MadVR - ArgyllCMS - Page 151 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 478Likes
Reply
 
Thread Tools
post #4501 of 4519 Old 01-30-2017, 02:18 PM
Newbie
 
Join Date: Jun 2015
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 0
Quote:
Originally Posted by madshi View Post
Something wrong with the 3dlut then. Did you use madTPG and did you tell ArgyllCMS/DisplayCAL that the 3dlut has to be video levels for input and output?
Yes I used these options (but the output range was greyed out to video levels, there was no way to change that).

Attached is a few seconds of video I used to test the WTW (rename the .doc to .mp4)
As soon as I enable the LUT the text becomes visible.
Attached Files
File Type: doc range.doc (147.6 KB, 12 views)
newpball is offline  
Sponsored Links
Advertisement
 
post #4502 of 4519 Old 01-30-2017, 03:05 PM
VBB
AVS Forum Special Member
 
VBB's Avatar
 
Join Date: May 2003
Location: Long Beach, CA
Posts: 2,205
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1009 Post(s)
Liked: 861
Quote:
Originally Posted by newpball View Post
Yes I used these options (but the output range was greyed out to video levels, there was no way to change that).

Attached is a few seconds of video I used to test the WTW (rename the .doc to .mp4)
As soon as I enable the LUT the text becomes visible.
You said your monitor expects full range. Did you then set the entire video chain to 0-255 (madVR, video driver, monitor)? That would be the only correct option, and the result should actually be the opposite: You should not be able to see BTB/WTW at all, because it would be clipped.
holdemphil likes this.
VBB is offline  
post #4503 of 4519 Old 01-30-2017, 03:20 PM
Newbie
 
Join Date: Jun 2015
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 0
Quote:
Originally Posted by VBB View Post
You said your monitor expects full range. Did you then set the entire video chain to 0-255 (madVR, video driver, monitor)? That would be the only correct option, and the result should actually be the opposite: You should not be able to see BTB/WTW at all, because it would be clipped.
Correct!

If I do not set the lut indeed the video levels are expanded to full range as it should!
It is only when I use the lut I get to see WTW, which I should not.
newpball is offline  
 
post #4504 of 4519 Old 01-30-2017, 04:42 PM
Member
 
Join Date: Jul 2016
Location: Finland
Posts: 19
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 1
Quote:
Originally Posted by newpball View Post
Correct!

If I do not set the lut indeed the video levels are expanded to full range as it should!
It is only when I use the lut I get to see WTW, which I should not.
What about creating the 3DLUT with the Input encoding TV RGB 16-235 (clip WTW)? Would that help here? I have the same chain as you and only see that WTW pattern when I switch levels in madVR, and not when I use a 3DLUT (though I haven't needed to clip WTW).
holdemphil is offline  
post #4505 of 4519 Old 01-30-2017, 06:22 PM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,401
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 731 Post(s)
Liked: 250
Quote:
Originally Posted by newpball View Post
Correct!

If I do not set the lut indeed the video levels are expanded to full range as it should!
It is only when I use the lut I get to see WTW, which I should not.
only WTW ?

did you corrected the white point before calibration?
mightyhuhn is offline  
post #4506 of 4519 Old 01-30-2017, 09:10 PM
Advanced Member
 
HDgaming42's Avatar
 
Join Date: Sep 2006
Location: Alberta, Canada
Posts: 714
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 19
Quote:
Originally Posted by madshi View Post
I'm not sure if it makes sense discussing "macroblocks" as long as your configuration is not correct.

Please first fix the configuration, then double check it by using a test pattern. If steps 16-235 have different brightness steps, but 0-15 are all black and 236-255 are all white, then black/white levels seems to be setup correctly. Please take extra care to set this up correctly, because everything else depends on that.
Switched to PC 0-255. Confirmed bolded. Reran LUT. Behold, what I refer to as "macroblocks".

Frustrating as all hell. I'm hoping you can help me troubleshoot this further...

LUT with PC 0-255


LUT Unity / Passthrough


LUT with PC 16-235


Passthrough / Unity LUT with TV Brightness Maxed


LUT with PC 0-255


LUT with Unity


The only configuration differences between the PC 0-255 LUT and the PC 16-235 LUT (beyond the video card output levels) was running fewer patches (154 vs 11K). All I did was drop the number of patches down. I didn't have to adjust brightness or contrast on the plasma--patterns looked good. DisplayCAL was happy with the white point RGB Balance and read 117cd/m2.

I did see something fly by in the logs along the lines of "black point hack failed to fire". Does that warrant looking into?

Last edited by HDgaming42; 01-30-2017 at 09:22 PM.
HDgaming42 is offline  
post #4507 of 4519 Old 01-30-2017, 09:49 PM
Newbie
 
Join Date: Jun 2015
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 0
Quote:
Originally Posted by mightyhuhn View Post
only WTW ?

did you corrected the white point before calibration?
Yes, there is no BTB, I checked that!

For Resolve and Scratch I generate luts without "Apply calibration (vcgt)" do I need to check that for the madVR lut?

I will check with a custom (6500) white level, perhaps that is indeed the issue.
newpball is offline  
post #4508 of 4519 Old 01-31-2017, 04:09 AM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,401
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 731 Post(s)
Liked: 250
@HDgaming42
can you create a 3D LUT for madVR from the reading and check the result with a proper black level test pattern.
if the results are correct you don't have to do the readings again which saves a lot of time.

for me it looks like all images are black clipping even the totally undersaturated one.

@newpball
i don't know it is on by default so leave it on.
and white level is still as measured right?
mightyhuhn is offline  
post #4509 of 4519 Old 01-31-2017, 05:32 AM
Senior Member
 
Join Date: Dec 2015
Posts: 224
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 138 Post(s)
Liked: 19
Hello! I wonder which settings are correct when watching movies via computer? I have a computer hooked to Pioneer SC-95 to the projector Epson HC-2040. Should the settings be RGB or ycbcr 0-255 or 16-235? Anything else I can do to make it correct and "cinemalike"?

Skickat från min SM-G920F via Tapatalk
westbergjoakim is online now  
post #4510 of 4519 Old 01-31-2017, 06:14 AM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,401
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 731 Post(s)
Liked: 250
depends on your playback device and the settings in it. test it with test file like the black clipping test file from AVS HD 709 - Blu-ray & MP4 Calibration
using YCbCr on a PC is nearly for sure not optimal.

Quote:
Anything else I can do to make it correct and "cinemalike"?
get a colorimeter and calibrate it.
that's the topic of this thread.
mightyhuhn is offline  
post #4511 of 4519 Old 01-31-2017, 06:25 AM
Advanced Member
 
HDgaming42's Avatar
 
Join Date: Sep 2006
Location: Alberta, Canada
Posts: 714
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 19
Quote:
Originally Posted by mightyhuhn View Post
@HDgaming42
can you create a 3D LUT for madVR from the reading and check the result with a proper black level test pattern.
if the results are correct you don't have to do the readings again which saves a lot of time.

for me it looks like all images are black clipping even the totally undersaturated one.
Heading to work--I can try this tonight.

The shots are from "Dark Knight" and "Stark Trek: Into the Darkness". Both scenes are VERY dark. I agree there seems to be no detail in the blacks, but that's how those scenes look out of my PS3 direct to the TV...PS3 being YPbPr and superwhite enabled, TV 16-235.
HDgaming42 is offline  
post #4512 of 4519 Old 01-31-2017, 06:45 AM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,401
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 731 Post(s)
Liked: 250
i don't have these movies so what ever and it is a camera.

just use test pattern and ignore the PS3 make sure it is working correctly on the PC.

BTW. AVR and other stuff in your chain can in theory do a range conversation.
i personally would start with a direct connection to the PC and calibrate from there. the readings are still correct so you don't have to do them again as long as the chain from the PC to the screen is correct and there is no easier chain than PC->TV.
mightyhuhn is offline  
post #4513 of 4519 Old 01-31-2017, 04:42 PM
Newbie
 
Join Date: Jan 2017
Posts: 3
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 0
Hi

I have some weird problem with Rec709 video, my LED widegamut monitor (AdoibeRGB 99%) and madVR.
I build madVR LUT3D in the usual way: madTPG, configure DisplayCAL to use madVR and measure XYZLUT, then make LUT3D (both 16-235 tp 16-235), abs colorimetric with WP scaling. MadVR set to 0-255. Computer to monitor with Displayport full range 10bpc (also tried with 8bpc) . I mean configuration is OK.

"Some videos", olny a few, show native gamut greenes and cians on scenes instead of Rec709 ones. Their neigbour colors are "almost" full saturation green or cian. It looks like when feed with such full green or cian, ArgyllCMS's LUT3D was broken or madVR has a serious bug.

I've tried making LUT3D from my desktop GPU-calibration profile, same results. I've tried setting as source profile "monitor's driver" ICM (factory synthetic profile with EDIT ganut and D65 white), same results. I've even tried set AdobeRGB profile as input to LUT3D, same results. I've tries also realtive colorimetric intent and it showed the same artifacts but over bigger area.
I've tried also 16-235 and 16-235 clip WTW, same result on those videos.

The only combination that stop this artifact is to build on OS console with ArgyllCMS a LUT3D with 0-255 in and 0-255 out.
I know that this setting is wrong, but visually it's almost equal to proper LUT3D "monitor to rec709" and does not show this ugly native gamut artifacts. When verifing that LUT3D (0-255 one) with HCFR showed 1.x-2dE oversatured primaries over Rec709 RGB primaries... but at least it has not such "AdobeRGB-like" green-cian patches where madVR shoud display Rec709 ones.

"MP4-2c\100% Color\4-100% Green.mp4" from a Test patch frm AVSForum llok gook, Rec709 green. Also Florian's "GrangerRainbow1920.avi" shows proper color, no oversaturated or AdobeRGB like cian or greens.

Any advice? Is it possible that madVR or ArgyllCMS LUT3D maker fails when some video with "WTW" green or cians is feed into their data path?
If somebody has a widegamut monitor calibrated to its full native gamut, have you noticed such artifacts?

I've seen it in Premiere CC2015 & CC2017 videos recompresed by me, or even in commercial DVD titles like "Focus" (Will Smith and Margot Robbie, the scene where Will wears a cian shirt or when ther ase some cars covered with a cian plastic), also in some MP4 anime files.

Here's a link to a Premiere CC2017 recompresed sample shortened to 2sec. I'll try to atach later an snapshot of MPC-HC with the "AdobeRGB cian" patch encoded in a JPG with my EDID/driver monitor profile embebed in that JPG.
https (double dot) (slash slash) 1fichier (dot) com (slash) ?idolmb5hge
(sorry, I cannot attach links or images)
The file is named HDR but it's not HDR (Premiere converted it to SDR) and this issue does not seem to be related to that conversion because "non-HDR" TS anime shows converted to MP4 showed teh same issue and commercial DVD (wich obviously are not HDR) showed the same issues. I mean, forget file name.

Thanks for your time and I hope somebody knows how to fix this issue!.
Hotarou is offline  
post #4514 of 4519 Old 01-31-2017, 04:45 PM
Newbie
 
Join Date: Jan 2017
Posts: 3
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 0
Sorry for the typos!
Hotarou is offline  
post #4515 of 4519 Old 02-02-2017, 02:20 AM
Newbie
 
Join Date: Jan 2017
Posts: 3
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 0
Hi

Nobody here? Does somebody see the oversaturaed blue on top left corner in the MP4 file that I linked with a widegamut monitor and a native to rec709 LUT3D with madVR?

Thanks in advance
Hotarou is offline  
post #4516 of 4519 Old 02-14-2017, 04:44 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 4,466
Mentioned: 22 Post(s)
Tagged: 1 Thread(s)
Quoted: 1071 Post(s)
Liked: 1540
Quote:
Originally Posted by HDgaming42 View Post
Switched to PC 0-255. Confirmed bolded. Reran LUT. Behold, what I refer to as "macroblocks".

Frustrating as all hell. I'm hoping you can help me troubleshoot this further...

LUT with PC 0-255


LUT Unity / Passthrough


LUT with PC 16-235


Passthrough / Unity LUT with TV Brightness Maxed


LUT with PC 0-255


LUT with Unity


The only configuration differences between the PC 0-255 LUT and the PC 16-235 LUT (beyond the video card output levels) was running fewer patches (154 vs 11K). All I did was drop the number of patches down. I didn't have to adjust brightness or contrast on the plasma--patterns looked good. DisplayCAL was happy with the white point RGB Balance and read 117cd/m2.

I did see something fly by in the logs along the lines of "black point hack failed to fire". Does that warrant looking into?
Just to inform, we have discussed that issue via email between us, usually the ways the user has to use to locate issues to a system is to evaluate the performance by taking post-calibration measurements with a meter, verify color reproduction patterns (this is why I have added about 150 Color Reproduction patterns to my calibration disk for that evaluation) or watch real content to see if there problems introduced.

What I did? since LightSpace CMS has tools and you can import any 3D LUT correction table, I can locate the errors in a cube generated correction in a few seconds, not only for those generated for my system but from any other user which has used any other calibration solution and it's experiencing problems.

There 3 ways to locate the problems with LightSpace, looking the: 1) 1D LUT Viewer, 2) Looking and rotating the 3D Cube Viewer, 3) using LUT Preview.



Looking the 1D LUT Viewer, I saw a strange (large) lift of near black details, is the the reason of the strange blocking of near black.



I used LightSpace's LUT manipulation filter to fix that problem (flatten near black) and user verified that blocking issue gone after loaded the fixed correction but it's not the solution, he has to find out why this is happening.

Also another strange issue is that the user had pre-calibrated his 100% White before starting the display profiling but as you see at top right end of the 1D LUT Viewer the RGB channels balance has changed (re-adjusted a lot).



The Pink line is the UNITY line, I added this manually using PhotoShop for better understanding of what the LUT is doing to the signal.

This shows that there something wrong not related with hardware used as 3D LUT table holder (eeColor 3D LUT Box), but something related to software or setup configuration, just it needs some further testing from the user.
holdemphil likes 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
ConnecTEDDD is online now  
post #4517 of 4519 Old 02-14-2017, 08:21 AM
AVS Forum Special Member
 
madshi's Avatar
 
Join Date: May 2005
Posts: 6,025
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 501 Post(s)
Liked: 474
Very nice!
madshi is offline  
post #4518 of 4519 Old Yesterday, 08:26 PM
Advanced Member
 
HDgaming42's Avatar
 
Join Date: Sep 2006
Location: Alberta, Canada
Posts: 714
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 19
Quote:
Originally Posted by ConnecTEDDD View Post
This shows that there something wrong not related with hardware used as 3D LUT table holder (eeColor 3D LUT Box), but something related to software or setup configuration, just it needs some further testing from the user.
Sadly my media room is disassembled ATM due to a carpet installation. Followed by baseboards when I have the time. I was trying to get a base profile of my plasma with Lightspace DPS and TED's disc when I had to tear it all down.

I'll be back to report my findings. Thanks for the help!

edit: Ted--can't remember if I emailed you that a run at 80nits showed exactly the same issue(s).
ConnecTEDDD likes this.
HDgaming42 is offline  
post #4519 of 4519 Old Today, 02:46 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 4,466
Mentioned: 22 Post(s)
Tagged: 1 Thread(s)
Quoted: 1071 Post(s)
Liked: 1540
Quote:
Originally Posted by HDgaming42 View Post
edit: Ted--can't remember if I emailed you that a run at 80nits showed exactly the same issue(s).
Hi, I haven't received your new data.

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  
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