Better to have 2.2 gamma or stable 2.3 gamma on a plasma? - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 135 Old 06-07-2012, 11:54 AM - Thread Starter
Member
 
plasma_fan's Avatar
 
Join Date: Jan 2012
Posts: 53
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
All things being equal would it be better to calibrae gamma to 2.22 and have it take a nose dive in the high ire patterns with the larger apl windows because of the plasma abl. Or calibrate to 2.31 and have a less flat curve but more stability with the larger aplwindows?
plasma_fan is offline  
Sponsored Links
Advertisement
 
post #2 of 135 Old 06-07-2012, 03:30 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,047
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 186
Quote:
Originally Posted by plasma_fan View Post

All things being equal would it be better to calibrae gamma to 2.22 and have it take a nose dive in the high ire patterns with the larger apl windows because of the plasma abl. Or calibrate to 2.31 and have a less flat curve but more stability with the larger aplwindows?

You could do both and see which looks better with reference material like a good BD movie. If you have Day/Night modes you can easily switch between the two and if not you can just use two inputs instead of one.
PlasmaPZ80U is online now  
post #3 of 135 Old 06-07-2012, 03:51 PM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
Sounds like the test patterns you’re using are too large for the display and activating the ABL.
Use the smallest window patterns that you can, ideally a square of 1% area. (144×144px at 1080p)
If your pattern generator scales by dimensions rather than area, use a 13% size pattern. (250×140px at 1080p)

Using anything else will give you inaccurate results on displays with an ABL circuit.
Chronoptimist is offline  
post #4 of 135 Old 06-07-2012, 03:54 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,047
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 186
Quote:
Originally Posted by Chronoptimist View Post

Sounds like the test patterns you’re using are too large for the display and activating the ABL.
Use the smallest window patterns that you can, ideally a square of 1% area. (144×144px at 1080p)
If your pattern generator scales by dimensions rather than area, use a 13% size pattern. (250×140px at 1080p)

I think the OP is using the AVS disc, which has windows, fields, and small and large APL patterns. I think the regular windows are 15% total screen area.
PlasmaPZ80U is online now  
post #5 of 135 Old 06-07-2012, 03:59 PM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
Quote:
Originally Posted by PlasmaPZ80U View Post

I think the OP is using the AVS disc, which has windows, fields, and small and large APL patterns.
The APL patterns are a nice idea, but correct calibration procedure uses the 1% patterns as described above.

You need to bypass the ABL (or at least do your best) for the true panel response. Attempting to compensate for it cannot work unless you know the absolute specifics on how the ABL system works in that display and specifically create patterns for it. (a fool’s errand)

Just checked and they’re slightly over 14% area (at least the one I loaded up was) which is considerably larger than ideal.
Chronoptimist is offline  
post #6 of 135 Old 06-07-2012, 06:37 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,047
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 186
Quote:
Originally Posted by Chronoptimist View Post

The APL patterns are a nice idea, but correct calibration procedure uses the 1% patterns as described above.

You need to bypass the ABL (or at least do your best) for the true panel response. Attempting to compensate for it cannot work unless you know the absolute specifics on how the ABL system works in that display and specifically create patterns for it. (a fool’s errand)
Just checked and they’re slightly over 14% area (at least the one I loaded up was) which is considerably larger than ideal.

I believe Doug Blackburn has mentioned in recent posts that a window size from 10% to 18% is best for plasmas. I don't think 1% patterns are what most calibrators recommend here.
PlasmaPZ80U is online now  
post #7 of 135 Old 06-07-2012, 06:41 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,047
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 186
PlasmaPZ80U is online now  
post #8 of 135 Old 06-07-2012, 08:37 PM
Advanced Member
 
Smackrabbit's Avatar
 
Join Date: Sep 2001
Location: Portland, OR, USA
Posts: 892
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 29
Quote:
Originally Posted by PlasmaPZ80U View Post

I believe Doug Blackburn has mentioned in recent posts that a window size from 10% to 18% is best for plasmas. I don't think 1% patterns are what most calibrators recommend here.

You're getting hung up on two ways to refer to the size. In this case 1% is saying it takes 1% of the screen, which is 144x144 or so then. 10% is saying it goes 10% wide and 10% high, so 192x108, which would also be 1% of the area. 18% would be the same, but 18% wide and tall, for 3.25% of the screen or so.

Really it's just two different terms for the same thing, and so then your range would be a pattern from 1-3% of the screen area ideally. At least I'm 99% sure what he's saying.

Chris Heinonen
Senior Editor, Secrets of Home Theater and High Fidelity, www.hometheaterhifi.com
Displays Editor, AnandTech.com
Contributor, HDGuru.com and Wirecutter.com
ISF Level II Certified Calibrator, ReferenceHomeTheater.com
Smackrabbit is offline  
post #9 of 135 Old 06-07-2012, 09:29 PM - Thread Starter
Member
 
plasma_fan's Avatar
 
Join Date: Jan 2012
Posts: 53
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Yes, I am using the AVSHD disc. 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?

I will try both settings on the same content and see what it looks like. Thanks
plasma_fan is offline  
post #10 of 135 Old 06-08-2012, 04:38 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
Quote:
Originally Posted by Smackrabbit View Post

You're getting hung up on two ways to refer to the size. In this case 1% is saying it takes 1% of the screen, which is 144x144 or so then. 10% is saying it goes 10% wide and 10% high, so 192x108, which would also be 1% of the area. 18% would be the same, but 18% wide and tall, for 3.25% of the screen or so.
Really it's just two different terms for the same thing, and so then your range would be a pattern from 1-3% of the screen area ideally. At least I'm 99% sure what he's saying.

When Doug says 10-18%, he means by screen area, not dimension. This is the "Joe Kane" standard based on CRT measurements.
zoyd is online now  
post #11 of 135 Old 06-08-2012, 04:53 AM
Advanced Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 522
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 211
Light Illusion is offline  
post #12 of 135 Old 06-08-2012, 05:10 AM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
The standard for broadcast is the patterns I have described: http://tech.ebu.ch/docs/tech/tech3325.pdf
I can’t think of any good reason to use larger patterns on a plasma display.

The exception to this rule would be local dimming LED backlit LCD displays where you actually need either a window of a larger size, or might even be better off using full field patterns.
Chronoptimist is offline  
post #13 of 135 Old 06-08-2012, 05:32 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
I agree with you Chronoptimist except in cases where the display is modifying gamma at low levels which is more and more common these days. In those cases I believe fixed APL patterns in a "stable gamma" range are more appropriate. At least they have led to better results on my display which exhibits this behavior.
zoyd is online now  
post #14 of 135 Old 06-08-2012, 12:28 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,452
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 206
A "10% window pattern" means that the window covers 10% of the surface area of the screen.

It would also mean the window contains about 10% of the display's pixels. So if there are 2 million pixels in the panel (2 million red, 2 million green, 2 million blue), the window pattern would contain 200,000 pixels. If the window pattern has 192x108 pixels, it only contains a bit more than 20,000 pixels so that would be much too small (roughly 1% of the area). A 10% window would have about 600x336 pixels for a 1.78 display. You'd end up with 3 of these across the screen, 3 down the screen with a bit of a border around that 3x3 grid that would also contain ~200,000 pixels (fo make up the 10th ~200,000 pixel area on the panel)

As said before, using a smaller window size will exaggerate the effect the plasma has on APL by putting the plasma in dark and light modes that will RARELY, if ever, occur with real video content.

It can take a LONG LONG time for a calibrator to determine exactly what the ABL "factor" is for any given brand/model when looking at one for the first time. WAY too much time to add to any normal calibration. You can be fooled by the panel if you don't use very small changes in window size while evaluating ABL. There are few, if any times where the content of a 1% (physical size) window would EVER define the performance of the panel with real-world video content. Furthermore, you can't assume that the ABL performance of any given brand/model manufactured in, say, June, will be the same as the same model manufactured in, say, September. Some kind of firmware or hardware change could very well change the ABL behavior so to be sure you're doing the right thing (as a calibrator) you'd have to add MANY hours to every plasma calibration to check each and every sample of every model you ever calibrate. That means a $300 calibration could become a $450 calibration pretty easily, just because of all the added time to investigate the ABL behavior, which might make a difference 1% of the time... maybe. It's also quite possible that using the fixed 10%-sized window will produce a calibration that's indistinguishable from the tiny windows or variable windows--in fact, I'd say this is the likely outcome 99% of the time or more.

You can drive yourself crazy with miniscule details that mean little or nothing for 99% of brands/models 99% of the time or you can futz to your heart's content with this "fanatical" level of detail that may return nothing... or if it does return "something" it might be visible 1% of the time on 1% of the displays. It's your choice... there is certainly no requirement for ANYONE, pro or amateur to use variable-size or tiny windows for calibrating a panel... in fact, if you don't know EXACTLY what you are doing, ultra-small windows can make a plasma look worse than it would if calibrated with windows that are 10%-18% of the surface area of the screen.

The only time a pro calibrator would be likely to mess with this is if a particular display brand/model had some badly screwed up characteristic that would appear in real-world content if the calibration was done the "normal" way (using a window size that's between 10% and 18% of the area of the display).

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #15 of 135 Old 06-08-2012, 03:21 PM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
Quote:
Originally Posted by zoyd View Post

I agree with you Chronoptimist except in cases where the display is modifying gamma at low levels which is more and more common these days. In those cases I believe fixed APL patterns in a "stable gamma" range are more appropriate. At least they have led to better results on my display which exhibits this behavior.
As with any rule, there are always exceptions...
Quote:
Originally Posted by Doug Blackburn View Post

A "10% window pattern" means that the window covers 10% of the surface area of the screen.
It would also mean the window contains about 10% of the display's pixels. So if there are 2 million pixels in the panel (2 million red, 2 million green, 2 million blue), the window pattern would contain 200,000 pixels. If the window pattern has 192x108 pixels, it only contains a bit more than 20,000 pixels so that would be much too small (roughly 1% of the area).
It really depends on the patterns or signal generator you're using. Ideally when you set "10%" you should get a 607x342 pattern, but with many generators (such as CalMAN's) setting "10%" will give you 192x108, which is 10% horizontally and vertically, or 1% area.
Quote:
Originally Posted by Doug Blackburn View Post

As said before, using a smaller window size will exaggerate the effect the plasma has on APL by putting the plasma in dark and light modes that will RARELY, if ever, occur with real video content.
I think you have this backwards. The ABL circuit is in effect the larger a test pattern gets. A 1% area pattern (ideally a 144x144px square for measurement purposes) should be small enough that the ABL does not get activated on the display at all, and you can accurately measure its native response. With larger patterns the ABL goes into effect, which gives higher gamma readings than expected for brighter patterns.

Quote:
Originally Posted by Doug Blackburn View Post

It can take a LONG LONG time for a calibrator to determine exactly what the ABL "factor" is for any given brand/model when looking at one for the first time. WAY too much time to add to any normal calibration. You can be fooled by the panel if you don't use very small changes in window size while evaluating ABL. There are few, if any times where the content of a 1% (physical size) window would EVER define the performance of the panel with real-world video content. Furthermore, you can't assume that the ABL performance of any given brand/model manufactured in, say, June, will be the same as the same model manufactured in, say, September. Some kind of firmware or hardware change could very well change the ABL behavior so to be sure you're doing the right thing (as a calibrator) you'd have to add MANY hours to every plasma calibration to check each and every sample of every model you ever calibrate. That means a $300 calibration could become a $450 calibration pretty easily, just because of all the added time to investigate the ABL behavior, which might make a difference 1% of the time... maybe. It's also quite possible that using the fixed 10%-sized window will produce a calibration that's indistinguishable from the tiny windows or variable windows--in fact, I'd say this is the likely outcome 99% of the time or more.
I could be misinterpreting what you mean here, but it's relatively easy to find the main effects of any ABL system. You set white to 100 nits with a 1% area pattern, and take brightness measurements from 1% to 100% area in 5% steps. Takes all of five minutes to do. This will give you a clear illustration of the point at which the ABL really starts to take over, and how much of a brightness drop to expect between 1% and 100% area patterns. (often as much as 60% with Plasma displays)

However, It is exactly because the effect of the ABL on any given image (test patterns aside) is essentially unpredictable that you should be using 1% area patterns to "bypass" it.
The ABL function is not something that can be "calibrated out" so there is no point in trying to use "ABL patterns" or finding a "better" size of window pattern to use for calibration. Anything else is going to give you inaccurate results.
On some displays you might find that a 15% area pattern gives exactly the same results as 1% area, but on others it may not.
Quote:
Originally Posted by Doug Blackburn View Post

You can drive yourself crazy with miniscule details that mean little or nothing for 99% of brands/models 99% of the time or you can futz to your heart's content with this "fanatical" level of detail that may return nothing... or if it does return "something" it might be visible 1% of the time on 1% of the displays. It's your choice... there is certainly no requirement for ANYONE, pro or amateur to use variable-size or tiny windows for calibrating a panel... in fact, if you don't know EXACTLY what you are doing, ultra-small windows can make a plasma look worse than it would if calibrated with windows that are 10%-18% of the surface area of the screen.
The only time a pro calibrator would be likely to mess with this is if a particular display brand/model had some badly screwed up characteristic that would appear in real-world content if the calibration was done the "normal" way (using a window size that's between 10% and 18% of the area of the display).
I agree that there's almost no reason to use variable size/"APL" patterns on a display, but I am curious as to why you think using 10–18% area patterns would be a better choice than 1%?

You might find that you want to set peak white with a 100% pattern (e.g. the panel dims from 100 nits at 1% to 40 nits at 100%) and let the display be brighter than it should be at lower APLs, but the rest of the display calibration (greyscale & gamma tracking) should still be performed with the 1% area patterns. (even more important in this case, as you are more likely to run into the ABL with contrast turned up)
Chronoptimist is offline  
post #16 of 135 Old 06-08-2012, 03:47 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
I really don't understand the resistance to using a fixed APL pattern for everything. Whatever window size you use you are calibrating different operating points of the display as you progress through the various levels. This used to be fine as long as you avoided ABL but increasingly displays have non-constant response at lower stimuli. Why not choose an operating point which is in a region of typical APL and calibrate there, it takes all of the guess work out of it. You can argue over what that typical level is but it's bound to give you better overall average response than trying to find the perfect window size. It's the same concept that is working well with color calibration at 75% saturation (for plasmas) instead of 100% saturation, you get lower average dE over the entire gamut.
chronitis likes this.
zoyd is online now  
post #17 of 135 Old 06-08-2012, 08:34 PM
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,355
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 204
Quote:
Originally Posted by zoyd View Post

I really don't understand the resistance to using a fixed APL pattern for everything. Whatever window size you use you are calibrating different operating points of the display as you progress through the various levels. This used to be fine as long as you avoided ABL but increasingly displays have non-constant response at lower stimuli. Why not choose an operating point which is in a region of typical APL and calibrate there, it takes all of the guess work out of it. You can argue over what that typical level is but it's bound to give you better overall average response than trying to find the perfect window size. It's the same concept that is working well with color calibration at 75% saturation (for plasmas) instead of 100% saturation, you get lower average dE over the entire gamut.
I can't speak for anyone else, but my reluctance is based on three factors.

The first is philosophical. Test patterns should be the reference against which the display is measured. Creating specialized test patterns to accommodate poor display design seems to turn this on its head. Also, it is not really like using 75% saturation patterns. All these do is use an area in the color space more typical of actual program material than the gamut boundary, which is unrelated to display design deficiencies.

The second objection I have is based solely on how the fixed APL pattern is designed. In particular, employing color APL windows that use multiple colors in the surround is not a good idea, as it has the likely effect of contaminating readings from non-contact devices. Also, using what is in effect a rectangular ramp of different grayscale levels in the surround has the likely effect of exceeding the plasma's voltage limitations at high levels of stimulus and, again, contaminating readings from non-contact devices with excess light spill. This concern can be resolved by simply designing better patterns.

Third, we need some objective criteria to determine whether the display in question is actually using some dynamic brightness circuitry that affects the accuracy of low stimulus readings. Perhaps this exists, but I have seen little discussion of this, which often leaves amateurs to simply use whatever test pattern that yields the best measured results, which only has the effect of covering up real problems that should be fixed during calibration. The only objective criteria that comes to mind is if a plasma goes totally black at 0%--shows no MLL at all, much like a local dimming LCD.

Tom Huffman
ChromaPure Software/AccuPel Video Signal Generators
ISF/THX Calibrations
Springfield, MO

TomHuffman is offline  
post #18 of 135 Old 06-09-2012, 03:39 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
Quote:
Originally Posted by TomHuffman View Post

I can't speak for anyone else, but my reluctance is based on three factors.
The first is philosophical. Test patterns should be the reference against which the display is measured. Creating specialized test patterns to accommodate poor display design seems to turn this on its head.

Windowed patterns are just as specialized as any other pattern, it just doesn't seem that way because they are the norm. In my mind they are even more contrived than a fixed APL pattern because they are further away from normal stimuli than a fixed APL pattern is. The windowed patterns achieved wide acceptance because they were the easiest to implement in pattern generators and they "work", there is nothing "reference" about them at all. In other words there is no derived color science parameter that requires any particular pattern shape. In the field of metrology, especially the one I work in, there is a rule of thumb, "test as you fly". You want to calibrate your device under operating conditions that are as close as possible to how it will be used in the field. Now I know that windowed patterns are here to stay because of the infrastructure built up around them but there are clearly cases in which they are not optimal and I've yet to hear of any similar cases when using a fixed APL pattern. They can be used to avoid luminance manipulation in those displays that have it.
Quote:
Also, it is not really like using 75% saturation patterns. All these do is use an area in the color space more typical of actual program material than the gamut boundary, which is unrelated to display design deficiencies.

It is exactly because of design deficiencies that this done, if the display's CMS had linear saturation tracking you can calibrate the gamut boundary and not worry about anything else. As is it you have to decide which area of the gamut to minimize errors in and so you choose one which is closer to it's normal operating conditions.
Quote:
The second objection I have is based solely on how the fixed APL pattern is designed. In particular, employing color APL windows that use multiple colors in the surround is not a good idea, as it has the likely effect of contaminating readings from non-contact devices. Also, using what is in effect a rectangular ramp of different grayscale levels in the surround has the likely effect of exceeding the plasma's voltage limitations at high levels of stimulus and, again, contaminating readings from non-contact devices with excess light spill. This concern can be resolved by simply designing better patterns.

I agree, that is precisely why I recommend full field backgrounds at 25% gray with small patterned windows in the center, which are optimal for contact meters. For non-contact meters a larger area could be used but this depends on the FOV and how far away you need to get. One should also use a light tunnel to block out-of-field contamination.
Quote:
Third, we need some objective criteria to determine whether the display in question is actually using some dynamic brightness circuitry that affects the accuracy of low stimulus readings. Perhaps this exists, but I have seen little discussion of this, which often leaves amateurs to simply use whatever test pattern that yields the best measured results, which only has the effect of covering up real problems that should be fixed during calibration. The only objective criteria that comes to mind is if a plasma goes totally black at 0%--shows no MLL at all, much like a local dimming LCD.

I disagree here, the closer you can get a pattern to emulate the display's normal operating point the less you have to worry about it's deficiencies. I would argue that you need to prove it doesn't have a dynamic brightness problem in order to reliably use windowed patterns.
zoyd is online now  
post #19 of 135 Old 06-09-2012, 07:38 AM
Member
 
Monitorman's Avatar
 
Join Date: Feb 2001
Location: New York, NY
Posts: 194
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by plasma_fan View Post

All things being equal would it be better to calibrae gamma to 2.22 and have it take a nose dive in the high ire patterns with the larger apl windows because of the plasma abl. Or calibrate to 2.31 and have a less flat curve but more stability with the larger aplwindows?

To get back to the original question. For plasma owners with a well designed home theater like environment with total light control for movie watching 2.4 would be the goal. The ITU (International Transmissions Union) one of our standards setting organizations has recently settled on 2.4 gamma as the reference for video mastering. So if you have a choice of 2.2 or 2.3, and you have the right environment you should opt for 2.3. Panels with the ability to have Day and Night modes like the Panasonic's with ISF ccc, and the Samsungs with Cal-Day and Cal-Night modes are desirable in that you can set gamma at 2.3 to 2.4 for the night mode, and somewhat lower for an effective brighter day mode.

Kevin Miller
ISF Technician
Co-Instructor ISF Level I & II Seminars
Display Product Development Consultant
kevin@isftv.com
Co-founder tweaktv.com
Monitorman is offline  
post #20 of 135 Old 06-09-2012, 07:57 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
Quote:
Originally Posted by Monitorman View Post

The ITU (International Transmissions Union) one of our standards setting organizations has recently settled on 2.4 gamma as the reference for video mastering.

Only for displays that can achieve black levels less than 0.01 cd/m^2 If that condition is not met then the functional form of the recommendation is more complicated and depends on both mll and peak white. The upshot for many displays is a target closer to 2.3 above 40% stimulus with a gradual decrease in gamma to 2.1 or 2.15 at 10%. The recommendation does not assume any particular viewing environment.
zoyd is online now  
post #21 of 135 Old 06-09-2012, 09:14 AM
AVS Special Member
 
turbe's Avatar
 
Join Date: Feb 2001
Location: Lake Tahoe, NV
Posts: 4,424
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 84
Quote:
Originally Posted by zoyd View Post

The recommendation does not assume any particular viewing environment.

It actually does and Kevin mentions this as well with 'video mastering'.

The ITU recommendation is for HDTV production / HDTV Studio Production (Studio Mastering).

and as Poynton said,
Quote:
we seek to mimic the image appearance at mastering

The environment used during mastering. What a lot of Home Theater owners fail to realize or cannot do much about in reality for various reasons even if they want, is how important also the environment is in achieving the creator's intent. Of course GeorgeAB and a few others have been pounding away at this for years here.

In regards to the recent ITU recommendation...
Ron Williams, Landmark CEO, said at a recent Hollywood Post Alliance first-ever Reference Monitor Symposium, sponsored by Walt Disney Studios, Dolby and Sony and several others:
Quote:
while the U.S. and European standards bodies wanted a gamma of 2.35, Japan insisted on 2.4. So 2.2 never really existed in anything, although some presentations mentioned it.

~

Rec.709 is just a recommendation and doesn't really pertain to a display, it's a production standard and has nothing to do with display. What does specify display is the recent ITU recommendation.



I also found comment interesting from Josh Pines, Technicolor VP of Imaging Research & Development:
Quote:
Can video display or "emissive" monitors be used for Digital Cinema mastering?
Pines touched on the Stevens effect (that perceived contrast decreases at lower luminance), Hunt effect (that perceived colorfulness also decreases at lower luminance) and display flare characteristics.
Quote:
The same content has to look good in a normal surround (office), dim surround (living room) and dark surround (theatrical)
Quote:
People will disagree with me, but the way you do that is different display gammas, from 2.2 to 2.4 to 2.6 for each of the above.

Need to find a Professional Calibrator? Click Here to PM me with your Display & City

Calibrator List - Pioneer ISFccc Interface

Calibration Reports - Pioneer

 

ControlCAL™
Designed by Calibrators for Calibrators™

No need to fumble through the Display's Menu with its Remote Control™

turbe is offline  
post #22 of 135 Old 06-09-2012, 09:41 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
Quote:
Originally Posted by turbe View Post

It actually does and Kevin mentions this as well with 'video mastering'.
The ITU recommendation is for HDTV production / HDTV Studio Production (Studio Mastering).

The actual quote within the body of the text as well as the recommendation title is:
Quote:
Reference electro-optical transfer function for flat panel
displays used in HDTV studio production

There is no discussion of requirements of what the studio production environment should look like, there are plenty of other places to go for that but this document isn't one of them.

You'll also notice from the above quote that the recommendation describes a function, not a particular number. To state the recommendation is "gamma = 2.4" is only true in the limit of zero black. If you'd like to see what the actual function looks like for displays that don't meet that criteria you can use the attached calculator. Note that the recommendation treats <0.01 cd/m^2 equivalent to zero black.


bt1886_calculator.xlsx.zip 38k .zip file
Attached Files
File Type: zip bt1886_calculator.xlsx.zip (38.5 KB, 32 views)
zoyd is online now  
post #23 of 135 Old 06-09-2012, 09:55 AM
AVS Special Member
 
turbe's Avatar
 
Join Date: Feb 2001
Location: Lake Tahoe, NV
Posts: 4,424
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 84
yep, like I posted at the top, I was still in the edit stage of the post, wanted to give additional information from those in the industry responsible / involved in the mastering..specifically Josh Pines and Ron Williams.... it's updated now.

More to add but I'll end the edit now..
Quote:
Originally Posted by zoyd View Post

There is no discussion of requirements of what the studio production environment should look like, there are plenty of other places to go for that but this document isn't one of them.
it is clear of the environment and that's also coming from conversations with many of these people including with some who attended that Symposium and EBU does specific the conditions for Studio Monitor measurement (of course not stated by EBU for grading), first being:
Quote:
Measurements should be made in a darkened room (VESA FPDM2 [6] specify this as less than 1 Lux room brightness)
There are documents from others for grading conditions which is different than for performance measurement (I'll get a few links and edit them here):

http://documentation.apple.com/en/color/usermanual/index.html#chapter=6%26section=2%26tasks=true
Quote:
Set Up Your Viewing Environment Carefully
The environment in which you view your monitor also has a significant impact on your ability to properly evaluate the image.

There should be no direct light spilling on the front of your monitor.

Ambient room lighting should be subdued and indirect, and there should be no direct light sources within your field of view.

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

There should be indirect lighting behind the viewing monitor that’s between 10–25% of the brightness of the installed monitor set to display pure white.

The ideal viewing distance for a given monitor is approximately five times the vertical height of its screen.

The color of the room within your working field of vision should be a neutral gray.

These precautions will help to prevent eye fatigue and inadvertent color biasing while you work and will also maximize the image quality you’ll perceive on your display.



If you really want to get info, join/check out http://tig.colorist.org/wiki/Main_Page




Of course, grading environments for film and those for projects intended for wide distribution (i.e. BD/DVD) will be different and color suites can tend to use lighting conditions more appropriate for projects to be seen under general living condition in normal homes. In fact, today's films go through many grading sessions depending on target dist. and even several just for Cinema...... blurb about Prometheus':

Stephen Nakamura, one of a handful of artists who helped to pioneer DI color grading, shares his experience on creating the look for Prometheus as envisioned by director Ridley Scott:
Quote:
My biggest challenge was maintain Ridley's vision throughout all these deliverables. We graded for different 3D projection systems, one that can put approximately both 4 foot-lamberts, the other 6 foot-lamberts, of light on the screen. We also mastered to 2D for digital cinema (a DCP version) and did another for film-out. We also had to create another one for IMAX, which has an entirely different aspect ratio. We also had to approve all IMAX prints, and there was IMAX digital and IMAX film. So that made for six versions of the film. Then of course there's the Blu-ray and DVD.

We started the workflow for grading the multiple 3D versions with the less-bright 4 foot-lambert version, grading that all the way through. Once that was done, then it's easy to do the 6 foot-lambert version; it's just about pushing more light through when projecting it onto the screen. Some shots that may be on the verge of being clipped, looking totally blown out, in 4 foot-lamberts won't look that way when projected at 6 foot-lamberts. Basically, anything that looks good projected at 4 foot-lamberts will generally look better projected at 6. It still requires some fine-tuning, but it certainly makes a lot more sense than grading for 6 first. So much of what looks good projected that way will look terrible at 4. We also based the 2D master on the 4 foot-lambert 3D master and made refinements for 2D's much brighter projection systems. For the IMAX version, the color remains the same, but it involves panning-and-scanning to accommodate the aspect ratio.

400




Edit (continued from my Post above):
Here is what Kevin Wines, Image Technology Director at THX said at the Symposium that I also found interesting:
Quote:
The viewing quality of video is dramatically affected by room lighting conditions. We have no control over what direction the windows face, and sunlight changes color and intensity throughout the day. Seldom is the light at that ideal level when we're finishing it in post production. No one is suggesting we change the standard for post production. But you do have to be ready to adjust playback for the environment. You may be better lowering the gamma to get the perception back, as opposed to less sophisticated behavior with adjusting brightness or contrast.

and one more from Ron Williams, Landmark CEO:
Quote:
Critical elements for display image interchangeability include color gamut, contrasts, black/cutoff, and picture size. Screen size matters when comparing images. A larger screen appears to be brighter at the same fL [foot lamberts]." How bright is the right bright? Even here the numbers varied wildly: SMPTE documents specifies 35 fL; CRT monitors varied from 28 to 33 fL; HD CRT monitors run to 19 to 21 fL; plasmas vary between 18 to 24 fL; and computer monitors are between 19 to 26 fL.

Lastly:
Quote:
Contrast ratio is the ratio of luminance between the brightest white and the darkest black of a particular device or a particular environment. Gamma is closely related to contrast, which is why it's important to work with a defined gamma ratio because if the gamma ratio is incorrect, the contrast along the response curve will be inaccurate. In a Rec. 709 environment, the EBU references a gamma curve of 2.4, recently changed from 2.3.
Quote:
Note: Rec. 709 was written for the camera, and EBU has taken over responsibility for developing and recommending guidelines for applying Rec.709 to the reference monitor. The EBU has created three grades of monitors: Grade 1 is the highest standard, to replace CRT; Grade 2 is where most of the current interim monitors fit; Grade 3 is where all the other monitors fall. The EBU has set very precise guidelines for reference monitors, available on its website as Technical Document 3320.

btw, if you want a Grade 1 reference monitor now, Dolby has one

338

Need to find a Professional Calibrator? Click Here to PM me with your Display & City

Calibrator List - Pioneer ISFccc Interface

Calibration Reports - Pioneer

 

ControlCAL™
Designed by Calibrators for Calibrators™

No need to fumble through the Display's Menu with its Remote Control™

turbe is offline  
post #24 of 135 Old 06-09-2012, 12:18 PM
AVS Special Member
 
turbe's Avatar
 
Join Date: Feb 2001
Location: Lake Tahoe, NV
Posts: 4,424
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 84
well, I'm done editing, I think I have posted enough on this above (many edits since original posts)..

my .02 to OP, forget 2.4 now, I doubt many have a plasma today that can do it... target 2.2 - 2.3 using 10% windows for your reference viewing.. I agree with what Josh Pines said about normal and dim surround.

Need to find a Professional Calibrator? Click Here to PM me with your Display & City

Calibrator List - Pioneer ISFccc Interface

Calibration Reports - Pioneer

 

ControlCAL™
Designed by Calibrators for Calibrators™

No need to fumble through the Display's Menu with its Remote Control™

turbe is offline  
post #25 of 135 Old 06-09-2012, 01:00 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
Quote:
Originally Posted by turbe View Post

well, I'm done editing, I think I have posted enough on this above (many edits since original posts)..
my .02 to OP, forget 2.4 now, I doubt many have a plasma today that can do it... target 2.2 - 2.3 using 10% windows for your reference viewing.. I agree with what Josh Pines said about normal and dim surround.

Thanks for the post, very informative. I don't agree that 10% windows are appropriate for all display technologies and I've explained why earlier but let's leave it at that.
zoyd is online now  
post #26 of 135 Old 06-09-2012, 01:16 PM
AVS Special Member
 
turbe's Avatar
 
Join Date: Feb 2001
Location: Lake Tahoe, NV
Posts: 4,424
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 84
not 10% windowed for all display types, but for this Thread's subject, Plasma.

I have seen the reference to the very small patterns Chronoptimist posts about (and linked document), THX tests their displays with very small patterns as well but larger than 1%..
Quote:
Originally Posted by zoyd View Post

Thanks for the post, very informative.

took a little time to go through my bookmarks smile.gif

Need to find a Professional Calibrator? Click Here to PM me with your Display & City

Calibrator List - Pioneer ISFccc Interface

Calibration Reports - Pioneer

 

ControlCAL™
Designed by Calibrators for Calibrators™

No need to fumble through the Display's Menu with its Remote Control™

turbe is offline  
post #27 of 135 Old 06-09-2012, 01:28 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,047
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 186
While we're on the subject of gamma, is it better to use the power law formula or BT.1886? It seems like the new formula should provide better shadow detail by making gamma lower at the low end and improve contrast at the top end by raising gamma there.
PlasmaPZ80U is online now  
post #28 of 135 Old 06-09-2012, 01:33 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
That's my opinion, emulate the functional form of bt.1886 and shift the curve lower for bright surround but never higher.
zoyd is online now  
post #29 of 135 Old 06-09-2012, 01:37 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,431
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 78 Post(s)
Liked: 299
Quote:
Originally Posted by turbe View Post

not 10% windowed for all display types, but for this Thread's subject, Plasma.

Not for plasmas which manipulate gamma at low luminance which include all 2011 Samsungs and Panasonics
zoyd is online now  
post #30 of 135 Old 06-09-2012, 03:01 PM
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,355
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 204
Quote:
Originally Posted by zoyd View Post

Windowed patterns are just as specialized as any other pattern, it just doesn't seem that way because they are the norm. In my mind they are even more contrived than a fixed APL pattern because they are further away from normal stimuli than a fixed APL pattern is. The windowed patterns achieved wide acceptance because they were the easiest to implement in pattern generators and they "work", there is nothing "reference" about them at all. In other words there is no derived color science parameter that requires any particular pattern shape. In the field of metrology, especially the one I work in, there is a rule of thumb, "test as you fly". You want to calibrate your device under operating conditions that are as close as possible to how it will be used in the field. Now I know that windowed patterns are here to stay because of the infrastructure built up around them but there are clearly cases in which they are not optimal and I've yet to hear of any similar cases when using a fixed APL pattern. They can be used to avoid luminance manipulation in those displays that have it.
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.
Quote:
Originally Posted by zoyd View Post

It is exactly because of design deficiencies that this done, if the display's CMS had linear saturation tracking you can calibrate the gamut boundary and not worry about anything else. As is it you have to decide which area of the gamut to minimize errors in and so you choose one which is closer to it's normal operating conditions.
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.
Quote:
Originally Posted by zoyd View Post

I disagree here, the closer you can get a pattern to emulate the display's normal operating point the less you have to worry about it's deficiencies. I would argue that you need to prove it doesn't have a dynamic brightness problem in order to reliably use windowed patterns.
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.

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.

Tom Huffman
ChromaPure Software/AccuPel Video Signal Generators
ISF/THX Calibrations
Springfield, MO

TomHuffman is offline  
Reply Display Calibration

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off