MadVR - ArgyllCMS - Page 148 - AVS Forum | Home Theater Discussions And Reviews

Forum Jump: 
 455Likes
Reply
 
Thread Tools
post #4411 of 4425 Old 12-02-2016, 06:49 AM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 407
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 29
I understand?
madvr options 10-bit and exclusive mode and madtpg fullscreen
I will get 10-bit patterns in dispcal for 3dlut?
For sure 10-bit?

but not 8-bit ?
anta1974 is offline  
Sponsored Links
Advertisement
 
post #4412 of 4425 Old 12-02-2016, 07:09 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 830
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 407 Post(s)
Liked: 266
Quote:
Originally Posted by anta1974 View Post
madvr options 10-bit and exclusive mode and madtpg fullscreen
I will get 10-bit patterns in dispcal for 3dlut?
For sure 10-bit?
IF your display actually supports it, yes. The internal processing of the display itself will probably dither down to 8-bit, even if it can receive and process 10-bit, because true 10-bit panels are rare AFAIK. You'll still get the benefit of finer gradation steps though, so this matters little in real-world situations.
anta1974 likes this.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.2.1 released 2016-11-25
Previous development snapshot (OUTDATED): 3.1.3.2 Beta (Windows/0install) released 2016-04-23 | Standalone | Changelog
DisplayCAL on Facebook
fhoech is offline  
post #4413 of 4425 Old 12-02-2016, 09:48 AM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 407
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 29
Of course.
I have 2 true 10-bit panels: samsung 55B7000 and panasonic 50VT50E.
anta1974 is offline  
post #4414 of 4425 Old 12-02-2016, 10:44 AM
Senior Member
 
anta1974's Avatar
 
Join Date: Jun 2009
Location: Ukraine
Posts: 407
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 56 Post(s)
Liked: 29
Dear Florian!
Edit testchart for 3dlut:
write patch 100% R=255 G=255 B=255, but
must be (for 10-bit) R=1024

Mistake?
anta1974 is offline  
post #4415 of 4425 Old 12-02-2016, 05:11 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 830
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 407 Post(s)
Liked: 266
Quote:
Originally Posted by anta1974 View Post
Dear Florian!
Edit testchart for 3dlut:
write patch 100% R=255 G=255 B=255, but
must be (for 10-bit) R=1024

Mistake?
The testchart values are always floating-point, no specific bitdepth. You're probably confused by the testchart editor status bar, which shows what the nearest 8-bit values of the selected test patch are (because people may be more familiar with 8-bit than floating point values). But these are purely informational.
anta1974 likes this.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.2.1 released 2016-11-25
Previous development snapshot (OUTDATED): 3.1.3.2 Beta (Windows/0install) released 2016-04-23 | Standalone | Changelog
DisplayCAL on Facebook
fhoech is offline  
post #4416 of 4425 Old 12-02-2016, 09:26 PM
VBB
AVS Special Member
 
VBB's Avatar
 
Join Date: May 2003
Location: Long Beach, CA
Posts: 2,165
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 984 Post(s)
Liked: 842
Quote:
Originally Posted by fhoech View Post
It would be nice to know at which point this file came into existence exactly, but that's probably impossible to find out after the fact. DisplayCAL does not call Argyll's synthcal utility, but it does `dispwin -s` to save the current videoLUT at certain points (although it'll always use a temporary directory for that, never the drive root).
Florian, I looked at the timestamp of the file and compared to which software I closed last that night, and it turns out it was HCFR. So, the file must have been created by HCFR after using madVR patterns to verify the 3D LUT created by DisplayCAL. I'll have to ask zoyd about this.
fhoech likes this.

Henry
Vizio M70-C3
Onkyo HT-RC360
Synology DS1813+ & DX513 x2
X-Rite i1Display Pro, ArgyllCMS, DisplayCAL, madVR, HCFR
HTPC running Kodi with MPC-HC + madVR, Plex for remote streaming
VBB is offline  
post #4417 of 4425 Old 12-06-2016, 09:27 AM
Member
 
Join Date: Jan 2015
Posts: 65
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 5
Last night I ran a standard 1500 patch run for MadVR. I used a white point of 'as measured' so the 3dlut corrects the white point. It looks good, apart from the AVSHD white clipping pattern - the bars flash green all the way to 254. Without the 3dlut, the bars are clipped at 235 as expected with an all PC levels chain (and the bars under 235 flash neutral grey). Does anyone know what could cause this, and if it's a problem? A 100% white test pattern displays as white the 3dlut enabled, and running 'report on calibrated display' from Displaycal reports that the white point is 6570k.
iSeries is offline  
post #4418 of 4425 Old 12-06-2016, 09:30 AM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,340
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 684 Post(s)
Liked: 225
try a clip WTW 3d lut.
mightyhuhn is online now  
post #4419 of 4425 Old 12-06-2016, 10:05 AM
Member
 
Join Date: Jan 2015
Posts: 65
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 40 Post(s)
Liked: 5
Quote:
Originally Posted by mightyhuhn View Post
try a clip WTW 3d lut.
Hi Huhn, sorry should have mentioned, also tried that - even with the clip WTW option the bars are still flashing all the way up to 254
iSeries is offline  
post #4420 of 4425 Old Yesterday, 07:02 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 830
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 407 Post(s)
Liked: 266
Hmm. If the 3D LUT is the only thing that corrects the whitepoint, then the white may have to be scaled down to fit within the destination gamut (this is what "Absolute colorimetric with whitepoint scaling" does, and the only alternatives are clipping, or using the display whitepoint as-is instead, i.e. relative colorimetric). This means part of the WtW range will be "pulled into" the 16..235 range. For normal video content, this shouldn't be a problem, but it can introduce color casts into WtW when the 3D LUT runs out of "headroom" (i.e. the WtW with the correct hue would exceed the maximum 8-bit value of 255 in one or several of the channels = clipping). I'm actually not sure what "clip WtW on input" does in such a case, Graeme may have to comment on this. Also, I'm not at all sure how WtW is handled in madVR when a 3D LUT is used (see http://forum.doom9.org/showthread.ph...14#post1709814 under "Known limitations and complications").
iSeries likes this.

DisplayCAL - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.2.1 released 2016-11-25
Previous development snapshot (OUTDATED): 3.1.3.2 Beta (Windows/0install) released 2016-04-23 | Standalone | Changelog
DisplayCAL on Facebook
fhoech is offline  
post #4421 of 4425 Old Yesterday, 02:13 PM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,340
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 684 Post(s)
Liked: 225
i still have problem with out of gamut colors created from chroma scaling.
and that ended with nothing at all.

madVR is still sending out of gamut information and the 3D LUT still can't handle them.
mightyhuhn is online now  
post #4422 of 4425 Old Yesterday, 02:24 PM
AVS Special Member
 
madshi's Avatar
 
Join Date: May 2005
Posts: 5,957
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 424 Post(s)
Liked: 431
I thought ArgyllCMS had a switch to clip BTB/WTW? If that doesn't work, you can probably use a custom pixel shader to clip BTB/WTW in madVR.
madshi is online now  
post #4423 of 4425 Old Yesterday, 02:27 PM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,340
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 684 Post(s)
Liked: 225
the WTW clipping doesn't work as i thought see above and the shader works if i remember correctly but eats a lot of processing power.
mightyhuhn is online now  
post #4424 of 4425 Old Yesterday, 03:46 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 903
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 231 Post(s)
Liked: 176
Quote:
Originally Posted by fhoech View Post
I'm actually not sure what "clip WtW on input" does in such a case, Graeme may have to comment on this. Also, I'm not at all sure how WtW is handled in madVR when a 3D LUT is used
It's hard to comment with nothing concrete to go on. Assumptions are being made that "the problem is WTW clipping", but it would be good to figure out whether this is what's actually going on.

What ArgyllCMS does: by default all Video encoded signals are assumed to be in the range 16..135, and any WTW input value is clipped by scaling the RGB towards black, to preserve hue. If no per channel input and output curves are being used (i.e. -ni -no or a 3dlut output format is selected), then this scaling is undone after color management in an attempt to pass WTW through to the TV. If the -e T input encoding option is used instead, the clipped output value remain clipped.

In creating the table for MadVR, you normally generate a video encoded 3d table (-et -Et).

Last edited by gwgill; Yesterday at 04:02 PM.
gwgill is offline  
post #4425 of 4425 Old Today, 03:44 AM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,340
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 684 Post(s)
Liked: 225
just to be sure -e T for collink correct?
mightyhuhn 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