MadVR - ArgyllCMS - Page 85 - AVS Forum
Forum Jump: 
 23Likes
Reply
 
Thread Tools
post #2521 of 2534 Old 08-12-2014, 09:12 AM
AVS Special Member
 
madshi's Avatar
 
Join Date: May 2005
Posts: 5,443
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 116
Feel free to create a bug entry in the madVR bug tracker, but you should be aware of that I won't have a multi-monitor setup to test with for probably the next 4 months, so I won't be able to do anything about it anytime soon.
madshi is online now  
Sponsored Links
Advertisement
 
post #2522 of 2534 Old 08-12-2014, 10:54 AM
Advanced Member
 
-Hitman-'s Avatar
 
Join Date: Nov 2004
Location: Manchester, UK
Posts: 660
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 34 Post(s)
Liked: 61
Quote:
Originally Posted by madshi View Post
Feel free to create a bug entry in the madVR bug tracker, but you should be aware of that I won't have a multi-monitor setup to test with for probably the next 4 months, so I won't be able to do anything about it anytime soon.


Ok no problem.


Thanks M.
-Hitman- is offline  
post #2523 of 2534 Old 08-12-2014, 05:28 PM
Newbie
 
IceB's Avatar
 
Join Date: Jan 2008
Location: Haifa, Israel
Posts: 10
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 10
Thumbs up 3DLUT report

madshi, msdVR is installed as well as the DispacalGUI and madTPG on the same PC. I do not remember the order of running the dispcalGUI and madTPG.
However, I have successfully created the 3DLUT following the guide.
I will share the workaround and some measurements charts.
First, I have installed the latest dispcalGUI (beta).
Then I have figured out that I have missed the "Dark region emphasis" change while creating the testchart - it was 0% on mine.
So I created the new testchart with 20% "Dark region emphasis" and passed the measurements without a problem.
I have rolled back to the 2.1 latest stable dispcalGUI and ran the same calibration again - passed it. But now I got an error during the profiling process : "Error - test_crt returned error code 3 ".
I do not know what that error means, so I have just restated the profiling and in a couple of hours it has passed successfully.
3DLUT was created with no issues.

So here are some charts from my setup :PC with AMD HD7970 -> Anthem MRX 300 -> JVC X30 (no CMS) -> Carada 96' Criterion Brilliant White Projection Screen.
All the measurements were taken with DTP-94 color meter with ColorHCFR with AVS HD 709 patterns (MP4).
The screenshots on the left are the measurements of the hardware calibrated projector without the 3DLUT. The screenshots on the right are the same projector with the created and applied 3DLUT on madVR.

Greyscale:


Primaries:


Luminance:


Gamma:


RGB:


Color temperatures:


CIE diagramm:


As you can see the grey scale delta E was decreased but the delta E of the primaries/secondaries was in fact increased but still to the acceptable level.
The gamma was increased as well. Overall the results with the 3DLUT applied are polished more than expected and the charts are almost flat.
The primaries/secondaries delta E might be the price paid for that.
Since I have used the controllers on my projector to calibrate it prior to the dispcalGUI calibration, I am curious to see how accurate is the dispcalGUI calibration and the madVR with 3DLUT on the less calibrated devices.

madshi and all the involved souls, thank you for the amazing work and the great tools you share with us. Thumbs UP !
IceB is online now  
post #2524 of 2534 Old 08-12-2014, 06:33 PM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 3,714
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 262
IceB.

If you are using Argyll/dispcal, you may want to use the measurement report found in dispcal>Tools.

I use the xl.ti1, measures 490 points.

ss
sillysally is offline  
post #2525 of 2534 Old 08-15-2014, 05:23 PM
Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 179
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 32
Quote:
If you are using Argyll/dispcal, you may want to use the measurement report found in dispcal>Tools.
Could we please establish that when talking about "dispcal", we should refer to the Argyll CMS dispcal.exe command line utility, and when talking about dispcalGUI, we refer to dispcalGUI (obviously)? Things get really confusing otherwise . If you don't want to type out "dispcalGUI" and want a (even) shorter form, I propose the abbreviation "DCG" (or "dcGUI", probably a little more concise).

--

On another note, I've updated the dispcalGUI Beta snapshot to 2.1.1.5. This continues the moderate visual overhaul I've started to give the user interface in 2.1.1.4, already has support for upcoming Argyll CMS 1.7 features (provisional), and I'm aiming for a stable release soon. Until then, feel free to try the snapshot (changes from last stable release) if you want. If you used the 0install setup, you can switch over to the "testing" version to get the snapshot. Alternatively, download a standalone version of the snapshot (for Windows) as zipfile.
fhoech is online now  
post #2526 of 2534 Old 08-15-2014, 07:34 PM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 3,714
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 262
Quote:
Originally Posted by fhoech View Post
Could we please establish that when talking about "dispcal", we should refer to the Argyll CMS dispcal.exe command line utility, and when talking about dispcalGUI, we refer to dispcalGUI (obviously)? Things get really confusing otherwise . If you don't want to type out "dispcalGUI" and want a (even) shorter form, I propose the abbreviation "DCG" (or "dcGUI", probably a little more concise).
Now I am confused.

So is the proper way to refer to what I posted above, dispcalGUI. Leave Argyll CMS out.
Or should I have said, Argyll CMS dispcal.exe command line utility when referring to dispcalGUI.?

ss
sillysally is offline  
post #2527 of 2534 Old 08-16-2014, 05:49 AM
Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 179
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 32
Quote:
Originally Posted by sillysally View Post
So is the proper way to refer to what I posted above, dispcalGUI.
Yes, that's what I meant (my remark wasn't aimed at you specifically, more as general comment. In your above post it was apparent that you meant dispcalGUI).
fhoech is online now  
post #2528 of 2534 Old 08-28-2014, 12:30 PM
Member
 
mystigiandoll's Avatar
 
Join Date: Mar 2013
Posts: 46
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 11
In the guide on the first post is it OK to use the 64 bit version of ArgyllCMS and then use the 64 bit version of MPC-HC or should I scrap what I have done so far and redo it all with the 32 bit version of ArgyllCMS and use the 32 bit version of MPC-HC? Sorry if it is a silly question, am new to this stuff but just wasn't sure if the first post has been updated since possible 64 bit versions may be supported through that chain.

Also I followed everything on that first page guide (with the exception I did use 64 bit ArgyllCMS) and everything seemed to go through. Only part I wasn't sure what to do was after the 'calibrate and profile' completed I wasn't sure what to hit, I think the options were 'Dont install, install and apply, cancel' ....something along those lines. I tried the Install and apply and gave me an error about applying to a virtual device so I just hit Dont Install.

Anyhow still reading up and learning, hopefully will be able to check out a report of the calibration tonight and see how it went. Great tools!
mystigiandoll is online now  
post #2529 of 2534 Old 08-28-2014, 01:59 PM
Advanced Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 567
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 228 Post(s)
Liked: 81
Quote:
Originally Posted by mystigiandoll View Post
In the guide on the first post is it OK to use the 64 bit version of ArgyllCMS and then use the 64 bit version of MPC-HC or should I scrap what I have done so far and redo it all with the 32 bit version of ArgyllCMS and use the 32 bit version of MPC-HC? Sorry if it is a silly question, am new to this stuff but just wasn't sure if the first post has been updated since possible 64 bit versions may be supported through that chain.
MPC HC has nothing to do with creating the 3d lut. and there is no 64 bit madVR version. 64 bit argyllcms works totally fine because there is a 64 bit madTPG. and the created 3d lut doesn't care about 64/32 bit it's not a program it's a LUT a 3D LUT. i created mine with a 64 bit version of argyllCMS.

Quote:
Also I followed everything on that first page guide (with the exception I did use 64 bit ArgyllCMS) and everything seemed to go through. Only part I wasn't sure what to do was after the 'calibrate and profile' completed I wasn't sure what to hit, I think the options were 'Dont install, install and apply, cancel' ....something along those lines. I tried the Install and apply and gave me an error about applying to a virtual device so I just hit Dont Install.
this guide is about creating 3D lut not for general use. so everything is working as intended. you just have to create the 3 D LUT from this.
mightyhuhn is offline  
post #2530 of 2534 Old 08-28-2014, 02:14 PM
Member
 
mystigiandoll's Avatar
 
Join Date: Mar 2013
Posts: 46
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 11
Yap sounds good. Just stumbled on some older posts when 64/32 bit had some compatibility issues so just double checking to make sure I don't shoot myself in the foot right away. As for the media player, I know it doesn't have anything to do with this set up, just another old thread that 64 bit player didn't work with madvr but I can deal with media players later, just good to know everything so far is right! Thanks!
mystigiandoll is online now  
post #2531 of 2534 Old 08-29-2014, 02:25 AM
Member
 
James Freeman's Avatar
 
Join Date: Sep 2013
Posts: 158
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 25
@ Florian, Graeme

Thank you VERY much for adding Black Output Offset to the 3DLUT creation in the latest Beta.
Now I can compress the blacks to my liking and watching environment.

Using Offset of 0% (aka BT.1886) gives me too elevated first few steps.
Using Offset of 100% completely crushed (turned to black) the first 3-4 steps.

I have created several 3D LUTs of 95%, 90%, 85%, 80% Offset and found the perfect black compression for my display.
Now I can see the first step without having the lows too bright.
I used the same % value to create a proper calibration ICM for my OS as well.


I also find that a Relative curve gives a proper gamma above 50% (probably even lower), but Absolute will not even reach 2.2 and remained to too bright/low.

A quick question:
In what circumstance should I use Absolute if it will not even reach proper gamma value at any region?
Maybe it's more linear and not over compressing ("bending") the dark region?
* I should point that the display is a Dell U2410 with 850:1 CR and 0.1 cd/m2 MLL.


Panasonic TC-P50ST60.
HTPC (MPC-HC, MadVR), PS3.
Studio Monitors + Polk Audio PSW110 Subwoofer.

Last edited by James Freeman; 08-29-2014 at 03:33 AM.
James Freeman is offline  
post #2532 of 2534 Old 08-29-2014, 04:43 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 596
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 75
Quote:
Originally Posted by James Freeman View Post
@ Florian, Graeme

Thank you VERY much for adding Black Output Offset to the 3DLUT creation in the latest Beta.
Now I can compress the blacks to my liking and watching environment.
Thanks for the feedback. Good to hear that the addition knob is useful.
Quote:
I used the same % value to create a proper calibration ICM for my OS as well.
I think there is an inconsistency in the matching control in dispcal (it is reversed). I'm not sure quite how to make that consistent with collink, without confusing people.
Quote:
I also find that a Relative curve gives a proper gamma above 50% (probably even lower), but Absolute will not even reach 2.2 and remained to too bright/low.
I'm not quite sure what you mean by relative/absolute in this context. Do you mean effective/technical gamma, or something else ?
gwgill is offline  
post #2533 of 2534 Old 08-29-2014, 04:54 PM
Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 179
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 32
Quote:
Originally Posted by gwgill View Post
I'm not quite sure what you mean by relative/absolute in this context. Do you mean effective/technical gamma, or something else ?
I guess (in dispcalGUI I went with "relative" = effective, "absolute" = technical). So "absolute" (technical) gamma will have a lower effective gamma due to being influenced by the black level more, but I think this is by design?
fhoech is online now  
post #2534 of 2534 Old 08-29-2014, 11:03 PM
Member
 
James Freeman's Avatar
 
Join Date: Sep 2013
Posts: 158
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 25
Quote:
Originally Posted by gwgill View Post
Thanks for the feedback. Good to hear that the addition knob is useful.
Its very useful.
Now I can create new 3DLUT in 5 minutes to test Black Out Offset, instead of waiting 1.5 Hours for a new calibration.

Quote:
Originally Posted by gwgill View Post
I think there is an inconsistency in the matching control in dispcal (it is reversed). I'm not sure quite how to make that consistent with collink, without confusing people.
Are you sure they are reversed from one another? For me it behaves as expected.
It looks exactly the same when using the same Black Output Offset (-f [degree]) percentage with 3DLUT and ICM profile (collink, dispcal).
I have tested it with black steps pattern and switched between 3DLUT/Off (OS On), using madVR.

Quote:
Originally Posted by gwgill View Post
I'm not quite sure what you mean by relative/absolute in this context. Do you mean effective/technical gamma, or something else ?
Absolute:
-G option is an alternative that allows the actual power to be specified instead, meaning that when combined with the displays non-zero black value, the response at 50% input will probably NOT match that of the ideal power curve with that gamma value.

Relative:
-g option is used to set and actual response curve that makes an allowance for the non-zero black of the actual display, and will have the same relative output at 50% input as the ideal gamma power curve, and so best matches typical expectations.

Only -g (Relative) gives me proper gamma with this test image (gamma 2.2), whether -G will not.




The question was: Why use -G if it can't even match the gamma I calibrate to (expected gamma)?
Don't we want the image to look like the mastering studio intended at least above the first black region?
If I can reach 2.2 at 10% and up with -g, why not? Whats the point of -G?
Why BT.1886 uses -G -f0 if most LCD displays will not ever reach 2.4 with this terms? Don't we want at least some region to be 2.4?

Thanks


Panasonic TC-P50ST60.
HTPC (MPC-HC, MadVR), PS3.
Studio Monitors + Polk Audio PSW110 Subwoofer.
James Freeman is offline  
post #2535 of 2534 Old 08-31-2014, 05:45 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 596
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 75
Quote:
Originally Posted by James Freeman View Post
Are you sure they are reversed from one another? For me it behaves as expected.
I think the sense is the same, but defaults are reversed.
Quote:
Absolute:
-G option is an alternative that allows the actual power to be specified instead, meaning that when combined with the displays non-zero black value, the response at 50% input will probably NOT match that of the ideal power curve with that gamma value.

Relative:
-g option is used to set and actual response curve that makes an allowance for the non-zero black of the actual display, and will have the same relative output at 50% input as the ideal gamma power curve, and so best matches typical expectations.
Right, so you mean technical gamma vs. effective gamma.

They both should be able to achieve exactly the same resulting curve. Internally, effective gamma is converted to a technical gamma number.
Quote:
The question was: Why use -G if it can't even match the gamma I calibrate to (expected gamma)?
Some people insist on using a fixed, technical power value and BT.1886 demands it. I find the effective gamma much more useable, because it is more closely related to the visual result in the face of a black offset.
Quote:
Don't we want the image to look like the mastering studio intended at least above the first black region?
One of the practical problems is that there is no certainty about how the mastering studio was setup, nor how the director and colorist wanted a film to look. The more the mastering setup is standardized, the better this situation will get. An issued that consumers face is that their gear is not necessarily as good as a mastering studio, nor is their viewing environment as controlled, so it's useful to be able to compensate for these differences.
Quote:
If I can reach 2.2 at 10% and up with -g, why not? Whats the point of -G?
Why BT.1886 uses -G -f0 if most LCD displays will not ever reach 2.4 with this terms? Don't we want at least some region to be 2.4?
I'm afraid I think log/log values can be quite useful in measuring what's going on, but assigning visual significance at any other point than 50% is fraught. It's not a visually linear scale.
gwgill is offline  
post #2536 of 2534 Old Today, 01:20 PM
Member
 
Francois76l's Avatar
 
Join Date: May 2014
Posts: 58
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hello everybody,
Maybe some of you have the same problem but when I enable the 3dlut in madvr (tv, projector of pc screen) I always observe this phenomena on the upper and lower black bands:


The blacks of the image where the 3dlut is working are not as dark as the rest.
Do you know how I can correct this?
I targeted a BT1886 without black point compensation. I used the settings posted on the first page
Thanks for your help!

Last edited by Francois76l; Today at 02:27 PM.
Francois76l is online now  
Reply Display Calibration

User Tag List

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