HDR Calibration & Discussion - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 424Likes
Reply
 
Thread Tools
post #1 of 759 Old 09-10-2015, 10:45 AM - Thread Starter
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,246
Mentioned: 54 Post(s)
Tagged: 1 Thread(s)
Quoted: 1565 Post(s)
Liked: 2013
HDR Calibration & Discussion

I'm opening this thread to add all the current and any upcoming information about the available software solutions that support HDR calibration/profiling, HDR calibration methods/workflows/hardware/pattern generators etc.

Here is the available solutions we have now:

CalMAN 5

HDR Calibration Support: ST.2084 Gamma available from CalMAN Ultimate/Studio 5.4.1 released @ 5 March 2015 (or later version).
ST.2084 Gamma available from CalMAN Enthusiast 5.6.0 RC1 released @ 7 October 2015 (or later version).

License Level Required for HDR Calibration: CalMAN Ultimate, CalMAN Studio, CalMAN Enthusiast.

ChromaPure 2.x

HDR Support: Not yet supported. Expected be to supported in ChromaPure 3.x.

License Level Required for HDR Calibration: To be announced.

HCFR

HDR Calibration Support: ST.2084 Gamma available from HCFR 3.3.0 released @ 6 June 2015 (or later version). *Supports HDR Parametric Gamma also.

License Level Required for HDR Calibration: Free (Open Source) Software

LightSpace

HDR Calibration Support: ST.2084 Gamma available from LightSpace 6.6.7.2061 released @ 25 March 2015 (or later version).

License Level Required for HDR Calibration: All license levels support it.

Notes: LightSpace supports HDR Parametric Gamma capability also. Users can download an example HDR Parametric Gamma profile from LightSpace website; this one for the Sony BVM-X300, which is used by some major Post Facilities, including Light Iron – who were one of the first to start using the Sony display.

dispcalGUI (Powered by ArgyllCMS)

HDR Calibration Support: ST.2084 Gamma available from dispcalGUI 3.0.3 released @ 6 July 2015 (or later version).

License Level Required for HDR Calibration: Free (Open Source) Software.
Chad B, spart1cle, Elix and 2 others like this.

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; 10-08-2015 at 12:06 AM.
ConnecTEDDD is online now  
Sponsored Links
Advertisement
 
post #2 of 759 Old 09-10-2015, 10:54 AM - Thread Starter
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,246
Mentioned: 54 Post(s)
Tagged: 1 Thread(s)
Quoted: 1565 Post(s)
Liked: 2013
Here is the list with the published Peak Luminance Range Limit for each meter available in market:

Colorimeters

X-Rite DTP-94 up to 1.000 cd/m2
X-Rite i1Display 2 up to 3.000 cd/m2
Sencore's CP-6000 up to 1.000 cd/m2
X-Rite Chroma 5 up to 1.000 cd/m2
X-Rite Hubble up to 1.350 cd/m2
Sencore OTC-1000 up to 1.350 cd/m2
X-Rite i1Display PRO (i1d3 OEM/Retail i1d3) up to 1.000 cd/m2
SpectraCAL C6 (Branded OEM i1d3) up to 1.000 cd/m2
SpectraCAL C6-HDR (Branded OEM i1d3) up to 1.300 cd/m2
Datacolor Spyder 2 up to 5.000 cd/m2
Datacolor Spyder 3 up to 5.000 cd/m2
Datacolor Spyder 4 up to 5.000 cd/m2
Datacolor Spyder 5 up to 5.000 cd/m2
BasICColor Discus up to 2.500 cd/m2
Colorimetry Research CR-100 up to 5.140 cd/m2
Klein K-80 up to 10.000 cd/m2
Klein K-10A up to 10.000 cd/m2
Minolta CS-100A up to 300.000 cd/m2
Minolta CA-210 up to 1.000 cd/m2
Minolta CA-310 up to 1.000 cd/m2
Minolta CS-200 up to 20.000.000 cd/m2

Spectroradiometers/Spectrophotometers

X-Rite ColorMunki up to 1.000 cd/m2
X-Rite i1PRO1 up to 300 cd/m2
X-Rite i1PRO2 up to 1.200 cd/m2
JETI 1201 up to 70.000 cd/m2 (using optional JETI filters... up to 75.000/250.000 cd/m2)
JETI 1211 up to 2.500 cd/m2 (using optional JETI filters... up to 10.000/25.000/50.000/75.000/250.000 cd/m2)
JETI 1501 up to 150.000 cd/m2
JETI 1511 up to 150.000 cd/m2
Colorimetry Research CR-250RH up to 154.180 cd/m2
Photoreseach PR-650 up to 5.000 cd/m2
Photoreseach PR-655 up to 15.000 cd/m2
Photoreseach PR-670 up to 8.566.000 cd/m2
Photoreseach PR-680 up to 17.130.000 cd/m2
Minolta CS-1000 up to 80.000 cd/m2
Minolta CS-2000 up to 500.000 cd/m2
Minolta CS-2000A up to 500.000 cd/m2

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; 08-27-2016 at 12:24 PM.
ConnecTEDDD is online now  
post #3 of 759 Old 09-10-2015, 10:55 AM - Thread Starter
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,246
Mentioned: 54 Post(s)
Tagged: 1 Thread(s)
Quoted: 1565 Post(s)
Liked: 2013
** Reserved For HDR 3D LUT Hardware Boxes **

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
ConnecTEDDD is online now  
 
post #4 of 759 Old 09-10-2015, 10:55 AM - Thread Starter
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,246
Mentioned: 54 Post(s)
Tagged: 1 Thread(s)
Quoted: 1565 Post(s)
Liked: 2013
** Reserved For HDR Pattern Generators **

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; 09-10-2015 at 11:48 AM.
ConnecTEDDD is online now  
post #5 of 759 Old 09-10-2015, 11:06 AM
Advanced Member
 
CalWldLif's Avatar
 
Join Date: Feb 2011
Location: South Bay'Los Angeles California.
Posts: 894
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 248 Post(s)
Liked: 69
Thanks for the info and break downs.

Loving D65
CalWldLif is offline  
post #6 of 759 Old 09-15-2015, 01:13 PM - Thread Starter
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,246
Mentioned: 54 Post(s)
Tagged: 1 Thread(s)
Quoted: 1565 Post(s)
Liked: 2013
CalMAN Studio is also supporting ST.2084 (HDR EOTF).

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
ConnecTEDDD is online now  
post #7 of 759 Old 09-15-2015, 03:15 PM
AVS Forum Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 5,903
Mentioned: 16 Post(s)
Tagged: 1 Thread(s)
Quoted: 1655 Post(s)
Liked: 2436
Thanks for all the work, Ted. I have an HDR ordered - LG 55EF9500 OLED and am looking forward to working with LightSpace.
ConnecTEDDD likes this.

Buzz
THX Certified Level II Video Calibrator



 
buzzard767 is offline  
post #8 of 759 Old 09-16-2015, 12:20 AM - Thread Starter
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,246
Mentioned: 54 Post(s)
Tagged: 1 Thread(s)
Quoted: 1565 Post(s)
Liked: 2013
Quote:
Originally Posted by buzzard767 View Post
Thanks for all the work, Ted. I have an HDR ordered - LG 55EF9500 OLED and am looking forward to working with LightSpace.
Hi Buzz, very good news, let us know how it performs/calibrates.

BTW here are the 2 HDR clips to test:



Exodus - Ultra-HD HDR



Life of Pi - Ultra-HD HDR

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
ConnecTEDDD is online now  
post #9 of 759 Old 12-10-2015, 04:01 AM - Thread Starter
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,246
Mentioned: 54 Post(s)
Tagged: 1 Thread(s)
Quoted: 1565 Post(s)
Liked: 2013
HDR News Updates

CalMAN 5.6.1 RC1 release added for CalMAN Expert, Professional & Ultimate license levels:

HDR-10 support for Quantum Data 780 and 804 pattern generators. (This requires firmware version 15092260 or higher)
HDR-10 support for Astro Design VG-876 & VG-877 Video Signal Generators.

HDR-10 is a standard used for mastering content, it has metadata content describing the mastering monitor peak brightness and native gamut. HDR-10 doesn't support consumer devices; displays/projectors etc.

LightIllussion added a page related with HDR here: http://www.lightillusion.com/hdr.html

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
ConnecTEDDD is online now  
post #10 of 759 Old 12-12-2015, 01:37 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 ConnecTEDDD View Post
HDR News Updates

CalMAN 5.6.1 RC1 release added for CalMAN Expert, Professional & Ultimate license levels:

HDR-10 support for Quantum Data 780 and 804 pattern generators. (This requires firmware version 15092260 or higher)
HDR-10 support for Astro Design VG-876 & VG-877 Video Signal Generators.

HDR-10 is a standard used for mastering content, it has metadata content describing the mastering monitor peak brightness and native gamut. HDR-10 doesn't support consumer devices; displays/projectors etc.

LightIllussion added a page related with HDR here: http://www.lightillusion.com/hdr.html
That LightIllusion page is wrong in so many ways. Everything from their explanation of useful dynamic range, to the impact of viewing distance, to the claim that 650 nits is excessive. Here is an obvious question: If 650 was excessive then why wasn't the standard limited to 650? And suggesting you can clip the 2084 curve to your display white luminance? That is a guaranteed way to destroy the image quality. Just, wow. They really don't get it.
EvLee is offline  
post #11 of 759 Old 12-12-2015, 01:44 PM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,276
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 429 Post(s)
Liked: 731
Quote:
Originally Posted by EvLee View Post
That LightIllusion page is wrong in so many ways. Everything from their explanation of useful dynamic range, to the impact of viewing distance, to the claim that 650 nits is excessive. Here is an obvious question: If 650 was excessive then why wasn't the standard limited to 650? And suggesting you can clip the 2084 curve to your display white luminance? That is a guaranteed way to destroy the image quality. Just, wow. They really don't get it.
Actually, the technical data is 100% correct based on the ST2084 standard - maybe you should read it?
It defines exactly those clipping points as shown.

And the comments about viewing distance are also based on well documented tests - the image shown is from one of those test. Do you sit that close to your display?
Additionally, resolution is part of such a discussion, but that's another story.

And the comments about what is an acceptable Nits level are from actual work performed on actual HDR displays - diretc experience!

Have you been able to run test on different HDR displays at different Nits levels?
That is what we did...

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is online now  
post #12 of 759 Old 12-12-2015, 02:15 PM
 
Join Date: Nov 2015
Posts: 57
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 44 Post(s)
Liked: 16
If your company is in charge of HDR (progress and development) don't think y'all are. Please don't kill it with your 650 nits.

Hoping Happy Consumer


Quote:
Originally Posted by Light Illusion View Post
Actually, the technical data is 100% correct based on the ST2084 standard - maybe you should read it?
It defines exactly those clipping points as shown.

And the comments about viewing distance are also based on well documented tests - the image shown is from one of those test. Do you sit that close to your display?
Additionally, resolution is part of such a discussion, but that's another story.

And the comments about what is an acceptable Nits level are from actual work performed on actual HDR displays - diretc experience!

Have you been able to run test on different HDR displays at different Nits levels?
That is what we did...

Steve
Amzie Williams is offline  
post #13 of 759 Old 12-12-2015, 02:28 PM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,276
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 429 Post(s)
Liked: 731
The problem with HDR is it is presently the most ill defined concept we have seen in a long while.

There are no standards that as yet are fully defined, and the various hardware manufacturers have gone off half cocked in an attempt to sell more displays.

As a 'colour management' company we have to research all the various proposals to make sure LightSpace CMS can manage them as required, and in doing that we obviously have to have a lot of direct access and experience of different displays.

Having sat in front of a large number of HDR displays we have our own 'feelings' on what works and what doesn't.
But we have to be able to deal with all variants.

For example, one of the points that often comes up when calibrating a display is 'fatigue' levels.
Excessive HDR causes (can cause?) serious eye fatigue.

There is also the issue of the way dynamic backlighting works.
ABL with plasmas was a real pain, and HDR is no different - potentially worse.
The last HDR display we played with had a real issue with the screen taking serious time to dim after a bright scene.

And while HDR is Hight Dynamic Range, UWG has also been brought into the mix.
This is causing yet another level of total standard failure.

This is really VHS vs. Betamax, but about 100 times worse...

Steve
mpgxsvcd likes this.

Steve Shaw
LIGHT ILLUSION


Last edited by Light Illusion; 12-12-2015 at 02:47 PM. Reason: SP.
Light Illusion is online now  
post #14 of 759 Old 12-13-2015, 03:00 AM
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 Light Illusion View Post
Actually, the technical data is 100% correct based on the ST2084 standard - maybe you should read it?
It defines exactly those clipping points as shown.

And the comments about viewing distance are also based on well documented tests - the image shown is from one of those test. Do you sit that close to your display?
Additionally, resolution is part of such a discussion, but that's another story.

And the comments about what is an acceptable Nits level are from actual work performed on actual HDR displays - diretc experience!

Have you been able to run test on different HDR displays at different Nits levels?
That is what we did...

Steve
2084 doesn't define reproduction at the home. It defines the EOTF for mastering, hence the title "High Dynamic Range EOTF of Mastering Reference Display". So you have that part wrong for starters. Home displays aren't expected to track the reference display exactly for a number of reasons, and the behavior of home displays was intentionally left out of that document. However, the one thing they definitely should not be doing is hard clipping the highlights. Unless the home display can match the luminance of the mastering display as reported in the 2086 metadata, it needs to perform some form of highlight compression. I'd love to hear what you are using as your reference display. My direct experience is actually working with filmmakers, sitting in color correction and guess what… 650 does not cut it. You can't just put up a picture of a bright sky and use that to set a bar on where peak luminance should fall. I've made this point in other threads, even a 100 nits picture can be uncomfortable to look at if it's poorly composed or edited. However, if you give a skilled DP and colorist the opportunity to tailor a composition for HDR, you can easily have values going well over 1000 nits without it being uncomfortable to view.
Elix likes this.
EvLee is offline  
post #15 of 759 Old 12-13-2015, 03:24 AM
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 Light Illusion View Post
The problem with HDR is it is presently the most ill defined concept we have seen in a long while.

There are no standards that as yet are fully defined, and the various hardware manufacturers have gone off half cocked in an attempt to sell more displays.

As a 'colour management' company we have to research all the various proposals to make sure LightSpace CMS can manage them as required, and in doing that we obviously have to have a lot of direct access and experience of different displays.

Having sat in front of a large number of HDR displays we have our own 'feelings' on what works and what doesn't.
But we have to be able to deal with all variants.

For example, one of the points that often comes up when calibrating a display is 'fatigue' levels.
Excessive HDR causes (can cause?) serious eye fatigue.

There is also the issue of the way dynamic backlighting works.
ABL with plasmas was a real pain, and HDR is no different - potentially worse.
The last HDR display we played with had a real issue with the screen taking serious time to dim after a bright scene.

And while HDR is Hight Dynamic Range, UWG has also been brought into the mix.
This is causing yet another level of total standard failure.

This is really VHS vs. Betamax, but about 100 times worse...

Steve
Remember regular old HD? It wasn't fully defined until um… let's see when was ITU-R 1886 published? 2011. But it seemed to work pretty well before then. Here's what you are missing. This new container was made both big (10k nits) and wide (2020 primaries) so it wouldn't have to be redefined in another 2, 5, 10 years down the road every time display technology advances. It was made as large as reasonable with the knowledge that only a limited portion of it would be used initially. This is where 2086 comes into play. You start with a smaller utilization and grow into the larger range over time. Even with HD, 100 nits for white was completely undocumented but informally agreed on in practice because that was all you could get out of the Sony reference CRTs. So what is going to happen with HDR and WCG? Well, there are two reference displays currently available to studios. Those will set a limit on what actually produced for distribution. By the nature of the industry and the need to interchange content between facilities, they will converge on a common (although not necessarily SMPTE standardized) target. But all you'll need to do is take a look inside the 2086 metadata to find out what that is. My suggestion is wait until CES, wait until test discs come out and then try again.

Also, as to your ABL comment… one of the most linear (flattest APL curve) response displays I have ever seen is HDR and intended for consumer market. I have also seen some very junk ones as well, and I think the biggest risk for HDR is that too many low tier displays, especially with low quality dimming like you described, will be forced into the market by manufacturers and advertised as "HDR". There is a real danger that these will compromise the consumer impression of the format, when it is capable of so much more when done right.
Elix likes this.
EvLee is offline  
post #16 of 759 Old 12-13-2015, 03:27 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,276
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 429 Post(s)
Liked: 731
Quote:
Originally Posted by EvLee View Post
2084 doesn't define reproduction at the home. It defines the EOTF for mastering, hence the title "High Dynamic Range EOTF of Mastering Reference Display". So you have that part wrong for starters. Home displays aren't expected to track the reference display exactly for a number of reasons, and the behavior of home displays was intentionally left out of that document. However, the one thing they definitely should not be doing is hard clipping the highlights. Unless the home display can match the luminance of the mastering display as reported in the 2086 metadata, it needs to perform some form of highlight compression. I'd love to hear what you are using as your reference display. My direct experience is actually working with filmmakers, sitting in color correction and guess what… 650 does not cut it. You can't just put up a picture of a bright sky and use that to set a bar on where peak luminance should fall. I've made this point in other threads, even a 100 nits picture can be uncomfortable to look at if it's poorly composed or edited. However, if you give a skilled DP and colorist the opportunity to tailor a composition for HDR, you can easily have values going well over 1000 nits without it being uncomfortable to view.
The idea of home calibration is to match the calibration/environment that was used for the grading - as far as is possible.
That is the whole idea - so ST2084 defines the requirements for all viewing of material that was graded using ST2084...

The 'hard clipping is indeed that way ST2084 is specified, but is also part of why HDR is so badly defined.
The actual application should 'probably' be to roll-off the clip, but that is not defined anywhere.
(But, that's easy to do with LightSpace CMS when used for calibration of HDR.)

However, as the peak brightness areas for ST2084 are just the main specular highlights, as shown in the histogram, clipping will actually not be as 'harsh' as first appears - but still not ideal.

As a company, we are attempting to manage and work to the HDR specifications, and they are all over the place.
Personally, I think HLG is a far better approach, as it is much more realistic in what it is attempting.

All the members of Light Illusion are actually professionals in the Film and TV industry.
I have personally graded (and supervised) many films, and we all continue to consults with many facilities (as well as manufacturers) helping define colour workflows and pipelines. As such we spend a lot of time doing this for real.

We have seen/worked with all the professional HDR displays presently available, and many of the consumer models.

All the comments I make are only ever based on direct experience.
(But often my personal thoughts on that experience as it helps others to define their own thoughts and views)

Steve

Steve Shaw
LIGHT ILLUSION


Last edited by Light Illusion; 12-13-2015 at 03:45 AM. Reason: SP - as pointed out!
Light Illusion is online now  
post #17 of 759 Old 12-13-2015, 03:42 AM
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 Light Illusion View Post
The idea of home calibration is to match the calibration/environment that was used for the grading - as far as is possible.
That is the whole idea - so ST2084 defines the requirements for all viewing of material that was graded using ST2084...

The 'hard clipping is indeed that way ST2084 is specified, but is also part of why HDR is so badly defined.
The actual application should 'probably' be to roll-off the clip, but that is not defined anywhere.
(But, that's easy to do with LightSpace CMS when used for calibration of HDR.)

However, as the peak brightness areas for HST2084 are just the main spectral highlights, as shown in the histogram, clipping will actually not be as 'harsh' as first appears - but still not ideal.

As a company, we are attempting to manage and work to the HDR specifications, and they are all over the place.
Personally, I think HLG is a far better approach, as it is much more realistic in what it is attempting.

All the members of Light Illusion are actually professionals in the Film and TV industry.
I have personally graded (and supervised) many films, and we all continue to consults with many facilities (as well as manufacturers) helping define colour workflows and pipelines. As such we spend a lot of time doing this for real.

We have seen/worked with all the professional HDR displays presently available, and many of the consumer models.

All the comments I make are only ever based on direct experience.
(But often my personal thoughts on that experience as it helps others to define their own thoughts and views)

Steve
Specular highlights. Not spectral. [Note: I see you corrected this. Thank you! ]

You still are misinterpreting 2084. You are not supposed to see hard clipping at the display. You need to have a limiter in your color grading system so that you do not feed the display with values that force it into the region of clipping behavior. This is nothing new. When you grade digital cinema, you don't do it directly in XYZ space do you? So if your display can not go over 1000 nits for example, you should not be sending it any 2084 code values above that level. Same way you don't send a digital cinema projector XYZ values that fall outside its native gamut. If you ignore this, you are going to screw up your clients' content. I guarantee it. I do this for real too.

At home, it is fine to roll off with a soft clip. At the simplest level, this is what the majority of consumer electronics manufacturers are going to be doing with HDR, but they have the advantage of being able to dynamically adjust their roll off as well as doing more sophisticated tone mapping, whereas calibrating with a LUT you have to pick one constant roll off for everything.
mpgxsvcd likes this.

Last edited by EvLee; 12-13-2015 at 04:04 AM.
EvLee is offline  
post #18 of 759 Old 12-13-2015, 04:38 AM
AVS Forum Club Gold
 
Carbon Ft Print's Avatar
 
Join Date: Nov 2009
Location: FL
Posts: 524
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 308 Post(s)
Liked: 278
Quote:
Originally Posted by EvLee View Post
Remember regular old HD? It wasn't fully defined until um… let's see when was ITU-R 1886 published? 2011. But it seemed to work pretty well before then. Here's what you are missing. This new container was made both big (10k nits) and wide (2020 primaries) so it wouldn't have to be redefined in another 2, 5, 10 years down the road every time display technology advances. It was made as large as reasonable with the knowledge that only a limited portion of it would be used initially. This is where 2086 comes into play. You start with a smaller utilization and grow into the larger range over time. Even with HD, 100 nits for white was completely undocumented but informally agreed on in practice because that was all you could get out of the Sony reference CRTs. So what is going to happen with HDR and WCG? Well, there are two reference displays currently available to studios. Those will set a limit on what actually produced for distribution. By the nature of the industry and the need to interchange content between facilities, they will converge on a common (although not necessarily SMPTE standardized) target. But all you'll need to do is take a look inside the 2086 metadata to find out what that is. My suggestion is wait until CES, wait until test discs come out and then try again.

Also, as to your ABL comment… one of the most linear (flattest APL curve) response displays I have ever seen is HDR and intended for consumer market. I have also seen some very junk ones as well, and I think the biggest risk for HDR is that too many low tier displays, especially with low quality dimming like you described, will be forced into the market by manufacturers and advertised as "HDR". There is a real danger that these will compromise the consumer impression of the format, when it is capable of so much more when done right.
I have a HDR capable JVC RS600 (9000) projector being delivered by the end of the month. What level of effect does HDR have on mid-tones when they grade movies for HDR? With today's blu rays, JVC gives me the controls on my 6710 projector to add pop to the mid tones, but the highlights may get clipped a little ... which is great for movies like aliens/Prometheus. Would HDR prevent highlights from being clipped while providing more pop to the mid-tones ... or do I have this all wrong?
Carbon Ft Print is offline  
post #19 of 759 Old 12-13-2015, 08:43 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,276
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 429 Post(s)
Liked: 731
HDR display configuration can only be used on material that has been mastered for HDR.
And only on material that has been mastered for the specific HDR format you display uses.

To attempt to use HDR on SDR material, or mix HDR formats (mastering with final display), will result in very inaccurate images.

This is another of the HDR issues - everything is so ill defined that it is almost impossible to guarantee that source matches destination.
And matching source to destination is key in accurate final image display, as defined by calibration.

And honestly, as TVs have been 100+ Nits for years, and have looked great, why would a projector that delivers around 100 Nits all of a sudden be HDR?

That is yet another HDR issue that has not been accurately resolved or defined.

Steve
mpgxsvcd likes this.

Steve Shaw
LIGHT ILLUSION

Light Illusion is online now  
post #20 of 759 Old 12-13-2015, 09:54 AM
AVS Forum Club Gold
 
Carbon Ft Print's Avatar
 
Join Date: Nov 2009
Location: FL
Posts: 524
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 308 Post(s)
Liked: 278
There has to be a reason why a director(s) wants to color grade both their films and 4k blu ray movies with HDR (besides increasing sales). Are they looking to provide more pop in the mid-tones and shadows where much of the emotion/envelopment resides? Or are they trying to provide more pop in the highlights to create more emotion/envelopment where it is difficult to do so (because it's so bright)? JVC projectors can create a small amount of pop in the mid-tones just by adjusting several of their controls. Is this what the directors want to do with films, provide more pop, but to do it while grading the film and applied it to the full range of shadows, mid-tones and highlights? ... something that the JVC projector cannot tweak unless it too has HDR capabilities. What do directors think?
Carbon Ft Print is offline  
post #21 of 759 Old 12-13-2015, 12:55 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 Carbon Ft Print View Post
There has to be a reason why a director(s) wants to color grade both their films and 4k blu ray movies with HDR (besides increasing sales). Are they looking to provide more pop in the mid-tones and shadows where much of the emotion/envelopment resides? Or are they trying to provide more pop in the highlights to create more emotion/envelopment where it is difficult to do so (because it's so bright)? JVC projectors can create a small amount of pop in the mid-tones just by adjusting several of their controls. Is this what the directors want to do with films, provide more pop, but to do it while grading the film and applied it to the full range of shadows, mid-tones and highlights? ... something that the JVC projector cannot tweak unless it too has HDR capabilities. What do directors think?
HDR provides directors a larger color palette to work with. It's not just about "pop", it lets them produce images that are fundamentally different from what you are accustomed to seeing on a standard display. What you see in projection, even at a Dolby Cinema location, is not full HDR. It is the bottom half (darks) of HDR plus a little extra range in the highlights. So they are making two different HDR versions of the movie when they do color grading. They make a theatrical HDR grade which is limited in the brights, and they make a home grade which is intended for viewing on a true HDR display. If you try to project a home HDR grade at 100 nits, even with a projector that can go fully black, it won't look right.
Carbon Ft Print likes this.
EvLee is offline  
post #22 of 759 Old 12-13-2015, 01:05 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 Light Illusion View Post
And honestly, as TVs have been 100+ Nits for years, and have looked great, why would a projector that delivers around 100 Nits all of a sudden be HDR?
I fully agree with you on this. HDR projection wasn't defined in any of the standards because there is no HDR projection technology available in the market. Unfortunately, the manufacturers of projectors see the buzz around HDR and want to be able to capitalize on it so you see products that are being sold as HDR when they really are not. They are "HDR-compatible" meaning they can recognize an HDR signal but don't actually reproduce it accurately. Sort of like a "4k-compatible" display that can receive 4k signal but has only 720 pixels native resolution.
EvLee is offline  
post #23 of 759 Old 12-14-2015, 01:14 AM
AVS Forum Club Gold
 
Carbon Ft Print's Avatar
 
Join Date: Nov 2009
Location: FL
Posts: 524
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 308 Post(s)
Liked: 278
So the new star wars is coming out on December 18 in Dolby HDR. Star Wars: The Force Awakens Will Be Shown in Dolby Vision HDR at Dolby Cinemas Scott Wilkinson has seen other HDR movies at El Capitan Theatre in Hollywood, CA and really liked it. The commercial theater's Ymax is 100nits (DCI) which falls into the range of these new JVC projectors along with 12bit and DCI. So why so cautious with HDR on home cinema projectors?
Carbon Ft Print is offline  
post #24 of 759 Old 12-14-2015, 01:58 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,276
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 429 Post(s)
Liked: 731
Quote:
Originally Posted by Carbon Ft Print View Post
So the new star wars is coming out on December 18 in Dolby HDR. Star Wars: The Force Awakens Will Be Shown in Dolby Vision HDR at Dolby Cinemas Scott Wilkinson has seen other HDR movies at El Capitan Theatre in Hollywood, CA and really liked it. The commercial theater's Ymax is 100nits (DCI) which falls into the range of these new JVC projectors along with 12bit and DCI. So why so cautious with HDR on home cinema projectors?
I actually see no difference with the professional so-called HDR projectors and consumer ones.

100 Nits is just not HDR in reality - it is just projection as it should always have been, to match standard home TVs when viewed in similar viewing environments - ie., dark.

Obviously, the size of the screen, and the resulting viewing subtend angle, means we do get close to the desired 45 degrees, which does make a difference to home viewing - but that also brings us back to some of my original points on HDR.

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is online now  
post #25 of 759 Old 12-17-2015, 12:41 PM
AVS Forum Club Gold
 
Carbon Ft Print's Avatar
 
Join Date: Nov 2009
Location: FL
Posts: 524
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 308 Post(s)
Liked: 278
Looks like what the director and colorists will be grading "with HDR" is shown in the 2 diagrams below that I grabbed from Scott Wilkinson talk on hdr - eotf: SMPTE Webinar: Dolby Vision PQ EOTF


These diagrams answer my question of where among Highlights, Mid-Tones or Shadows would one see improvements in HDR technology for commercial and Home Cinema. Instead of the 3 relative luminance ranges which are color graded today (Highlights, Mid-Tones and Shadows), Dolby redefines them as 6 absolute luminance ranges: Super Highlights, Highlights, White Range, "Most Typical Objects", Shadows and Blacks. At 100nits PQ for projectors, one could extrapolate that there are major improvements in the shadows and mid-tones at the expense of highlights. To use Dolby's vernacular at 100nits, there are no Super Highlights, no Highlights and the White Range content is sacrificed to insure liberal distribution of the luminance range among the 3 lower Dolby luminance ranges: "Most Typical Objects", Shadows and Blacks. So for Commericial and JVC (hehehe) HT, one will still see a great picture and it will play into the strengths of the projector ... in a light controlled room. From what I've seen in Dolby's HDR demos, the shadows and mid-tones will take on slightly brighter colors than what we are used to ... this will make objects in the shadows and mid-tones easier for your eyes to see and more colorful ... a different kind of color pop ... but still based on luminance. JVC has 3 controls (Bright Level, Picture Tone, Dark Level) to fine tune the effect of HDR.


JJ


****



Source Code
The last part of the webinar illustrated how different ranges of luminance values are allocated in a gamma-based display and a PQ-based display. In the graphs below, the allocation of luminance values (code words) are shown for different peak light levels.


A gamma-based system allocates very few luminance values (code words) to shadow detail and progressively fewer values to the brightness of most typical objects and whites as the peak brightness increases. In this graph, each cylinder represents a different peak light level (500 nits, 1000 nits, etc.), increasing from right to left, with a gamma of 2.4. (Courtesy of Dolby Laboratories)


In a PQ-based display, more values are allocated to shadow detail at all peak light levels, and the range of values allocated to typical objects and whites remains more constant as the peak brightness increases. In this graph, each cylinder represents a different peak light level (500 nits, 1000 nits, etc.), increasing from right to left. (Courtesy of Dolby Laboratories)
Manni01 and Elix like this.

Last edited by Carbon Ft Print; 12-17-2015 at 12:54 PM.
Carbon Ft Print is offline  
post #26 of 759 Old 12-17-2015, 01:30 PM
AVS Forum Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 7,500
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 770 Post(s)
Liked: 660
Quote:
Originally Posted by EvLee View Post
You still are misinterpreting 2084. You are not supposed to see hard clipping at the display. You need to have a limiter in your color grading system so that you do not feed the display with values that force it into the region of clipping behavior. This is nothing new. When you grade digital cinema, you don't do it directly in XYZ space do you? So if your display can not go over 1000 nits for example, you should not be sending it any 2084 code values above that level. Same way you don't send a digital cinema projector XYZ values that fall outside its native gamut. If you ignore this, you are going to screw up your clients' content. I guarantee it. I do this for real too.

At home, it is fine to roll off with a soft clip. At the simplest level, this is what the majority of consumer electronics manufacturers are going to be doing with HDR, but they have the advantage of being able to dynamically adjust their roll off as well as doing more sophisticated tone mapping, whereas calibrating with a LUT you have to pick one constant roll off for everything.
I don't think you are taking sufficiently seriously the issues with proper home reproduction that HDR has introduced. To my knowledge this is the first gamma specification that requires a defined luminance value at 100%--in this case 10,000 nits. In all previous specifications, all luminance targets were definable as a percentage of 100% video without that level being specified as a hard target. Tell me what the display outputs at 100% and at black, then BT.1886--for example--tells you what the display should output at any video level. HDR does the same thing except that it defines 100% as 10,000 nits. The HDR formula then proceeds to recommend levels of output at anything above about 80% video that are IMPOSSIBLE for current HDR displays.

The analogy to color is not a good one. Home video sources are not mastered in DCi-P3, and even if they were manufacturing displays with a DCi-P3 gamut poses no special problems. Many are capable of this already and have been for a long time. Not so with HDR. The consumer displays now labeled as "HDR" generally are capable of no more than 10%--10%!--of the output REQUIRED by the HDR spec. Furthermore, there are no established guidelines that I know of for how to adapt source material designed for peak output of 10,000 nits for displays capable of barely 1/10 of that. I assume that this will eventually be straightened out, but right now it is the wild west.
Michael Osadciw and Manni01 like this.

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

TomHuffman is offline  
post #27 of 759 Old 12-17-2015, 08:41 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 TomHuffman View Post
I don't think you are taking sufficiently seriously the issues with proper home reproduction that HDR has introduced. To my knowledge this is the first gamma specification that requires a defined luminance value at 100%--in this case 10,000 nits. In all previous specifications, all luminance targets were definable as a percentage of 100% video without that level being specified as a hard target. Tell me what the display outputs at 100% and at black, then BT.1886--for example--tells you what the display should output at any video level. HDR does the same thing except that it defines 100% as 10,000 nits. The HDR formula then proceeds to recommend levels of output at anything above about 80% video that are IMPOSSIBLE for current HDR displays.

The analogy to color is not a good one. Home video sources are not mastered in DCi-P3, and even if they were manufacturing displays with a DCi-P3 gamut poses no special problems. Many are capable of this already and have been for a long time. Not so with HDR. The consumer displays now labeled as "HDR" generally are capable of no more than 10%--10%!--of the output REQUIRED by the HDR spec. Furthermore, there are no established guidelines that I know of for how to adapt source material designed for peak output of 10,000 nits for displays capable of barely 1/10 of that. I assume that this will eventually be straightened out, but right now it is the wild west.
This post is going to ramble a bit because there is a lot of history behind it...

When I was referring to color, I wan't talking about P3. I was talking specifically about XYZ, or more specifically the digital cinema X'Y'Z' container. Since you didn't catch it the first time around, let me explain the analogy. Digital cinema has a container larger than the display gamut. Both in terms of primaries (which are XYZ vs P3) and luminance (X'Y'Z' encoding has headroom built into it above the 48 nits that the projectors are supposed to hit). You can't calibrate a digital cinema projector to 100% signal value because it's been beyond the peak luminance of the projector. If you naively try to calibrate a digital cinema projector using a 100% signal, it will be incorrect. If you use a test signal whose value matches the content white level (which was defined by DCI to be 48 nits), then the calibration will be correct. Is this starting to sound like HDR a little bit?

In the early days, studios were creating digital cinema packages that accidentally had X'Y'Z' values outside of the native projector gamut. They could even run into white clipping issues if they thought 100% was supposed to be white. Hands were slapped and now they should know better. Unfortunately, there was, and still is, no metadata to track how much of X'Y'Z' is being used. So these early DCP's only play correctly if shown on the same projector they were mastered on. Not an issue in the beginning because there was only one projector out there, but it is a problem now that laser projectors exist and also because of the wide variance in luminance for stereo projection. Fortunately, HDR has 2086 metadata to go along with 2084 container which is where digital cinema would have set 48 nits, except now it is variable and set according to the mastering display (none of which do 10,000 nits).

So HDR is not the first absolute luminance specification. Digital cinema was the first and was only broken years later because they did not consider how to handle the case of displays that do not match the reference specification. If you don't get that, you aren't going to get this encoding and why it was designed the way it is.

2084 should not be taken separately from 2086. One is the container, the other defines what peak luminance is being utilized and where you should be checking your calibration. If you unwrap any of the HDR content that is out there now, if it has been created correctly, you will not find any values at 10,000 nits nor a peak luminance reported in the metadata anywhere near that. Of course there may be some bad packages released and hopefully they will be reported and any mistakes corrected. Yes, it is a bit like the wild west because manufacturers are competing with each other to innovate in this field and establish their position at the front. That's why a fixed luminance number could not be locked down. If they had to agree on a single number it would be the very bottom minimum that everybody could hit with even the cheapest display. It would be 400, maybe 500 nits at most. Leaving it open encourages the development of better displays, and provide a way to create content for those displays without continually revising the encoding. The standard is intended to grow with the technology, but certainly not be utilized all at once. That would be like drinking from a fire hose. In time it will converge on a sweet spot.

I'd like to add that even though it is early days you can find people on this forum (even those who were skeptical initially) reporting on how great HDR looks and that it is unlike anything they have seen before and never want to go back. All without even being able to manually calibrate the setting! Not bad for a broken standard.
Carbon Ft Print likes this.

Last edited by EvLee; 12-17-2015 at 09:06 PM.
EvLee is offline  
post #28 of 759 Old 12-17-2015, 08:55 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 Carbon Ft Print View Post
Looks like what the director and colorists will be grading "with HDR" is shown in the 2 diagrams below that I grabbed from Scott Wilkinson talk on hdr - eotf: SMPTE Webinar: Dolby Vision PQ EOTF


These diagrams answer my question of where among Highlights, Mid-Tones or Shadows would one see improvements in HDR technology for commercial and Home Cinema. Instead of the 3 relative luminance ranges which are color graded today (Highlights, Mid-Tones and Shadows), Dolby redefines them as 6 absolute luminance ranges: Super Highlights, Highlights, White Range, "Most Typical Objects", Shadows and Blacks. At 100nits PQ for projectors, one could extrapolate that there are major improvements in the shadows and mid-tones at the expense of highlights. To use Dolby's vernacular at 100nits, there are no Super Highlights, no Highlights and the White Range content is sacrificed to insure liberal distribution of the luminance range among the 3 lower Dolby luminance ranges: "Most Typical Objects", Shadows and Blacks. So for Commericial and JVC (hehehe) HT, one will still see a great picture and it will play into the strengths of the projector ... in a light controlled room. From what I've seen in Dolby's HDR demos, the shadows and mid-tones will take on slightly brighter colors than what we are used to ... this will make objects in the shadows and mid-tones easier for your eyes to see and more colorful ... a different kind of color pop ... but still based on luminance. JVC has 3 controls (Bright Level, Picture Tone, Dark Level) to fine tune the effect of HDR.


JJ


****



Source Code
The last part of the webinar illustrated how different ranges of luminance values are allocated in a gamma-based display and a PQ-based display. In the graphs below, the allocation of luminance values (code words) are shown for different peak light levels.


A gamma-based system allocates very few luminance values (code words) to shadow detail and progressively fewer values to the brightness of most typical objects and whites as the peak brightness increases. In this graph, each cylinder represents a different peak light level (500 nits, 1000 nits, etc.), increasing from right to left, with a gamma of 2.4. (Courtesy of Dolby Laboratories)


In a PQ-based display, more values are allocated to shadow detail at all peak light levels, and the range of values allocated to typical objects and whites remains more constant as the peak brightness increases. In this graph, each cylinder represents a different peak light level (500 nits, 1000 nits, etc.), increasing from right to left. (Courtesy of Dolby Laboratories)
What we consider to be highlights, midtones and shadows are scene dependent. You can't specify the luminance of white because perceptually, white is judged relative to a light source. The higher the luminance of the light source in a scene, the higher the luminance of your white and corresponding midtones and shadows will be. If a colorist keeps their "whites" at the same luminance they are used to, like 100 nits, then it allows bright colors to hold saturation and be more colorful. It also allows the reproduction of luminous objects that are brighter than white. However, instead of keeping everything normalized around 100 nits they can also push the entire scene up to a higher luminance range, or a lower luminance range since the shadow detail is better now, and use that to build a very convincing effect like the transition from a dark interior into a bright exterior. Imagine walking from inside an unlit cabin out into a field at midday. This has implications about how display controls need to behave because making adjustment with fixed pivot positions is not well-suited to HDR. The controls should ideally scale with the scene luminance.

I think you are correct that with a 100 nits projector it is near impossible to produce the effect of luminous colors, aka super highlights, but you can produce pretty good highlights if the projection is done with whites held down to the 48 nits we are used to seeing in standard projection.

Last edited by EvLee; 12-17-2015 at 08:58 PM.
EvLee is offline  
post #29 of 759 Old 12-17-2015, 10:00 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 937
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 256 Post(s)
Liked: 192
Quote:
Originally Posted by TomHuffman View Post
Not so with HDR. The consumer displays now labeled as "HDR" generally are capable of no more than 10%--10%!--of the output REQUIRED by the HDR spec. Furthermore, there are no established guidelines that I know of for how to adapt source material designed for peak output of 10,000 nits for displays capable of barely 1/10 of that. I assume that this will eventually be straightened out, but right now it is the wild west.
Agreed - the SMPTE ST 2084:2014 is touted as an HDR EOTF for Mastering Reference Displays, but now seems to have been adopted as a down-stream delivery format by the media and display manufacturers in their haste to push the "next big thing".
It seems to be seriously flawed in this role. The use of an absolute luminance scale with the recommendation that "The EOTF does not impart a preferred rendering appearance for any particular viewing environment. Image modifications needed for viewer contrast, colorfulness, highlight details, and visible detail in shadows at any particular output level must be chosen as part of the mastering process." simply ignores all the practicalities of delivering media to the end consumer. The fact is that end consumer display capabilities and viewing environments are wide and varied, and that therefore encoding a preferred rendering appearance (including such information as the nominal diffuse white point encoding) is essential in being able to automatically adapt the media to the actual viewing environment. To expect such things to be done "as part of the mastering process" is simply unbelievably naive. The fact that an absolute format may be mastered to different viewing environments and then have no formal way of actually labeling what environment they are mastered to, is equally astounding.
gwgill is offline  
post #30 of 759 Old 12-17-2015, 10:45 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
2084 doesn't define an ecosystem. I don't understand why everybody here keeps acting like it does. It's one component. You can't analyze it as being anything more than what it defines itself to be. A reference EOTF. It astounds me that absolutely nobody has talked, or even asked, about the implications of 2086 or dynamic metadata. I agree that 2084 completely by itself does not work for distribution, but that is not what is being rolled out despite what many here proclaim.

Graemme, what are the parameters that you think constitute a viewing environment? Luminance, black level? These are transferred via 2086. Diffuse white can not be encoded statically because it depends on the scene content. It can be captured in either dynamic metadata (2094) for color appearance rendering, or be calculated on the fly by a display / color management system. 2084 encodes JND's, not appearance.

http://www.screenplaysmag.com/2015/1...uality-issues/

"That intent has been bolstered by the development of SMPTE 2094, a draft standard moving to adoption that introduces more dynamism into the color transformation process than provided by SMPTE 2086, otherwise known as the “Master Display Color Volume Metadata Supporting High Luminance and Wide Color Gamut.” SMPTE 2086 serves as the metadata component referenced by the SMPTE 2084 Electro-Optical Transfer Function (EOTF), an alternative to the traditional Gamma or Opto-Electric Transfer Function (OETF) that is now part of the BDA’s UHD standard and the HDR specifications recommended by the Consumer Electronics Association.

By providing a means of assigning a dynamic brightness dimension to the rendering of colors by TV displays, SMPTE 2094, an adaptation of the metadata instruction set used in Dolby Vision, brings the full HDR experience into play with 10-bit as well as 12-bit sampling systems. With SMPTE 2084 and ITU Rec-2020 color gamut now specified in both the BDA’s and the CEA’s HDR profiles, a fairly clear HDR roadmap is in place, leaving it to individual industry players to determine whether they want to utilize 12-bit formats like Dolby Vision and the one developed by Technicolor or the 10-bit formats offered by Samsung and others."
Carbon Ft Print likes this.

Last edited by EvLee; 12-17-2015 at 10:56 PM.
EvLee is offline  
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