R.Masciola's HDR-10 UHD Test Patterns - Page 2 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 327Likes
Reply
 
Thread Tools
post #31 of 874 Old 06-17-2016, 04:58 AM
Member
 
carneb's Avatar
 
Join Date: Nov 2006
Location: Sydney, Australia
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 35
Quote:
Originally Posted by mascior View Post
Color primaries BT2020 w/p3 mastering display color primaries vs BT2020 w/BT2020 mastering display color primaries...not really much of a difference...at least with the displays I tested
You would think there would be a difference at the lower saturation tracking points (50% and 25%). I'll borrow a diagram from the HDTVTest.co.uk review of the Panasonic TX-58DX902B HERE

You can see on the following diagram showing tracking of BT2020 that even though the 100% points are still in roughly the same place as P3 primaries, the lower saturation points have been re-aligned to have the same hue as the 2020 primaries.



I suppose it depends on the TV but you would think that your testing would show something similar. It might be a silly question, but are the RGB values for the BT2020 w/p3 mastering display color primaries windows different to the RGB values for the BT2020 w/BT2020 mastering display color primaries? For example the red 100% saturation/100% stimulus window for 2020 mastering primaries would be 940,64,64 but for P3 mastering primaries I would think that the RGB values would be roughly 727, 104, 64?
carneb is online now  
Sponsored Links
Advertisement
 
post #32 of 874 Old 06-17-2016, 05:31 AM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Quote:
Originally Posted by carneb View Post
You would think there would be a difference at the lower saturation tracking points (50% and 25%). I'll borrow a diagram from the HDTVTest.co.uk review of the Panasonic TX-58DX902B HERE

You can see on the following diagram showing tracking of BT2020 that even though the 100% points are still in roughly the same place as P3 primaries, the lower saturation points have been re-aligned to have the same hue as the 2020 primaries.

I suppose it depends on the TV but you would think that your testing would show something similar. It might be a silly question, but are the RGB values for the BT2020 w/p3 mastering display color primaries windows different to the RGB values for the BT2020 w/BT2020 mastering display color primaries? For example the red 100% saturation/100% stimulus window for 2020 mastering primaries would be 940,64,64 but for P3 mastering primaries I would think that the RGB values would be roughly 727, 104, 64?
The master display primaries metadata may still be up in the air...I'm going to do multiple RGBYCM saturation sweeps today. I may end up utilizing:
-
  • Mastering display color primaries : R: x=0.708000 y=0.292000, G: x=0.170000 y=0.797000, B: x=0.131000 y=0.046000, White point: x=0.312700 y=0.329000
and let the display map the colors to the best of it's abilities...this may be a better testing method.

Ryan
mascior is offline  
post #33 of 874 Old 06-17-2016, 08:09 PM
AVS Forum Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 7,495
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 765 Post(s)
Liked: 659
This suggests that there is no consistency among display manufacturers regarding the display gamut. Given that, how can you develop one set of test patterns designed to be used for all displays?

BTW, to determine which gamut the display uses the green hue line is critical, as it is by far the most distinguishing between a 2020 and DCi gamut.


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

TomHuffman is offline  
 
post #34 of 874 Old 06-18-2016, 09:16 PM
Advanced Member
 
EvLee's Avatar
 
Join Date: Dec 2009
Posts: 577
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 332 Post(s)
Liked: 300
Quote:
Originally Posted by mascior View Post
The master display primaries metadata may still be up in the air...I'm going to do multiple RGBYCM saturation sweeps today. I may end up utilizing:
-
  • Mastering display color primaries : R: x=0.708000 y=0.292000, G: x=0.170000 y=0.797000, B: x=0.131000 y=0.046000, White point: x=0.312700 y=0.329000
and let the display map the colors to the best of it's abilities...this may be a better testing method.

Ryan
The professional displays that are currently used for mastering are only capable of P3 so your test content will more accurately reflect what is going out to market if you stick with:

Mastering display color primaries................R: x=0.680000 y=0.320000, G: x=0.265000 y=0.690000, B: x=0.150000 y=0.060000, White point: x=0.312700 y=0.329000
WiFi-Spy likes this.
EvLee is offline  
post #35 of 874 Old 06-19-2016, 07:52 PM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Quote:
Originally Posted by EvLee View Post
The professional displays that are currently used for mastering are only capable of P3 so your test content will more accurately reflect what is going out to market if you stick with:

Mastering display color primaries................R: x=0.680000 y=0.320000, G: x=0.265000 y=0.690000, B: x=0.150000 y=0.060000, White point: x=0.312700 y=0.329000
I would agree...however, the mastering display color primaries metadata doesn't seem to make any difference one way or the other...at least with the displays I tested. As a matter of fact, I thought the MaxCLL/MaxFALL metadata had an effect...however that turned out to be inaccurate. The only metadata besides the main flags Color Primaries BT2020, Transfer Characteristics ST2084, and the Matrix Coefficients BT2020nc is the minimum and maximum display luminance. I also found the recommended minimum luminance of 0.05 for LCD and 0.0005 for OLED to be accurate...with a max luminance of 1,000 nits for both.

Ryan
WiFi-Spy likes this.
mascior is offline  
post #36 of 874 Old 06-19-2016, 08:19 PM
Member
 
carneb's Avatar
 
Join Date: Nov 2006
Location: Sydney, Australia
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 35
Quote:
Originally Posted by mascior View Post
I would agree...however, the mastering display color primaries metadata doesn't seem to make any difference one way or the other...at least with the displays I tested.
Did you try mixing and matching colour space RGB values and mastering metadata values? What I mean by that is, for example, you have a red 2020 window (xy 0.708,0.292) with 2020 mastering metadata, then the same window with P3 Mastering metadata. Then try a red P3 window (xy 0.68,0.32) with P3 mastering metadata, then with 2020 mastering metadata to see what affect the metadata has?

Wish I had some way to generate the test windows myself as I'm vey curious to see what will happen.
carneb is online now  
post #37 of 874 Old 06-20-2016, 03:43 AM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Quote:
Originally Posted by carneb View Post
Did you try mixing and matching colour space RGB values and mastering metadata values? What I mean by that is, for example, you have a red 2020 window (xy 0.708,0.292) with 2020 mastering metadata, then the same window with P3 Mastering metadata. Then try a red P3 window (xy 0.68,0.32) with P3 mastering metadata, then with 2020 mastering metadata to see what affect the metadata has?

Wish I had some way to generate the test windows myself as I'm vey curious to see what will happen.
Ok...so this is what I've put together regarding this entire "P3/BT2020" mass confusion. There is no consumer display trying to achieve any percentage of P3, only a percentage of BT2020. Now with that being said, 100% of P3 is roughly close to 75% of BT2020 with different x/y coordinates, especially in green. So my thinking is if the mastering display is utilizing P3 primaries (which is what everything is currently mastered in) it is up to the display to interpret that metadata and calculate that to a percentage of BT2020. I feel this is why changing the mastering metadata between P3 and BT2020 has no effect on my patterns during testing because in the end...the display will always scale to a percentage of BT2020. Now this is only my assumption from what I could gather...thoughts from anyone?

Ryan
mascior is offline  
post #38 of 874 Old 06-20-2016, 05:50 AM
AVS Forum Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 7,495
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 765 Post(s)
Liked: 659
Quote:
Originally Posted by mascior View Post
Ok...so this is what I've put together regarding this entire "P3/BT2020" mass confusion. There is no consumer display trying to achieve any percentage of P3, only a percentage of BT2020. Now with that being said, 100% of P3 is roughly close to 75% of BT2020 with different x/y coordinates, especially in green. So my thinking is if the mastering display is utilizing P3 primaries (which is what everything is currently mastered in) it is up to the display to interpret that metadata and calculate that to a percentage of BT2020. I feel this is why changing the mastering metadata between P3 and BT2020 has no effect on my patterns during testing because in the end...the display will always scale to a percentage of BT2020. Now this is only my assumption from what I could gather...thoughts from anyone?

Ryan
The CIE94 between 75% of Rec. 2020 green and 100% of P3 green is over 6.

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

TomHuffman is offline  
post #39 of 874 Old 06-20-2016, 06:36 AM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Quote:
Originally Posted by TomHuffman View Post
The CIE94 between 75% of Rec. 2020 green and 100% of P3 green is over 6.
Precisely. So for example...if a display reads the mastering display primaries metadata of P3, then it knows it has to map the gamut to an equivalent BT2020 percentage. So my thoughts are the metadata isn't forcing the display to map to a different gamut (P3), just to the correct percentage of BT2020 gamut.

Ryan
mascior is offline  
post #40 of 874 Old 06-20-2016, 09:37 AM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
I'm also wondering if it would be beneficial to have two sets of pattern suites. One for LCD with a master min luminanace of 0.05 and one for OLED master min luminance of 0.0005?

Ryan
Byronic Man likes this.
mascior is offline  
post #41 of 874 Old 06-20-2016, 03:12 PM
Member
 
carneb's Avatar
 
Join Date: Nov 2006
Location: Sydney, Australia
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 35
Quote:
Originally Posted by mascior View Post
Ok...so this is what I've put together regarding this entire "P3/BT2020" mass confusion. There is no consumer display trying to achieve any percentage of P3, only a percentage of BT2020. Now with that being said, 100% of P3 is roughly close to 75% of BT2020 with different x/y coordinates, especially in green. So my thinking is if the mastering display is utilizing P3 primaries (which is what everything is currently mastered in) it is up to the display to interpret that metadata and calculate that to a percentage of BT2020. I feel this is why changing the mastering metadata between P3 and BT2020 has no effect on my patterns during testing because in the end...the display will always scale to a percentage of BT2020. Now this is only my assumption from what I could gather...thoughts from anyone?

Ryan
I've posted a number of times about RGB values and haven't recieved any comments specifically about that. Maybe I'm not making my comments/questions clear so I'll start from the beginning to explain what I mean.

When a movie is made and finished on a mastering display with P3 primaries, a 100% sat/100% stimulus red pixel will have an RGB value of 940,64,64. Because consumer gear can't handle DCI-P3, this has to be converted to 2020, or to say it another way, put into a 2020 container. Once this conversion is done that red pixel now has an RGB value of roughly 727,104,64, and the metadata says that it was mastered on a P3 display.

When the movie is then played back on one of the current HDR/2020 capable TV's the tv reads the metadata and if the the mastering display primaries are completely inside the TV's native colour gamut the TV will just display that pixel as P3 red xy 0.68,0.32. with no conversion or mapping nescesary. If the mastering display primaries are outside the TV's native colour gamut then the TV should do some sort of colour mapping, but this may not be done depending on what the manufacturer has decided to do.

Now when you make your test windows, if you want to produce a 2020 red window xy 0.708,0.292, then you produce a window with RGB value 940,64,64, and whatever metadata you put in probably won't make a difference, so even if the metadata says P3 mastering primaries, the TV will still try to show the window as 2020 red using whatever mapping it has. If you want to produce a P3 red window xy 0.68,0.32 then you produce a window with RGB value of roughly 727,104,64 and the metadata says P3 mastering primaries. The TV will show this as a red P3 window if it falls within it's native gamut or otherwise will use colour mapping to display it.

This is my understanding of how things work.
carneb is online now  
post #42 of 874 Old 06-20-2016, 04:49 PM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Quote:
Originally Posted by carneb View Post
I've posted a number of times about RGB values and haven't recieved any comments specifically about that. Maybe I'm not making my comments/questions clear so I'll start from the beginning to explain what I mean.

When a movie is made and finished on a mastering display with P3 primaries, a 100% sat/100% stimulus red pixel will have an RGB value of 940,64,64. Because consumer gear can't handle DCI-P3, this has to be converted to 2020, or to say it another way, put into a 2020 container. Once this conversion is done that red pixel now has an RGB value of roughly 727,104,64, and the metadata says that it was mastered on a P3 display.

When the movie is then played back on one of the current HDR/2020 capable TV's the tv reads the metadata and if the the mastering display primaries are completely inside the TV's native colour gamut the TV will just display that pixel as P3 red xy 0.68,0.32. with no conversion or mapping nescesary. If the mastering display primaries are outside the TV's native colour gamut then the TV should do some sort of colour mapping, but this may not be done depending on what the manufacturer has decided to do.

Now when you make your test windows, if you want to produce a 2020 red window xy 0.708,0.292, then you produce a window with RGB value 940,64,64, and whatever metadata you put in probably won't make a difference, so even if the metadata says P3 mastering primaries, the TV will still try to show the window as 2020 red using whatever mapping it has. If you want to produce a P3 red window xy 0.68,0.32 then you produce a window with RGB value of roughly 727,104,64 and the metadata says P3 mastering primaries. The TV will show this as a red P3 window if it falls within it's native gamut or otherwise will use colour mapping to display it.

This is my understanding of how things work.
I completely understand what you are trying to say and I'm not saying that I totally disagree...It's that I've heard multiple explanations regarding this matter.

Ryan
mascior is offline  
post #43 of 874 Old 06-20-2016, 06:02 PM
AVS Forum Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 7,495
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 765 Post(s)
Liked: 659
Quote:
Originally Posted by carneb View Post
I've posted a number of times about RGB values and haven't recieved any comments specifically about that. Maybe I'm not making my comments/questions clear so I'll start from the beginning to explain what I mean.

When a movie is made and finished on a mastering display with P3 primaries, a 100% sat/100% stimulus red pixel will have an RGB value of 940,64,64. Because consumer gear can't handle DCI-P3, this has to be converted to 2020, or to say it another way, put into a 2020 container. Once this conversion is done that red pixel now has an RGB value of roughly 727,104,64, and the metadata says that it was mastered on a P3 display.

When the movie is then played back on one of the current HDR/2020 capable TV's the tv reads the metadata and if the the mastering display primaries are completely inside the TV's native colour gamut the TV will just display that pixel as P3 red xy 0.68,0.32. with no conversion or mapping nescesary. If the mastering display primaries are outside the TV's native colour gamut then the TV should do some sort of colour mapping, but this may not be done depending on what the manufacturer has decided to do.

Now when you make your test windows, if you want to produce a 2020 red window xy 0.708,0.292, then you produce a window with RGB value 940,64,64, and whatever metadata you put in probably won't make a difference, so even if the metadata says P3 mastering primaries, the TV will still try to show the window as 2020 red using whatever mapping it has. If you want to produce a P3 red window xy 0.68,0.32 then you produce a window with RGB value of roughly 727,104,64 and the metadata says P3 mastering primaries. The TV will show this as a red P3 window if it falls within it's native gamut or otherwise will use colour mapping to display it.

This is my understanding of how things work.
How do you get 727,104,64? If you assume a Rec. 2020 gamut and start with a CIE value of 0.6800, 0.3200, 0.2290, then the equivalent linear RGB value is 835, 282, 64. This assumes a 2.22 gamma. A 2.4 gamma is slightly different. The method is to convert the xyY value into R'G'B' using an inverted Rec. 2020 XYZ matrix and then to apply the relevant degamma to get the linear RGB value in %. Finally convert from %RGB to code values.

Also, what do you mean when you write "Because consumer gear can't handle DCI-P3"? Consumer gear is already achieving a gamut of very close to this. It is Rec. 2020 that consumer gear is incapable of reproducing.

Finally, are you saying that the target color coordinates that the consumer should aim for are the DCi-P3 coordinates, assuming that the display is capable of displaying them? This passage certainly seems to say this:
"When the movie is then played back on one of the current HDR/2020 capable TV's the tv reads the metadata and if the the mastering display primaries are completely inside the TV's native colour gamut the TV will just display that pixel as P3 red xy 0.68,0.32."

If this is so, it is unclear what role Rec. 2020 plays in this at all. The content is mastered on a DCi-P3 display and it is reproduced in the home of a TV with DCi-P3 primaries.

In order to maintain Rec. 2020 as the reference gamut on consumer displays, it seems to me the approach should be to calibrate to the most saturated xy coordinate that fits within the display gamut AND is also consistent with Rec. 2020 hue lines.
For example, the most saturated green that fits within the DCi-P3 gamut that also maintains the Rec. 2020 hue line is approximately 50% Rec. 2020 saturation. This is an RGB value of approximately 486, 902, 486. Ironically, this results in a LESS saturated green than you would have if the standard had simply been DCi-P3.


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


Last edited by TomHuffman; 06-20-2016 at 06:37 PM.
TomHuffman is offline  
post #44 of 874 Old 06-20-2016, 09:38 PM
Member
 
carneb's Avatar
 
Join Date: Nov 2006
Location: Sydney, Australia
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 35
Quote:
Originally Posted by TomHuffman View Post
How do you get 727,104,64?
I took xyY 0.68, 0.32, 0.2292, coverted to XYZ 0.487, 0.229, 0, then using the formulas on Bruce Lindblooms site did the Matrix conversion to BT2020 Linear rgb 773, 47,0. I then converted to video legal rgb values and got 727,104,64. I wasn't sure it was correct, and after reading your reply I realise I haven't taken into account the 'gamma'. Using ST2084 I work out that the companded RGB values are 994,683,0 or in video legal values 914,648,64.

To be honest, I wouldn't bet my house on those numbers, but the point of posting the numbers wasn't to be accurate(that's why I said "roughly"), it was to show that they are NOT 940,64,64.

Quote:
Also, what do you mean when you write "Because consumer gear can't handle DCI-P3"? Consumer gear is already achieving a gamut of very close to this. It is Rec. 2020 that consumer gear is incapable of reproducing.
I've read on this forum a number of times that there is no consumer standard for getting P3 into a TV. That's why it has to first be converted to 2020. I meant that the TV can't accept a direct P3 input.

Quote:
Finally, are you saying that the target color coordinates that the consumer should aim for are the DCi-P3 coordinates, assuming that the display is capable of displaying them? This passage certainly seems to say this:
"When the movie is then played back on one of the current HDR/2020 capable TV's the tv reads the metadata and if the the mastering display primaries are completely inside the TV's native colour gamut the TV will just display that pixel as P3 red xy 0.68,0.32."
I'm not talking about targeting any colour coordinates, I'm simply saying that for my example movie that has been mastered for P3 colour the TV should show that as a P3 colour. In my example I used red 0.68, 0.32. If that colour falls within the TVs native gamut then it should be displayed as red 0.68,0.32, otherwise if it is not within the TVs native gamut the TV will map the colour to some other colour within it's native gamut.

Quote:
If this is so, it is unclear what role Rec. 2020 plays in this at all. The content is mastered on a DCi-P3 display and it is reproduced in the home of a TV with DCi-P3 primaries.
I think it would be easier if the standard would have been P3, but we have to deal with the standard that is in place and everything has to be converted to 2020.

Quote:
In order to maintain Rec. 2020 as the reference gamut on consumer displays, it seems to me the approach should be to calibrate to the most saturated xy coordinate that fits within the display gamut AND is also consistent with Rec. 2020 hue lines.
For example, the most saturated green that fits within the DCi-P3 gamut that also maintains the Rec. 2020 hue line is approximately 50% Rec. 2020 saturation. This is an RGB value of approximately 486, 902, 486. Ironically, this results in a LESS saturated green than you would have if the standard had simply been DCi-P3.
I'm not sure if that would work when using 2084 EOTF as when using this and plotting the 25,50, and 75% saturation points on the CIE diagram they are not linearly spaced unlike what is shown on your attached diagram.
carneb is online now  
post #45 of 874 Old 06-20-2016, 09:50 PM
AVS Forum Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 7,495
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 765 Post(s)
Liked: 659
Quote:
Originally Posted by carneb View Post
I took xyY 0.68, 0.32, 0.2292, coverted to XYZ 0.487, 0.229, 0, then using the formulas on Bruce Lindblooms site did the Matrix conversion to BT2020 Linear rgb 773, 47,0. I then converted to video legal rgb values and got 727,104,64. I wasn't sure it was correct, and after reading your reply I realise I haven't taken into account the 'gamma'. Using ST2084 I work out that the companded RGB values are 994,683,0 or in video legal values 914,648,64.
The Lindbloom site does not include RGB conversion matrices for Rec. 2020. You must have used a matrix for a different gamut.

Quote:
Originally Posted by carneb View Post
I've read on this forum a number of times that there is no consumer standard for getting P3 into a TV. That's why it has to first be converted to 2020. I meant that the TV can't accept a direct P3 input.
P3 is just a series of color coordinates. There is nothing to prevent a consumer TV from displaying those. In fact, many do so nearly now. I think what you are referring to has less to do with a technical limitation than it does with some legal restrictions regarding the use of this particular gamut, which was originally developed for commercial cinema. The inability of TVs to display Rec. 2020 is a technical limitation.

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

TomHuffman is offline  
post #46 of 874 Old 06-20-2016, 11:01 PM
Member
 
carneb's Avatar
 
Join Date: Nov 2006
Location: Sydney, Australia
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 35
Quote:
Originally Posted by TomHuffman View Post
The Lindbloom site does not include RGB conversion matrices for Rec. 2020. You must have used a matrix for a different gamut.
I didn't use the online calculator. I calculated the transformation matrix [M] for BT2020 then [rgb]=[M]^-1*[XYZ]. I did the calcs in excel.

Quote:
Originally Posted by TomHuffman View Post
P3 is just a series of color coordinates. There is nothing to prevent a consumer TV from displaying those. In fact, many do so nearly now. I think what you are referring to has less to do with a technical limitation than it does with some legal restrictions regarding the use of this particular gamut, which was originally developed for commercial cinema.
That's what I meant
carneb is online now  
post #47 of 874 Old 06-20-2016, 11:56 PM
Advanced Member
 
EvLee's Avatar
 
Join Date: Dec 2009
Posts: 577
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 332 Post(s)
Liked: 300
Quote:
Originally Posted by mascior View Post
Ok...so this is what I've put together regarding this entire "P3/BT2020" mass confusion. There is no consumer display trying to achieve any percentage of P3, only a percentage of BT2020. Now with that being said, 100% of P3 is roughly close to 75% of BT2020 with different x/y coordinates, especially in green. So my thinking is if the mastering display is utilizing P3 primaries (which is what everything is currently mastered in) it is up to the display to interpret that metadata and calculate that to a percentage of BT2020. I feel this is why changing the mastering metadata between P3 and BT2020 has no effect on my patterns during testing because in the end...the display will always scale to a percentage of BT2020. Now this is only my assumption from what I could gather...thoughts from anyone?

Ryan
P3 is the target for minimum gamut to receive UHD Alliance certification. So there are quite a few consumer displays out there that are trying to achieve P3.

One possibly reason you may not be seeing any effect from altering the mastering display metadata could be that the displays are simply ignoring it. There is no specification for how the mastering display metadata should be used, so one option (and the simplest to implement) is just ignore it. Said another way - The color processing in the current tv's is not particularly sophisticated.
EvLee is offline  
post #48 of 874 Old 06-21-2016, 01:47 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,263
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 425 Post(s)
Liked: 721
The Ultra HD specification, while using Rec2020 as the target (envelope) colours space, actually specifies that any Ultra HD display only has to reach 90% of DCI-P3 to be accepted as a UHDTV display - and 90% of DCI-P3 is not not that much larger than Rec709.

See: http://www.lightillusion.com/uhdtv.html
There is an example Gamut Coverage graph towards the bottom of the page.

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is online now  
post #49 of 874 Old 06-21-2016, 03:11 AM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,275
Mentioned: 89 Post(s)
Tagged: 0 Thread(s)
Quoted: 1316 Post(s)
Liked: 1064
R.Masciola's HDR-10 UHD Test Patterns

Quote:
Originally Posted by mascior View Post
I would agree...however, the mastering display color primaries metadata doesn't seem to make any difference one way or the other...at least with the displays I tested.


Did you look at what it does with a fully saturated green input that's specified to have BT.2020 primaries? If it's color managed it might clip it at the same hue instead of the more yellow P3 location, or somewhere along the gamut edge between 50% BT.2020 and P3.

Last edited by zoyd; 06-21-2016 at 03:39 AM.
zoyd is offline  
post #50 of 874 Old 06-21-2016, 06:07 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,218
Mentioned: 53 Post(s)
Tagged: 1 Thread(s)
Quoted: 1555 Post(s)
Liked: 1998
Quote:
Originally Posted by carneb View Post
Now when you make your test windows, if you want to produce a 2020 red window xy 0.708,0.292, then you produce a window with RGB value 940,64,64

Hi, 100% Red Saturation with 100% Stimulus Level for a 10.000nits peak output display has 940.64.64 RGB Triplet.

Things are very complex for HDR this time. Seems the current displays don't follow the primary's info of the metadata and they are doing their own, using an unknown for each model internal gamut/tone mapping method.



A good example is that HDR video from Sony that plays without problem to any HDR display, but if you look the metadata is problematic:

Color range: Limited
Color primaries: BT.2020
Transfer characteristics: SMPTE ST 2084
Matrix coefficients: BT.2020 non-constant
Mastering display color primar: R: x=1.000000 y=1.000000, G: x=1.000000 y=1.000000, B: x=1.000000 y=1.000000, White point: x=1..000000 y=1.000000
Mastering display luminance: min: 0.1000 cd/m2, max: 0.5000 cd/m2


I believe that that secret/unknown mapping of each display that has been calculated from each display model from the factory based to the display's performance (uncalibrated status) at it's HDR mode, any usage of calibration controls to calibrate the HDR mode will produce poorer color performance, since the display don't know if you have calibrated or not...what clipping point you used... To make things more calibration-friendly, my idea that may work better for the companies is to introduce a menu to each model where the user will enter the measured (calibrated) primaries (xyY) and Peak/Black level output, to help the display make better internal mapping. Also to add the option to disable the internal mapping, to be easier for external 3D LUT devices combined with calibration software to do the whole job.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5

Last edited by ConnecTEDDD; 06-21-2016 at 06:25 AM.
ConnecTEDDD is online now  
post #51 of 874 Old 06-23-2016, 10:28 AM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
I just wanted to give a quick update on the project and bullet a few things. I'm near completion with the last of the encoding. I did decide to go with two separate pattern suites/sets...one for OLED with a min of 0.0005 and one for LCD with a min of 0.05 with the following metadata:

OLED
Color primaries : BT.2020
Transfer characteristics : SMPTE ST 2084
Matrix coefficients : BT.2020 non-constant
Mastering display color primaries : R: x=0.680000 y=0.320000, G: x=0.265000 y=0.690000, B: x=0.150000 y=0.060000, White point: x=0.312700 y=0.329000
Mastering display luminance : min: 0.0005 cd/m2, max: 1000.0000 cd/m2
Maximum Content Light Level : 1000 cd/m2
Maximum Frame-Average Light Level : 400 cd/m2

LCD
Color primaries : BT.2020
Transfer characteristics : SMPTE ST 2084
Matrix coefficients : BT.2020 non-constant
Mastering display color primaries : R: x=0.680000 y=0.320000, G: x=0.265000 y=0.690000, B: x=0.150000 y=0.060000, White point: x=0.312700 y=0.329000
Mastering display luminance : min: 0.05 cd/m2, max: 1000.0000 cd/m2
Maximum Content Light Level : 1000 cd/m2
Maximum Frame-Average Light Level : 400 cd/m2

I also pickup a very slick software from Florian Friedrich called "SEIedit" which allows me to modify the SEI/VUI metadata without recoding all of my files! Big thumbs up to Florian...as I feel this is an essential tool when dealing with HDR encoded files and the constant change of metadata! This project has taken a little longer than expected to launch...however, this is not a project that one would want to rush!

Ryan

Last edited by mascior; 06-23-2016 at 10:34 AM.
mascior is offline  
post #52 of 874 Old 06-26-2016, 01:30 AM
Member
 
carneb's Avatar
 
Join Date: Nov 2006
Location: Sydney, Australia
Posts: 89
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 35
Quote:
Originally Posted by TomHuffman View Post
In order to maintain Rec. 2020 as the reference gamut on consumer displays, it seems to me the approach should be to calibrate to the most saturated xy coordinate that fits within the display gamut AND is also consistent with Rec. 2020 hue lines.
For example, the most saturated green that fits within the DCi-P3 gamut that also maintains the Rec. 2020 hue line is approximately 50% Rec. 2020 saturation. This is an RGB value of approximately 486, 902, 486. Ironically, this results in a LESS saturated green than you would have if the standard had simply been DCi-P3.

Quote:
Originally Posted by carneb View Post
I'm not sure if that would work when using 2084 EOTF as when using this and plotting the 25,50, and 75% saturation points on the CIE diagram they are not linearly spaced unlike what is shown on your attached diagram.



My apologies, I was wrong about this. A software bug led me to believe that the points moved significantly when ST2084 is selected as the EOTF

On another topic, what calibration software will be compatible with the test patterns? If I want to use them with HCFR will it need an option to select these patterns (as there is for the current MCD) or will it work with HCFR as is?
carneb is online now  
post #53 of 874 Old 06-28-2016, 11:40 AM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Quote:
Originally Posted by carneb View Post
On another topic, what calibration software will be compatible with the test patterns? If I want to use them with HCFR will it need an option to select these patterns (as there is for the current MCD) or will it work with HCFR as is?
I'll have three available workflows to accommodate hcfr, Chromapure, and Calman.

Ryan
WiFi-Spy and BlueChris like this.
mascior is offline  
post #54 of 874 Old 06-29-2016, 05:59 AM
Advanced Member
 
DaMacFunkin's Avatar
 
Join Date: Mar 2008
Location: Manchester, Not the one by the sea
Posts: 846
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 611 Post(s)
Liked: 415
Put my name on the list for CP workflow, cheers.
DaMacFunkin is offline  
post #55 of 874 Old 07-04-2016, 10:27 AM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Happy 4th of July to everyone here on the forum! Just a quick update.
-
  • Basic Setup Patterns - 100% Complete
  • Advanced Setup Patterns - 100% Complete
  • Calman Workflow - 95% Complete
  • ChromaPure Workflow - 75% Complete
  • HCFR Workflow - 65% Complete
  • Miscellaneous Patterns - 100% Complete
Quite a few patterns have been added, a few modified, and some deleted from the project. However, all patterns 100% focused on HDR-10, with current display limitation in mind, and the future. The request for percentages of BT2020(being the main focus of the project) have been added...especially 50% and/or 100 cd/m2. I have also added P3 in BT2020 container patterns which has also been requested. The workflows are coming along nicely and track the software modules perfectly. Thank you everyone for the continued interest/support and hope to have this project launched soon!

Ryan
mascior is offline  
post #56 of 874 Old 07-05-2016, 11:52 PM
Member
 
Dimitri_ISF's Avatar
 
Join Date: Jan 2011
Posts: 97
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 11
Very interested in the Calman workflow!
Dimitri_ISF is offline  
post #57 of 874 Old 07-06-2016, 11:31 AM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Quote:
Originally Posted by Dimitri_ISF View Post
Very interested in the Calman workflow!
The Calman workflow will go as followed:
-
  1. Pre Calibration View Module
  2. White Balance Module
  3. Grayscale Module
  4. CMS Adjust Module
  5. ColorChecker Module(along with DCI Red, Green, & Blue)
  6. Saturation Sweeps Module
  7. Post Calibration View Module
Ryan
mascior is offline  
post #58 of 874 Old 07-06-2016, 12:05 PM
Member
 
Dimitri_ISF's Avatar
 
Join Date: Jan 2011
Posts: 97
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 11
Ok, thank you.

How will I know when the disc(download) is ready?.

Dimitri.
Dimitri_ISF is offline  
post #59 of 874 Old 07-06-2016, 12:17 PM - Thread Starter
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 633
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 259 Post(s)
Liked: 347
Quote:
Originally Posted by Dimitri_ISF View Post
Ok, thank you.

How will I know when the disc(download) is ready?.

Dimitri.
I will post when the project is completed and ready for download. However, you mentioned "disc(download)"...these patterns will only be available in .ts format/container for playback on HDR-10 enabled displays via usb. Thanks.

Ryan
mascior is offline  
post #60 of 874 Old 07-07-2016, 08:28 AM
AVS Forum Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,588
Mentioned: 52 Post(s)
Tagged: 0 Thread(s)
Quoted: 551 Post(s)
Liked: 1317
Any way we could get a disc based version?


Sent from my iPhone using Tapatalk

touring ISF/THX calibrator with Jeti 1211 and Klein K-10
Latest reviews:JVC RS500 and 600, JVC RS400, LG 65C7P, JVC RS420
Copied settings measured, older LG OLED calibration tips, Review index
Chad B is online now  
Sponsored Links
Advertisement
 
Reply Display Calibration

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