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 2

post #31 of 135
FWIW, SMPTE RP-166 "Critical Viewing Conditions for Evaluation of Color Television Pictures" goes into the issue of proper measuring environment in great detail.
post #32 of 135
I couldn't find the text in the public domain, since you need to purchase the document I didn't feel comfortable quoting due to copyright.. it used to be $20, think it's more now.. there was also a Widescreen special edition several years ago with it printed...

I thought GeorgeAB posted that on here or another Forum but I can't find it (and not sure what's up with the search here after conversion).
post #33 of 135
Quote:
Originally Posted by turbe View Post

I couldn't find the text in the public domain, since you need to purchase the document I didn't feel comfortable quoting due to copyright.. it used to be $20, think it's more now.. there was also a Widescreen special edition several years ago with it printed...
I thought GeorgeAB posted that on here or another Forum but I can't find it (and not sure what's up with the search here after conversion).
Here's the link to order the long out of print, nearly 400 page, PDF download of the 'Imaging Science Theatre 2000' special edition from 'Widescreen Review:' http://shop.widescreenreview.com/products/Imaging-Science-Theatre-2000-%28Digital-Download%29.html . It costs $5.00 and includes the complete SMPTE RP 166-1995 document titled: 'Critical Viewing Conditions For Evaluation Of Color Television Pictures.' It also includes the companion SMPTE RP-167, 'Alignment Of NTSC Color Picture Monitors.'

The entire issue is a fascinating read for anyone wanting to understand the historical turning point in consumer television that resulted in the display calibration community of today. Joe Kane's work with SMPTE is described, as well as the formation of the Imaging Science Foundation. Other early luminaries are interviewed and their writings featured in a relatively simple format that belongs in anyone's library who is serious about understanding the fundamentals of video picture quality.

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

"Advancing the art and science of electronic imaging"
post #34 of 135
Quote:
Originally Posted by TomHuffman View Post

Window patterns have been around as long as there has been display measurements. CRTs, which used to be the only studio mastering standard, have the same voltage limitations as plasmas. This is not a design deficiency, so much as a consequence of the display technology itself.

My point was that there is no technical reason that I know of that makes windowed patterns any better than any other pattern. People use them because they've always used them under the assumption that other than for ABL limited technologies, it doesn't matter. Well, it does matter in some cases. Is there a pattern that truly let's us assume "it doesn't matter"? Probably not but my contention is that this assumption will be valid more often when using a fixed APL pattern, that's all.
Quote:
The presence or absence of a CMS is not the primary issue here. Displays are not perfect. Even without a CMS they may have non-linear color errors, just like many have non-linear grayscale errors. Just like grayscale measurements sample multiple instances of gray throughout the color space, measuring multiple instances of a color or colors throughout the color space is similarly advisable. This has been the standard in professional environments that use very sophisticated LUTs for, well, forever. The only reason that this is only recently become a focus in the consumer arena is that issues of cost and a lack of suitable (and affordable) hardware and software have prevented it.
This dynamic black phenomenon is a design deficiency in a completely different sense. It is an attempt on the part of some manufacturers to goose up the measured contrast because people on this forum and elsewhere have shown such a single-minded fanaticism about achieving low black levels--almost to the exclusion of all other concerns--that the manufacturers have pulled out all the stops to try and satisfy demand. In doing so, some have apparently implemented a completely useless "feature" that makes the image worse, instead of better, and has caused this sudden concern to come up with an entirely new set of test patterns.

I agree it was a bad choice of words. My point here was the metrology solution in both cases is similar. We have a limited number of control points and it is better to locate those control points within boundaries that are most beneficial to actual perceived content.
Quote:
Gee, isn't this just a little over the top? I had never even heard of this issue until a couple of years ago, and it was only relevant with a few models. Now we need to assume it as a default for all displays, regardless of whether they have this problem or not? I am perfectly willing to consider the idea of a 25% surround, if it doesn't cause other problems, such as the ones I describe above, but only because it probably does better simulate actual program material. I do not support the use of tiny 1% windows. It makes using non-contact devices impossible. Also, I still await instructions on how we can reliably and objectively test for the existence of this phenomenon, other than seeing zero output at 0% stim.

See 1st response above, the whole point is you don't have to assume anything, it makes your life easier.smile.gif Think of APL patterns as a superset of windowed, they will work as well as windowed for technologies in which "it doesn't matter" and also for the cases that have been identified as problematic. And I do think it's a real issue for plasma calibration, all the 2011 samsungs and pannys have this problem, I would guess LG as well although I've not measured one. ChadB says the 2012 pannys have it but it's weaker than last year and I'm sure samsung has kept it this year.
Quote:
BTW, on a related note, my suggestion earlier to just use full field test patterns below 40% is perhaps not a good idea after all. These actually measure differently than window patterns, even on displays with no dynamic black "feature" implemented.

Interesting, how do you typically check for the effect?
Edited by zoyd - 6/10/12 at 5:17am
post #35 of 135
This has been said else where, but the 'majority' of professional grading environments used within the post-production environment use 2.2 gamma as the standard for 'tv' grading work, and as the target for film work re-mastered for tv. after a film grade (using print film LUT emulation, or a DCI P3 environment).

This is not definitive, as other gamma values do get used, but it is the most common, but 2.2 gives a better level of shadow detail than higher gamma values.

We know this as LightSpace CMS is used by many of the top post-production companied for their internal calibration, and we perform a lot of calibration throughout the world for many other post facilities.

This may change if the ITU-R BT.1886 standard gets adopted - but it hasn't yet.

Further, plasmas are impossible to calibrate accurately, due to the ABL function that is active within all plasmas.

You can see this in post 49 on this thread: http://www.avsforum.com/t/1412947/plasma-owners-how-do-you-calibrate-saturation/30

Using RGB Separation as a way to check a display for its ability to be calibrated accurately is a very good tool. The closest you will ever get is with a free-form 3D LUT, but even that will not be accurate due to this RGB Separation issue caused by the ABL function.

We recommend to all our clients to not use plasmas if accurate calibration is required...
post #36 of 135
Quote:
Originally Posted by plasma_fan View Post

From the sounds of it the truly correct fashion is to use the small 1% area windows set the gamma to 2.22 and not worry about any ill effects caused by the abl in real content?

Just in case it's not completely clear by the replies in this thread, the fact of the matter is that there is no agreed on method to accurately calibrate gamma and grayscale across various displays. One complication in trying to come up with a very simple method of adjusting gamma and grayscale is simply that different displays don't necessarily perform similarly. If you were to assemble various displays (CRT, multiple Plasmas, fixed backlit LCDs, LCDs with variable backlighting, etc.) and critically view them side by side there would be some variations in how they display identical inputs. You could select one single image and calibrate to get all of the displays to show that image similarly, but once you started to compare brighter or darker images the displays would tend to diverge to some extent, simply because various displays are not built to operate similarly as brightness changes. Regardless if you choose to "calibrate" to window patterns of a selected size or use patterns with a fixed Average Picture Level, bottom line it's probably a crapshoot if you are coming nearer to or moving farther from the originally intended picture, unless you happen to know how your display performs relative to the production monitor.
Edited by alluringreality - 6/11/12 at 5:56am
post #37 of 135
Quote:
Originally Posted by Light Illusion View Post

Further, plasmas are impossible to calibrate accurately, due to the ABL function that is active within all plasmas.

I gave you a counter-example to this statement in the saturation thread using the vendor provided internal controls which you chose not to respond to. I agree you might not want to use a plasma in post-production due to the ABL capped output but they are perfectly capable of reference color reproduction without the use of an outboard 3D-LUT.



edit: removed RGB tracking plot as it is not relevant to what we were discussing. The counter-example still applies.
Edited by zoyd - 6/10/12 at 9:09am
post #38 of 135
I don't think you will find that is RGB Separation...

That looks to just be 'tracking'. Not the same I think you will find.

(Oh, and I didn't ignore your other post - I just don't follow threads very often - just don't have the time!)

biggrin.gif
post #39 of 135
Right, this is tracking, let's take this back to the other thread as it is more appropriate in the context of saturation.
post #40 of 135
Quote:
Originally Posted by zoyd View Post

Interesting, how do you typically check for the effect?
I measured fields and windows at several levels of stim below the voltage limiting point. The plasma always measured differently. I tried this on two plasmas (Pioneer 5020 and a Panny ST30). The level of stim had no effect on this, which I would not expect if I were looking for some dynamic brightness circuit. However, when I repeated the same test on a couple of Samsung LCDs, windows and fields measured almost exactly the same. Again, level of stim made no difference. How a plasma measures is affected a variable other than a dynamic brightness feature or voltage limitations. There is something else going on.
post #41 of 135
The effect that I'm most familiar with can quickly be checked for by using the dynamic brightness pattern on the AVSHD disk. If the lower levels in the gray step patterns in the bottom right change with background level, it's active. Put a probe on the 10% bar if you want to measure the variation.

I thought you had said earlier that you had a plasma that did not show the effect above AND showed differences between full field and windowed luminances? If that is the case then depending on the magnitude it could be that the ABL circuit is active at lower levels than we have been assuming.
post #42 of 135
Quote:
Originally Posted by zoyd View Post

The effect that I'm most familiar with can quickly be checked for by using the dynamic brightness pattern on the AVSHD disk. If the lower levels in the gray step patterns in the bottom right change with background level, it's active. Put a probe on the 10% bar if you want to measure the variation.
I thought you had said earlier that you had a plasma that did not show the effect above AND showed differences between full field and windowed luminances? If that is the case then depending on the magnitude it could be that the ABL circuit is active at lower levels than we have been assuming.
I have not checked using the AVS disc. However, I think I'll create my own test pattern that should make this much easier to test for. A standard 10% window, with different levels of surround, say black up to 50%. If there is a dynamic black feature then the window pattern will measure lower with the lower surround and higher with the higher surround. I have already shown that on LCDs that windows and fields measure the same. Thus, the level of surround shouldn't matter so long as you are below the voltage limiting point.
post #43 of 135
Quote:
Originally Posted by TomHuffman View Post

I have not checked using the AVS disc. However, I think I'll create my own test pattern that should make this much easier to test for. A standard 10% window, with different levels of surround, say black up to 50%. If there is a dynamic black feature then the window pattern will measure lower with the lower surround and higher with the higher surround.

That type of pattern works very well, I have already done it. I've actually measured Gamma vs. APL at constant center stimulus this way. On the Samsungs the lower stimulus levels (excluding 5%) get darker (gamma rises) as APL rises until about 20% and then flatten out, this plot is what drove me to use the 25% surround.

The only reasoning I could dig up to implement something like this was a description in one of the patent documents that went something like "increase dark detail at low APL for bright environments" which is basically what's happening below.


ScreenShot2012-03-15at123704PM.png


Here is a summary of all the APL measurements I made. Each curve was measured after the display was calibrated to a flat 2.3 gamma with the pattern variety listed in the legend. Here you can clearly see where ABL kicks in at 50%. Prior to that there is a fairly steep slope to 20% and then a relatively stable region with a slow rise until the ABL circuit takes over. Just to clarify the x-axis is APL of the surround, it's not what you normally see plotted as % stim.


ScreenShot2012-06-11at92853PM.png
Edited by zoyd - 6/18/12 at 11:11am
post #44 of 135
I tested my Pioneer Kuro 5020 using a fixed 10% stimulus signal. I used a standard 11% window with different surrounds, 0%, 5%, 15%, 20%, 30%, 40%, and 50%.

I measured a very small effect. Using a 0% surround the gamma was 2.31 and using a 50% surround the gamma was 2.28. This is almost certainly below the threshold of visibility. I only tested at 10% stimulus because this is where the effect, if there was one, would presumably be the strongest.

316

I have a couple of other plasmas here--including a VT30--that I'll test in the next couple of days, but it seems pretty clear that no such effect plagues the Pioneer plasmas.
post #45 of 135
Thanks for the info. The effect is strongest above 10%, the 20-30% patterns had the steepest slopes on the samsung (see plot 1 above). 10% was weaker, it varied from 2.28 to 2.31 at (5-20% surround) and then dropped back to 2.29 at 50%. 30% went from 2.17 to 2.36

I've only measured this one and the GT30 which showed similar behavior. Other users have reported seeing it in the dynamic brightness patterns on the D7000 and D550.
Edited by zoyd - 6/12/12 at 3:40am
post #46 of 135
Quote:
Originally Posted by turbe View Post

Ambient room lighting should match the color temperature of your monitor (6500K in North and South America and Europe, and 9300K in Asia).

This is bull. Contrary to the often repeated but never backed up claim that the strange people is Asia use D.94 they in fact use D.65 just like the rest of us ( I also love the way that in the article they talk in terms of color temperature).

Various theories I've heard for this nonsense are;

Its to make yellow skin tones more natural.

Its to better balance the redder colour of the sun in these areas of the world.

Its to better match the color of sunlight ; look at the sun man its obviously D93! (said to my face)

These asians all use Fuji stock which is greener than Kodak so they need a higher white ref.

All displays in Asia are based on office workstations because they work such long hours so broadcast TV is balanced the same way to avoid eye strain at home.



Go off and find any real evidence of anyone in Asia using anything other than D65 for broadcast. You won't be able to because its a myth.
I also take the supposed gamma difference with a pinch of salt.
post #47 of 135
Totally agree with Mr.D - total bull!!!

for TV globally it is D65.
Today the whole work uses Rec709, for HD and SD.
This may change, but it's the way it is right now.
post #48 of 135
I don't personally have any interest in the Asia 93 statement, I just quoted the that entire section from apple in regards to documented environments.. would be best to quote the text that Tom mentions or you can purchase that special edition from GeorgeAB's site...

If anyone does have an interest, there is an old Thread here on AVS: http://www.avsforum.com/t/1120246/d93-9300k


.
post #49 of 135
Tom - I added a fixed 25% window with variable background to my mll level test disk. That's where the effect was strongest on the GT30 and my display.
post #50 of 135
Here's what I got on a Panasonic ST30

317

There is about a 20% drop from 2.5% to 16.2%, then it remains quite stable until 90% where the display's voltage limiting effect comes into play.
post #51 of 135
So you see why I like region above 20% on these displays, not only are the levels stable there, but it is more representative of average APL of viewed material. that initial drop in luminance is a change in gamma from 2.3 to 2.4 relative to a 120 cd/m^2 white. The APL of 10% windows is very small (1% - 10%) so if you calibrate with those, gamma for the mid-APL scenes will be roughly 0.1 higher than your target (2.3 instead of 2.2, 2.4 instead of 2.3, etc.)

Quote:
Originally Posted by TomHuffman View Post

then it remains quite stable until 90% where the display's voltage limiting effect comes into play.

The ABL in your plot looks like it's active beginning at 40.6% but you'd have to repeat to confirm. It's much less aggressive than the samsung. I think even in 2011 the panny phosphors are more efficient.

239
Edited by zoyd - 6/17/12 at 4:46am
post #52 of 135
Quote:
Originally Posted by zoyd View Post

So you see why I like region above 20% on these displays, not only are the levels stable there, but it is more representative of average APL of viewed material. that initial drop in luminance is a change in gamma from 2.3 to 2.4 relative to a 120 cd/m^2 white. The APL of 10% windows is very small (1% - 10%) so if you calibrate with those, gamma for the mid-APL scenes will be roughly 0.1 higher than your target (2.3 instead of 2.2, 2.4 instead of 2.3, etc.)
I don't think it is as simple as APL. When I tested using a 10% window (11% area) with a 50% surround--that's an APL of about 45--I measured virtually no effect at all. There's something else going on here.

BTW, so time ago I sampled several films looking for the typical APL. As you might imagine, It varied between films, 18-22% seemed fairly common.
post #53 of 135
Quote:
Originally Posted by TomHuffman View Post

I don't think it is as simple as APL. When I tested using a 10% window (11% area) with a 50% surround--that's an APL of about 45--I measured virtually no effect at all. There's something else going on here.
BTW, so time ago I sampled several films looking for the typical APL. As you might imagine, It varied between films, 18-22% seemed fairly common.

The only complexity is that each level is affected differently, this is shown in my plots above. The luminances are redistributed as a function of APL for whatever "enhancement" they are going after. When you average levels 5%-50% the net effect is if you calibrate for gamma=2.3 at low APL you'll get (on average) gamma=2.4 for that 18-22% range typical of films on all samsungs(2011,2012) and 2011 panny's (ST and GT). I don't know if the 2011 VT does it or any of the 2012 models.
post #54 of 135
Quote:
Originally Posted by zoyd View Post

So you see why I like region above 20% on these displays, not only are the levels stable there, but it is more representative of average APL of viewed material. that initial drop in luminance is a change in gamma from 2.3 to 2.4 relative to a 120 cd/m^2 white. The APL of 10% windows is very small (1% - 10%) so if you calibrate with those, gamma for the mid-APL scenes will be roughly 0.1 higher than your target (2.3 instead of 2.2, 2.4 instead of 2.3, etc.)
http://www.avsforum.com/image/id/704956/width/490/height/239
If you have proven that the ABL is in effect at 25% APL, why on earth would you then use a 25% APL pattern? Your reasoning seems completely backwards.
post #55 of 135
It's not ABL, it's dynamic contrast and the reason you would want to use fixed APL patterns if you are following the discussion is that the display response is stable between surrounds of 20-40%. At other APLs including those normally associated with windowed patterns (<10%) and high ABL (>50%) gamma is unstable on these displays. What we've shown is that just like the ABL effect at high APL, there is luminance manipulation occurring at low APL that causes unstable gamma.
Quote:
Originally Posted by zoyd View Post

So you see why I like region above 20% on these displays, not only are the levels stable there, but it is more representative of average APL of viewed material. that initial drop in luminance is a change in gamma from 2.3 to 2.4 relative to a 120 cd/m^2 white. The APL of 10% windows is very small (1% - 10%) so if you calibrate with those, gamma for the mid-APL scenes will be roughly 0.1 higher than your target (2.3 instead of 2.2, 2.4 instead of 2.3, etc.)

Think of it in terms of luminance. We want the display to have a constant output for a given stimulus level independent of scene content. On these displays that condition is only satisfied when the APL is in the range of approximately 20%-40%, luckily that also corresponds to more typical average video levels in content.
Edited by zoyd - 6/17/12 at 1:20pm
post #56 of 135
Quote:
Originally Posted by zoyd View Post

It's not ABL, it's dynamic contrast and the reason you would want to use fixed APL patterns if you are following the discussion is that the display response is stable between surrounds of 20-40%. At other APLs including those normally associated with windowed patterns (<10%) and high ABL (>50%) gamma is unstable on these displays. What we've shown is that just like the ABL effect at high APL, there is luminance manipulation occurring at low APL that causes unstable gamma.
Think of it in terms of luminance. We want the display to have a constant output for a given stimulus level independent of scene content. On these displays that condition is only satisfied when the APL is in the range of approximately 20%-40%, luckily that also corresponds to more typical average video levels in content.
Sorry, I misinterpreted what it is you were showing with that graph.

How do those results compare to measuring something like these patterns?
http://www.filedropper.com/abl-test
post #57 of 135
Quote:
Originally Posted by Chronoptimist View Post

Sorry, I misinterpreted what it is you were showing with that graph.
How do those results compare to measuring something like these patterns?
http://www.filedropper.com/abl-test

Those are display loading patterns to test ABL. The effect measured above does not occur for peak white so the loading patterns should show constant luminance until APL ~40% and then fall off.
post #58 of 135
Quote:
Originally Posted by zoyd View Post

Those are display loading patterns to test ABL. The effect measured above does not occur for peak white so the loading patterns should show constant luminance until APL ~40% and then fall off.
It seems like you should be referencing that data against the display's ABL effects, rather than simply claiming it's dynamic contrast in effect.

Even if there is some dynamic contrast adjustment going on, which I am not disputing, surely the ABL should also be taken into account if you are going as high as 90% average video level.
post #59 of 135
The 50-90% levels are there so you can see where ABL cuts-on and how severe it is , it doesn't change the conclusion that something is going on at the low APL end. Btw, the patterns I used were only made to measure the effect, (it sounds like you may be thinking these are for calibration, they are not). All I'm concerned about is defining a stable operating range where luminance does not vary with surround for any input level. That range is where you can build a pattern to calibrate for your target gamma and it will not change. Another way to think of it is to ask the questions, Why do I use 10-18% windows to calibrate gamma on a plasma?answer: To avoid ABL distorting the measurement. Why would I use windows in a constant APL background? answer: To avoid dynamic contrast distorting the measurement. As long as the constant APL luminance is also below the ABL cut-on you've solved both problems.


10-18% windows on a black background will give you a gamma calibration that very seldom represents your target value when viewing real video content on the displays measured above. As I mentioned earlier, the measurements indicate that the average "error" will be about 0.1, so a 2.3 gamma calibration will for most material really give you 2.4 response. This is not a huge error but I think it's large enough to pay attention to.
Edited by zoyd - 6/17/12 at 6:34pm
post #60 of 135
Quote:
Originally Posted by zoyd View Post

The 50-90% levels are there so you can see where ABL cuts-on and how severe it is , it doesn't change the conclusion that something is going on at the low APL end. Btw, the patterns I used were only made to measure the effect, (it sounds like you may be thinking these are for calibration, they are not). All I'm concerned about is defining a stable operating range where luminance does not vary with surround for any input level. That range is where you can build a pattern to calibrate for your target gamma and it will not change. Another way to think of it is to ask the questions, Why do I use 10-18% windows to calibrate gamma on a plasma?answer: To avoid ABL distorting the measurement. Why would I use windows in a constant APL background? answer: To avoid dynamic contrast distorting the measurement. As long as the constant APL luminance is also below the ABL cut-on you've solved both problems.
10-18% windows on a black background will give you a gamma calibration that very seldom represents your target value when viewing real video content on the displays measured above. As I mentioned earlier, the measurements indicate that the average "error" will be about 0.1, so a 2.3 gamma calibration will for most material really give you 2.4 response. This is not a huge error but I think it's large enough to pay attention to.
You have not yet shown anything that proves there is something "going on" with small area patterns on that display. It's certainly possible from the measurements you have posted, but the graph you have posted looks rather similar to the general ABL response I have measured on a number of Plasma displays, which would suggest that you're simply seeing the ABL in effect, rather than the display doing anything untoward.

Most of the Plasmas I have measured have high peak numbers using a 1% pattern because they are not power limited (this is how they can put high brightness & contrast numbers on spec sheets) which drops off at about 5–10%, stays constant to about 25–30% on most models (the KRPs go to 50%) and then starts to drop considerably as APL goes above that.

That's why you can't simply look at the results you have posted in isolation, the need to be referenced against the display's ABL response to have any kind of meaning.

Considering that 25% APL is usually right at the upper-end of where the ABL tends to "level out" with most Plasma displays, it seems like a bad idea to be using patterns that bright.

It seems like you are choosing patterns that give you the results that you want from your display, rather than patterns which are best for taking accurate measurements of the display's response.


Considering that the average picture level is something that is going to fluctuate depending on the content you are viewing, why would you intentionally calibrate your display in a power limited state?

I've often seen numbers in the 20–30% APL range thrown around the forums here as being the average level for most content, but whenever I actually analyse the content I watch, most scenes are not in that range. They are usually either quite a bit lower than that, or quite a bit higher than that. (averaging those numbers might end up at 20–30% though)

Just for fun, I decided to pick three films at random, go through all the chapter markers, and grab the APL of them. (72 images)
Michael Clayton: Average, 16 APL. Range, 1–51 APL.
Wall-E: Average, 24 APL. Range, 1–60 APL.
Almost Famous: Average, 20 APL. Range, 1–63 APL.

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%.
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?