MadVR - ArgyllCMS - Page 146 - AVS Forum | Home Theater Discussions And Reviews
Baselworld is only a few weeks away. Getting the latest news is easy, Click Here for info on how to join the Watchuseek.com newsletter list. Follow our team for updates featuring event coverage, new product unveilings, watch industry news & more!



Forum Jump: 
 441Likes
Reply
 
Thread Tools
post #4351 of 4371 Old 07-29-2016, 05:29 PM
AVS Addicted Member
 
Stereodude's Avatar
 
Join Date: Jan 2002
Location: Detroit Metro Area
Posts: 13,606
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 2904 Post(s)
Liked: 2085
Is it possible to create a meter correction matrix so I can profile my TV as if the 3D glasses are between my photometer and the display without actually having them there? I was able to do this for HCFR, but it's not clear if I can do the same thing with DisplayCAL. If so, how?
Stereodude is offline  
Sponsored Links
Advertisement
 
post #4352 of 4371 Old 07-29-2016, 07:49 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 882
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 216 Post(s)
Liked: 173
Quote:
Originally Posted by Stereodude View Post
Is it possible to create a meter correction matrix so I can profile my TV as if the 3D glasses are between my photometer and the display without actually having them there?
Yes, should be possible. Put the spectrometer behind the glasses, and the colorimeter directly on the display.
Quote:
I was able to do this for HCFR, but it's not clear if I can do the same thing with DisplayCAL. If so, how?
I'd imagine that "Tools->Create colorimeter correction..." would be an obvious place to start...
gwgill is offline  
post #4353 of 4371 Old 07-30-2016, 08:05 AM
AVS Addicted Member
 
Stereodude's Avatar
 
Join Date: Jan 2002
Location: Detroit Metro Area
Posts: 13,606
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 2904 Post(s)
Liked: 2085
Quote:
Originally Posted by gwgill View Post
Yes, should be possible. Put the spectrometer behind the glasses, and the colorimeter directly on the display.

I'd imagine that "Tools->Create colorimeter correction..." would be an obvious place to start...
Well, I don't have a spectro, just a colorimeter. However, I'll check it out before asking asking any more questions.
Stereodude is offline  
post #4354 of 4371 Old 07-30-2016, 11:24 AM
Member
 
Sunset1982's Avatar
 
Join Date: Jun 2006
Posts: 71
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 18
is it possible to do a 2.4 gamma autocalibration with displaycal/argyIIcms with a gamma of 2.3 at 5% and the rest at 2.4?
Sunset1982 is offline  
post #4355 of 4371 Old 08-02-2016, 06:00 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 814
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 400 Post(s)
Liked: 258
I'm happy to announce that I've released DisplayCAL 3.1.5.

This release brings HDR 3D LUT (BT.2020 & ST.2084) improvements, namely roll-off according to BT.2390 (hermite spline at the top end, black level lift similar to BT.1886 but with tapering factor at the dark end) and a simple form of tone mapping.

Example screen capture of Sony "Camp" HDR demo @ 400 cd/m2 peak luminance (TV gamut roughly Rec. 709):


How to set up DisplayCAL to create a HDR to SDR 3D LUT:

Set HDR peak luminance according to the capability of your display. I wouldn't recommend below 265 cd/m2.
Black output offset controls if there's roll-off at the dark end (0% black output offset = BT.2390), which is similar to BT.1886, or if the whole curve is offset and scaled by the black level (100% output offset).

I've also tweaked the perceptual intent gamut mapping to give better preservation of saturated colors than normally, but I would still recommend colorimetric as it should preserve saturations even better (the above screen capture is from a colorimetric 3D LUT). Note that despite colorimetric normally clipping, there's now always a form of tone mapping when SMPTE 2084 roll-off is selected.

Enjoy!
Attached Thumbnails
Click image for larger version

Name:	DisplayCAL-315-HDR-Setup.png
Views:	610
Size:	62.9 KB
ID:	1580841   Click image for larger version

Name:	Sony_HDR_Camp_00-01-28_DisplayCAL_HDR_400_3DLUT.jpg
Views:	117
Size:	606.5 KB
ID:	1580849  
N3W813, VBB, 10k and 1 others like this.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.1.6 released 2016-08-25
Previous development snapshot (OUTDATED): 3.1.3.2 Beta (Windows/0install) released 2016-04-23 | Standalone | Changelog
DisplayCAL on Facebook

Last edited by fhoech; 08-02-2016 at 06:34 PM.
fhoech is online now  
post #4356 of 4371 Old 08-25-2016, 04:48 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 814
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 400 Post(s)
Liked: 258
DisplayCAL 3.1.6 has been releasaed.

This release brings further improvements in HDR 3D LUT tonemapping.

  • Overall saturation preservation should be better, as well as slightly too light near-black tones have been fixed.
  • Also supported is madVR 0.90.24, where HDR 3D LUTs must be set in specific slots and require an adjusted 3D LUT header that is now generated automatically by DisplayCAL if the tone curve is set to SMPTE 2084. Note that installation of HDR 3D LUTs into these slots from within DisplayCAL is currently not possible (you'll be asked to save the 3D LUT instead) due to missing API in madVR, so you have to select the 3D LUT in madVR manually.

Two new features have been added for HDR/SMPTE 2084, both only visible if you enable "Show advanced options" in the options menu:

  • Maximum content light level (maxCLL). Adjusting the maximum content light level to that of the actual video material will gain you a slightly(!) better utilization of your display's maximum brightness capabilities, but note that this needs to be set carefully because setting it wrongly can cause visual artifacting, and HDR metadata is not a reliable source to determine the actual encoded maxCLL. Therefore, I would strongly(!) suggest to leave the maxCLL settting at its default value of 10000 cd/m2.
  • Content colorspace. For all currently available HDR content (that I know of), this should be left at the default (DCI P3). The content colorspace setting influences perceptual gamut mapping, less so colorimetric, and colorimetric is still the recommended choice for generating HDR 3D LUTs.

Enjoy!
N3W813, VBB and chros73 like this.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.1.6 released 2016-08-25
Previous development snapshot (OUTDATED): 3.1.3.2 Beta (Windows/0install) released 2016-04-23 | Standalone | Changelog
DisplayCAL on Facebook
fhoech is online now  
post #4357 of 4371 Old 08-25-2016, 05:44 AM
AVS Special Member
 
madshi's Avatar
 
Join Date: May 2005
Posts: 5,925
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 406 Post(s)
Liked: 410
Many thanks - and great job!

FWIW, I can say that the new DisplayCAL build produces very nice HDR tone mapping 3dluts, which are at least comparable, probably slightly superior to what madVR does with its real time shader math. Furthermore, using a 3dlut instead of shader math saves GPU power. So overall highly recommended! The only disadvantage is that things like video and display peak luminance are burned into the 3dlut, while when using the shader math, you can adjust those values on the fly.

@fhoech , maybe it would make sense to create a selection of "neutral" 3dluts (meaning multiple 3dluts for different display peak luminance values), and make them available either on your or my server? Might be useful for non-calibration users, who just want to save some GPU performance, and get the best possible HDR quality, without having to dive into ArgyllCMS/DisplayCAL installation and usage. Just a thought, though. What do you think?
madshi is online now  
post #4358 of 4371 Old 08-25-2016, 03:38 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 814
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 400 Post(s)
Liked: 258
Quote:
Originally Posted by madshi View Post
@fhoech , maybe it would make sense to create a selection of "neutral" 3dluts (meaning multiple 3dluts for different display peak luminance values), and make them available either on your or my server? Might be useful for non-calibration users, who just want to save some GPU performance, and get the best possible HDR quality, without having to dive into ArgyllCMS/DisplayCAL installation and usage. Just a thought, though. What do you think?
Sure, we can do that. Which specific peak luminance levels do you have in mind? Also by "neutral", assume Rec. 709 gamma 2.2 as the destination or something along those lines? I could host the files, although I'm on shared hosting.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.1.6 released 2016-08-25
Previous development snapshot (OUTDATED): 3.1.3.2 Beta (Windows/0install) released 2016-04-23 | Standalone | Changelog
DisplayCAL on Facebook
fhoech is online now  
post #4359 of 4371 Old 08-26-2016, 12:45 AM
AVS Special Member
 
madshi's Avatar
 
Join Date: May 2005
Posts: 5,925
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 406 Post(s)
Liked: 410
Ouch, I forgot about BT.709 vs DCI-P3, and gamma 2.2 vs 2.4. I think we'd need a whole bunch of 3dluts to cover all areas... Can you give me a short instruction on how to create neutral 3dluts myself, then I'll do that when I find some time. I just tried, but didn't really manage to do it. In DisplayCAL itself, when pressing on 3dlut (without profiling) the button at the bottom to create the 3dlut stayed gray, and I couldn't make it enable. Then I tried using the 3dlut-maker exe directly, but somehow the created 3dlut changes doesn't appear to be neutral. Probably I messed something up somewhere...
madshi is online now  
post #4360 of 4371 Old 08-26-2016, 03:32 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 814
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 400 Post(s)
Liked: 258
A DCI-P3/SMPTE431 profile comes with Argyll CMS (in the "ref" folder). You can generate a Rec709 gamma 2.2 profile with DisplayCAL's synthetic ICC profile creator. When these profiles are to be used as a substitute for an actual display profile, they need to be set under "Target profile" in the 3D LUT maker.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.1.6 released 2016-08-25
Previous development snapshot (OUTDATED): 3.1.3.2 Beta (Windows/0install) released 2016-04-23 | Standalone | Changelog
DisplayCAL on Facebook
fhoech is online now  
post #4361 of 4371 Old 08-26-2016, 03:36 AM
AVS Special Member
 
madshi's Avatar
 
Join Date: May 2005
Posts: 5,925
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 406 Post(s)
Liked: 410
Ok, thanks, I'll give that a try.
madshi is online now  
post #4362 of 4371 Old 08-27-2016, 05:43 AM
Member
 
Bloodwound's Avatar
 
Join Date: Apr 2007
Posts: 133
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 21
A question to gwgill and/or fhoech.

Displaycal is unable to get the whitebalance right on my laptop. Since there are no hardware RGB controls on most laptops, it's all in software.

Would it be possible to have 3 sliders for RGB to set the (8/10bit depending on the display) the whitebalance for 100% in software in the initial measurement dialong akin to adjusting the hardware controls? I'm not sure this is the right way, but I am unable to match the whitebalance on my laptops (and to each other) compared to my external monitor. It would be easier to adjust my luminance too. (Argyllcms will display this when run from the commandline, but it would be nice to know what your luminance would be before you start the calibration/profiling process)

Last edited by Bloodwound; 08-27-2016 at 05:47 AM.
Bloodwound is online now  
post #4363 of 4371 Old 09-21-2016, 09:47 AM
Newbie
 
markanini's Avatar
 
Join Date: Sep 2008
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 0
From https://hub.displaycal.net/forums/to...ng-up-with-2-2
Quote:
Hi,

it is important to note that BT. 1886 (= 2.4 absolute gamma, black output offset 0%) is <em>not</em> the same as a 2.4 power curve – it takes into account the black level (and white level) of the display, thus in practice the overall gamma may end up closer to 2.2 than 2.4, as you have observed. BT. 1886 will only match a 2.4 power curve if the display black level is zero (e.g. OLED).

The tone curve parameters are highly configurable though: Setting the gamma from “absolute” to “relative” will make the curve match the desired gamma at 50% input, and setting the black output offset to a non-zero value will make the curve shape more resemble a power curve near black (at the expense of less even gradation out of black).

If you’d like to play with the tone curve parameters and create additional 3D LUTs from your existing profile, disable “Create 3D LUT after profiling” on the “3D LUT” tab.
I was looking for a helpful summary like this to wrap my head around the black output offset and absolute/relative options. Now, correct me if I'm wrong, I'm just taking the information on face value. Lets say you have a display device with a poor contrast ratio like 1:300, the effective gamma using vanilla BT.1886 is around 2.0. Using a non-0 offset in absolute mode will basically crush the blacks leaving higher tones washed out same as before, which is pretty useless. Relative mode focuses on the midtones, I assume when used with my example device (1:300 CR)2.4 relative gamma with the slider at 50% produces an effective gamma of 2.2?


EDIT:Just made a quick calibration run with 2.4 relative 50% and got 2.4 effective gamma on my low contrast display. I have further testing ahead of me.

Last edited by markanini; 09-21-2016 at 11:55 AM.
markanini is offline  
post #4364 of 4371 Old 09-21-2016, 03:06 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 814
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 400 Post(s)
Liked: 258
Quote:
Originally Posted by Bloodwound View Post
Displaycal is unable to get the whitebalance right on my laptop.
On a Laptop, the videoLUTs are the only means to adjust the whitepoint. This can make it difficult to match the Laptop whitepoint to another display's white, especially if the panel and/or backlighting technology are different. One option is to match the other display's whitepoint to the Laptop instead. I will have a better solution to this in a future version of DisplayCAL.

Quote:
Originally Posted by markanini View Post
From https://hub.displaycal.net/forums/to...ng-up-with-2-2


I was looking for a helpful summary like this to wrap my head around the black output offset and absolute/relative options. Now, correct me if I'm wrong, I'm just taking the information on face value. Lets say you have a display device with a poor contrast ratio like 1:300, the effective gamma using vanilla BT.1886 is around 2.0. Using a non-0 offset in absolute mode will basically crush the blacks leaving higher tones washed out same as before, which is pretty useless. Relative mode focuses on the midtones, I assume when used with my example device (1:300 CR)2.4 relative gamma with the slider at 50% produces an effective gamma of 2.2?
Relative = effective, meaning it will produce exactly the chosen gamma value (in your example 2.4) at 50% input. If you want an effective gamma of 2.2, set it to 2.2 relative. With an effective gamma target, output offset influences what happens below the 50% point.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.1.6 released 2016-08-25
Previous development snapshot (OUTDATED): 3.1.3.2 Beta (Windows/0install) released 2016-04-23 | Standalone | Changelog
DisplayCAL on Facebook
fhoech is online now  
post #4365 of 4371 Old 09-21-2016, 04:57 PM
Newbie
 
markanini's Avatar
 
Join Date: Sep 2008
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 0
Quote:
Originally Posted by fhoech View Post
Relative = effective, meaning it will produce exactly the chosen gamma value (in your example 2.4) at 50% input. If you want an effective gamma of 2.2, set it to 2.2 relative. With an effective gamma target, output offset influences what happens below the 50% point.
I tried 2.2 relative 0% and I'm happy with the results. Before that I used 2.4 absolute 88% which looked similar, with more linear highlights, but darker shadows. Overall I'm happier than I was with vanilla BT.1886. I get the impression that a BT.1886-based curve has superior tonal presentation compared to 2.2 pure power or sRGB and could take their place for general use too.
Attached Thumbnails
Click image for larger version

Name:	2.2relative0.jpg
Views:	24
Size:	53.9 KB
ID:	1674465   Click image for larger version

Name:	2.4absolute88.jpg
Views:	22
Size:	53.5 KB
ID:	1674473  
markanini is offline  
post #4366 of 4371 Old 09-22-2016, 10:10 AM
Newbie
 
markanini's Avatar
 
Join Date: Sep 2008
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 0
I found a third way to achieve a 2.2 gamma at the midtones, by setting a custom black point and white point. I assume 2.2 relative 0% is the recommended setting though?
markanini is offline  
post #4367 of 4371 Old 09-23-2016, 07:02 AM
Member
 
Join Date: Jan 2015
Posts: 52
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 33 Post(s)
Liked: 4
Hi,

After many attempts using HCFR with all kinds of window sizes and background patterns, I finally managed to get a good result on my plasma / bluray player with Mascior's 4% EE-style APL pattern. The relief was immense lol.

I'd now like to move back to calibrating from my HTPC with a 3dlut, using the MadTPG generator. For the sake of consistency, what would be the closest linear light background % with a 4% window, to the 4% patterns found on Mascior's disc?
iSeries is online now  
post #4368 of 4371 Old Yesterday, 12:14 PM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 4,717
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 794 Post(s)
Liked: 819
I see Argyll CMS has released version 1.9.0.

Graeme am I missing something, I don't see support for the Lumagen.??
eeColor processor - ArgyllCMS

ss

"Don't worry be happy"
Setup dispcalGUI, 2nd link settings for 65EF9500
http://www.avsforum.com/forum/40-ole...l#post35174522
http://www.avsforum.com/forum/40-ole...l#post39870898

Last edited by sillysally; Yesterday at 02:00 PM.
sillysally is offline  
post #4369 of 4371 Old Yesterday, 05:07 PM
Newbie
 
markanini's Avatar
 
Join Date: Sep 2008
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 0
After trying different gamma settings I've settled on gamma 2.35 relative 0% black output offset. Google tells me 2.35-2.4 gamma is used by most colorists, 2.35 is specified by european broadcast standards. US broadcast standards specify 2.2 and some broadcast monitors measure 2.2 out of the factory but this is mostly ignored/overridden. Recent Apple displays measure 2.35-2.4 out of the factory. Hopefully this means I have good grounds for my choice of gamma settings besides looking good subjectively. This wouldn't be practical on a low-contrast display without Graeme and Florians work so I want to express thanks for their work!

Last edited by markanini; Yesterday at 05:21 PM.
markanini is offline  
post #4370 of 4371 Old Yesterday, 07:47 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 882
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 216 Post(s)
Liked: 173
Quote:
Originally Posted by sillysally View Post
I see Argyll CMS has released version 1.9.0.
Graeme am I missing something, I don't see support for the Lumagen.??
eeColor processor - ArgyllCMS
Unfortunately it is still on the "to do" list. There are some preparations to support it,
but I simply ran out of time for it and the Prisma. The spectraval is supported though.
gwgill is offline  
post #4371 of 4371 Old Today, 07:50 AM
Member
 
Join Date: Jan 2015
Posts: 52
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 33 Post(s)
Liked: 4
Hi,

The guide says to set the target white point in Displaycal as 'as measured', and to use the interactive adjustment to set the white point using the TV controls. I am just wondering why this is better than not using any of the TV controls and just choosing a white point target of 6500k in Displaycal?
iSeries is online now  
Sponsored Links
Advertisement
 
Reply Display Calibration



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