AVS Forum banner
Status
Not open for further replies.
1 - 20 of 55 Posts

·
Registered
Joined
·
909 Posts
Discussion Starter · #1 ·
To my shock and horror, Russ Brookes measured my viewing brightness at only 2.75 ftlamberts. He used the updated colorfacts version whereby measurements were taken off my Stewart Studiotek Screen.


After removing the FLD Filter off my Plus Piano, my brightness when up tp almost 6 ftLamberts. Russ believed that the brighter image was the more realistic one, however I preferred the image with the FLD filter on as it produced a smoother, more realistic skin tones and better blacks.


I find grey blacks much more distracting then a less bright image.


When the results were revealed to me I was very surprised because I was thinking " man you guys are arguing between 12 and 16 ft lamberts...and I only had 2.75 and I get totally and fully immersed in my movie watching.."


None of the my guest (must be numbering at about 60 now) have commented that the projected image seemed dim at all.


My point is that if you don't have a point of reference, you can actually be happy and live with an image for a long time.


I am much happier with this image at 2.75 lamberts than when compared to the 1000 ansi lumens my previous NEC VT540 was producing.


Contrast is king fellas...no point arguing.
 

·
Registered
Joined
·
1,481 Posts
lyndolim,

sorry if I might look rude, but these posts are those that may confuse people.


What do you mean with "2.75 fl"? I guess at 100 IRE, but you didn't say it...


What screen dimension we are talking about? If you use a 80" wide this could heart stopping, but if you have a 130" one with low contrast and calibrated conditions, this may or could appear as normal...


I've briefly tested the Piano, and I totally agree that the image was pretty and I appreciate it a lot: not a first class, but an enjoyable pj.


But I'm ready to bet that if you try to measure your CR, the number won't be so high... IMHO, what you find good in your pj is its color balance and depth, more than a simple number as CR... I've measured a 7" CRT (Maxivideo) cheap and honest, with a CR that didn't exceed 488:1, but with good blacks, a film-like image and 3D reconstruction: better than every LCD I've seen so far!
 

·
Registered
Joined
·
909 Posts
Discussion Starter · #4 ·
Andrea,


I believe 2.75 was measured at 100 IRE with a 80 inch wide screen. The projector is fully calibrated and does have a FLD daylight filter.


The contrast ratio measured off the screen was higher with the brighter image without the filter so you are right that the color balance and blacks were better not so much the contrast ratio. However, when the contrast is measured straight off the light coming from the projector, the contrast ratio is higher when the filter is in place. So it must be the environment which is affecting the measurements when readings are taken off the screen. There is absolutely no loss of image detail with the filter on and the overall image just takes on a creamier film like appeal with the filter on.
 

·
Registered
Joined
·
1,481 Posts
lyndonlim,

thanks for your reply and sorry again for being somehow rude... ;)


What you say once confirms to me that CR is a... number! What really matters is that you see something on your screen, I'd better say reflected from your screen, and this has a RGB levels and spectral response.


That is what your (=my) eyes really get.


I can't expect a high CR with only 2.75 fL on a 80" screen, but, if it so (as I guess), tell me why you find your image so good?


Why do you prefer this setup to a NEC VT540 that surely has a better CR?


For the CR values? Naaahhh... ;)
 

·
Registered
Joined
·
467 Posts
2.75 :eek:


Wow. That is one very dim projector. I calculate 52 lumens.


If you are happy with it that's all that matters. Personally I couldn't stand to watch a picture that dim, even with no point of reference to compare. I just don't get drawn into the movie unless it has a certain level of brightness.
 

·
Registered
Joined
·
23,188 Posts
Quote:
Originally posted by lyndonlim
To my shock and horror, Russ Brookes measured my viewing brightness at only 2.75 ftlamberts. He used the updated colorfacts version whereby measurements were taken off my Stewart Studiotek Screen.
If you have a chance to do it again it would be interesting to see what you would measure at the screen if the sensor was turned around toward the projector. I just think it would be a good test of the sensor/test accuracy. You know the screen is greater than 1.0, so you should measure less than 2.75 ft-lamberts that way. If you measure something like 5 ft-lamberts then you know that your old reading was because of some measurement problem (like you had to angle the sensor and were actually measuring the off-axis performance).


Okay, I guess it shows that I am a little skeptical of the measurement and would want to run some tests to try to rule out measuring error.


I just got this new sensor, but haven't tried this test myself.


--Darin
 

·
Registered
Joined
·
1,319 Posts
what is the formula for calculating brightness?


I have an Infocus X1, it is supposed to output about 600 lumens in video mode. I am using an ND4 filter and 3 other filters (skylight, yellow 1 and warm filter. My screen is 5 ft wide. I am guessing that I am also below 10 ft L. I find the picture bright enough though.
 

·
Registered
Joined
·
909 Posts
Discussion Starter · #9 ·
Darin,


I am quite confident that the sensor was measuring the brightness accurately because it takes a reading in a targeted fashion. There is a small area in which you can lock in the reading off the screen.


And Andrea, I really don't believe that the VT540 could exceed the Plus Piano in contrast ratio. Its whites would be brighter but the blacks would really sux. I guess I have always associated excellent blacks with high contrast ratio and that isn't always the case. So what I am trying to say is that getting good black levels and shadow detail is very important to me in not being distracted when viewing a movie.


I don't think those with HD2 machines will have too much to complain in the area of brightness vs good blacks because HD2 is at a level where you can pretty much sit back and enjoy the movie without being overly distracted by grey black....unless you want to start getting extra critical.
 

·
Registered
Joined
·
1,481 Posts
Quote:
Originally posted by lyndonlim


And Andrea, I really don't believe that the VT540 could exceed the Plus Piano in contrast ratio. Its whites would be brighter but the blacks would really sux. I guess I have always associated excellent blacks with high contrast ratio and that isn't always the case. So what I am trying to say is that getting good black levels and shadow detail is very important to me in not being distracted when viewing a movie.


I don't think those with HD2 machines will have too much to complain in the area of brightness vs good blacks because HD2 is at a level where you can pretty much sit back and enjoy the movie without being overly distracted by grey black....unless you want to start getting extra critical.
lyndonlim,

once again, CR is... a ratio!

So, let me reason in terms of cd/mq (it's easier because we don't have too much zeroes like the fL case), here 2 examples (not real, only figured):


Pj A: 0 IRE = 0.02; 100 IRE 45 => CR 2250:1


Pj B: 0 IRE = 0.2; 100 IRE 450 => CR 2250:1


Pj C: 0 IRE = 0.04; 100 IRE 45 => CR 1125:1


From the experiences I made with instruments (and according to my taste, that's trivial, ;) !), and assuming all the pjs had same calibrated color balance, Pj A and C will look very similar when watching a movie.

I would throw Pj B out of the window...


I've measured various HD2s in the 0.03 - 0.08 range, and, even if noticeable, the difference is tiny.

On the other side, high emissions at 100 IRE are normally what it leads to white crushing. Too much light on the screen to be handled properly.


And, more: let's focus on the 100 IRE value.


Pure white, isn't it?


How many times, making spatial and temporal averages, you have a pure white?

I'm encouraging you to think to this matter considering the images you are used to watch at.


Here's an actual example:

http://www.htprojectors.com/htprojec...DSC09467_b.JPG


The area on the left part of the girl's face is white. If not a 100 IRE, close to it.


How much, in terms of surface? 2%? 5% to be generous?


And how much do you think this value will affect your image?


We are considering here something that nearly never occurs to help us evaluating a pj's performance...


What is really important is black level, as you stated.

IMHO, add color accuracy.

3D rendering.

Resolution.

Numbers that can get an ABSOLUTE value, not a RELATIVE one like CR... ;)
 

·
Registered
Joined
·
909 Posts
Discussion Starter · #11 ·
Andrea,


You are good, my friend. What you say sounds very logical. Doesn't a high contrast ratio contribute to 3D rendering though ? or is it a combination of good black level and contrast ratio. I do agree with you that the sum of the parts is what makes a picture appealling not just evaluating the parts by itself.


If I lived in Rome, I would not bother owning a home theatre as I would be outdoors all the time admiring its beauty : ) Rome has got to be within the top 2 most amazing city in the world to live in.
 

·
Banned
Joined
·
3,781 Posts
Lyndonlim,


The meter on the CF6500 is not accurate for ft/l when compared to a Minolta CS-100. Do not misunderstand me, it is GREAT meter with excellent color readings and is probably the next best thing besides a Photo Research photo spectraradiometer for color readings. But it's main purpose is not ft/l and its readings vary greatly compared to the Minolta. The ft/l feature is for relative readings and not absolute ones. Maybe, Mark can chime in and tell us the actual differences, I know they make software that works with the Minolta as well.


Bob
 

·
Registered
Joined
·
10,201 Posts
Despite Andreas intimations to the contrary, I don't know anyone who judges a projector solely by its CR---and, in particular, on/off CR. He seems to continue to believe this, despite pages and pages of discussion with him in other threads on this forum. In fact he's suggested to me, jokingly I'm sure, that it is some sort of American obsession :)


The truth is that we Americans don't worship at the altar of on/off CR. We fully understand that it is a ratio that is, by itself, useless. It needs to be combined with other information about the projector, such as its brightness (white level), to give any sort of meaninful interpretation. And even then, it is only part of the PQ story. But it is hardly a useless number to have even in the context of this discussion.


I suspect that part of this belief may stem from a continued confusion on the difference between on/off CR and instantaneous CR---that is, the amount of contrast a projection system can reveal in a single scene. The truth is, for most scenes we'll watch on our projectors, instantaneous contrast is determined not by the projector itself but by the viewing environment: the screen, the color of the walls and ceiling, and so forth.


The ANSI checkerboard contrast measurement is a measurement of instantaneous contrast. From reading this forum, the best number I've seen for that is 300:1---and that was a nearly perfect test room: black walls, black furniture, even black clothes on the viewers. A single white shirt in that test room brought that number up to 200:1. I'll bet that most reasonable rooms come in around 100:1 or less.


So in the screenshot above, I fully agree that projectors A and C should indeed look the same if they have the same greyscale tracking and color fidelity. In this case, on/off CR doesn't matter at all.


When does it ever matter, then? Simple: dark scenes! In a dark scene, that 100:1 instantaneous CR "window" gets shifted down, and the black level of a projector with lower CR will likely fit "within" that window. Suddently, the projector's poor on/off CR is limiting the instantaneous contrast. Put a sufficiently dark scene up, and I guarantee that the lower on/off CR of projector C would be apparent.


My point is this: properly interpreted, and properly integrated with other measurements of picture quality, on/off CR is plenty important, even in these scenarios.


P.S. I wouldn't "throw projector B out the window" either. That sounds like a great projector to have. First, spend some of that extra brightness on a bigger screen! :) Then, put an ND filter on it to bring the white level down for critical movie viewing in a dark room. You'll match the quality of PJ A if all else were equal. Then, when you're watching the football championships (American or European, doesn't matter) and you've got some lights on so everyone can find their beer and peanuts, take the ND filter off and enjoy that extra brightness. :)
 

·
Registered
Joined
·
1,481 Posts
Quote:
Originally posted by lyndonlim
Andrea,


You are good, my friend. What you say sounds very logical. Doesn't a high contrast ratio contribute to 3D rendering though ? or is it a combination of good black level and contrast ratio. I do agree with you that the sum of the parts is what makes a picture appealling not just evaluating the parts by itself.


If I lived in Rome, I would not bother owning a home theatre as I would be outdoors all the time admiring its beauty : ) Rome has got to be within the top 2 most amazing city in the world to live in.


lyndonlim,

come here, I'll find you a good hotel and we could spend time sightseeing and watching pjs with good black level and poor CR... ;) ;) ;)

Yes, I think CR can help 3D rendering: maybe combined to resolution and DMD speed... ;)
 

·
Registered
Joined
·
1,481 Posts
Quote:
Originally posted by BobL
Lyndonlim,


The meter on the CF6500 is not accurate for ft/l when compared to a Minolta CS-100. Do not misunderstand me, it is GREAT meter with excellent color readings and is probably the next best thing besides a Photo Research photo spectraradiometer for color readings. But it's main purpose is not ft/l and its readings vary greatly compared to the Minolta. The ft/l feature is for relative readings and not absolute ones. Maybe, Mark can chime in and tell us the actual differences, I know they make software that works with the Minolta as well.


Bob
Bob,

I don't have a Minolta CS-100, so I can't throw judgments so something I don't know personally.


What I can tell you is that I have a DeltaOhm DO9847 data logger, and I'm quoting myself here from the 6 DLPs shoot out on HTProjectors:

Quote:


Accuracy of the instruments

Although I think nothing is perfect (and soon there will be some news in this sector), the Gretag EyeOne’s performance (the Colorfacts CF-6000’s spectroradiometer) has been compared to the data reading’s values provided by the DeltaOhm DO9847K, a data logger made by DeltaOhm ( www.deltaohm.com for further details) featuring a probe (LP471 LUM2) for all luminance measurements. This device has a price of 1.200 € (about $ 1,400), has been individually calibrated, and it has replaced the HD9221 (still by DeltaOhm) we previously used, for a superior accuracy in its measurements.

Its main specifications include a measurement field ranging from 0.1 to 1999*10^6 cd/m2, with a 0.1 cd/m2 and a reading angle of 2° according to the photopic standard vision curve. The calibration’s inaccuracy is
All measured values have proven comparable.

Here is a practical example: a 31.35 cd/m2 value of the Gretag means a 31.8 cd/m2 value by the DO9847K, a 0.05 value in the first case coincides with a value between 0.0 and 0.1 in the second case.

This means I am “rather” sure about the numbers I obtained from this test.
I agree with you that it has some shifts, that can be minimized repeating the measures and making a dark reading before each measure (what I do when using it).


I frankly don't know where you got the information "The ft/l feature is for relative readings and not absolute ones".

What you say is true for the CA-1 (at least the version I used 2 years ago), according to what Cliff himself told. But not for the Gretag.


Once again, we bought the DeltaOhm (that's NOT a cheap lux or fL meter...) and found a good match. I do agree with you the Gretag was not made to be used as a fL meter, but it seems it could work as well... ;)
 

·
Registered
Joined
·
1,481 Posts
Michael,

:p :p ! I wish I had your sense of humor and could express myself in English with the lightness you show!

I couldn't stop laughing when reading about the "American obsession"!!!

Quote:
Originally posted by Michael Grant


I suspect that part of this belief may stem from a continued confusion on the difference between on/off CR and instantaneous CR---that is, the amount of contrast a projection system can reveal in a single scene. The truth is, for most scenes we'll watch on our projectors, instantaneous contrast is determined not by the projector itself but by the viewing environment: the screen, the color of the walls and ceiling, and so forth.


The ANSI checkerboard contrast measurement is a measurement of instantaneous contrast. From reading this forum, the best number I've seen for that is 300:1---and that was a nearly perfect test room: black walls, black furniture, even black clothes on the viewers. A single white shirt in that test room brought that number up to 200:1. I'll bet that most reasonable rooms come in around 100:1 or less.
I totally agree, and maybe I'll quote you when trying to kill CR in a upcoming article! ;)


Quote:
Originally posted by Michael Grant


When does it ever matter, then? Simple: dark scenes! In a dark scene, that 100:1 instantaneous CR "window" gets shifted down, and the black level of a projector with lower CR will likely fit "within" that window. Suddently, the projector's poor on/off CR is limiting the instantaneous contrast. Put a sufficiently dark scene up, and I guarantee that the lower on/off CR of projector C would be apparent.
Here I don't follow you... :( but maybe it's up to me and I'm not catching your point.


A dark scene is... dark! So, what should matter is: i) black level; ii) low IRE levels; iii) low level linearity (i.e., gamma at low levels), to avoid compressions or black crushing.


If you refer to ANSI CR, I guess we're introducing in our measurement system what I can call an ambient factor, thus loosing any chance to have a method. If this varies in every theater, it's valid only in MINE. So, practically useless.


If I'm correctly interpreting what you've wrote,
Quote:
instantaneous contrast is determined not by the projector itself but by the viewing environment: the screen, the color of the walls and ceiling, and so forth.


The ANSI checkerboard contrast measurement is a measurement of instantaneous contrast.
we are talking here about FOfo CR.

So, in a dark scene, since there's no light (or a slight amount of it, spatially and temporally), CR tells you: "This pj has this level of black" (Thanks for the info, it's useful!) and "This pj has this level of white" (I couldn't care less in this situation... :( ).

So, what's CR up for?


But I'm very intrigued by your position: how could we measure this "instantaneous contrast"? Suggestions?


I would like to find correlations between CR and PQ: but Michael, believe me, I know it's highly subjective, but until now I found NO correlation between the CR values and the quality of the projectors I've been testing all over the years... :(


Thanks, Michael, I loved this post!!! This

Quote:
Originally posted by Michael Grant


P.S. I wouldn't "throw projector B out the window" either. That sounds like a great projector to have. First, spend some of that extra brightness on a bigger screen! :) Then, put an ND filter on it to bring the white level down for critical movie viewing in a dark room. You'll match the quality of PJ A if all else were equal. Then, when you're watching the football championships (American or European, doesn't matter) and you've got some lights on so everyone can find their beer and peanuts, take the ND filter off and enjoy that extra brightness. :)
is a masterwork!!! ;)
 

·
Registered
Joined
·
16,014 Posts
Andrea,


Some of the people over at the CRT forum have been asking for HT1000 screenshots. I tried to find the ones you've posted but cannot. Can you post some over on that forum?


Thanks,


Bob
 

·
Registered
Joined
·
6,227 Posts
Quote:
but until now I found NO correlation between the CR values and the quality of the projectors I've been testing all over the years...
Apologies Andrea for not subscribing to read the complete text of your shootout but just for the record what are your 3 favorite digital projectors ever ??
 

·
Registered
Joined
·
1,481 Posts
Bob, yes! But maybe I ve too many...


These are interlaced component from Denon 3800:

http://www.htprojectors.com/htprojec.../dsc07607R.jpg


http://www.htprojectors.com/htprojec...C07604_med.jpg



Here we have DVI, Radeon 9700, Catalyst 3.0, TT 1.5:

http://www.htprojectors.com/htprojec.../dsc07556R.jpg


D-VHS tape,

http://www.htprojectors.com/htprojec.../dsc08213R.jpg

http://www.htprojectors.com/htprojec.../dsc08223R.jpg


I must admit that PQ of these photo has been improved later (and still far from perfection!), but these are original and coming from the NEC HT1000 single test.


Maybe you'd like these...

http://www.htprojectors.com/htprojec...C07561_med.jpg

http://www.htprojectors.com/htprojec...C07562_med.jpg



Calibration values were:

VGA

Brightness = -5

Contrast = 32

Color = 48

Hue = 30

Sharpness = 14


DVI

Reference setting = Movie

Gamma correction = Natural

Red = -16

Green = 6

Blue = 10

Yellow = 2

Magenta = -16

Cyan = 5

Color gain = 0

Color temperature = 6500°K

White balance = N.A.

Picture options: N.A.


I'm now reconsidering, since I'm performing a mini-shoot with the Domino 20, the Eco-mode, that has less light but a better PQ (IMHO, obviously!) and puts the HT1000 in a more quiet mode.


I don't want to overflow this Forum like I'm doing with HTProjectors, so I'll stop here ;)
 

·
Registered
Joined
·
10,201 Posts
Quote:
A dark scene is... dark! So, what should matter is: i) black level; ii) low IRE levels; iii) low level linearity (i.e., gamma at low levels), to avoid compressions or black crushing.
Obviously, Andrea, this is all correct. But the black level is nothing more than the white level divided by the on/off CR, so that's why I'm saying that the CR matters.


So actually, Andrea, I do agree with you that it's the absolute value of black level that we ultimately care about. But a projector manufacturer can't possibly give us that number on their spec sheets, because black level depends on screen size and screen gain, just like white level does. So the manufacturer gives us brightness and on/off CR instead, and leaves it to us to compute black level from that. In other words, on/off CR is important only because we can use it to compute black level.
Quote:
But I'm very intrigued by your position: how could we measure this "instantaneous contrast"? Suggestions?
I'm not sure what you mean. The ANSI contrast measurement is basically a measurement of worst-case instantaneous contrast. Isn't that acceptable? Just throw up an ANSI checkerboard and take some measurements.


But to be honest, ANSI contrast is a useless measurement to make for digital projectors, because all you'll end up measuring is the room. As I said above, ANSI contrast is determined almost exclusively by the screen gain, the screen color, the wall color, the ceiling color, and so forth---not the projector.
Quote:
I would like to find correlations between CR and PQ: but Michael, believe me, I know it's highly subjective, but until now I found NO correlation between the CR values and the quality of the projectors I've been testing all over the years...
You're telling me that you can't tell the difference between a 1125:1 projector and a 2251:1 projector, with all other factors being equal? For all possible movie scenes? With the projector set to provide a reasonable white level? I don't believe you. Sure, there would be no apparent difference for most material. But put up some very dark scenes, with a significant amount of 0IRE black, and the difference will be obvious. The difference may not bother you, but it will still be obvious.


Honestly I think that we may have nothing but a language difference here. You seem to agree that black level matters. Well, black level equals white level divided by on/off CR. So if black level matters, then on/off CR matters.
 
1 - 20 of 55 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Status
Not open for further replies.
Top