Since you can never achieve a totally calibrated picture anyway, why not just eyeball it until it looks "good" to you?"
Bob Sorel

I do not disagree that projector setup close to standards should give the best picture quality. As the film makers are using the standards when mastering the disc and want you to like the film. Equipment can make this alot easier and faster to do, and give you confidence real or imaged that the image is as good as your display is capable of.

The major problems I see with just going by eye is you may never finish, adjusting the image every time you watch a film. Unless you have read about the subject you may not understand how various settings you adjust work or interract. Wandering about may find the way, but using standards and equipment gives you a map and compass. The major advantage being that the map and compass tell you when you have arrived at your destination and reduce your urge to carry on wandering about.

Another way I look at it, is it depends what you put your trust in. When you want to get somewhere you do not know but might recognize when you arrive do you:
A: Just leave and rely on yourself. Tune by eye. Trust yourself.
B: Read up how to get there and buy a map. Self calibration with equipment trying to meet standards. Trust yourself when reassured by a sensor readout.
C: Phone for a cab. Hire a certified calibration. Trust someone else.

I was not saying you should not use equipment to help you get a image close to standards. Just agreeing with the OP that the word calibration is used somewhat loosely. Saying a projector is calibrated implies to me that it meets all the standards. My projector can not, so I would describe it as tuned for the best image quality I can get it to do. Since it can not meet all the standards simultaneously, I set it up with a compromise, rather than strictly sticking to meeting one standard D65 while ignoring another Gamma.

A reference quality display can pass calibration as it can meet all the required standards to be used as a reference display. I would use the description calibrated to refer to a calibrated reference display.

Gary Lightfoot

03-03-09, 02:59 PM

Hi Dover,

I agree that there are specs for video monitors in the industry (they have to be for accurate mastering), but we're not calibrating those monitors, just projectors, so those specs can't be applied stringently here. If you have a pj with less than say 2000:1 on/off CR for example, higher gammas (above 2.2) will crush a lot of shadow detail, so a lower gamma (say around 2) may be better for that particular pj. The contrast measures are for a direct view set in a specific viewing environment (dim room), not a projector in an ideal environment. Different environments will also benefit from different but suitable gammas too.

If we followed those specs, we'd have to have some ambient lighting in the room too (probably a specific fL measured at the screen), and that would certainly reduce the contrast capability of any projector in that room and call for a different gamma to the one we'd use in a fully light controlled room. Following those specs would limit the performance capabilities of a lot of projectors and that would be a shame, especially if you'd spent a lot of money on it for the extra contrast performance.

Gary

Art Sonneborn

03-03-09, 03:07 PM

The major problems I see with just going by eye is you may never finish, adjusting the image every time you watch a film. Unless you have read about the subject you may not understand how various settings you adjust work or interract. Wandering about may find the way, but using standards and equipment gives you a map and compass. The major advantage being that the map and compass tell you when you have arrived at your destination and reduce your urge to carry on wandering about.

And this is no small thing IMO. When you put a disc in you can just say yep this is green or blue or soft or lacks detail in the shadows etc. In fact you can just let go.

Art

Gary Lightfoot

03-03-09, 03:14 PM

Couldn't agree more - once you know things are set correctly, you know that what you are seeing is what's coming from the source. If something looks that way then it's supposed too, so you don't need to tweak.

A lot can be said for that! :)

Gary

dovercat

03-04-09, 07:42 AM

I agree about the white, black level being dependent on display type and viewing environment - ambient light. I think SMPTE has projecor recommendations.
The contrast specs are minimum specs, so should not limit the performance of a projector.
The gamma specs are for 10% to 90% white so should not adversely affect shadow detail in theory. In practice most displays like mine do not have gamma correction so you have to set it for the whole curve - greyscale, which will as you say effect shadow detail.

My understanding of gamma is that the mastering standards were designed for an end user with a crt display. So the idea of gamma on non-crts is to mimic a ideal crt display. But there seems to be alot of different opinions on what display gamma should be.
TV cameras with an end gamma of 1.2 is equal to 2.2222 display gamma
SMPTE 170M NTSC 2.2
SMPTE REC 470BG PAL 2.8 (The BBC - UK I think uses EBU spec below)
EBU-Tech 3320 (PAL) reference monitors 2.35, with crt being 2.2 to 2.5
High Definition BT 709 with an end gamma of 1.2 is aproximately equal to 2.35
High Definition BT 709 with an end gamma of 1.125 is aproximately equal to 2.22
THX certification requires 2.22
Charles Ponyton "Gamma" and its disguises: The nonlinear mappings of intensity in perception, CRTs, film and video, SMPTE J., December 1993, 1099-1108. says crt gamma is 2.5
DVE Joe Kane I think recommends 2.5
ISF I think also recommends 2.5, but I think say between 2.2 and 2.5 is ok.
Denis Pelli, Pixel independence: Measuring spatial interactions on a CRT display, Spatial Vision 10 (1997), 443-446. Gives CRT as 2.3
I have also read some crt results where gamma varies with phosphors Iiyama HM204DT CRT monitor R 2.35 G 2.36 B 2.53
From looking at PAL dvds with a display set to 2.2 then 2.35, it looks to me as if most dvds are mastered to 2.2. With 2.35 color looks less nice - smooth - natural, and on some over saturated, while the image does benefit from more contrast.

Many digital displays have several gamma presets to adjusted it to suit viewing conditions. The more ambient light the lower gamma so shadow detail can be seen and the average image is brighter. The lower ambient light the higher gamma to increase perceived contrast to compensate for the eyes contrast compresion when the display has a dark surround. Some displays with low contrast ratios may have gamma as low as 1.9. While other displays with poor - high black level may have a high gamma value to compensate. So on digital displays it seems to be all over the place.

However gamma too far out of specification I find adversely affects color saturation and hue. If you have seperate color management this is probably not an issue.

Out of intrest what specs do you use, apply stringently.

Gary Lightfoot

03-04-09, 04:25 PM

I don't think I apply anything stringently other than trying to get the best out of the pj with my limited knowledge and experience. Setting the white and black levels with a test disk is probably the most important thing you can do with any display since a lot of how the image looks relies on that being done correctly. Setting D65 greyscale and possibly the gamut of your choice are about all you can aim for with many projectors unless you have a vp and I think D65 is very important for the reasons Art concurred with.

I do aim for a specific fL level though - digital theater is 12fL +- 1fL and I prefer something around 9fL since it looks more cinematic and minimises source artifacts and image noise from SD material. Commercial theaters tend to have less than 12fL so doing the same in the home works for me in a light controlled space. Joel Silver is recommending 20fL for HD material though. 12fL is around my max and then it will reduce as the lamp dims with age. I'll use CC filters with recalibration to D65 to help improve contrast and reduce lumen output to get close to my fL preference. Even with todays high on/off machines I'll still do this just out of curiosity or use an ND filter (or iris) if needed.

I won't aim for a specific gamma since it could have an adverse effect on the image, so setting something that works well seems to me to be the best option if you have the choice. Like you say most pjs only have presets to choose from so that's a limitation and you have to pick what works or make compromises elsewhere. 2.2 does seem like a good guide though.

Other specs for commercial theater are things like center to edge brightness and seating distance/viewing angles. Some will aim for a CIH set up to emulate the correct implementation of scope material and others are happy with CIW. Loads to choose from and not many will necessarily be within our control. :) There may even be specs for escape lighting but I don't think using using those is a good idea. ;)

I wonder what projectors people here would class as 'reference'?

Gary

GeorgeAB

03-05-09, 12:07 AM

I wonder what projectors people here would class as 'reference'?
"The 'Baraka' Blu-ray with its 65mm source and 8K UltraDigital HD restoration process was intended to be a benchmark for the burgeoning Blu-ray industry. I needed a 1080p reference display that would allow me to do the critical quality assurance of the Blu-ray encoding and authoring. Pixel discrete flat screen monitors in the 42-55" were the first step, but I needed something bigger, but with reference quality. The Samsung SP-A800B, with its 100" diagonal picture gave me that next step up for our target high-end Home Theater market. The SP-A800B provided me with accurate color, deep blacks, superb sharpness, and surprising brightness uniformity. This projector gave me the best pixel for pixel reference for detecting artifacts in the image processing of 'Baraka.' This tool allowed me to push the delivered quality of the 'Baraka' Blu-ray to it's highest potential." Christopher M. Reyna - Restoration Producer

Best regards and beautiful pictures,
G. Alan Brown, President
CinemaQuest, Inc.
A Lion AV Consultants Affiliate

"Advancing the art and science of electronic imaging"

Gary Lightfoot

03-05-09, 07:39 AM

Thanks for that George - what impresses me the most about the Samsung (so far) is the fact that it's actually affordable (for me anyway)!! I was expecting something along the lines of a Sim2 HT5000. :)

I wonder how many other projectors he looked at before deciding the Samsung qualified as reference? It would be interesting to see which ones didn't come up to scratch.

Cheers

Gary

Lawguy

03-05-09, 08:32 AM

In the interest of full disclosure, Joe Kane Productions apparently did the QC on the transfer of Baraka and is also affiliated with Samsung on marketing its line of projectors. If JKP had anything but a ringing endorsement for the SP-800B, I expect that it would be sued for breach of contract.

Samsung projectors are also not priced for value in the US. They can be had for a song in Korea, a bit more in the UK, but are much more expensive in the US.

I have no opinion on whether the SP-800B is, in fact, a reference quality projector because I have not seen one. I would just not take JKP's word for it because they have a pecuniary interest in the project.

Gary Lightfoot

03-05-09, 09:06 AM

I knew JK and Samsung were affiliated, but wasn't aware of the Baraka link. I'd still like to know how Christopher M. Reyna decided on the Sammy unless it was totally a JK endorsement rather than a genuine image quality based decision.

Cheers

Gary

Cam Man

03-05-09, 10:25 AM

I thought it might be fun to ring in on this. After two decades on feature and TV sets, watching dailies, studio screenings, and supervising telecine transfer sessions for projects I've shot, I have found a way to be at peace with this subject.
Most movie directors, cinematographers, and directors of photography work in a film medium. Their focus is how a finished piece will look on film, not what it looks like once the thing is telecined and crammed into a limited 8 bit color gaumet. I doubt any DP has told a director "we better not film it that way it won't look good when it gets released on DVD." When shooting film for feature release, the DP is relying on his experience to "know" what he is going to see in dailies, and has some idea where that is going to fall in the telecine. The DP may or may not be shooting in film these days. If shooting film for TV (episodic, MOW, or commercial), the DP certainly is thinking about what he is capturing and he will be making decisions about how he is shooting but he won't be telling the director. It's just part of his job to make those decisions. The director communicates his vision to the DP, and generally the DP has a fair amount of autonomy.

If shooting HD, you actually can have a pretty darn accurate display on set. As I started shooting HD I began to do the calibration of that monitor in the same way as ISF. I just invested in the tools so that I knew that I was seeing what I should.

The best we can hope for at home is to have our display reproduce the source material as faithfully as that display's technology can. Ideally, we will have accurate color, accurate white/gray, reference luminance (white), accurate black, and gamma. The toughest, and maybe most controversial, is color (primaries and secondaries). Those are great to have accurate, but the other items will result in maybe the most important result; nothing will be missing (shadow detail and white detail), and you won't see things you shouldn't (BTB). If lucky, gamma will be linear between the two and will render contrast faithfully. A technique I like to use is to turn down color saturation completely to evaluate black, white, and gamma. I watch something I know well. Color can be a great distractor to evaluation of those factors. After those are in good shape, add color and evaluate using good references.

That is what I need from my reference HD monitor on set. Most DPs who come from film will light by eye and meter. He knows how much light he will need in that night background to see some shadow detail. I take a look at the reference monitor rather late in the game to see how video is seeing it. Imagine if the black level is set wrong and I let it trick me into thinking I don't have enough light in the shadow areas. If I pump a bunch of extra light into the shadows, then I am only reducing the contrast. That is a matter of experience, but illustrates what can happen at home as we evaluate our images. How do we know if the blacks are crushed? How do we know if we are missing detail that is photographed to be seen?

Is there an interpretive factor to calibration? I contend there is for the purpose of finding a compromise when necessary at home. If a particular technology or model doesn't have greatest sequential (on/off) contrast, the "blacks" might not appear too impressive despite the fact that it is calibrated to a proper standard (generator or disc). You have to find a priority. If the black level setting can result in a minor loss of shadow detail, but improve the apparent blacks, then that may be an acceptable compromise for a consumer. There may be similar compromises with regards to color.

Intent? Hard to define, and pretty much impossible to assign to the calibration function. The best I can do is look at it with the eye of a DP, and accomplish a reasonable calibration based on the technology and model of the display.

So, I am a strong proponent of calibration, so much so that I actually also now do consumer ISF calibrations as well as professional displays. I think there is a real benefit, but it has its limitations. As with most things, the trick is finding balance. There are some things that, IMO, you can't compromise without significantly compromising faithful reproduction of the material. Others you can if done with care. In other words, minor tweaks after the "calibration" may be okay, but it is important to start from a known place; the settings that rendered the calibration.

I have even been told that DCI cinema projectors are not calibrated to 6500k, but rather 5500k. How many of us calibrate to 5500k?This is to accomodate the lamp house source color temp. It renders accuracy on projection in the cinema. It would not at home...unless we had DCI calibration in a memory. When those movies come home, they will have been converted to D65. The same is done with film. The color balance of our daylight light sources on set are ~5500K.

GeorgeAB

03-05-09, 10:48 AM

Full disclosure? Define "full." The endorsement quoted above was not from Joe Kane, but the restoration producer of the film. Joe Kane was engaged minimally in the QC process for the Blu-ray release of 'Baraka,' because he knows what reference imaging is supposed to look like, has an excellent reputation for perfectionism in the motion imaging industry, and has considerable experience in cutting edge disc mastering. He is proud to have been asked to contribute to this ground breaking project. His name, or that of JKP, Samsung, or Da-Lite appear no where in the credits on the disc, due to this minimal role.

The producers of the film restoration and transfer wanted the best that current technology has to offer for their project. So far, it's the only feature film scanned to 8K from a native 65mm negative and transferred to Blu-ray. The process takes much more time, and therefore expense, to use this methodology. To insinuate that they used an inferior projector to save a few thousand dollars and risk compromising the integrity of the project is ludicrous and ignorantly cynical.

Art Sonneborn

03-05-09, 11:56 AM

I thought it might be fun to ring in on this. After two decades on feature and TV sets, watching dailies, studio screenings, and supervising telecine transfer sessions for projects I've shot, I have found a way to be at peace with this subject.
When shooting film for feature release, the DP is relying on his experience to "know" what he is going to see in dailies, and has some idea where that is going to fall in the telecine. The DP may or may not be shooting in film these days. If shooting film for TV (episodic, MOW, or commercial), the DP certainly is thinking about what he is capturing and he will be making decisions about how he is shooting but he won't be telling the director. It's just part of his job to make those decisions. The director communicates his vision to the DP, and generally the DP has a fair amount of autonomy.

If shooting HD, you actually can have a pretty darn accurate display on set. As I started shooting HD I began to do the calibration of that monitor in the same way as ISF. I just invested in the tools so that I knew that I was seeing what I should.

The best we can hope for at home is to have our display reproduce the source material as faithfully as that display's technology can. Ideally, we will have accurate color, accurate white/gray, reference luminance (white), accurate black, and gamma. The toughest, and maybe most controversial, is color (primaries and secondaries). Those are great to have accurate, but the other items will result in maybe the most important result; nothing will be missing (shadow detail and white detail), and you won't see things you shouldn't (BTB). If lucky, gamma will be linear between the two and will render contrast faithfully. A technique I like to use is to turn down color saturation completely to evaluate black, white, and gamma. I watch something I know well. Color can be a great distractor to evaluation of those factors. After those are in good shape, add color and evaluate using good references.

That is what I need from my reference HD monitor on set. Most DPs who come from film will light by eye and meter. He knows how much light he will need in that night background to see some shadow detail. I take a look at the reference monitor rather late in the game to see how video is seeing it. Imagine if the black level is set wrong and I let it trick me into thinking I don't have enough light in the shadow areas. If I pump a bunch of extra light into the shadows, then I am only reducing the contrast. That is a matter of experience, but illustrates what can happen at home as we evaluate our images. How do we know if the blacks are crushed? How do we know if we are missing detail that is photographed to be seen?

Is there an interpretive factor to calibration? I contend there is for the purpose of finding a compromise when necessary at home. If a particular technology or model doesn't have greatest sequential (on/off) contrast, the "blacks" might not appear too impressive despite the fact that it is calibrated to a proper standard (generator or disc). You have to find a priority. If the black level setting can result in a minor loss of shadow detail, but improve the apparent blacks, then that may be an acceptable compromise for a consumer. There may be similar compromises with regards to color.

Intent? Hard to define, and pretty much impossible to assign to the calibration function. The best I can do is look at it with the eye of a DP, and accomplish a reasonable calibration based on the technology and model of the display.

So, I am a strong proponent of calibration, so much so that I actually also now do consumer ISF calibrations as well as professional displays. I think there is a real benefit, but it has its limitations. As with most things, the trick is finding balance. There are some things that, IMO, you can't compromise without significantly compromising faithful reproduction of the material. Others you can if done with care. In other words, minor tweaks after the "calibration" may be okay, but it is important to start from a known place; the settings that rendered the calibration.

Thanks very much for the post. I think you hit a balance of goals vs reality which recommends calibration as home then venture out at your own risk perspective. I agree with this completely.

Art

Cam Man

03-05-09, 02:33 PM

Thanks very much for the post. I think you hit a balance of goals vs reality which recommends calibration as home then venture out at your own risk perspective. I agree with this completely.

Art My pleasure.

HoustonHoyaFan

03-05-09, 03:32 PM

... There are some things that, IMO, you can't compromise without significantly compromising faithful reproduction of the material...That is the term that we should use, "faithful reproduction of the material". Yes that does include the Matrix green push and Speed Racer's overblown colors!! :)

GeorgeAB

03-05-09, 11:40 PM

As a supplement to post #199, here's a quote from the latest edition of 'Home Theater' magazine I just received today. It's from page 38 of Fred Kaplan's article: "A New Criterion For Blu-ray- The movie lover's label moves into the next gen." The Criterion label has been known for decades now as a premier label that offers superior transfers of unique and classic films.

"Part emboldened and part simply curious Kline produced a high-def disc of clips from 10 Criterion films to see what they'd look like on Blu-ray. He showed it to me in the company's custom-built screening room, which sports a Samsung SP-A800B projector, designed in conjunction with HD video pioneer Joe Kane, and a 10-foot-wide screen. (Klien and his crew scour trade shows for the best equipment. They do much of their quality-control work on Sony's BVM professional-grade CRTs.)"

Best regards and beautiful pictures,
G. Alan Brown, President
CinemaQuest, Inc.
A Lion AV Consultants Affiliate

"Advancing the art and science of electronic imaging"

GlenC

03-07-09, 07:23 PM

It is rather simple, electronic devices are generally "calibrated" an dit has carried over to the digital TVs. This can be debated till the cows come home, do you calibrate the grayscale and tune-up the TV or is the entire system calibrated or tuned? Can you get the same displays to look the same, it's possible, but not by copying the settings from one to the other.

If DCI cinema projector is calibrated to D55, that is fine as long as the media is mastered to the same specs. Flesh tones should look the same whether they are viewed in a Digital Cinema or from a DVD, on a Home Theater projector calibrated to REC 709. This will happen because the consumer DVDs are mastered to REC 709 and a white point of D65. If either system is "OFF" from the specifications, the images will not look as "intended".

"Calibration...are we just fooling ourselves? " With the image improvements in image quality and color fidelity I have seen, as the result of a "calibration", I would say no. I do think many are fooling themselves, thinking there is nothing to be gained by having their TV calibrated.

While all displays cannot be calibrated to accurately reproduce the current standards, a professional calibrator should be able to maximize the displays performance by getting it as close to the standards as possible. Just as many displays perform differently, many "calibrators" perform differently as well. Do your homework, buy the right display and get the right calibrator and really enjoy your investment.

Otto J

03-09-09, 05:34 AM

What about frame interpolation? With frame interpolation, we are actually manufacturing new frames that are nowhere to be seen in the original.

Some people seem to like it and some don't but I have never heard anyone make an argument that it is not accurate to have frame interpolation enabled even though it seems like one of the most destructive things one can do to a source.

It is not accurate to have frame interpolation enabled. There, now you have. I agree, it is very destructive, and I have never heard any filmmaker endorsing it, just like no filmmaker in their right mind would endorse Philips Ambilight. Some would like higher frame rates in the source, but that's not the same as saying they want frame interpolation.

Black frame insertion, now that's another story.

Otto J

03-09-09, 06:58 AM

I'll make some comments, although most of my arguments have already been made. I will try to be as little one-sided as I can, although obviously due to my choice of line of work, I believe in calibration.

Tryg: If I'm understanding you correctly, what you're trying to say is that the word "calibration" should imply that you are calibrating towards a certain goal, WITHIN A SPECIFIED TOLERANCE. I think this part is misunderstood by many that is arguing against you. We have standards to calibrate to, but in a lot of cases, there is no specified tolerance, you are just expected to get as close as possible, how close you can get depends on the hardware. By that definition, yes maybe you're right, the word "calibration" may indeed imply a bit more than what's actually being done. However, in reality I believe that there is no real word or phrase that covers it better than calibration. You are suggesting tuning, but in my opinion (english isn't my first language, but both words are basically used as such in Danish as well) "tuning" suggest an improvement over what's originally offered. And by this, I don't mean what the hardware manufacturer set the display to, but an improvement over the original content. Calibration is not improving the quality of the content material, therefore I don't think "tuning" is a good description of what is being done. I believe "calibration" pretty much covers it, although because of the lack of tolerances in the specifications, it's hard to quantify the accuracy of the calibration.

Re: JVC's, which are hard to get around in a thread started by you, yes there are definately a lot of displays outthere that will provide post-calibration pictures that are less accurate than a non-calibrated JVC. Neither that nor the fact that the JVC's could be even better if they had proper calibration controls, changes the fact that you should always calibrate the grayscale of a JVC to reach it's maximum performance. It just proves that most displays available are rather poor. Which is why we should always try to get the most out of whatever display we have to settle with.

Lawguy: I don't think you are dense enough to not understand the difference between achieving the director's intent, and TRYING to achieve the director's intent. Knowing that we can't reach perfection, does not mean that we shouldn't strive for it. If you strive for perfection, you end up having to settle for the best. What I THINK you're trying to say is that calibration is being marketed as perfection. Like any marketing phrases, you can probably find examples of calibration marketing that is superficial as opposed to what is actually being offered. But ask any serious calibrator, and I believe noone would tell you that calibration guarantees perfection. If this is what you're trying to say, I think you're blowing it out of proportions. I do think most calibrators are very realistic about what can be achieved when explaining to their customers what they get out of a calibration. Marketing HAS to be superficial, firstly because you need to start out by getting the potential customer's attention, before you can list out every little detail of what you will provide, second because due to the nature of calibrating displays, people will get different experiences out of having their display calibrated. Often people mention a calibrated picture as dimmer, duller, with less color etc. But fairly often I find people have been over-adjusting their set in trying to reach a watchable picture, so that sometimes the end result is MORE color, MORE sharpness etc. It's different every time. So, as a calibrator we HAVE to use fairly broad terms, and we cannot (or rather should not) for instance show before and after screenshots, although they actually work very well in marketing terms, because it creates certain expectations in the customers mind that may be very different to what is actually being offered. That's why we can't tell you _exactly_ what you will get out of a calibration.

Regarding some hardware not being built properly for calibration: Personally, I offer my customers "no cure no pay" terms, and just turn down customers that I feel I won't be able to help out, knowing their equipment. It happens very rarely that people need to take me up on that offer. What I'm saying is, while I certainly agree that some displays really don't have the adjustments needed to make a calibration worthwhile, I don't think that's a good argument to not calibrate those displays that are. The same goes for the color blindness argument - if you're not colorblind, who cares if colorblinded people see the difference or not? But, I have calibrated for several red/green colorblinded people, who saw significant differences, this will of course vary depending on the product. But note that while red/green colorblinded people may have difficulty seeing what actual color is being displayed, they WILL be sensitive to the amount of red and green as opposed to blue in, say, skin tones, in the sense of whether the skin tone looks natural or not (or whatever the filmmaker tried to achieve). The point of the matter is, yes you will probably be able to find people who shouldn't pay for a calibration. But I don't think anyone within the business is denying that.

Another note: Calibrating is important. Who carries out the calibration doesn't really matter, as long as it's done properly. If you are better at calibrating than whoever you would otherwise hire to do it, obviously you shouldn't pay for a calibration. I don't pay for calibrations on my system, when working on my own system, I'm a DIY'er just like you. I just happen to have experience from my line of work that I can use in my DIY system. Calibration services is being offered to those who don't feel they can do a good enough job of it themselves. Noone is forcing you to pay for a calibration if you're confident your picture is already as good as it can be.

One more thing: Most of the time I feel that people arguing against calibration is missing the point of the standards. I very often see the argument "who cares about all this technical stuff about standards, why focus on nitpicking about reaching standards, I just want to watch pictures". That's completely missing the point. Having your display calibrated is HOW you get to that point where you don't care about the picture quality anymore and just watch pictures.

The goal of a calibration IS to make sure you don't have to worry about the picture anymore. If you adjust your picture every now and then, something is wrong (either with the picture or with you). Watching accurate pictures is not about marvelling over how accurately the color of red matches rec.709 standards. The reason we have the standards, is because we need a crossover-point from the creation of the material, to the reproduction of that material. If you watch something live, say in a live theatre, you don't need that crossover-point, you will inherently see the action as it is. With film, we need a way to allow the filmmaker to display his art to the viewer.

And no, Lawguy, it will never be perfect, and yes, some filmmakers don't make the full use of that possibility, whether it be due to budget issues, or just sloppiness. But we need to have that crossover-point, to allow YOU to decide how YOU want to watch whatever the filmmaker chose to present to you. You have the OPTION of trying to match the standard, to more accurately see the material as the creator intended (though, for the billionth time, we know perfection will not be achieved). Even if we acknowledge that some people for some reason like bluish skin tones, we need a standard that you can deviate from if you want. Otherwise, you WILL have to adjust the picture for each film, which in turn forces you to think about picture quality while watching films. Because of this, I strongly disagree that calibrating to the standards as close as possible, is putting picture quality over creative content. It is quite the opposite, it is a necessety in allowing you to see the creative content, WITHOUT having to worry about picture quality.

To enjoy films, you need to be able to seperate the strive for quality from the actual enjoyment of films. It's a two-piece process. First, you achieve the best possible performance, then you enjoy your movies in whatever quality you managed to achieve. This has just as much to do with your way of thinking as it has to do with the actual quality of the image, but in my opinion, the more accurate the display is, the easier it is to get immersed in the picture without focusing on picture quality.

The funny thing is, when I demonstrate pictures, I can confidently say that the more accurate the picture is (i.e. the better the hardware I have at hand), the less people are talking about technical stuff during the presentation. Most of the time, we end up discussing how the movie was shot, rather than how the projector is displaying the image. Which is _exactly_ what I'm trying to achieve. If someone asks "why did they make his face that red in that scene" rather than "isn't the skin tones too red on that projector?", I reached my goal.

Lastly: The purpose is to get a good picture, which, as someone else has already argued, equals accurate pictures. If someone is able to put out a product that delivers a good enough accuracy to make people happy without the need for calibration, all power to them. Calibration services are available to those who invest in products that don't.

Otto J

03-09-09, 07:28 AM

I think image accuracy to source and perceived image quality are two different things. They are not necessarily mutually dependent. The film Matrix may give the matrix-world a green hue or another film may give night scenes a blue hue. If you tune the display to remove the hue you improve perceived image quality. Many people prefer a warmer image than D65, the eyes can completely adapt to other color temperatures. Many people prefer a high gamma for a more pop out of the screen image. Many people prefer higher color saturation. Many people prefer a brighter image. Many people prefer using contrast enhancement and detail enhancement to alter the image.

...

What calibration gives you is that all sources should look ok or better.

I don't agree. What calibration (or rather, the standards that we calibrate to) gives you is the possibility of the filmmaker to decide whether skin tones should be warm or not, whether gamma should be high or not. Calibrating the display to D65 does not mean that "white" in any particular movie is D65. Someone mentioned a different white point for DCI - that doesn't mean that "white" within a certain picture will look different. It's just a fixation point that allows the filmmaker to decide what color of white he wants you to see. If the film is in DCI, "white" will look a certain way when displayed on DCI equipment, if it is in HD, that "white" will look just the same, displayed on a display calibrated to Rec.709. And in turn, if the movie was originally made using a different white point, you should not calibrate your display differently as long as the source has been made for HD displays. To spell it out, calibrating your display to D65 doesn't mean that the color of D65 will necessarily ever be displayed on-screen.

Within the range of the standard at hand, ANY adjustment that you can make to the picture, could have been made by the filmmaker if he thought it would improve the look. Most new action flics makes heavy use of high gamma and contrast enhancements, just like those effects that we try to avoid in the display. It's not that we don't necessarily dislike high gamma, contrast enhancement, we just want to allow the filmmaker to decide whether those enhancements should be applied, instead of leaving it up to the display manufacturer. That's why these enhancements should be done by the filmmaker and not the display. If you don't like the look of the picture, it's a part of your complete movie watching experience, just like if the actors are good or not. That's why there are technical academy awards as well as creative ones. If anything, most new pictures have an "enhanced" look inherently, so having the display do the same is essentially putting sugar in your sugar.

By the way, while we are fairly good at adjusting for wrong color temperature, the white point can be off towards or away from green, even if it is at 6500 kelvin. Errors in green is actually at least as, if not more, important as reaching the accurate color temperature. Therefore, our sensitivity to color temperature is not directly related to the need to calibrate to D65. Which is why we calibrate to D65 and not 6500 kelvin in the first place. Anyone calibrating to 6500 kelvin and not D65 should be spanked. Hard.

I DO agree, though, that the standards themselves cause a limitation of the creativity of the filmmaker, and hence the picture quality that you will be able to achieve. But that is not something you can change on the consumer side, that can only be changed by settling on a new standard - like what DCI is trying to do. As long as the material has been created within the limitations of the HD standards, that's what we have to live with. You will never be able to improve on that in one area, without sacrificing something in another.

takisot

03-09-09, 08:11 AM

Having your display calibrated is HOW you get to that point where you don't care about the picture quality anymore and just watch pictures.

Quote of the Year... Well said Otto!

dovercat

03-09-09, 08:30 AM

Was not saying it was a good idea, just pointing out that perceived picture quality and accurate picture quality are not allways the same thing. Although they often are as the film maker usually wants you to like the film.

I like most people on the forum prefer a accurate image but some people must prefer an inaccurate picture, after all some people set their picture up that way. Displays often have inaccurate picture enhancements as selling points like dlp projectors with brilliant and vivid color or blu-ray players with detail enhancement and contrast enhancement. High definition is marketed in the UK as having more vivid colors than standard defintion, I do not see how this is true but it shows what marketers think people want. Display manufactures in general do not market their displays as being the most accurate to standards so they must think most people prefer sharper, higher contrast, more vivid color, brighter, etc... They sell their display as having a better image quality than the competition they do not equate this to a more accurate to standards picture.

The standards seem to be written for crt home displays. But we now have a variety of different display technologies that all tend to look inherently different from one another. Some people prefer dlp, others lcos, others lcd, or for flat panels some lcd others plasma. This is often because in comparison to crt or each other the image looks sharper or brighter or higher contrast or colors more vivid. Should we all be buying crt because it can give a more accurate image, or are we all willing to sacrifice some accuracy for a improvement in perceived image quality.

Tryg

03-09-09, 09:41 AM

R Harkness

03-09-09, 09:46 AM

Do "calibrators" give the customer some kind of paperwork telling them where their projector has been adjusted to a standard and how close? NO

Actually I think quite a few do supply such papers, e.g. print outs of the before and after calibration results. The calibrator who did my plasma
did.

Art Sonneborn

03-09-09, 09:52 AM

.

I say the word "calibrated" should not be used because it implies that you are taking a device that is capable of meeting certain specifications AND putting it within certain tolerances. Neither of which are often true. AND because of this different displays should be almost indistinguishable once "calibrated".

I think "tuning" is a much more appropriate word, although obviously not as scientific sounding, because that is what "calibrators" generally are doing.

Do "calibrators" all do the adjustments the same? NO.
Do "calibrators" give the customer some kind of paperwork telling them where their projector has been adjusted to a standard and how close? NO
Would on "calibrators" work be identical to others? NO

Like a musical instrument the projector is merely "tuned".

"Tuners" probably like to use the word "calibration" because it actually sounds more scientific and justifies another $400 to their service:o

The term calibrate is used when getting balances and time pieces closer to the reference device even when they aren't as accurate as the reference. Some devices (projectors) can't be brought as close to the reference as others,not a reason to throw out the process ,effort or term.

Art

dovercat

03-09-09, 10:08 AM

Otto J

03-09-09, 10:22 AM

Was not saying it was a good idea, just pointing out that perceived picture quality and accurate picture quality are not allways the same thing. Although they often are as the film maker usually wants you to like the film.

This is why some calibrators like to make the distinction between picture quality, and accurate pictures. Accurate pictures doesn't mean that the picture will look good. 300 is the obvious example, does it look "good"? To many, it doesn't - if you don't "get" the stylized approach to the look of the film, it may annoy you. But in no way can you make that film look "natural" by adjusting your set. The point being, the stylization is there for a reason, and perhaps you should consider why the director chose that style, instead of wondering if your picture is too red? If an accurate picture looks wrong, then it's the fault of the filmmaker, not the set, the calibrator or you. (i'm deliberately saying "accurate", not "calibrated" - a "calibrated" picture is only as accurate as the hardware allows - just because you have seen a "calibrated" display, doesn't mean you have seen accurate pictures)

I like most people on the forum prefer a accurate image but some people must prefer an inaccurate picture, after all some people set their picture up that way.

That's *NOT* the same thing. I can't stress that enough. There are several reasons to why people choose to not use accurate settings. I have done a lot of demonstrations of fairly accurate pictures, and every single time everyone I talk to say that it looks great, and that they would like that at home. Not a single person, out of maybe a couple of thousands that have seen my demos, have stated that they would like more color. Not one. Yet, most people have too much color in their pictures. Most of them acknowledge that when their set is delivered, it has too much color, so they turn it down - but do they turn it down far enough? In an A/B test, as is what you are basically seeing when you are adjusting manually, you tend to get fooled into seeing things differently than what they actually look like. Just like the "echalk optical illusions" website shows (just google echalk...). What you think you see, is not always what you see.

Then there are those few that have measuring gear, or had their set calibrated, and still prefer something else. I divide those into two categories: One half that just owns displays that can't get anywhere near accurate whatever you do, so you may end up choosing between two inaccurate choices - for example, choose between accurate grayscale and light output - and then you may want to sacrifice some color accuracy, to get luminance accuracy. The other half, in my opinion, is so focused on "picture quality" that they see sharpness, saturation, contrast etc, instead of seeing a movie. The more you focus on the creative content, the more you will like an accurate picture over an inaccurate one.

But, definately there are displays where compromises have to be made, which is why calibrating is not just buying a sensor and getting to work. You need experience in the effects of different anomalies, to figure out how to reach the best compromise. For instance, if one or more of the primaries are wrong, you should change the target of the secondaries accordingly, to get the most pleasing picture, and not just blindly work towards the Rec.709 points. I suspect that most cases where people say that they like inaccurate pictures more, is just a matter of them not fully understanding all aspects of image accuracy, so what they think is more accurate, may in fact actually be more inaccurate. An example could be the JVC's. Because the primaries are so oversaturated, if you adjust color brightness (the color control) to make the SMPTE pattern look "right", the picture will look wildly wrong. So, you need to deliberately create an error in color brightness, that takes you to a more watchable compromise overall.

Calibration is not about making the measurements look accurate. It's about making the picture look accurate.

Displays often have inaccurate picture enhancements as selling points like dlp projectors with brilliant and vivid color or blu-ray players with detail enhancement and contrast enhancement. High definition is marketed in the UK as having more vivid colors than standard defintion, I do not see how this is true but it shows what marketers think people want. Display manufactures in general do not market their displays as being the most accurate to standards so they must think most people prefer sharper, higher contrast, more vivid color, brighter, etc... They sell their display as having a better image quality than the competition they do not equate this to a more accurate to standards picture.

Again, the echalk illusions will tell you why. When I demo just one display with an accurate picture, everyone thinks it looks great. If I demo _the exact same_ display next to five inaccurate displays, most will think the accurate display looks horrible. In theory, you should never, ever, ever, evaluate a display while other displays in the room are turned on. In reality, you often don't have much choice, but you WILL end up fooling your eyes if you do.

The standards seem to be written for crt home displays. But we now have a variety of different display technologies that all tend to look inherently different from one another. Some people prefer dlp, others lcos, others lcd, or for flat panels some lcd others plasma. This is often because in comparison to crt or each other the image looks sharper or brighter or higher contrast or colors more vivid. Should we all be buying crt because it can give a more accurate image, or are we all willing to sacrifice some accuracy for a improvement in perceived image quality.

The standard most certainly IS written for CRT displays. That's not a secret at all. However, as I said we cannot fix that on the consumer side alone. But what we SHOULD do is try to emulate what a CRT looks like. The reason different display types look different, is because they are not equally good at doing this. And I most certainly will not try to make you buy a CRT, just because it may technically be more accurate - a CRT monitor is, in most setups, inaccurate in one very important area - image size. Anyway: If the picture from two different technologies look different, it is because they are NOT both accurate. If it looks different, but doesn't measure differently, you didn't all areas of performance. For instance, just because 75% stimulus provides accurate primaries, doesn't mean that 25% stimulus will. So, if two displays have exactly the same saturation at 75% stimulus, but different at 25%, real-life pictures will look different because real-life pictures consist of a mix of those points (and a whole lot of other stuff). If _every_ aspect of color performance, gamma, contrast, light output and image size of two displays were equal, I guarantee you that the pictures would look very close. The difference will be down to technology-specific issues (RBE, screendoor, dithering, flickering etc.). We of course want these issues to be minimized, and of course they could potentially be so bad that even if all other areas were perfect, I still wouldn't want to buy the product.

You mention that some display technologies might have "more vivid colors", even though both are accurate. That is simply not possible - if one is more vivid than the other, they won't measure the same. But a _lot_ of people, including a lot of those offering calibration services - quality of service is different in any business, so why shouldn't it be in calibration - don't realize for instance the importance of accuracy of color brightness (see for instance the recent series in Widescreen Review to learn more - I'm sure there are threads on this here on AVS as well). Just because the CIE measurements of two projectors look alike, doesn't mean that they are equally accurate.

Now, if Tryg's point basically was that "calibration" doesn't equal perfection, I wholeheartedly agree, and one of the arguments of that is that different calibrators will reach varying results, depending on the knowledge and experience of that calibrator. I'm sure there are people outthere that can reach better results than I do, I'm very much still a newcomer in this business compared to some of the guys outthere. So, if people believe that the word "calibration" is a guarantee of perfection, then yes, Tryg, we are fooling ourselves. I just don't think very many people actually believe that, and because of that I don't think it's right to undermine the whole concept of calibration, just because a few people don't seem to have fully understood what it is. So, Tryg: We shouldn't stop using the word "calibration", and we most certainly shouldn't stop calibrating. But "we" should most certainly stop thinking that "calibrated" equals "accurate", if "we" ever did.

To take it from the top, I think that this discussion should be divided into two parts: 1: What are we trying to achieve, and 2: How do we achieve it? It doesn't make sense to discuss how we reach our goal, if we don't agree on what the goal is in the first place. If we really wanted to reach an agreement (but who says we should), we should start by discussing what goals we are trying to achieve. I'm not even sure that is what Tryg wanted to discuss, I think the issue was more about whether or not too much is being promised to calibration customers, but a short summary of my opinion:

1: The best possible picture quality obtainable, is whatever the filmmaker put on the disc - whether that's good or bad, you can't get anything better than what he/they created, because the purpose of watching a movie IS to see what they created for us. If he created something bad, it will and should look bad, no matter what you try to do about it. If you think it looks bad, watch another movie. Art was never supposed to please every single person that lays eyes on it, it's supposed to provoke a reaction, good or bad.

2: The filmmaker "painted" his movie on a canvas. That canvas happens to be HD. So, what we need to do is match the HD standards, which basically means Rec.709, to see what he has painted. To do this, we need one of two things: A display system that is delivered factory-tuned to match Rec.709, or a display system that allows field-calibration to Rec.709. Either way, we are limited by the quality of the hardware. We cannot calibrate to Rec.709, if the hardware doesn't allow it. However, in reality it is rarely practical or economical to try to achive accuracy out-of-the-box, due to a number of reasons (burn-in time, unit variations, source conciderations, room conciderations etc). Therefore, in reality we need to calibrate the display system in the field, to make sure it is as accurate as possible. It is easier and most likely cheaper to manufacture a calibratable display and calibrate it in the field, than to manufacture a display system that will work without calibration.

The reason displays look so different, even though getting good, accurate pictures isn't actually that difficult (at least better than what we're being offered in general today), is that there are extremely few hardware companies that are trying to do so, for the simple reason that it's not as easy to sell. It's easier to explain to the average consumer that he should look for vivid pictures and provide that, than it is to explain to the average consumer why he should want accurate pictures. But "easy" doesn't equal "true".

Otto J

03-09-09, 10:38 AM

Maybe slightly off topic but on some displays they have markers for ISF day and ISF night settings, that a ISF calibrator can save settings to. What is the difference between the two, are they actually used for different sources, tv stations and blu-ray, or for different settings of lcd flat pannel backlight. Or as the names imply settings for different viewing conditions. The displays capabilities and the standards do not as far as I know change with the time of day.

In reality, that depends on the display and the calibrator. I agree that the capabilities of the display doesn't change according to the room lighting, but the theory behind those settings is to allow a compromised "day" setting to allow a higher light output, and maybe gamma, setting in bright rooms (two settings per input). But not all displays have properties that actually allows you to make much use of it. If the set allows, it could be used for different settings of backlight as you mention.

I don't agree with everything ISF does. Please remember that ISF is not the standard we're calibrating to, their job is to improve awareness of the standards that are available. Having the ISF logo on a set, tells you that there are certain controls available on that set, but it doesn't guarantee you good performance. So I don't think you should judge the need for calibration by what the ISF does or does not do. The standards that we are trying to reach, includes the room as well as the display - I'm not sure very many people have fully understood this. So, if you change one variable - the room - the other parts of the standard really no longer applies.

Time pieces and balances I have used were accurate when set to the reference, and regularly checked to make sure they stayed accurate. Their level of accuracy had to be within set tolerances, depending on what they were used for. The set tolerances for display calibration seem to be whatever the dislpay can manage, so there is no minimum level of accuracy.

That is correct, there is no minimum level of accuracy in the standards. Which is why THX was established, for whatever that's worth, they are trying to change that by establishing certain tolerances for those areas of performance where they aren't established by the SMPTE standards. So, basically you could say that a "THX calibration" would match your criteria of a calibration (except that the tolerances aren't openly available to my knowledge), while an "ISF calibration" as some would call it, does not (why then JVC has a product with a "THX" mode, that you cannot calibrate, still blows my mind...) .

So, I guess the discussion is down to the definition of the word "calibration", more than a discussion about whether we should try to achieve accuracy or not, or?

Otto J

03-09-09, 10:58 AM

Otto, some good feedback.

I say the word "calibrated" should not be used because it implies that you are taking a device that is capable of meeting certain specifications AND putting it within certain tolerances. Neither of which are often true. AND because of this different displays should be almost indistinguishable once "calibrated".

I didn't read your post before writing the last post, I guess it is about the definition of the word then. Please note this quote from "the free online dictionary":

"calibration - the act of checking or adjusting (by comparison with a standard) the accuracy of a measuring instrument"

This falls in line with other definitions I find across the net. Please note that "calibration" doesn't actually have to mean that you are adjusting anything, or that you need to match certain criteria. Just checking what the accuracy is, would fulfill the definition of the word "calibration". So, as long as you deliver a set of measurements that tells you how far you're off, you don't even have to adjust anything - it's a calibration! So, your definition isn't quite accurate, as far as I'm concerned, you don't need to have defined tolerances to use the term "calibration". The definition of calibration is that you have a set standard that you work towards. There is no implication of how strict the tolerances you use actually are. I believe the word "tuning" does not necessarily imply that you have a set standard (it would when tuning a musical instrument, but not when tuning a car). Also, the word "calibration" implies that you use a measuring device, whereas "tuning" does not (necessarily).

EDIT: Actually, if you look at thefreedictionary under "tuning", tuning of a musical instrument is defined as "calibrating" to a certain tone, whereas tuning a car is defined as "adjusting" for maximum performance.

By that definition, I believe the word "calibration" covers what is being done when adjusting a television set, much better than the word "tuning". Basically, if you are measuring your display device and trying to match the standards, you are in fact calibrating, no matter how close you actually get. If you grab the remote and adjust the picture by eye, you are tuning.

Art Sonneborn

03-09-09, 10:58 AM

So, I guess the discussion is down to the definition of the word "calibration", more than a discussion about whether we should try to achieve accuracy or not, or?

Reread the original post. That was to throw out the concept because of the term.... very very bad IMO.

Art

TomHuffman

03-09-09, 11:00 AM

That is correct, there is no minimum level of accuracy in the standards. Which is why THX was established, for whatever that's worth, they are trying to change that by establishing certain tolerances for those areas of performance where they aren't established by the SMPTE standards.SMPTE has set some tolerance standards for the Digital Cinema Initiative, which I think are useful:

Color accuracy: 4 dE units in CIELAB. This applies both to grayscale and to primary/secondary colors.
Measuring equipment: The luminance meter must have a minimum accuracy of ±0.2 fL for white field measurements and ±0.002 fL for black field measurements. The chromaticity must be measured with a spectroradiometer with a minimum accuracy of ±0.002 for the measurement of the x and y chromaticity coordinates at luminances above 10 cd/m².
Output: Peak output of 14 fL.
Contrast: Minimum sequential contrast of 2000:1 and checkerboard contrast of 150:1.

anirbana

03-09-09, 11:12 AM

Otto,

Wonderful posts. Thank you very much. Your posts offer not only information but well articulated content that offers reasoning and purpose. :)

Great job.

Otto J

03-09-09, 11:20 AM

SMPTE has set some tolerance standards for the Digital Cinema Initiative, which I think are useful:

I agree, that is very useful, and I think it could have been useful to have such a definition in the consumer world as well. Like I said, if i.e. the primaries get too far off, you can't use some of the other target points anymore. Having openly defined tolerances might have urged the manufacturers to be a bit less loose about them, which in turn would make it much easier to get a good picture.

Tryg

03-09-09, 11:24 AM

I think the issue was more about whether or not too much is being promised to calibration customers

More great work Otto! You are getting to the jist of my original post.

I think the word "Calibration" connotates precision. But if you took 10 different displays even of similar technology and compared most would feel there's not much precision at all with the end results of what the picture looks like. They all would look very different.

Otto J

03-09-09, 11:24 AM

Otto,

Wonderful posts. Thank you very much. Your posts offer not only information but well articulated content that offers reasoning and purpose. :)

Great job.

I do my best to be as articulate as possible, to make sure people understand me correctly, but it is a bit more work than writing in my native language, so thanks for noticing :-)

Otto J

03-09-09, 11:39 AM

More great work Otto! You are getting to the jist of my original post.

Good. In that case, as I read your post you think calibrators are promising too much, in the very implications of the word "calibration". I think I've established that this is not the case, so if you basically want to discuss good vs bad calibrations and calibrators, I don't think discussing the term is the right way to go. But I very much agree that we (and I mean both "we" as a calibrator community", and "we" in general terms) need to be very aware not to, for instance, create the assumption than anyone with a measuring device and an ISF logo on the door, will be able to make good calibrations. The ISF services of Best Buy certainly increases the need for this awareness. Customers definately need to be aware that they should try to get a calibrator that they trust can do a good job.

I just don't think that collides with the use of the word "calibration", it just means that just like a projector isn't just a projector, a calibration isn't just a calibration. What I'm trying to say is, I think most of us calibrators actually agree with you that the term "calibration" doesn't guarantee you any specific quality of the service, but at least I, and I suspect most others, just don't think changing the word "calibration" would in any way change that. If it was called "tuning", you would be able to buy "ISF tuning" of good and bad quality, and have the exact same issue. So why not leave it at "calibration", and discuss the real issue instead?

Tryg

03-09-09, 11:57 AM

Otto, that is part of the real issue. If a display is not even capable of achieving any of the standards is it still "calibration"?

If so, like my original post, I believe we are simply using the term too loosely.

Art Sonneborn

03-09-09, 12:08 PM

At best about all you can do is "tune" the projector. Maybe that is even too much to claim.

This is your original post. Even if you had said that it seems some projectors can't be calibrated it would have been less incediary.

Your inference is not to use the term in a blanket statement...even to say tuning is too much.

Art

Otto J

03-09-09, 12:18 PM

I think the word "Calibration" connotates precision. But if you took 10 different displays even of similar technology and compared most would feel there's not much precision at all with the end results of what the picture looks like. They all would look very different.

Well, that's your opinion, but to put it bluntly, then it's you who needs to change your impression of the word, rather than us who needs to use another word just because some people may have a wrong assumption of what the word actually means. To quote your original post, which I just re-read to follow Art's suggestion:

"When you calibrate an instrument it generally means you will change it to a specified standard (or be able to quantify the difference). In other words it will do exactly what the other instrument will do."

This is a wrong definition. Calibrating doesn't mean matching _exactly_, and it never did. It means matching, to some degree or another.

In any case, using a measuring device, hence "calibrating" in the sense of the word, is more precise than not using a measuring device. So, I still think it's right to use a word that is linked to some degree of precision.

Otto, that is part of the real issue. If a display is not even capable of achieving any of the standards is it still "calibration"?

Yes, it is.

Do you understand the concept of "exactly" vs "as close as possible"?

If so, like my original post, I believe we are simply using the term too loosely.

Then you need to loosen up your impression of the word, because we are not. It's the word that specifically matches what we're doing, so any use of another word would be skewing the meaning of what's being said. Just like you shouldn't use the term S-VHS for an S-video connection, just because the majority of people think it's called an S-VHS connection, I do think we should use the most accurate term available for calibrating displays, because that's what we're doing. If people in general think that two calibrated displays will look exactly identical, then that's the issue we need to adress, by educating people that no display can be calibrated better than the hardware allows. Trying to change the definition of the term, doesn't solve that problem. But I do agree that there are some who have a misguided opinion about what a calibration is, which means that you may see those people saying something like "my picture is perfect, it was calibrated!". But I don't believe those were who your rant was pointed at.

Basically, you're blaming calibrators for the fact that some forum users don't know the definition of the word "calibration". Do you really think that's fair?

Tryg

03-09-09, 12:27 PM

Basically, you're blaming calibrators for the fact that some forum users don't know the definition of the word "calibration". Do you really think that's fair?

I do not blame "calibrators".

I'm trying to find the relationship between "calibration" and accuracy to some standard. I'm having a hard time with this given my observations over the years.

Otto J

03-09-09, 12:33 PM

I'm trying to find the relationship between "calibration" and accuracy to some standard. I'm having a hard time with this given my observations over the years.

OK, here's the deal - sorry if I'm sounding like your old grammar-teacher, but it's the easiest way to explain it: Calibration is a noun. Accurate is an adjective. You can have good calibrations, bad calibrations, accurate calibrations, and inaccurate calibrations. But they're all still calibrations. The noun "calibration" doesn't have any implication of accuracy, unless it is linked to an adjective.

Hope this helps.

anirbana

03-09-09, 12:40 PM

I think if we look at it this way it will make more sense

1. Calibration is trying to achieve a known or set standard. It has nothing to do with personal preference or likings.
2. Tuning is actually more related to personal taste and hence should not be confused with Calibration.
3. A perfect calibration to lab / clinical standards is dependent on the hardware, person doing it and also importantly on the environment where it is used for a display device like a PJ. And in reality it is bit difficult to attain. And hence calibrated display does not guarantee a accurate picture (as per source) always but once calibrated in whatever form or accuracy it is still near to a standard
4. Whereas tuning is always a deviation – either from / to a standard, presets and vary from person to person making the word tuning too distant from calibration.
5. Tuning has a wide range of deviation between various audiences and their preferences whereas calibration even considering the limitations, is close to a standard being set.

So do we tune or calibrate?

I think the answers lies with individuals from a crude point of view. But given the nature of awareness we have nowadays, it is easier to appreciate standards in everything in life as a start point and then go from there to where our own preference takes us.

To me an ideal case will be to have a calibrated setting which I know is set as a standard and if I like it I will be happy to see it as it is. If I don’t like it and want some of my own deviation, I ‘Tune‘it from the calibrated point and save it in other settings. And I would know at any given time I have a reference set to go back to if I mess up the tuning.

I remember a project we did for a customer and we had to change 4 different shades of colour on a complete website when the CEO looked at it and he was very particular about the exact shades his company uses in every material. To him it was the standard. He asked to change it as it looks like a deviation from the standard. And we noticed it is one or two point deviation from the colour code and very minor. This is not a best analogy but the point is him being a customer we had to do what he wanted and paid us to do.

Similarly in films and movies, the director or whoever it is uses a specific palette and shades of colour as an artist and those colours are actually derived from standard set of colours. Now when those colours get plotted on a frame the nearest it can faithfully match the artists eye will depend on the frame and fabric and the lighting / environmental conditions. If you consider, Film, DVD, Blu ray and other medium as the fabric for source then your hardware and environment will make up the frame and other factors. So what we see is in real life, though we have all the same DVDs and Blu rays (even if they are wrongly mastered) we still get different images as the frame, and lighting or environmental conditions are too dynamic.

So how do we fix it? (If we really want to fix)
1. Tune it ourselves by our own eye – becoming the artist ourselves and hence could lead to further deviation from the source
2. Calibrate our display – up to the limit of the display so that we know the frame of reference is set and know that we are closer to seeing stuff in the way the artist intended. How close depends on the factors other than the fabric itself.

Otto J

03-09-09, 12:43 PM

I do not blame "calibrators".

I do, by the way, think that this:

"Tuners" probably like to use the word "calibration" because it actually sounds more scientific and justifies another $400 to their service

qualifies as "blaming calibrators".

But skimming through the thread again, you actually gave a very good analogy yourself:

Just because she may have all the parts that make her a woman, it doesn't mean she's beautiful.

This is exactly the point. You don't have to be beautiful to be a woman, and you don't have to be accurate to be a calibration.

Tryg

03-09-09, 12:47 PM

The noun "calibration" doesn't have any implication of accuracy.

Thanks. I understand now as this meets my observations.

Otto J

03-09-09, 12:50 PM

Thanks. I understand now as this meets my observations.

Great. Now just don't make the mistake of bashing the whole concept of calibration just because there are bad ones available.

dovercat

03-09-09, 12:52 PM

Yes I have been arguing semantics when I have not even taken the trouble to look up the word in a dictionary. I have misunderstood the meaning of the word calibration.

"calibration - the act of checking or adjusting (by comparison with a standard) the accuracy of a measuring instrument"

Under that definition you can calibrate a display just by taking a measurement with an insturment and comparing it to a standard. Without adjusting anything regradless of the reading your display is now calibrated. Or you could adjust anything in any direction making it more or less accurate to the standard and call it calibrated because not only have you taken a measurement, compared it to a standard, you have also adjusted its accuracy to that standard.

I have been wrongly assuming the word is usually used in the context "calibration to a standard." Incorrectly assuming a calibrated display is synonymous to an accurate display. Then questioning how accurate does it have to be to qualify as calibrated. I guess most people will not be so foolish as to make those assumptions.

Tunning at least means improving performance.

Tryg

03-09-09, 01:00 PM

dovercat gets it.

I think a lot of people are fooling themselves.

Otto J

03-09-09, 01:04 PM

I have been wrongly assuming the word is usually used in the context "calibration to a standard." Incorrectly questioning how this can be if the standards can not be met.

Tunning at least means improving performance.

I think it's pretty obvious to most people by now that what you and Tryg are doing is badmouthing the whole concept of calibration. I have no idea why this is so important to you. I don't give a rats ass if you call it calibration, tuning, hitchhiking or monkeyhumping. Shouldn't we be discussing why the displays aren't built for accurate monkeyhumping, instead of discussing grammar?

Otto J

03-09-09, 01:10 PM

dovercat gets it.

I think a lot of people are fooling themselves.

I don't argue that, I fully agree. I would argue that if you think you can get the best possible picture without calibrating your display, you are fooling yourself. However, you seem to imply that if you calibrate your display, you are by definition fooling yourself. That doesn't have to be the case, so why are you (any of you) mocking the whole concept of calibration (which you obviously are, so don't even bother), just because some people have a wrong perception of it?

Noshandy

03-09-09, 01:14 PM

This is why some calibrators like to make the distinction between picture quality, and accurate pictures. Accurate pictures doesn't mean that the picture will look good. 300 is the obvious example, does it look "good"? To many, it doesn't - if you don't "get" the stylized approach to the look of the film, it may annoy you. But in no way can you make that film look "natural" by adjusting your set. The point being, the stylization is there for a reason, and perhaps you should consider why the director chose that style, instead of wondering if your picture is too red? If an accurate picture looks wrong, then it's the fault of the filmmaker, not the set, the calibrator or you. (i'm deliberately saying "accurate", not "calibrated" - a "calibrated" picture is only as accurate as the hardware allows - just because you have seen a "calibrated" display, doesn't mean you have seen accurate pictures)

That's *NOT* the same thing. I can't stress that enough. There are several reasons to why people choose to not use accurate settings. I have done a lot of demonstrations of fairly accurate pictures, and every single time everyone I talk to say that it looks great, and that they would like that at home. Not a single person, out of maybe a couple of thousands that have seen my demos, have stated that they would like more color. Not one. Yet, most people have too much color in their pictures. Most of them acknowledge that when their set is delivered, it has too much color, so they turn it down - but do they turn it down far enough? In an A/B test, as is what you are basically seeing when you are adjusting manually, you tend to get fooled into seeing things differently than what they actually look like. Just like the "echalk optical illusions" website shows (just google echalk...). What you think you see, is not always what you see.

Then there are those few that have measuring gear, or had their set calibrated, and still prefer something else. I divide those into two categories: One half that just owns displays that can't get anywhere near accurate whatever you do, so you may end up choosing between two inaccurate choices - for example, choose between accurate grayscale and light output - and then you may want to sacrifice some color accuracy, to get luminance accuracy. The other half, in my opinion, is so focused on "picture quality" that they see sharpness, saturation, contrast etc, instead of seeing a movie. The more you focus on the creative content, the more you will like an accurate picture over an inaccurate one.

But, definately there are displays where compromises have to be made, which is why calibrating is not just buying a sensor and getting to work. You need experience in the effects of different anomalies, to figure out how to reach the best compromise. For instance, if one or more of the primaries are wrong, you should change the target of the secondaries accordingly, to get the most pleasing picture, and not just blindly work towards the Rec.709 points. I suspect that most cases where people say that they like inaccurate pictures more, is just a matter of them not fully understanding all aspects of image accuracy, so what they think is more accurate, may in fact actually be more inaccurate. An example could be the JVC's. Because the primaries are so oversaturated, if you adjust color brightness (the color control) to make the SMPTE pattern look "right", the picture will look wildly wrong. So, you need to deliberately create an error in color brightness, that takes you to a more watchable compromise overall.

Calibration is not about making the measurements look accurate. It's about making the picture look accurate.

Again, the echalk illusions will tell you why. When I demo just one display with an accurate picture, everyone thinks it looks great. If I demo _the exact same_ display next to five inaccurate displays, most will think the accurate display looks horrible. In theory, you should never, ever, ever, evaluate a display while other displays in the room are turned on. In reality, you often don't have much choice, but you WILL end up fooling your eyes if you do.

The standard most certainly IS written for CRT displays. That's not a secret at all. However, as I said we cannot fix that on the consumer side alone. But what we SHOULD do is try to emulate what a CRT looks like. The reason different display types look different, is because they are not equally good at doing this. And I most certainly will not try to make you buy a CRT, just because it may technically be more accurate - a CRT monitor is, in most setups, inaccurate in one very important area - image size. Anyway: If the picture from two different technologies look different, it is because they are NOT both accurate. If it looks different, but doesn't measure differently, you didn't all areas of performance. For instance, just because 75% stimulus provides accurate primaries, doesn't mean that 25% stimulus will. So, if two displays have exactly the same saturation at 75% stimulus, but different at 25%, real-life pictures will look different because real-life pictures consist of a mix of those points (and a whole lot of other stuff). If _every_ aspect of color performance, gamma, contrast, light output and image size of two displays were equal, I guarantee you that the pictures would look very close. The difference will be down to technology-specific issues (RBE, screendoor, dithering, flickering etc.). We of course want these issues to be minimized, and of course they could potentially be so bad that even if all other areas were perfect, I still wouldn't want to buy the product.

You mention that some display technologies might have "more vivid colors", even though both are accurate. That is simply not possible - if one is more vivid than the other, they won't measure the same. But a _lot_ of people, including a lot of those offering calibration services - quality of service is different in any business, so why shouldn't it be in calibration - don't realize for instance the importance of accuracy of color brightness (see for instance the recent series in Widescreen Review to learn more - I'm sure there are threads on this here on AVS as well). Just because the CIE measurements of two projectors look alike, doesn't mean that they are equally accurate.

Now, if Tryg's point basically was that "calibration" doesn't equal perfection, I wholeheartedly agree, and one of the arguments of that is that different calibrators will reach varying results, depending on the knowledge and experience of that calibrator. I'm sure there are people outthere that can reach better results than I do, I'm very much still a newcomer in this business compared to some of the guys outthere. So, if people believe that the word "calibration" is a guarantee of perfection, then yes, Tryg, we are fooling ourselves. I just don't think very many people actually believe that, and because of that I don't think it's right to undermine the whole concept of calibration, just because a few people don't seem to have fully understood what it is. So, Tryg: We shouldn't stop using the word "calibration", and we most certainly shouldn't stop calibrating. But "we" should most certainly stop thinking that "calibrated" equals "accurate", if "we" ever did.

To take it from the top, I think that this discussion should be divided into two parts: 1: What are we trying to achieve, and 2: How do we achieve it? It doesn't make sense to discuss how we reach our goal, if we don't agree on what the goal is in the first place. If we really wanted to reach an agreement (but who says we should), we should start by discussing what goals we are trying to achieve. I'm not even sure that is what Tryg wanted to discuss, I think the issue was more about whether or not too much is being promised to calibration customers, but a short summary of my opinion:

1: The best possible picture quality obtainable, is whatever the filmmaker put on the disc - whether that's good or bad, you can't get anything better than what he/they created, because the purpose of watching a movie IS to see what they created for us. If he created something bad, it will and should look bad, no matter what you try to do about it. If you think it looks bad, watch another movie. Art was never supposed to please every single person that lays eyes on it, it's supposed to provoke a reaction, good or bad.

2: The filmmaker "painted" his movie on a canvas. That canvas happens to be HD. So, what we need to do is match the HD standards, which basically means Rec.709, to see what he has painted. To do this, we need one of two things: A display system that is delivered factory-tuned to match Rec.709, or a display system that allows field-calibration to Rec.709. Either way, we are limited by the quality of the hardware. We cannot calibrate to Rec.709, if the hardware doesn't allow it. However, in reality it is rarely practical or economical to try to achive accuracy out-of-the-box, due to a number of reasons (burn-in time, unit variations, source conciderations, room conciderations etc). Therefore, in reality we need to calibrate the display system in the field, to make sure it is as accurate as possible. It is easier and most likely cheaper to manufacture a calibratable display and calibrate it in the field, than to manufacture a display system that will work without calibration.

The reason displays look so different, even though getting good, accurate pictures isn't actually that difficult (at least better than what we're being offered in general today), is that there are extremely few hardware companies that are trying to do so, for the simple reason that it's not as easy to sell. It's easier to explain to the average consumer that he should look for vivid pictures and provide that, than it is to explain to the average consumer why he should want accurate pictures. But "easy" doesn't equal "true".

Heres a comparison of my projector with CMS at REC709 and one thats not.

No CMS
http://img19.imageshack.us/img19/4701/hd750215at5.jpg
Rec709
http://i734.photobucket.com/albums/ww343/Noshandy/Kill%20Bill%202%20Bluray/DSC00792-1.jpg

Initially when I saw them side by side I questioned my calibration as my shot looks oversatured and less natural. Looking at the two I would choose the more natural one.

Thats until I noticed the eye colour. Uma has green eyes. Now upon knowing this which would you prefer? The more accurate one to the material or the more natural one?

I know screenshots are not the best way to show this as ideally you would need to see them physically side by side but it does in someway illustrate what Otto was saying.

Otto J

03-09-09, 01:23 PM

Thats until I noticed the eye colour. Uma has green eyes. Now upon knowing this which would you prefer? The more accurate one to the material or the more natural one?

The whole point of the argument is that it's not possible to make that judgement based on ssuch an example, especially not via a screenshot that I'm watching on a very inaccurate LCD screen on my laptop. I would have no idea what that particular scene was supposed to look like, so I wouldn't be able to say which of those two are "natural". It doesn't matter if the skin tone looks natural in that particular scene, if it was _supposed_ to look unnatural. If that's the case (which I don't know), other scenes that were supposed to look natural, would look un-natural if set to the "no cms" setting.

lcaillo

03-09-09, 01:25 PM

Her eyes have always appeared blue to me. Look at a few close ups of her again.

HogPilot

03-09-09, 01:36 PM

There's multiple definitions of the word. The fact that it has many meanings doesn't mean that all apply to what's inferred when talking about displays.

Here's the most applicable definition:

"To make corrections in; adjust"

And since we're splitting hairs about the proper usage of terms, note that the goal of display calibration is accuracy - not precision - as the two are related but very different concepts. A display can be just as precise when set to Warm 2 as when set to Cool 1; it can only be accurate when properly set to D65.

To literally accept that the end result of a display calibration could be as simple as determining its deviation from a standard is just silly. After reading pages of arguing semantics, I fail to see where the problem lies with calibration or what its stated goal is.

GlenC

03-09-09, 01:40 PM

I think the word "Calibration" connotates precision. But if you took 10 different displays even of similar technology and compared most would feel there's not much precision at all with the end results of what the picture looks like. They all would look very different.This can also be true for a collection of micrometers, dial indicators or vernier scales. You can get very accurate or so-so accurate. They can be "calibrated" within their respective tolerances, and can all return different measurements from the same object, depending in the degree of precision desired.

The one place where "color" may look more vivid could possibly be in 3-chip displays vs. single chip. You have full bit depth of color with a 3-chip that you cannot achieve with a single chip.

Another problem the calibration industry has, IMO, is that many calibrators don't have or have never seen a truly reference system/display. Many lack the ability to see that their instrument readings are not returning accurate results.

No matter how we feel about the "standards", consumers are stuck with media sources that are supposed to be mastered to those standards, currently REC 709 for HD. If your display is adjusted (calibrated) to reproduce REC 709 as accurately as possible, minimizing visible errors, you have the best opportunity to view content as it was mastered. As others mentioned, this is no guarantee it will look "natural".

The comment, by Otto, regarding the "color blind" is very appropriate. We as individuals and or calibrators, have no idea what others actually see, color, sharpness, luminance, etc. The more accurate we can make the display, the more natural it will look to anyone viewing.

As for the ISF Day/Night, this seems to fit more in the rear projection displays as well as LCD with back lighting. Then some with front projection want a mode for having some ambient lighting..... Plasmas, being a direct viewing source are not as affected by ambient light and the resulting elevated black levels.

great work Otto!

GlenC

03-09-09, 01:48 PM

As for the term "Calibration", there is nothing that can be done about it now. Just as many/most still believe Plasmas have image burn and short life issues..... We eould never be able to change the reference.... for marketing purposes, calibration has a better sound than a tune-up. The biggest issue is getting the information/idea that there is a way/procedure for consumers to get their TV to produce a more accurate picture, and, that they will enjoy the change.

Noshandy

03-09-09, 01:51 PM

The whole point of the argument is that it's not possible to make that judgement based on ssuch an example, especially not via a screenshot that I'm watching on a very inaccurate LCD screen on my laptop. I would have no idea what that particular scene was supposed to look like, so I wouldn't be able to say which of those two are "natural". It doesn't matter if the skin tone looks natural in that particular scene, if it was _supposed_ to look unnatural. If that's the case (which I don't know), other scenes that were supposed to look natural, would look un-natural if set to the "no cms" setting.

I agree to what your saying, its not hard proof by any means.

I find REC 709 gives you a consistancy across all material. Thats why now I don't have to adjust any settings anymore and just enjoy what I see.

dovercat

03-09-09, 01:55 PM

I am not against calibration, I am in favour of getting the most accurate display you can get. The only way to do this reliably is to use instruments to take measurements and compare these to the standards, adjusting the display to get it more accurate. In other words to calibrate the display towards a standard.

What I do fine annoying is misrepresentation unintentional as it maybe. That a calibrated display meets any strict criteria. Along with the word calibrated I find the word certified also misleading because of the connotation it has to me, rather than its meaning. Perhaps it is just me being annoyed at my poor grasp of the English language.

"so why are you (any of you) mocking the whole concept of calibration (which you obviously are, so don't even bother)"

The only thing I would mock is that the display calibration profession has no minimum standards. Institutions like ISF could make their accreditation mean something if they applied strict minimum standards of competence and minimum standards of calibrated display accuracy. THX it seems maybe somewhat better, I do not know.

If a certified calibrator guaranteed me a certain specified minimum level of competance and a calibrated display guaranteed me a certain specified minimum level of accuracy to standards, I would be happy they used the terms. As it is I think the terms are used because of what they imply, then their use defended by using their strict definitions, which renders them pretty meaningless statements.

Otto J

03-09-09, 02:13 PM

Another problem the calibration industry has, IMO, is that many calibrators don't have or have never seen a truly reference system/display. Many lack the ability to see that their instrument readings are not returning accurate results.

I couldn't agree more, I feel that watching JK's demos and attending his classes, as much as owning a similar system myself, has made major improvements to my work as a calibrator. Being able to tell if the picture "appears right" or not, makes it a lot easier especially when you need to make compromises, which you sadly often need to.

My main gripe with this discussion is: If a display cannot be calibrated fairly accurately, should we blame the design of the display, or the concept of calibration? And, even if it can, why wouldn't a calibration to whatever accuracy CAN be achieved, be an advantage? If the image post-calibration gets worse, I am willing to bet that it is due to a poor calibration, not an error with the concept of calibration itself. I mentioned the JVC example: You effectively have to create an error to minimize the negative effect of another error, to make sure the overall performance is more accurate. Put another way, 20% error in one area and 20% in another, may be much better than 5% error in one area and 50% in another. That's why I'm saying that it's the picture and not the measurements that need to be accurate. In my experience, more accurate _overall_, is always, always, _always_, better in the customers point of view, at the very least after a week or so of viewing. But, more accurate overall may mean compromising specific areas of performance, which many calibrators (and especially DIY'ers) have a very hard time understanding. And that's why I really dont care much for the "I saw a calibrated set once and it looked bad", or "I calibrated my set and it got worse" arguments. It is more than likely that it was calibrated poorly then.

Tryg, I'd like to make an analogy here: When we talk about "calibration" as a means of improving the viewing experience, it should be fairly obvious that we mean "calibration to the standards, as closely as possible", not "messing up the display and passing it on as calibration". Likewise, when you recommend people to use a JVC/HP combo, you inherently mean a properly set up JVC/HP combo, not a setup with the JVC at max zoom, max lens shift - right? Or, when you recommend 3-chip over 1-chip, you inherently mean a good 3-chip, not a cheapo LCD presentation device - etc. Why would calibration be any different?