MadVR - ArgyllCMS - Page 109 - 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: 
 285Likes
Reply
 
Thread Tools
post #3241 of 3848 Old 04-03-2015, 02:43 AM
Member
 
RajinderGill's Avatar
 
Join Date: Aug 2007
Posts: 37
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 16
Quote:
Originally Posted by sillysally View Post
Ran a Full calibration and a very large profile.
Used 2.9.0.6 and argyllCMS 1.7b dated 03/25.
madVR 0.87.14
Used a K10-A

Had the same issue with black level bars, as is being reported.
When I turn off my eecolor box the 17 and up bars all appeared as they should.

ss
That means the lut calibrated the lower levels at values your meters can see but the eye cannot. Is this a dark room? If it is, then you have a similar situation to us.

We can try what GW suggested and move the meters further back.

Last edited by RajinderGill; 04-03-2015 at 02:49 AM.
RajinderGill is offline  
Sponsored Links
Advertisement
 
post #3242 of 3848 Old 04-03-2015, 03:57 AM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 4,187
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 418 Post(s)
Liked: 470
Quote:
Originally Posted by RajinderGill View Post
That means the lut calibrated the lower levels at values your meters can see but the eye cannot. Is this a dark room? If it is, then you have a similar situation to us.

We can try what GW suggested and move the meters further back.
Yes I am viewing my 65VT60 in my man cave, no lights.
From looking at the Gamma curve (BT.1886), it looks as if it is reversed from high to low.
Other than the issue with the flashing bars, every thing looks very good.
I did use on screen mode for the K10, but I have done off screen with the same results.
Did a 493 point testchart report, see screen shot.

ss
Attached Thumbnails
Click image for larger version

Name:	Screen shot.jpg
Views:	46
Size:	178.4 KB
ID:	643729  
sillysally is offline  
post #3243 of 3848 Old 04-03-2015, 04:08 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,811
Mentioned: 63 Post(s)
Tagged: 0 Thread(s)
Quoted: 949 Post(s)
Liked: 807
Quote:
Originally Posted by sillysally View Post
Ran a Full calibration and a very large profile.
Used 2.9.0.6 and argyllCMS 1.7b dated 03/25.
madVR 0.87.14
Used a K10-A

Had the same issue with black level bars, as is being reported.
When I turn off my eecolor box the 17 and up bars all appeared as they should.

ss
In your case it is probable that your plasma can't reproduce the required light levels.
zoyd is online now  
post #3244 of 3848 Old 04-03-2015, 04:33 AM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 763
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 144 Post(s)
Liked: 123
Quote:
Originally Posted by RajinderGill View Post
I have never overridden the black levels and it still happens.
If you email me your dispcal command line, .cal file, colprof command line & .ti3 file, and your collink command line, I'll look into it.
gwgill is offline  
post #3245 of 3848 Old 04-03-2015, 04:34 AM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 763
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 144 Post(s)
Liked: 123
Quote:
Originally Posted by sillysally View Post
Ran a Full calibration and a very large profile.
Used 2.9.0.6 and argyllCMS 1.7b dated 03/25.
madVR 0.87.14
Used a K10-A

Had the same issue with black level bars, as is being reported.
When I turn off my eecolor box the 17 and up bars all appeared as they should.
What has changed since it was last working well for you ?
gwgill is offline  
post #3246 of 3848 Old 04-03-2015, 05:21 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 612
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 279 Post(s)
Liked: 175
Quote:
Originally Posted by RajinderGill View Post
I have never overridden the black levels and it still happens.
Not knowingly of course, but if you used DCG 2.9.x beta, then it will use the newly introduced -b switch for collink that Graeme mentioned.
fhoech is online now  
post #3247 of 3848 Old 04-03-2015, 06:03 AM
Senior Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 312
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 120
Ok, so i'm bring my issue to the main thread as I am at a complete loss here! I would like to thank Fhoech, Zoyd, and N3W813 in assisting me on the side in the troubleshooting of this anomaly. It is obvious there is extreme color clipping going on if you reference the image below and I feel as though I have exhausted every avenue to remedy the issue. Here is a link to my file if anyone is interested in looking at the data. Also, here is my gear/software info:

  • My display is a Sharp Elite Pro-70x5FD
  • eeColor Box
  • I am utilizing isf night/day
  • i1 display profiled against my i1 pro 2
  • Dell laptop Latitude E6540 (Intel Graphics 4600 hdmi out)
  • HDMI output forced 0-255 with madVR tweak
  • MadVR configured 16-235 with my display expecting 16-235
  • I ran a custom 10,000 patch measure "Profile only" However, I did a preliminary calibration to bring my display close within specs
  • I Used Dispcal 2.9.0.6 and argyllCMS 1.7b dated 03/25.
  • I used madVR 0.87.14

I have tried multiple trouble shooting methods, surpassing my eeColor and hooking directly into my display and same results. I tried letting my laptop output 16-235 and set madVR tpg to 0-255...same results. I not sure where to go from here and any help would be greatly appreciated. Thanks in advance!

Ryan



Last edited by mascior; 04-03-2015 at 06:09 AM.
mascior is offline  
post #3248 of 3848 Old 04-03-2015, 06:55 AM
AVS Special Member
 
Brian Hampton's Avatar
 
Join Date: Apr 2000
Posts: 7,376
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 360 Post(s)
Liked: 193
Went through this process for the second time today.

(I think something went wrong the first time... I left it running and my projector turned itself off at some point and the resulting LUT file was close to 100MB.)

So, anyway.... I bet I don't know 3% of what I'm doing with this process and yet the results are spectacular! I'm so happy.... Thank you to everyone involved with all these programs... I've never seen my projector throw an image like this ... beautiful!

=Brian
Brian Hampton is offline  
post #3249 of 3848 Old 04-03-2015, 06:57 AM
Member
 
RajinderGill's Avatar
 
Join Date: Aug 2007
Posts: 37
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 16
Quote:
Originally Posted by gwgill View Post
If you email me your dispcal command line, .cal file, colprof command line & .ti3 file, and your collink command line, I'll look into it.
I'm using dispcal gui 2.6.0

Let me know what you need and I'll send it over.

Currently I've managed to work around the issue by setting a black level of 0.02 in a custom source file.
RajinderGill is offline  
post #3250 of 3848 Old 04-03-2015, 07:17 AM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 4,187
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 418 Post(s)
Liked: 470
Quote:
Originally Posted by zoyd View Post
In your case it is probable that your plasma can't reproduce the required light levels.
Yes I would agree, but when I turn off eecolor and just have my pre internal setting, all bars (17 and up) show. That's what I find strange.

Quote:
Originally Posted by gwgill View Post
What has changed since it was last working well for you ?
I wish I knew for sure.
I am sorry for this vague reply but I really don't know.

ss
sillysally is offline  
post #3251 of 3848 Old 04-03-2015, 07:22 AM
Senior Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 312
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 120
Quote:
Originally Posted by sillysally View Post
Yes I would agree, but when I turn off eecolor and just have my pre internal setting, all bars (17 and up) show. That's what I find strange.
My display exhibits the same behavior. With the eeColor box on, 18 and up flash. However, on bypass I can get 17 and up to flash. A Pure power gamma 2.22 I set black offset to 98% and can get 17 to flash. However, with bt.1886 17 flashes?!?

Ryan
mascior is offline  
post #3252 of 3848 Old 04-03-2015, 07:32 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,811
Mentioned: 63 Post(s)
Tagged: 0 Thread(s)
Quoted: 949 Post(s)
Liked: 807
Quote:
Originally Posted by sillysally View Post
Yes I would agree, but when I turn off eecolor and just have my pre internal setting, all bars (17 and up) show. That's what I find strange.




ss

If you set 17 manually to the first available dither level using your brightness control that may actually be too bright for a 2.4 transfer curve when you have very deep blacks
zoyd is online now  
post #3253 of 3848 Old 04-03-2015, 07:37 AM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 4,187
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 418 Post(s)
Liked: 470
Quote:
Originally Posted by mascior View Post
My display exhibits the same behavior. With the eeColor box on, 18 and up flash. However, on bypass I can get 17 and up to flash. A Pure power gamma 2.22 I set black offset to 98% and can get 17 to flash. However, with bt.1886 17 flashes?!?

Ryan
Yes I have tried what you suggested with no luck.

btw, this is just a shot in the dark about your profile report. Have you tried using the setting in the screen shot.

ss

Edit, I used Custom not 2.2 as the screen shot shows.
Attached Thumbnails
Click image for larger version

Name:	Making 3DLUT.jpg
Views:	41
Size:	137.5 KB
ID:	644073   Click image for larger version

Name:	Profile info.jpg
Views:	44
Size:	117.6 KB
ID:	644081  

Last edited by sillysally; 04-03-2015 at 07:45 AM.
sillysally is offline  
post #3254 of 3848 Old 04-03-2015, 07:50 AM
Senior Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 312
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 120
Quote:
Originally Posted by sillysally View Post
Yes I have tried what you suggested with no luck.

btw, this is just a shot in the dark about your profile report. Have you tried using the setting in the screen shot.

ss
Thanks ss, that is exactly how I configure my LUT. After a 10,000 path run, I still average 0.5, 4.0 max, and testing with my color clipping patterns...i'm flashing on all colors way above 235. The issue I have is sub par artifacts and color banding with normal viewing content. Something just doesn't add up.

Ryan
mascior is offline  
post #3255 of 3848 Old 04-03-2015, 08:48 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 612
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 279 Post(s)
Liked: 175
Quote:
Originally Posted by RajinderGill View Post
I'm using dispcal gui 2.6.0

Let me know what you need and I'll send it over.

Currently I've managed to work around the issue by setting a black level of 0.02 in a custom source file.
If you look into C:\Users\YourUsername\AppData\dispcalGUI\storage\N ame_Of_Profile, you'll find the files Graeme mentioned. The commandline used can be seen in the Name_Of_Profile.log file in the same folder.
fhoech is online now  
post #3256 of 3848 Old 04-03-2015, 09:41 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 612
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 279 Post(s)
Liked: 175
Quote:
Originally Posted by mascior View Post
Thanks ss, that is exactly how I configure my LUT. After a 10,000 path run, I still average 0.5, 4.0 max, and testing with my color clipping patterns...i'm flashing on all colors way above 235. The issue I have is sub par artifacts and color banding with normal viewing content. Something just doesn't add up.
Have you run a measurement report with a few hundred patches? I'm sure the problem will show up in reports as well if enough points (with enough density in the problem regions) are measured.
I've looked at a few test images through the 3D LUTs and the problem is mostly apparent with saturated blues as well as in the saturated yellow/green region (and to some extent in orange/red as well). It's not apparent with all content that I looked at (in-gamut material actually looks fine), but everything that's in the out-of gamut regions has severe artifacting.

DCG (dispcalGUI) - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.0.3 (released 2015-07-06) | DCG on Facebook

Previous development snapshot (outdated!): 3.0.2.1
(released 2015-06-30) | Changelog
fhoech is online now  
post #3257 of 3848 Old 04-03-2015, 12:02 PM
Senior Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 312
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 120
Quote:
Originally Posted by fhoech View Post
Have you run a measurement report with a few hundred patches? I'm sure the problem will show up in reports as well if enough points (with enough density in the problem regions) are measured.
I've looked at a few test images through the 3D LUTs and the problem is mostly apparent with saturated blues as well as in the saturated yellow/green region (and to some extent in orange/red as well). It's not apparent with all content that I looked at (in-gamut material actually looks fine), but everything that's in the out-of gamut regions has severe artifacting.
What you have described, is exactly what I see...In gamut material looks fantastic and near/out of gamut looks horrible. I guess the million dollar question is...can it be fixed. @ N3W813, if possible, could you send me copy or a link to your last successful run on your Elite? I would like to cross reference with my data. I want to see if I can rule out my laptop, display, or software...IDK Thanks for everyone's help.

Ryan
mascior is offline  
post #3258 of 3848 Old 04-03-2015, 12:05 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,811
Mentioned: 63 Post(s)
Tagged: 0 Thread(s)
Quoted: 949 Post(s)
Liked: 807
zoyd is online now  
post #3259 of 3848 Old 04-03-2015, 12:22 PM
Senior Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 312
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 120
Quote:
Originally Posted by zoyd View Post
I thought the elite native gamut was large enough to avoid clipping
My thoughts exactly...I know it's not the perfect display when it comes to color gamut accuracy...but it's not exactly a budget display and one would expect it to at least hit rec709! This is why I think either I have a faulty display or my laptop is distorting the RGB signal in some way.

Ryan
mascior is offline  
post #3260 of 3848 Old 04-03-2015, 12:30 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,811
Mentioned: 63 Post(s)
Tagged: 0 Thread(s)
Quoted: 949 Post(s)
Liked: 807
zoyd is online now  
post #3261 of 3848 Old 04-03-2015, 12:55 PM
Senior Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 312
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 120
Quote:
Originally Posted by zoyd View Post
Measure the gamut edges using BD and compare to your profile
With the 3DLUT loaded into my eeColor box, I have ran measurements against my disc and grayscale, gamma, all color points, and saturation look great! Red is dE of 0.5, Green dE 0.5, Blue dE 1.7, Yellow dE 0.6, Cyan dE 0.8, and Magenta dE 0.6 all this is at 100% sat/100%amp. Even referencing against my color checker patterns...all under a dE of 1.0 and my color clipping pattern...all past 235?!? Do I need to measure a specific pattern in my problem areas of my gamut?

Ryan
mascior is offline  
post #3262 of 3848 Old 04-03-2015, 01:07 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,811
Mentioned: 63 Post(s)
Tagged: 0 Thread(s)
Quoted: 949 Post(s)
Liked: 807
In that case I don't understand where the clipping is coming from, maybe Florian can check your measures file for misreads
zoyd is online now  
post #3263 of 3848 Old 04-03-2015, 03:16 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 612
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 279 Post(s)
Liked: 175
Quote:
Originally Posted by mascior View Post
With the 3DLUT loaded into my eeColor box, I have ran measurements against my disc and grayscale, gamma, all color points, and saturation look great! Red is dE of 0.5, Green dE 0.5, Blue dE 1.7, Yellow dE 0.6, Cyan dE 0.8, and Magenta dE 0.6 all this is at 100% sat/100%amp. Even referencing against my color checker patterns...all under a dE of 1.0 and my color clipping pattern...all past 235?!? Do I need to measure a specific pattern in my problem areas of my gamut?
Primaries and secondaries as well as less saturated colors (like some of the ColorChecker ones) don't seem to be problematic, it's the saturated areas between primaries and secondaries. Here the forward check through the profile vs. Rec 709 (100%, dE*00):

White = 0.21
Red = 2.1
Green = 1.35
Blue = 2.3
Cyan = 0.56
Magenta = 0.56
Yellow = 1.25

Now e.g. some problem areas vs. Rec. 709 w/ 2.2 gamma (R% G% B%, dE*00):

100 75 0 = 6.80
100 50 0 = 4.60
100 25 0 = 5.62
25 100 0 = 3.85
50 100 0 = 5.50
75 100 0 = 7.45
0 25 100 = 12.60
0 50 100 = 14.95
0 75 100 = 11.74

Visual check through a BT.1886 3D LUT from the profile:
Click image for larger version

Name:	GrangerRainbow1920_16bit_TVlevels-madVR-2015-03-30-S-XYZLUT.Rec709.png
Views:	65
Size:	163.1 KB
ID:	644913

Quote:
Originally Posted by zoyd View Post
In that case I don't understand where the clipping is coming from, maybe Florian can check your measures file for misreads
The measurements seem reasonable, from the two data sets I have here from @mascior , one 679 and one 10 000, but some of the measured colors seem oddly desaturated (at the locations where the "dents" are in the gamut graph @mascior posted above, and the measurement VRML shows the same "dents").

DCG (dispcalGUI) - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.0.3 (released 2015-07-06) | DCG on Facebook

Previous development snapshot (outdated!): 3.0.2.1
(released 2015-06-30) | Changelog

Last edited by fhoech; 04-03-2015 at 03:29 PM. Reason: Added measurement VRML link
fhoech is online now  
post #3264 of 3848 Old 04-03-2015, 03:58 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 763
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 144 Post(s)
Liked: 123
Quote:
Originally Posted by RajinderGill View Post
I'm using dispcal gui 2.6.0
Let me know what you need and I'll send it over.
See my previous post.
gwgill is offline  
post #3265 of 3848 Old 04-03-2015, 04:19 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,811
Mentioned: 63 Post(s)
Tagged: 0 Thread(s)
Quoted: 949 Post(s)
Liked: 807
@mascior perhaps try using the dispcalgui window instead of madtpg to rule out the pattern generator. You'll need to add the -E switch to dispread for video level output
zoyd is online now  
post #3266 of 3848 Old 04-03-2015, 04:54 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 763
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 144 Post(s)
Liked: 123
Quote:
Originally Posted by mascior View Post
My display exhibits the same behavior. With the eeColor box on, 18 and up flash. However, on bypass I can get 17 and up to flash. A Pure power gamma 2.22 I set black offset to 98% and can get 17 to flash. However, with bt.1886 17 flashes?!?
It's hard off hand to know if the first step above black should be visible or not, since it depends on a lot of factors. A power curve is extremely shallow near zero, and there is no surprise if a pure power curve is being used, that the first step is invisible. For an input offset (BT.1886) style curve, it depends on the power and what the offset is. If the black point is quite dark, then again the step might be too small to see, even if it is being displayed correctly.

For instance, let's say you have a black of 0 and a white of 200 cd/m^2 and a gamma of 2.3.
The first step is (1/220)^2.3 * 200 = 0.000004 cd/m^2.
I suspect you would have trouble distinguishing that step, or even several above it, even
if it was displayed accurately.

Let's look at an input offset/BT.1886 case: Black 0.02 cd/m^2, white 200 cd/m^2, gamma 2.4.
Input offset to match the black will be 0.022018
The first step will be ((1/220 + 0.022018)/1.022018)^2.4 * 200 = 0.03138, a step of about 0.01 cd/m^2.
I guess this may be visible under good circumstances, but it's hard to be sure - having patches > 10 cd/m^2 in your view may make this impossible - i.e. this step may only be visible if you just display the first few steps in isolation on an otherwise black screen. Against full white, its's a ratio of 0.01/200 = 20000:1, which may be beyond human capability - internal glare in our eyes limits our static contrast range to about 1000:1.

With manual control and a sharp enough transition to adjust, then you can force 17 to flash - but being visible doesn't make it correct.
fhoech likes this.
gwgill is offline  
post #3267 of 3848 Old 04-03-2015, 05:08 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 763
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 144 Post(s)
Liked: 123
Quote:
Originally Posted by fhoech View Post
Visual check through a BT.1886 3D LUT from the profile:
Attachment 644913
If the profiling and linking is working correctly, then transitions like that are symptoms of non-monotonicity (reversals) in the apparent display behavour. If it is not a problem with the display itself, then it is likely to be problems in the measurement process - either patch display or measurement drift or faults, or display drift. Patch delay time problems can be a source of measurement faults. If such random faults are present, then running large patch sets increases the chances of such faulty readings being included in the profile.
gwgill is offline  
post #3268 of 3848 Old 04-03-2015, 05:09 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 612
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 279 Post(s)
Liked: 175
Quote:
Originally Posted by zoyd View Post
@mascior perhaps try using the dispcalgui window instead of madtpg to rule out the pattern generator. You'll need to add the -E switch to dispread for video level output
Good suggestion.

I just stumbled on something:
MadVR - ArgyllCMS
MadVR - ArgyllCMS
@N3W813 had issues with his Sharp Elite Pro 70x5FD some while back that at least sound similar, and he was able to solve with a custom testchart. I've created a testchart with about the same parameters. Maybe this'll help? I am really curious now how the gamut shape of @N3W813 profile looked back then (in L*a*b*).

DCG (dispcalGUI) - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.0.3 (released 2015-07-06) | DCG on Facebook

Previous development snapshot (outdated!): 3.0.2.1
(released 2015-06-30) | Changelog
fhoech is online now  
post #3269 of 3848 Old 04-03-2015, 05:19 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 612
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 279 Post(s)
Liked: 175
Quote:
Originally Posted by gwgill View Post
If it is not a problem with the display itself, then it is likely to be problems in the measurement process - either patch display or measurement drift or faults, or display drift. Patch delay time problems can be a source of measurement faults. If such random faults are present, then running large patch sets increases the chances of such faulty readings being included in the profile.
Good point. I have two data sets of very different sizes (679 and 10000) from @mascior , the resulting 3D LUTs show very similar artifacting. My initial thought was that the underlying issue could be related to internal processing in the TV due to the TV's settings. E.g. in the past I've occasionally seen some very weird gamut shapes from measurements of displays with wide gamut, when they were driven in a restricted gamut mode via one of their presets. This was in the few cases that I know of then fixed by setting the display to "native" gamut mode.

DCG (dispcalGUI) - Graphical front-end for Argyll CMS display calibration and characterization
Current stable version 3.0.3 (released 2015-07-06) | DCG on Facebook

Previous development snapshot (outdated!): 3.0.2.1
(released 2015-06-30) | Changelog
fhoech is online now  
post #3270 of 3848 Old 04-03-2015, 07:57 PM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 4,187
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 418 Post(s)
Liked: 470
Quote:
Originally Posted by fhoech View Post
Good suggestion.

I just stumbled on something:
MadVR - ArgyllCMS
MadVR - ArgyllCMS
@N3W813 had issues with his Sharp Elite Pro 70x5FD some while back that at least sound similar, and he was able to solve with a custom testchart. I've created a testchart with about the same parameters. Maybe this'll help? I am really curious now how the gamut shape of @N3W813 profile looked back then (in L*a*b*).
In your testchart I see you have set "% Dark region emphasis 0"
Also you don't check "Centered steps" but you use Multidmensional 17.

What is your thinking with those settings?

I have run a profile using Lightspace and a Lumagen 2041 as pattern source. I also used one of my custom argyllCMS testcharts Ti1 converted for Lightspace. Same meter profile matrix, on screen.
The result with the black bars was the same with Lightspace's 3DLUT.

The one thing that troubles me, is why when I turn off eecolor do I see bars 17 and up easily?

Thanks,
ss
Attached Thumbnails
Click image for larger version

Name:	screen shot.jpg
Views:	24
Size:	301.0 KB
ID:	645241  
sillysally 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