or Connect
AVS › AVS Forum › Display Devices › Display Calibration › Better to have 2.2 gamma or stable 2.3 gamma on a plasma?
New Posts  All Forums:Forum Nav:

Better to have 2.2 gamma or stable 2.3 gamma on a plasma? - Page 3

post #61 of 135
Quote:
Originally Posted by Chronoptimist View Post

You have not yet shown anything that proves there is something "going on" with small area patterns on that display.

I didn't say that something is "going on" with small area patterns, only that they may not be the best choice for this particular problem. The effect itself is present for any shape/size of pattern, it's inherent in the display. The question is how best to optimize the calibration given this behavior.

The easiest way to see what's going on is this plot. This plot shows you that independent of the various pattern sizes used to do this measurement the shape of gamma vs. APL very consistently shows that it's lower at low APL than it is at mid-APL.

Each of the curves below was measured after calibrating the display to a flat 2.3 gamma using the pattern type shown in the legend. So if you look at the red curve (15% windowed) it tells you that even though you think you've got a nice flat 2.3 gamma you don't. Your actual display performance will vary from a gamma of 2.22 at APL=5% up to 2.33 at APL=40%. We cannot change the shape of this curve, that's how the display drive works, all we can do is move it up and down. So we can argue where that ought to be but the effect is real and measurable.


ScreenShot2012-06-11at92853PM.png
Quote:
Originally Posted by Chronoptimist View Post

While they average out at 20 APL, the range is far greater, and most scenes were either below 15% APL (often in the 5–10% range) or above 30% APL. There were very few in the 20–30 range.
I have yet to see any compelling argument made for going against the display's native response—how it was designed to operate—and intentionally calibrating it in a power-limited state. While it's obviously a very limited selection, this seems to be the case with most content I watch, and if you play games at all, APL is considerably higher than 25%.

Thanks for that data, it's very interesting but limited sample. I agree that more data here would be nice to define a typical APL. Maybe someone has done this in more detail already but you would need a histogram of a variety of different films to do a good job. If it has a peak then that is where you want to calibrate gamma if your display has APL dependent gamma. I am not suggesting we calibrate in a "power limited" state, I'm suggesting we calibrate in an operating mode that avoids both ABL (like we do now) and low APL luminance manipulation. The latter effect is not an engineering limitation of the display (as Tom has shown, pioneers don't do it) but rather a choice by the manufacturer to achieve some unknown picture advantage but only succeeds in skewing our gamma measurements.[edit: subsequent measurements by Tom show that his pioneer does exhibit the behavior at certain stimulus levels]

Regarding games you are correct, but I would gladly sacrifice some accuracy there in favor of a film-based APL. By using small windows you have already made a choice to calibrate the display under very low APL conditions, why? This is not the typical operating APL of the device. It works fine for displays that don't muck with low level APL but for others it will give you a gamma reading that only applies at that very low APL.
Edited by zoyd - 6/20/12 at 3:17pm
post #62 of 135
Quote:
Originally Posted by zoyd View Post

ScreenShot2012-06-11at92853PM.png
I think I understand what you're trying to represent here, but could you clarify this graph for me please? I just want to be sure that I'm understanding it correctly.

Firstly, there's no reference line. You claim that the panel was calibrated to exactly 2.3 gamma, but it's virtually impossible to have a completely flat 2.3 across the range regardless of the display/patterns used. (most displays don't offer fine enough controls for a flat 2.30) Can you include that, and state what kind of patterns were used when arriving at that 2.3 result? Or is that what the "15% windowed" line is?

If the "15% windowed" line is the reference line, then it seems to be showing a far more stable response than the higher APL patterns. I would suggest not using any kind of spline curve/smoothing options when posting this kind of graph though, as the interpolation can muddy the results.

Am I right in understanding that the "40% APL Patterns" and "27% APL Patterns" data means you have a surround of 40/27% APL, respectively, or does it mean that the overall pattern averages to 40/27% APL, using a different surround for each stimulus level?

But then I have no idea what is meant by "10pt off." That would suggest to me that you have disabled 10pt gamma correction, but then there's no indication of what kind of patterns were used with these measurements.

Your X-axis is labelled APL (stimulus) but the only way I can read it so the graph makes sense, is that you actually meant %stim. (as in 10% grey pattern, 20% grey pattern, 30% etc.) but then I am confused as to why you only have measurements up to 70%stim, as you need a 100% measurement to calculate gamma.

I wish I had a PDP at my disposal right now, as I have a few ideas going around in my head of things that I would be interested to see measured. For example, is it better to keep pattern size constant, and alter the surround to keep the average picture level the same, or are you better to keep APL constant by varying the size of the patterns used. (so a 10% pattern would be considerably larger than a 100% one)

I've just put together some patterns to test this: http://www.filedropper.com/10apl
While they're at a lower APL than you would prefer, there's no way to do this comparison from 10% through to 100% stimulus at anything higher than 10% APL.

Ideally, the two sets of patterns should measure exactly the same, but I suspect you might find the "surround" patterns measure a bit lighter due to crosstalk. (part of the reason I generally avoid these "APL" style patterns)

And I still feel that when you are doing any kind of discussion/measurement of this kind of data, it's key to at least have a reference of what the display's overall ABL performance is like. (these patterns) You have 40% APL patterns in your graph, but I have no idea how power-limited your display is when operating at 40% APL.
post #63 of 135
Quote:
Originally Posted by Chronoptimist View Post

I think I understand what you're trying to represent here, but could you clarify this graph for me please? I just want to be sure that I'm understanding it correctly.

I forget that seeing this plot for the first time can be confusing (it was for me) due to the way we typically plot gamma vs. stimulus. I think the easiest way to explain it is to go through the measurements used to generate just one of those curves. So I will use the one labeled "15% windowed".

Gamma vs. APL measurement sequence
1. Calibrate the display using 15% area windows to obtain a flat 2.3 gamma from 10% to 90% stimulus. Achieving this with the 10pt. controls on my display is easy. (I can post a plot if you really want or take my word for it)
2. With the settings obtained from 1. measure the response of the display using patterns with 1% central windows at 8 different stimulus levels (5%, 10%,15%, 20%, 30%, 40%, 50%, 100%) of the central window and 9 different stimulus levels of the surround (5%, 10%,15%,20%,30%,40%,%50,%60,%70) <-These values are what you see on the x-axis.
3. Step 2 generates the plot below where gamma is the log/log ratio of the normalized luminance (to white) and the normalized central window stimulus.

ScreenShot2012-03-15at123704PM.png

4. Average the gammas obtained from each of the 7 central window levels relative to the eighth to form the averages that you see plotted in the previous post and the dotted line above at each APL.

I repeated those steps 3 more times replacing step 1 with the AVSHD large APL patterns (labeled "40% APL"), my own patterns using a 1% window 27% surround (labeled 27% APL), and skipping step 1 altogether with the 10 pt. control disabled.

So each of those curves represents 4 different possible settings of the 10 pt. control (including the off position) and each generates a different gamma vs. APL response even though 3 of them would show a flat 2.3 on a calibration report.
Edited by zoyd - 6/18/12 at 8:06pm
post #64 of 135
Here's some data on the APL of several films I sampled some time ago.

284

275

292

291

320

299

297

279

292

285

312

305

303

303
Edited by TomHuffman - 6/18/12 at 6:01pm
post #65 of 135
Unfortunately I just lost my post there (hit preview and nothing came up) so this is going to be briefer than my original reply.

It sounds like what you're doing here is simply using regular test patterns with a bright surround (rather than patterns which actually average 27/40 %stim at all levels) and measuring the level of contamination into the measurement area. This is a contrast related issue, not gamma or some sort of dynamic contrast processing. You will find it on all sorts of displays, there are a number of things that can cause this (including the meter you're using and how it's set up) and the EBU even has specific test patterns and details on what this means for broadcast monitors in various technical papers.

That you have managed to find a pattern size and surround brightness where the effects are constant relative to the levels you are trying to measure, would seem to be a happy coincidence. I dread to think what the results would be like if you tried to measure a CRT or any kind of projection-based display with those sort of patterns.

The best results for calibration are going to be obtained when the display is operating under optimal conditions; that is to say, the pattern size is small enough to avoid any power-limiting issues, but large enough for your meter to read accurately, especially if it's a non-contact meter—in which case you should be using a stray light elimination tube. Simply pulling the meter back from the display is not sufficient, especially if you have patterns with a non-zero surround.


It's certainly an interesting discussion, and if you still feel the need to calibrate your display in this manner, so be it, but I would not recommend it.
I do think it would be interesting, however, to see what sort of results you get from measuring the display with the two different sets of 10% APL patterns I posted though, possibly referencing them against standard 10% area patterns.


EDIT: Thanks Tom, that's interesting data. Can I ask how it was measured?
It seems to show what I have observed, which is that the average APL over an entire film might be in the 15–25% range, but the majority of scenes are 5–10% above or below that.
post #66 of 135
22.4% +/- 4% 1 sigma. This is the APL you would choose if you wanted to minimize the errors in display gamma while watching those films given a display with this type of APL dependent gamma. You'll also notice that all of the films are asymmetrically skewed to the high side of their averages (except Casablanca which is relatively high already). That's a good thing for this display because it's gamma is also more stable on the high side as opposed to the low side of that average.
Edited by zoyd - 6/18/12 at 7:46pm
post #67 of 135
Quote:
Originally Posted by Chronoptimist View Post

It sounds like what you're doing here is simply using regular test patterns with a bright surround (rather than patterns which actually average 27/40 %stim at all levels) and measuring the level of contamination into the measurement area.

I don't know what you are talking about, there is no cross-talk or contamination in these measurements. The measurements were obtained on a plasma using a contact meter. That might be a problem for LCD but not here. Also, on this PDP gamma varies with APL independent of how the APL is generated, you can use multiple levels to generate the average or a single level, it doesn't matter. The measurements just recently posted by Tom and I used multiple levels (starfield patterns), they show identical behavior to the single level APL patterns in the measurement sequence I described. If you were to view the dynamic brightness pattern from the AVSHD disk on any samsung or panasonic from 2011, you would immediately see the effect I'm talking about. You are trying to somehow negate the data so you can conclude that it's not telling you anything important, I can assure you there is nothing wrong with the methods or patterns used to obtain the data, or how I've interpreted it. If you don't understand what's been presented or if this type of behavior is not something you've seen before that's fine, but don't throw out WAGs as to why you think it's wrong. I obviously am not talking about CRTs or projectors - this is a discussion of one type of measured behavior on a PDP using appropriate patterns and measurement techniques for that display.
Quote:
Originally Posted by Chronoptimist View Post

he EBU even has specific test patterns and details on what this means for broadcast monitors in various technical papers.

The EBU test patterns are designed to quantify display error in order to classify the display against a standard. They are not designed to optimize calibration for any particular non-linearity in display performance, which is what I'm doing here.
Quote:
Originally Posted by Chronoptimist View Post

The best results for calibration are going to be obtained when the display is operating under optimal conditions; that is to say, the pattern size is small enough to avoid any power-limiting issues

This is where I disagree. The display is never used in it's optimal condition so why would I want to calibrate it there?

I have shown that with the nature of how APL dependent gamma works on this display, the smallest pattern set (lowest APL) will generate the least desirable results. The difference between 15% area windows and 27% APL is probably not large enough to worry about but I would not recommend anything smaller than that. As stated above the theoretical optimal gamma calibration given the data that I have presented and using Tom's film data set as the target operating condition will be obtained with APL patterns anywhere within the range 18.4-26.4% (1-sigma).
Edited by zoyd - 6/19/12 at 7:02pm
post #68 of 135
I have some interesting data.


  1. I created my own simple constant APL test patterns that are just standard 11% windows with a varying surround that keeps the APL at about 25 through the entire range.
  2. I then created a standard set of window patterns with a constant black surround.
  3. Then I measured both on my Pioneer plasma. The results show that there is virtually no difference between the two at 10% and at 50% and above. However between 20%-40% there are significant differences. What's more, the differences are not just in gamma, but in chromaticity as well.

So now that we have established that they measure differently, the question is which is more "correct." I think that the differences in chromaticity offer a way to determine this. Since it is at the low end of the range, I can also measure full field test patterns and use those as a reference. Here's the data.

322

What this clearly shows is that the constant APL patterns are considerably closer in chromaticity to the full field test patterns between 20-40%. I am now a believer.

Comments?

The constant APL test patterns can be downloaded at
http://www.chromapure.com/constant.zip
post #69 of 135
That's a nice set of measurements Tom, I hadn't thought of looking at the chromaticity. I have a couple of comments:

1. There appears to be some sort of transition region between 20-40% APL that might be endemic to plasma. As they all share a 10 multiple sub-field pixel addressing scheme, that is where I would look first.
2. I don't think we should use full field APL as a reference since this is not a typical state of the device. Multi-level surrounds would be a better choice.

I will check for chromaticity shifts on my display. I can't emphasize enough that when calibrating a non-linear system you want to emulate as best you can normal operating conditions "in the field" to achieve a calibration which minimizes errors where it's most important to do so. This applies to CMS calibrations as well.
Edited by zoyd - 6/19/12 at 8:32am
post #70 of 135
Quote:
Originally Posted by zoyd View Post

1. There appears to be some sort of transition region between 20-40% APL that might be endemic to plasma. As they all share the 10 sub-field pixel addressing scheme, that is where I would look first.
Actually they don't. Pioneer PDPs use 14 sub fields.

Its going to be interesting to see what you choose as your final conclusion when it comes to PDPs and what type of patterns should be used for calibration. I played around with what you have been doing about two years ago. Very curious to see if your decisions match mine.

I will give you a bit of advice....

Your logic is one dimensional. All PDPs don't act the same way. That includes different sizes from the same manufacturer and model year.
post #71 of 135
Quote:
Your logic is one dimensional. All PDPs don't act the same way.

No, I've been extremely careful to reference any conclusions to the data at hand and not extrapolate to all plasmas. Also, I'm not concerned as much with how the effect may or may not manifest in other panel brands or sizes. I am not trying to "fix it". I'm taking an approach which says that given an imperfect system what patterns best simulate actual usage so that one can distribute any errors in a more beneficial way relative to assuming the display is linear. As a calibration philosophy it does not matter where the non-linearity exists, it doesn't even need to exist because a linear system will measure the same using any surround. In other words you will do no harm by using a scene referenced APL pattern set on a linear display system. The measurements Tom and I (and you) have done can quantify the benefits on a particular display of this approach and hopefully they are positive, of a magnitude that is perceptually significant, and exist for enough cases to warrant their usage over standard windows for general usage.
post #72 of 135
Quote:
Originally Posted by zoyd View Post

No, I've been extremely careful to reference any conclusions to the data at hand and not extrapolate to all plasmas. Also, I'm not concerned as much with how the effect may or may not manifest in other panel brands or sizes. I am not trying to "fix it". I'm taking an approach which says that given an imperfect system what patterns best simulate actual usage so that one can distribute any errors in a more beneficial way relative to assuming the display is linear. As a calibration philosophy it does not matter where the non-linearity exists, it doesn't even need to exist because a linear system will measure the same using any surround. In other words you will do no harm by using a scene referenced APL pattern set on a linear display system. The measurements Tom and I (and you) have done can quantify the benefits on a particular display of this approach and hopefully they are positive, of a magnitude that is perceptually significant, and exist for enough cases to warrant their usage over standard windows for general usage.
As I've said, I am curious to see what conclusion you arrive at based on an imperfect system.
post #73 of 135
Why do not you answer?
Or pleasure as others suffer trying to find an answer - so you feel god?

On your laurels (completely deserved), and no one encroaches (short arm), but also suggest how to calibrate a home for yourself can be a plasma?

one question - area pattern and % surround (APL).

THX recommends for PDP 12% area.

and what should be % surround (APL)?

Zoyd

I think it is necessary to adjust CMS the start of the capabilities of hardware pattern generators QuantumData 882E.
there is no possibility of any inference % surround (APL)?
Edited by anta1974 - 6/19/12 at 8:30am
post #74 of 135
Quote:
I think it is necessary to adjust CMS the start of the capabilities of hardware pattern generators QuantumData 882E.
there is no possibility of any inference % surround (APL)?

I had originally thought that APL was not a significant driver for either color or white point chromaticity which is why I was concentrating on gamma, but Tom's data indicates otherwise. I will revisit that issue on my display with my next set of measurements.
Edited by zoyd - 6/19/12 at 9:12am
post #75 of 135
Quote:
Originally Posted by anta1974 View Post

Why do not you answer?
Or pleasure as others suffer trying to find an answer - so you feel god?
On your laurels (completely deserved), and no one encroaches (short arm), but also suggest how to calibrate a home for yourself can be a plasma?
I will take it that this was directed towards me as your post. Show me a question directed towards me in this thread before you claim I "don't answer".
Quote:
Or pleasure as others suffer trying to find an answer - so you feel god?
Feel god? Huh? I don't understand this part of your post. My curiosity has nothing to do with you. I'm curious as to the conclusion zoyd is going to arrive. He is very intelligent and I am almost positive he will come to the same conclusion I have.
post #76 of 135
Quote:
Originally Posted by zoyd View Post

Zoyd
I had originally thought that APL was not a significant driver for either color or white point chromaticity which is why I was concentrating on gamma, but Tom's data indicates otherwise. I will revisit that issue on my display with my next set of measurements.
Chronoptimist's posts in this thread do have merit.... hint wink.gif
post #77 of 135
Why not give?
so Zoyd wasted no time?


one question - area pattern and % surround (APL)
post #78 of 135
Why do not you want to say right?

because of this, and I think that because you feel God - immeasurably superior to all other.

with its separate board of Pioneer 101FD ( further reduces the black level.) - the exact MLL is unknown, because no one sensor does not see as deeply.
But it is guaranteed lower than the threshold - 0.0002fL (0.0007 cd/m2)
post #79 of 135
Quote:
Originally Posted by anta1974 View Post

Why not give?
Why not give what?
Quote:
so Zoyd wasted no time?
From my observations/interactions, zoyd likes puzzles and problems (same as me). You don't just give people answers who love puzzles and problems. If zoyd wants to ask me a question, he can easily PM me.... as he has done in the past.
Quote:
one question - area pattern and % surround (APL)
I have no desire to answer any of your questions. Why? Take a look at your posts you directed towards me. you attempt insult me, yet have the audacity to ask me to help you?Sit and wait for that to happen.
post #80 of 135
Quote:
Originally Posted by zoyd View Post

I don't think we should use full field APL as a reference since this is not a typical state of the device. Multi-level surrounds would be a better choice.
The only reason I looked at full fields at 20-40% was that I had a discrepancy between the standard and constant APL chromatacities. The only way to know which is correct is to compare to some reference. Since these different window/surround designs is precisely the issue in question, I can't think of what else to use. If you use several levels for the surround at 20% and get different results, again, which is the correct one?

Here's the data for a 20% window with varying surrounds. I don't really know what to make of it. The biggest change is in the y-axis, but I have no way of knowing if this difference is real or an artifact of the test pattern. I am inclined to trust the full field results.

136
post #81 of 135
Quote:
Originally Posted by D-Nice View Post

I have no desire to answer any of your questions. Why? Take a look at your posts you directed towards me. you attempt insult me, yet have the audacity to ask me to help you?Sit and wait for that to happen.
You do not Alexander the Great (Macedonian).
and I'm not a soldier your army.
and this is not 323 BC.
and your answer confirmed my opinion about you.
Truth is always impartial.
Proud of their knowledge to keep to yourself so why write in the forum where people are trying to solve the problem of self-calibration.
Afraid to lose the title of best in the world - very strange?
Or the least afraid of competition?
Because I live in the Evil Empire - USSR - there generally is no certified calibrator.
If I am asked on the forum (not that) and I know the answer then the answer is not the writing that has solved this problem 2 years ago
Edited by anta1974 - 6/19/12 at 11:48am
post #82 of 135
Quote:
Originally Posted by anta1974 View Post

You do not Alexander the Great (Macedonian).
and I'm not a soldier your army.
and this is not 323 BC.
and your answer confirmed my opinion about you.
Truth is always impartial.
Proud of their knowledge to keep to yourself so why write in the forum where people are trying to solve the problem of self-calibration.
Afraid to lose the title of best in the world - very strange?
Or the least afraid of competition?
Because I live in the Evil Empire - USSR - there generally is no certified calibrator.
If I am asked on the forum (not that) and I know the answer then the answer is not the writing that has solved this problem 2 years ago
.... Uh, Ok. Your location is irrelevant to me. As is the color of your skin, creed, sexal preference, etc. None of that matters. specifically stated I will not answer any of your questions based on the posts you directed towards me. I'm not fluent in Russian culture but here in the States, don't expect assistance from someone you attempted to insult.

BTW, there is nothing to solve. There is only a conclusion that can be derived.
post #83 of 135
Quote:
Originally Posted by TomHuffman View Post

I created my own simple constant APL test patterns that are just standard 11% windows with a varying surround that keeps the APL at about 25 through the entire range.
For some reason, I'm just getting a black image with your pattern disc.

If you're wanting to create patterns of any size/APL combination (rather than being roughly at the target size/APL) here's an excel spreadsheet I threw together to simplify the process: http://www.filedropper.com/patterncalculator
Even if you don't want a constant APL, it's useful for calculating the exact pattern size for any given area. (e.g. 12% would be 665×374)

A few things to note:
  1. The calculator assumes that the patterns are going to be 1080p.
  2. The RGB values for the surround should be more accurate than the percentages.
  3. I'd recommend you actually average the patterns out just to be sure.
    Due to some rounding involved, sometimes you'll find that one or two of the patterns will average slightly higher or lower than your target APL (±1 RGB) which just means you need to edit the surround slightly.

As mentioned previously, I also put together some patterns that keep APL constant by changing the pattern size on a black background, rather than brightening the surround. To me, it seems like this should actually produce better results, but the downside is that the maximum APL you can use is limited by your lowest measurement. So if you're doing 10–100%stim, the highest APL you can use is 10%. If you wanted 5–100%stim, the maximum APL you could use would be 5% for example, so it makes anything over 10% APL impractical if you're wanting a constant APL throughout the whole range.
post #84 of 135
Quote:
Originally Posted by Chronoptimist View Post

For some reason, I'm just getting a black image with your pattern disc.
I just tested it. Downloaded, unzipped, and burned with Nero burning ROM. It worked fine. It is NTSC. Are you using PAL?

Here's a PAL version.

http://www.chromapure.com/constantAPL_PAL.zip
post #85 of 135
Quote:
Originally Posted by TomHuffman View Post

The only reason I looked at full fields at 20-40% was that I had a discrepancy between the standard and constant APL chromatacities. The only way to know which is correct is to compare to some reference. Since these different window/surround designs is precisely the issue in question, I can't think of what else to use. If you use several levels for the surround at 20% and get different results, again, which is the correct one?
Here's the data for a 20% window with varying surrounds. I don't really know what to make of it. The biggest change is in the y-axis, but I have no way of knowing if this difference is real or an artifact of the test pattern. I am inclined to trust the full field results.
136

Confirmed that chromaticity shift is occurring on the samsung as well and is also primarily along the y-axis. 25% central window, random surround.


166
post #86 of 135
Quote:
Originally Posted by D-Nice View Post

From my observations/interactions, zoyd likes puzzles and problems

That is an accurate observation.smile.gif

I've burned a bunch of new patterns and will be measuring them over the next couple of days, probably won't have any further data to report until the weekend.
Edited by zoyd - 6/21/12 at 4:03am
post #87 of 135
Quote:
Originally Posted by Chronoptimist View Post

I've just put together some patterns to test this: http://www.filedropper.com/10apl

Thanks, I will add the variable area ones to the hopper, that's a neat idea.
post #88 of 135
Quote:
Originally Posted by zoyd View Post

Thanks, I will add the variable area ones to the hopper, that's a neat idea.
I'm very interested to see what results you get from comparing the two sets of patterns.

Ideally they should be the same, but I can think of at least a couple of reasons why they might not be.
post #89 of 135
Quote:
Originally Posted by Chronoptimist View Post

I'm very interested to see what results you get from comparing the two sets of patterns.
Ideally they should be the same, but I can think of at least a couple of reasons why they might not be.


The 40.jpg pattern from the APL Size is ~47.5% instead of 40%. All the other ones check out.
post #90 of 135
Quote:
Originally Posted by zoyd View Post

The 40.jpg pattern from the APL Size is ~47.5% instead of 40%. All the other ones check out.
You know, when creating patterns, I've run into a number of weird issues with Photoshop doing stuff like that, and it's not been limited to this version, it happened back when I was running CS2 on a Mac, and now CS5 on Windows. It's little things like if you try to fill an area with 40% brightness—which it says is 102,102,102 RGB, it will actually fill the area with 103,103,103. I have to fill it with 101,102,102 and then 102,102,102 for it to fill the correct values. If I had to guess, it's because I set the values as 40% rather than 102,102,102, even though it says 40% is 102,102,102.

I tried recreating that pattern again, and even though the patch was the correct value, it saved as 121,121,121 RGB again. (~47.5%) It's not a colour profile/management issue, I really don't know what the cause is. Normally I double-check this, but hadn't run into the issue when saving files from CS5 yet (only when creating them) so I assumed everything was OK. It's really quite concerning that issues like that still persist to this day.

Here's the fixed pattern.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › Better to have 2.2 gamma or stable 2.3 gamma on a plasma?