or Connect
AVS › AVS Forum › Display Devices › Display Calibration › Gamma with normal windows or with APL?
New Posts  All Forums:Forum Nav:

Gamma with normal windows or with APL?

post #1 of 193
Thread Starter 
Last night I checked the calibration of my 65VT60. The TV has about 450 hours on it. Meter used was a new i1 D3 profiled against an i1 Pro spectro whose accuracy recently has been verified in a lab with a certified Konica Minolta LS-100.

The initial calibration was done with 4% windows on a black background. (A run using 6.5% normal windows yields about identical curves as those with the 4%.)




The resulting setting determined by the 4% calibration were checked using 10% windows and a 22% APL screen.





There is a distinct difference with the 10% APL -- which, of course, can be corrected. However, after the correction there is a slight visible visual difference. This is not unexpected given the blue boost required specifically at the 30% stimulus point.


Technically, why the difference? Which is correct -- the 4% normal windows or a corrected 10% with 22% APL?


EDIT: Both Ryan Masciola's disk and the GCD disk were used as sources to further check. No difference between the two. Probably, they are the same patterns?



Just for completeness, here is the outcome from a quick calibration of the above 10% on 22% APL runs. The charts look as good as those of the normal 4%. but there is a visible blue tint to the resulting picture.



Larry
Edited by LarryInRI - 8/11/13 at 4:56pm
post #2 of 193
Quote:
Technically, why the difference? Which is correct -- the 4% normal windows or a corrected 10% with 22% APL?

Welcome to the gordian knot that is known as Plasma. Both and neither are correct ... it appears to be a constantly moving target.

IMHO, when you reach the point where you essentially have to "guess" what type of target signal will produce "the" desired outcome, then the whole calibration process becomes suspect. Yet apparently, Plasma is what studios are currently using as so-called "reference" monitors.

Pick the one that "looks the best" to you and drive on down the road. Perhaps the professional "experts" can provide "better" advice ... but I think you'll just drive yourself crazy trying to analyze this with scientific method. wink.gif
post #3 of 193
post #4 of 193
The problem is... NEITHER method produces what you'd call an "accurate" gamma. You cannot measure plasma gamma EITHER way. And you have no way of knowing what APL sequence would produce the best end result because your plasma will display images with APLs of 0-100%. Which one are you going to choose? So does it REALLY matter if you use constant size patterns or variable APL patterns for grayscale measurements? Not enough to worry about.
post #5 of 193
Thread Starter 
@HDTVChallenged: Thank you. However, I am not new to plasma or calibration -- almost seven years now. I performed this particular experiment on both my Samsung 64D7000 and now my 65VT60 with the same results. I have always used normal windows but after reading multiple times the virtues of using an APL pattern, I decided to give it a test ride.

@airscapes: Thanks. I've followed those threads and have read just about all of Poynton's publications. (Yes, I have a strong background in physics.)

@Doug: Thanks. That's the same conclusion that I came to.


@zoyd and Tom: Technically why are you proponents of using the APL approach for calibrating gamma? Chad B, I think you too use APL patterns. If so, why?



Thanks again, all.
Larry
post #6 of 193
Quote:
Originally Posted by Doug Blackburn View Post

The problem is... NEITHER method produces what you'd call an "accurate" gamma. You cannot measure plasma gamma EITHER way. And you have no way of knowing what APL sequence would produce the best end result because your plasma will display images with APLs of 0-100%. Which one are you going to choose? So does it REALLY matter if you use constant size patterns or variable APL patterns for grayscale measurements? Not enough to worry about.

Quote:
Originally Posted by LarryInRI View Post

@Doug: Thanks. That's the same conclusion that I came to.
Larry
Agree. Differences are fairly minor assuming you're using reasonably appropriate patterns for PDP. Then it pretty much becomes a personal preference.
post #7 of 193
Quote:
Originally Posted by LarryInRI View Post

Chad B, I think you too use APL patterns. If so, why?

Larry

I found that low end gamma and shadow detail matched what I was seeing more closely, in reference to an LCD with no dynamic backlight action or other similar stable displays.

With regular windows, plasmas might show a low gamma at the low end (around 2 or so), yet they would look very neutral.
post #8 of 193
Thread Starter 
Thanks, Chad. Yeah, for a few years now I have prefered to calibrate plasmas purposely with a very slight drop in gamma at the low end. Like this:


The detail that is gained -- lapels on dark suits, texture on dark leather jackets, etc -- is justified in my opinion.

Don't want to repeat all that has been thrashed over in other threads. Just wanted to demonstrate that real life is more than numbers alone. biggrin.gif

Larry
post #9 of 193
Plasma's Gamma is floating up/down upon displayed content.

The fact is that @ Professional World, they don't use APL Patterns...

CineTAL DAVIO's or Pantora Pluto's (both can work as pattern generators) that are installed to major DI Studios worldwide they are generating/using only Window/Full Field Patterns, there no option for APL there...

Also Professional Profiling software like LightSpace/TrueLight/CineSpace don't have option to generate APL Patterns using their internal pattern generators....

This is happening for over a decade in that professional world...

Anyone can use whatever pattern size he has found to his setup that he liked the final results more, not based @ dE Reports, but viewing actual movie content.

BTW with LightSpace + eeColor, I have generated from one full 17-Point Cube Profiling Measurements, 6 different Gammas for Rec.709: 2.18, 2.20, 2.22, 2.25, 2.27. 2.29 in a few seconds, not by taking new measurements for each gamma target, and i'm selecting realtime what gamma seems better to my eyes for each movie during the first minutes of each movie...

Since there no standard to gamma (typical 2.20) or studios they don't follow this 100%, these 6 gamma memory slots solves that problem yo my setup with my KURO.
Edited by ConnecTEDDD - 8/12/13 at 4:56pm
post #10 of 193
Thread Starter 
Ted,

Thank you. Thank you. Thank you.

I have been using, in the same way as you, three different gammas on my VT60. One in the Custom mode, one in Pro 1, and one in Pro 2. They are 2.19, 2.22, and 2.26 respectively.

Thank you. Now I don't feel so crazy.

Larry
post #11 of 193
Quote:
Originally Posted by LarryInRI View Post

Thanks, Chad. Yeah, for a few years now I have prefered to calibrate plasmas purposely with a very slight drop in gamma at the low end. Like this:


The detail that is gained -- lapels on dark suits, texture on dark leather jackets, etc -- is justified in my opinion.

Psst: You're kind of on the way to a BT.1886(ish) curve there. Might be worth trying to approximate it for real if you have a working 10pt (or more) GS.
post #12 of 193
Maybe BT.1886 gives a gamma that's too high for his liking. I'm sure Larry is aware of this option.
post #13 of 193
Quote:
Originally Posted by rahzel View Post

Maybe BT.1886 gives a gamma that's too high for his liking.

That would depend entirely on his measured black-level.
post #14 of 193
The VT60's mll is ~0.0012/0.0013fL, and I know Larry targets 28fL for his night mode. That and I see his gamma is ~2.2. That's the reasoning for my guess.
post #15 of 193
Quote:
Originally Posted by rahzel View Post

The VT60's mll is ~0.0012/0.0013fL, and I know Larry targets 28fL for his night mode. That and I see his gamma is ~2.2. That's the reasoning for my guess.

Taking data from Cleveland Plasma's review black level = 0.00146ftL (0.005 Nits,) assuming 35 ftL (119 Nits) for white then my BT1886 calculator outputs the following:
Code:
Comparison of Effective “Gamma” Values (e.i. what the power law exponent/gamma would be)             

% Stimulus      BT1886  
5%                     2.20     
10%             2.27    
15%             2.30    
20%             2.31    
25%             2.32    
30%             2.33    
35%             2.34    
40%             2.34    
45%             2.35    
50%             2.35    
55%             2.35    
60%             2.35    
65%             2.36    
70%             2.36    
75%             2.36    
80%             2.36    
85%             2.36    
90%             2.36    
95%             2.36

So that's what the actual BT1886 standard says for the VT60's black level. smile.gif So ... use at your own risk. For your entertainment and amusement only. wink.gif
post #16 of 193
HCFR has a BT.1886 option for gamma.

And FWIW, all of D-Nice's calibration reports show that the VT60 consistently measures 0.0012/0.0013fL using his Klein.

But ya, I figured the gamma would mostly be ~2.3 or higher if BT.1886 was used on the VT60 with a white level of 28-35fL.
Edited by rahzel - 8/13/13 at 2:03am
post #17 of 193
Quote:
Originally Posted by rahzel View Post

HCFR has a BT.1886 option for gamma.

And FWIW, all of D-Nice's calibration reports show that the VT60 consistently measures 0.0012/0.0013fL using his Klein.

I just scanned the CP review for the black level, but it appears that there are three possible settings related to this and only the middle one (as opposed to the darkest) yielded "stable" results. I'll have to go back and study it closer. It could be that the meter they used just hit it's minimum.
Quote:
But ya, I figured the gamma would mostly be ~2.3 or higher if BT.1886 was used on the VT60 with a white level of 28-35fL.

Well ... depending on how you (mis)interpret BT1886, you *could* go ahead and call that BL zero and you'd wind up with 2.4 across the board. smile.gif

PS: Why are we still using ft-L? We actually calibrate using nits (cd/m^2,) it's beyond annoying to have to convert back and forth. wink.gif

PPS: I'm just saying that instead of "guessing" at these gamma-ish things, perhaps if we all started using the actual standard, we might start getting more consistent and repeatable results. I know, a shocking concept. wink.gif
post #18 of 193
Quote:
Originally Posted by HDTVChallenged View Post

I just scanned the CP review for the black level, but it appears that there are three possible settings related to this and only the middle one (as opposed to the darkest) yielded "stable" results. I'll have to go back and study it closer.

Chad B does the CP reviews and for whatever reason his mll readings always come out not quite as dark, not questioning his skill mind you but have seen an inconsistency on other sets
post #19 of 193
Quote:
Originally Posted by HDTVChallenged View Post

I just scanned the CP review for the black level, but it appears that there are three possible settings related to this and only the middle one (as opposed to the darkest) yielded "stable" results. I'll have to go back and study it closer. It could be that the meter they used just hit it's minimum.
Possibly. Chad uses a C6 for his black level measurements. Although the C6/D3 usually can read down to 0.001 fL (or 0.0034cdm2 smile.gif) some C6/D3s might have issues reading that low. I would trust D-Nice's Klein measurements.
Quote:
PS: Why are we still using ft-L? We actually calibrate using nits (cd/m^2,) it's beyond annoying to have to convert back and forth. wink.gif
Dunno, seems like most people in North America use ftL and people in Europe use cd/m2.
Quote:
PPS: I'm just saying that instead of "guessing" at these gamma-ish things, perhaps if we all started using the actual standard, we might start getting more consistent and repeatable results. I know, a shocking concept. wink.gif
I'm not trying to be argumentative and I'm not disagreeing with you, I was just guessing why Larry chose not to use BT.1886. smile.gif

But I do wonder why BT.1886 isn't used more often, both by DIYers and pros. Obviously some displays simply don't have the controls needed to effectively calibrate to BT.1886, but I rarely see people use it.
Edited by rahzel - 8/13/13 at 11:24am
post #20 of 193
Thread Starter 
Quote:
Originally Posted by rahzel View Post

Maybe BT.1886 gives a gamma that's too high for his liking. I'm sure Larry is aware of this option.

Yes, Aware of it since it was announced. I tried it out and after awhile found it not to my liking.

bt1886 for 30 ftL and 0.0014 ftL


Larry
Edited by LarryInRI - 8/13/13 at 2:42pm
post #21 of 193
Quote:
Originally Posted by Chad B View Post

I found that low end gamma and shadow detail matched what I was seeing more closely, in reference to an LCD with no dynamic backlight action or other similar stable displays.

With regular windows, plasmas might show a low gamma at the low end (around 2 or so), yet they would look very neutral.

Uh, that's not possible. "Low end gamma and shadow detail" are as much under the influence of APL as the rest of the image. If the APL of images is in the 80%-100% range, your shadows are going to look different than how your shadows will look if the APL is in the 1%-20% range. It doesn't matter what type of calibration patterns you use, in the end you can only calibrate each grayscale step for some specific APL... doesn't really matter if you use, I don't know, 50% APL for example, for every step or variable APL... when you examine shadow performance, plasmas will NEVER look the same with high APL images and low APL images.

I understand the logic, but I think the prevailing "logic" for using constant APL patterns is just as flawed as the logic for using 10% (size) windows with a black background. And there's no way to come up with any calibration logic/scheme/method that is any better than "traditional" windows or constant APL windows. Every thing you try will look a little different, but none of them will be more right or less wrong than a different approach (assuming you aren't TRYING to make a bad choice for calibration).
post #22 of 193
Quote:
Originally Posted by LarryInRI View Post

I have been using, in the same way as you, three different gammas on my VT60. One in the Custom mode, one in Pro 1, and one in Pro 2. They are 2.19, 2.22, and 2.26 respectively.

perhaps I'm overlooking something, but it seems those three gammas are nearly identical to one another (especially the first two)... are you really seeing a meaningful difference as you toggle between them?
post #23 of 193
Quote:
Originally Posted by LarryInRI View Post

Yes, Aware of it since it was announced. I tried it out and after awhile found it not to my liking.

bt1886 for 30 ftL and 0.0014 ftL


Larry

Cool ... I was going to post that graph last night, but I was afraid it would be too ginormous and eat the entire forum. wink.gifbiggrin.gif
post #24 of 193
Thread Starter 
Quote:
Originally Posted by PlasmaPZ80U View Post

perhaps I'm overlooking something, but it seems those three gammas are nearly identical to one another (especially the first two)... are you really seeing a meaningful difference as you toggle between them?


The ones I use are not "nearly identical to one another." They are quite different. Read Ted's post above (#9). He uses "6 different Gammas for Rec.709: 2.18, 2.20, 2.22, 2.25, 2.27. 2.29".

There are definite visible changes with small gamma changes.


Larry
post #25 of 193
Larry, you should ask your buddy Silly Sally,I'am pretty sure he will give you the correct answer on which pattern is correct.biggrin.gif
post #26 of 193
Quote:
Originally Posted by ConnecTEDDD View Post

Plasma's Gamma is floating up/down upon displayed content.

The fact is that @ Professional World, they don't use APL Patterns...

CineTAL DAVIO's or Pantora Pluto's (both can work as pattern generators) that are installed to major DI Studios worldwide they are generating/using only Window/Full Field Patterns, there no option for APL there... .

Actually that's not true.

The Pluto does APL under CalMAN. Studios are moving more and more to using APL patterns for their plasmas.

Also Plasmas are typically secondary monitors, where the primary displays are typically LCDs.
post #27 of 193
Quote:
Originally Posted by sotti View Post


The Pluto does APL under CalMAN. Studios are moving more and more to using APL patterns for their plasmas.

I was talking of what is happening the last decade...

You have Add Support for APL Patterns for Pluto @ CaMAN Studio released @ May 27, 2013.

It's cool feature for Pluto users, Now think from the thousands of Movie Industry Studios around the world, how many have CalMAN Studio installed?

Because i'm watching various proffesional/colorist forums (RED,ARRI etc) I haven't seen any post/topic related with CalMAN Studio from people that are coming from professional world. (or @ SpectraCAL Forum). Users there they mainly talking for TruLight/CineSpace/LightSpace.

I have asked again to privide us a list with Studios that are using CalMAN Studio and I got a reply from you: ''Technicolor''.

Yeap, Technicolor TV/Broadcast Division has some CalMAN licencies, not the Film Division...

Any other Film Related Studio?

Can you name them?

Is this list secret?

There is a list of CalMAN Clients here, but the 2-3 there are TV/Broadcast Studios, Not Film...

ID2 Films = user's ''Manni'', who is reporting bugs/problems daily to your forum... (With Lumagen Radiance)

Quote:
Originally Posted by sotti View Post


Also Plasmas are typically secondary monitors, where the primary displays are typically LCDs.

Exactly, Plasmas are not recommended for Critical Color Applications @ Proffesional World. Grade 1 LCD's or DCI Projectors do that job better.
Edited by ConnecTEDDD - 8/14/13 at 9:37pm
post #28 of 193
Quote:
Originally Posted by LarryInRI View Post

Yes, Aware of it since it was announced. I tried it out and after awhile found it not to my liking.

bt1886 for 30 ftL and 0.0014 ftL

Larry

Actually, something looks quite wrong with that graph compared to what I get on the final (unreleased) version of my spreadsheet. BT1886 should "bottom out" at ~2.19 at 5% stimulus ... decreasing even more (brighter) below .... Edit to come.
post #29 of 193
Larry,

I've heard this debate between using large window patterns and 5% APLs with either a 18% or 25% APL..

As a DIY calibrator who has settled on APL patterns, I thought the picture looked kind of dark in the shadows when using the large windows pattern..

So I calibrated my wife's LCD display and put it next to the plasma. Idea being that you don't have the ABL thing going on with an LCD so it's gamma calibration would likely be fairly accurate.

The gamma derived from using a large window on the plasma was a lot closer to the LCD's "look" than the 5% APLs. Deep shadows were a touch darker on the plasma than on the LCD. That would be simple enough to correct.. In comparison, the APL pattern calibrations looked washed out and lacked a certain fidelity to them which makes sense and shows up moreso when doing a side by side comparison. I never really noticed it with the plasma by itself.

Now the real question is do I really like it that way.

Well.....it's growing on me.
Edited by JimP - 8/15/13 at 12:42am
post #30 of 193
Quote:
Originally Posted by ConnecTEDDD View Post

Plasma's Gamma is floating up/down upon displayed content.
--- SNIP---
BTW with LightSpace + eeColor, I have generated from one full 17-Point Cube Profiling Measurements, 6 different Gammas for Rec.709: 2.18, 2.20, 2.22, 2.25, 2.27. 2.29 in a few seconds, not by taking new measurements for each gamma target, and i'm selecting realtime what gamma seems better to my eyes for each movie during the first minutes of each movie...

Since there no standard to gamma (typical 2.20) or studios they don't follow this 100%, these 6 gamma memory slots solves that problem yo my setup with my KURO.

Ted, what at-least one poster may not understand is you are using the same large profile (4913 points) and from those 4913 measurements/profile, that one profile LS has created using the different gammas you type in.to create separate LUT's (calibration). So in theory even frailly small differences (2.18-2.22) would be because of the use of the gamma.
Rather than 6 different sets of measurements (calibrations) on a plasma, each using different Gammas. imo to much room for errors, therefore is it actually the Gamma or the difference in the readings when doing 6 calibrations.

As for LLM on my 65VT60 that can range from 0.0011 to 0.0014, depending on my pre calibration settings using custom, pro 1&2 modes.

As for APL verse non APL or window size, I am not going to go there except to say if I could do what JimP suggest for APL with a 5% window I would love to try it. For now I have only three choices about 2%, about 11% and full screen all non APL. I use about 11% with meter backed off about 18", oh I think you guys may be over looking on screen and off screen modes. smile.gif

ss
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › Gamma with normal windows or with APL?