MadVR - ArgyllCMS - Page 124 - AVS Forum | Home Theater Discussions And Reviews

Forum Jump: 
 450Likes
Reply
 
Thread Tools
post #3691 of 4404 Old 05-28-2015, 01:28 PM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,331
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 683 Post(s)
Liked: 225
Quote:
Originally Posted by Masharak View Post
I made a 3DLUT using 8bit depth setting in nVidia CP and 8bit depth setting in madVR, but later found out that my TV supports 12bit depth setting via nVidia CP. madVR was set to use 10bit+ setting when 12bit depth setting was activated in nVidia CP. Should another 3DLUT be made using 12bit depth setting if one wishes to view films using that 12bit depth setting? Or would previous 3DLUT, made with 8bit depth setting in nVidia CP and 8bit depth setting in madVR, would be just as good/accurate/smooth for playback using 12bit depth setting in nVidia CP and 10bit+ depth setting in madVR?

If another 3DLUT should be made using 12bit depth setting, then I assume madVR should be set to 10bit+ depth during profiling/calibration and dithering should be set to Ordered Dithering.., correct?
and you need to run madTPG in fullscreen exclusive to get 10 bit to the gpu driver.

and yes the old 3d LUT should be totally fine.
mightyhuhn is offline  
Sponsored Links
Advertisement
 
post #3692 of 4404 Old 05-28-2015, 03:11 PM
Member
 
Join Date: Jan 2015
Posts: 143
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 4
Ah! Thanks! So it is safe to assume that even when 12bit depth is selected in nVidia CP, only 255 grayscale levels can be calibrated? I thought with 8bit display, only 255 levels can be adjusted, but with 12bit, there would be levels in-between the 0-255. I guess I was wrong and 255 levels is all that can be changed even for 12bit displays (at least on Windows OS with current videocards).
Masharak is offline  
post #3693 of 4404 Old 05-28-2015, 03:25 PM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,331
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 683 Post(s)
Liked: 225
Quote:
Originally Posted by Masharak View Post
Ah! Thanks! So it is safe to assume that even when 12bit depth is selected in nVidia CP, only 255 grayscale levels can be calibrated? I thought with 8bit display, only 255 levels can be adjusted, but with 12bit, there would be levels in-between the 0-255. I guess I was wrong and 255 levels is all that can be changed even for 12bit displays (at least on Windows OS with current videocards).
the results are dithered you can you use dither to show more than 256 steps that's the hole idea behind that.
nearly all displays support 12 bit input no display a normal human being can buy is 12 bit and 99.9 % of all display are not 10 bit either.
mightyhuhn is offline  
post #3694 of 4404 Old 05-28-2015, 03:34 PM
Senior Member
 
BlueChris's Avatar
 
Join Date: May 2014
Location: Athens, Greece
Posts: 354
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 191 Post(s)
Liked: 81
Im confused... bottom line is better in nvidia cp to say 12bit or not?
BlueChris is offline  
post #3695 of 4404 Old 05-28-2015, 03:56 PM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,331
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 683 Post(s)
Liked: 225
Quote:
Originally Posted by BlueChris View Post
Im confused... bottom line is better in nvidia cp to say 12bit or not?
there is no clear answer to this.

8 bit is for sure more save and it's known to work.
BlueChris likes this.
mightyhuhn is offline  
post #3696 of 4404 Old 05-29-2015, 05:32 AM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,293
Mentioned: 86 Post(s)
Tagged: 0 Thread(s)
Quoted: 1314 Post(s)
Liked: 1044
@fhoech

When using these settings for a synthetic profile:



I'm getting a profile with a 2.2 gamma tone curve.
Attached Files
File Type: zip Rec1886_24.zip (525 Bytes, 11 views)
zoyd is offline  
post #3697 of 4404 Old 05-29-2015, 10:02 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 823
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 404 Post(s)
Liked: 262
Quote:
Originally Posted by zoyd View Post
@fhoech

When using these settings for a synthetic profile:



I'm getting a profile with a 2.2 gamma tone curve.
Good catch, thanks. When creating a synthetic profile with Rec. 1886 and black level of zero, the fallback gamma value of 2.2 was used instead of 2.4. As a workaround you can switch to custom gamma, or update to 3.0.0.5 beta.
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 #3698 of 4404 Old 05-29-2015, 12:20 PM
Senior Member
 
DrFaxe's Avatar
 
Join Date: Nov 2012
Posts: 209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 84
Thumbs up

Hello,

so I thought is was time to test the latest Argyll and DispCalGui versions. As always I'm really impressed by the results, but more on the new DispCalGui. It was really the first time I used it completely (hardcore command line user before ). The new workflow is really nothing but great. Everything is clear from the first step to the last and runs flawless.

I'm really happy You integrated my suggestion of the meter profile overview at the end of a colorimeter correction creation. This is so helpful as You see immediately if You have a good profile or not. Doing this by command line was always a little pita.

Congratulations und kudos to You and Your great work Florian and Graeme.
Donation will be sent to both of You.

Attached the meassurement report.
Attached Files
File Type: pdf Measurement Report.pdf (437.5 KB, 45 views)
DrFaxe is offline  
post #3699 of 4404 Old 05-29-2015, 02:11 PM
AVS Special Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 2,630
Mentioned: 82 Post(s)
Tagged: 0 Thread(s)
Quoted: 1497 Post(s)
Liked: 507
there's a slightly curious shape to my gamma curve after the latest run, the report can be found in https://www.dropbox.com/s/qey7nd511n...8-48.html?dl=0 and the pic from the report is attached

This is a jvc x3 with a fairly old bulb (~1200hrs) measured with an i1d3 (corrected by an i1pro) in a v dark but not batcave room. The PQ at the end looks outstanding to me tbh but the droop at the high and low end seems odd so curious if this is something to be corrected or is perhaps a measurement artefact or something else entirely? The shape at the low end looks like black output offset 0% rather than the 100% requested doesn't it?
Attached Thumbnails
Click image for larger version

Name:	gamma.PNG
Views:	97
Size:	98.3 KB
ID:	743889  

Last edited by 3ll3d00d; 05-29-2015 at 03:38 PM.
3ll3d00d is offline  
post #3700 of 4404 Old 05-30-2015, 09:10 AM
Member
 
Join Date: Jan 2015
Posts: 143
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 4
Quote:
Originally Posted by mightyhuhn View Post
the results are dithered you can you use dither to show more than 256 steps that's the hole idea behind that.
nearly all displays support 12 bit input no display a normal human being can buy is 12 bit and 99.9 % of all display are not 10 bit either.
Regarding 8bit vs 12bit:

I have tested 8bit and 12bit settings in nVidia CP. I created a high quality 1D LUT with i1Display Pro and ArgyllCMS 1.7.0. I loaded up a grayscale gradient and I could see some banding when I used that 1D LUT and 8bit setting in nVidia CP. When I selected 12bit, that banding was completely eliminated without dithering. This is not a case of 8bit + FRC dithering, but true 12bit depth, which I am certain my TV supports. It is also obvious when I load film content in WMP (without madVR rendering!) and compare the way it looks in 8bit mode and 12bit mode. 8bit mode shows obvious banding in darker/black areas, but 12bit does not. I use a Standard Speed HDMI cable, which allows 12bit depth to work only @ 23Hz (48Hz mode) , but @ 60Hz 12bit depth creates severe artifacts, most likely due to lack of bandwidth. 8bit mode works fine @ 23Hz (48Hz mode) and @ 60Hz. I think if I get a High-Speed 10.2GBPS HDMI cable, I can get 60Hz mode to work with 12bit depth.

Assuming that my TV does support true 12bit without dithering, would a 3DLUT made with 8bit depth setting work just as well with 12bit depth setting OR should that 3DLUT be re-created using 12bit depth setting? Ultimately, the question is this - does ArgyllCMS measure more grayscale levels for 3DLUT's if 12bit depth setting is selected (in driver settings / nVidia CP, not madVR) or does it still use only 256 levels for measurements?

Last edited by Masharak; 05-30-2015 at 09:13 AM.
Masharak is offline  
post #3701 of 4404 Old 05-30-2015, 10:24 AM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,331
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 683 Post(s)
Liked: 225
try this: http://forum.doom9.org/showpost.php?...ostcount=30280

with 8 and 12 bit.

what mastering monitor is your true 12 bit device?

and try your 1d lut with madVR overlay mode (works with nvidia and intel). the 1d lut is know to produce banding on nvidia.
mightyhuhn is offline  
post #3702 of 4404 Old 05-30-2015, 10:52 AM
Senior Member
 
BlueChris's Avatar
 
Join Date: May 2014
Location: Athens, Greece
Posts: 354
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 191 Post(s)
Liked: 81
Sorry for asking but what are the optimum settings in madtpg as matter window size and background for an IPS LG edgelit TV?
BlueChris is offline  
post #3703 of 4404 Old 05-30-2015, 10:54 AM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,331
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 683 Post(s)
Liked: 225
Quote:
Originally Posted by BlueChris View Post
Sorry for asking but what are the optimum settings in madtpg as matter window size and background for an IPS LG edgelit TV?
it shouldn't matter for LCD in general.
BlueChris likes this.
mightyhuhn is offline  
post #3704 of 4404 Old 05-30-2015, 04:10 PM
AVS Forum Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 4,740
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 805 Post(s)
Liked: 828
When I make a 3DLUT I get see below for screen shot.

If I use Relative colorimetric, I don't get popup for 2.2 or 2.4. If I use BT.1886 I do get the below popup.

Absolute colorimetric with white point scaling, I get the below popup no matter what gamma I use.

Also the black offset bar is missing, for all of the above.

My setup is Rec.709 and as measured. slider black offset bar is 100%.

Any Ideas why this is?

ss
Attached Thumbnails
Click image for larger version

Name:	Overwrite.jpg
Views:	107
Size:	160.4 KB
ID:	746689  

"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
sillysally is offline  
post #3705 of 4404 Old 05-30-2015, 04:35 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 823
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 404 Post(s)
Liked: 262
If you've already created a 3D LUT with the chosen settings, then it'll ask if you want to re-create it (normally there is no need).

Quote:
Originally Posted by sillysally View Post
Also the black offset bar is missing, for all of the above.
When the profile black point is zero, there's no need to show the output offset slider (and gamma relative/absolute choice).
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 #3706 of 4404 Old 06-02-2015, 04:50 PM
Senior Member
 
BlueChris's Avatar
 
Join Date: May 2014
Location: Athens, Greece
Posts: 354
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 191 Post(s)
Liked: 81
Im trying to understand something, in nowhere in dispcalgui i have choosed bt1886 but the 3dlut that i got if i use hcfr with madtpg unchecked the 2 disables, then the grayscale result that i got is exactly like bt1886 and not like 2.2 gamma. The scale at 20ire and below is getting luminance.
How can i eliminate this or at least to control it
BlueChris is offline  
post #3707 of 4404 Old 06-02-2015, 04:52 PM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 823
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 404 Post(s)
Liked: 262
Quote:
Originally Posted by BlueChris View Post
Im trying to understand something, in nowhere in dispcalgui i have choosed bt1886 but the 3dlut that i got if i use hcfr with madtpg unchecked the 2 disables, then the grayscale result that i got is exactly like bt1886 and not like 2.2 gamma. The scale at 20ire and below is getting luminance.
How can i eliminate this or at least to control it
How do your 3D LUT tone curve settings look like?

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 #3708 of 4404 Old 06-02-2015, 05:15 PM
Member
 
Join Date: Jan 2015
Posts: 143
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 4
Quote:
Originally Posted by mightyhuhn View Post
try this: http://forum.doom9.org/showpost.php?...ostcount=30280

with 8 and 12 bit.

what mastering monitor is your true 12 bit device?

and try your 1d lut with madVR overlay mode (works with nvidia and intel). the 1d lut is know to produce banding on nvidia.

OK, I tried that command, but the last line said that "Unable to determine Video LUT entry bit depth" when I used 12bit mode. When I used 8bit mode, it told me "Video LUT depth seems to be 8bit". Obviously, ArgyllCMS is limited to 10bits.

Banding is not due to nVidia or AMD or anything like that. Its due to slight measurement inaccuracies when you measure with a non-perfect-accuracy device like i1Display Pro. It may be that less banding was seen with AMD on 10bit devices since AMD supported 10bit before nVidia could support 12bit+.

I also learned that while 8bit uses only 0-255 levels, 12bit uses 0-4000+ levels! This makes me wonder whether ArgyllCMS would measure all of those 4000+ levels when 12bit depth mode is enabled...

I also purchased a High-Speed HDMI and it allowed me to use 12bit even @ 60Hz! And no High-Speed HDMI is not a gimmick, its not the same as HDMI versions, which are determined by the devices themselves and not the cable. High-Speed HDMI cables have much higher bandwidth compared to Standard HDMI cables, which were actually made for 720p and 1080i, but they surely handled more than that, yet were limited by bandwidth.

BD content was supposed to use 12bit depth, but it didn't...

Last edited by Masharak; 06-02-2015 at 05:25 PM.
Masharak is offline  
post #3709 of 4404 Old 06-02-2015, 05:41 PM
AVS Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 1,331
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 683 Post(s)
Liked: 225
Quote:
Originally Posted by Masharak View Post
Banding is not due to nVidia or AMD or anything like that. Its due to slight measurement inaccuracies when you measure with a non-perfect-accuracy device like i1Display Pro. It may be that less banding was seen with AMD on 10bit devices since AMD supported 10bit before nVidia could support 12bit+.
nvidia can send 12 bit since dx 10 cards so card like gtx 260 or stuff like this. the option in the driver is cosmetic it is only useful to force 8 bit with high bit deep input that's it.

nvidia is adding a ton of banding when used with a simply ICM profile and output is at 8 bit it's just a flaw in the nvidia driver. use madVR overlay mode should look worlds better.
Quote:
I also learned that while 8bit uses only 0-255 levels, 12bit uses 0-4000+ levels! This makes me wonder whether ArgyllCMS would measure all of those 4000+ levels when 12bit depth mode is enabled...
how? your colorimeter can't see the difference and how to send 12 bit to the driver in the first place it is not even possible. you could technically send 16 but...
that's for a pixel you can use dither to show more bit deep over multiply pixel that the idea of dithering in the first place.
12 bit is 2~12-1=4095 so 0-4095 but a display that can display this doesn't exist.

and feel free to calculate how many measuring are needed to make even use of 10 bit. 8 bit has already over 16 million possibilities.
the test chart is using % from maximum color so they are not based on a bit deep.

the test i linked you test how deep the colorimeter think's your "display" is
Quote:
I also purchased a High-Speed HDMI and it allowed me to use 12bit even @ 60Hz! And no High-Speed HDMI is not a gimmick, its not the same as HDMI versions, which are determined by the devices themselves and not the cable. High-Speed HDMI cables have much higher bandwidth compared to Standard HDMI cables, which were actually made for 720p and 1080i, but they surely handled more than that, yet were limited by bandwidth.
standard speed HDMI cable still exist? RGB 1080p60 hz with 12 bit is nothing special.
Quote:
BD content was supposed to use 12bit depth, but it didn't...
never heard this before. and a 12 bit panel is already very unrealistic in the first place when the BD spec was planned.
mightyhuhn is offline  
post #3710 of 4404 Old 06-02-2015, 06:08 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 897
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 228 Post(s)
Liked: 175
Quote:
Originally Posted by Masharak View Post
I also learned that while 8bit uses only 0-255 levels, 12bit uses 0-4000+ levels! This makes me wonder whether ArgyllCMS would measure all of those 4000+ levels when 12bit depth mode is enabled...
Taking 1 second per reading, measuring patches with all combinations of 12 bits would take about 2200 Years.
Have you got that long ?
gwgill is offline  
post #3711 of 4404 Old 06-03-2015, 04:29 AM
Senior Member
 
BlueChris's Avatar
 
Join Date: May 2014
Location: Athens, Greece
Posts: 354
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 191 Post(s)
Liked: 81
Quote:
Originally Posted by fhoech View Post
How do your 3D LUT tone curve settings look like?
I did something yesterday that indeed fixed the things for the better and here are the results. I understand that the whole problem is because my tv has bad black because its edgelit so if i leave the setting in dispcalguy at default its 0.000. I measured my total black with HCFR and it is 0.020 and i had puted there and the result is total different for the better..

Here is my response curve with my latest 3dlut after i did the above edit.


Another thing i need a clarify... i see my CIE Diagram a bit better which is fine but my Gamma and Luminance now have deviations in my eyes... see the below screens...its triplet images and the 1st is the tv total uncalibrated results, the 2nd is the calibrated result that i performed and the 3nd is the Results after i told to HCFR to use the 3dlut...
Is this logical? i mean i am not expert in color but this is the result of argyll trying to fix errors in colors by altering a bit the grayscale balance? if that is the case im ok with it






And here are my DispCalGuy settings...






Last edited by BlueChris; 06-03-2015 at 09:25 AM.
BlueChris is offline  
post #3712 of 4404 Old 06-03-2015, 01:54 PM
Member
 
Join Date: Jan 2015
Posts: 143
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 132 Post(s)
Liked: 4
Quote:
Originally Posted by mightyhuhn View Post
nvidia can send 12 bit since dx 10 cards so card like gtx 260 or stuff like this. the option in the driver is cosmetic it is only useful to force 8 bit with high bit deep input that's it.

nvidia is adding a ton of banding when used with a simply ICM profile and output is at 8 bit it's just a flaw in the nvidia driver. use madVR overlay mode should look worlds better.
how? your colorimeter can't see the difference and how to send 12 bit to the driver in the first place it is not even possible. you could technically send 16 but...
that's for a pixel you can use dither to show more bit deep over multiply pixel that the idea of dithering in the first place.
12 bit is 2~12-1=4095 so 0-4095 but a display that can display this doesn't exist.

and feel free to calculate how many measuring are needed to make even use of 10 bit. 8 bit has already over 16 million possibilities.
the test chart is using % from maximum color so they are not based on a bit deep.

the test i linked you test how deep the colorimeter think's your "display" is
standard speed HDMI cable still exist? RGB 1080p60 hz with 12 bit is nothing special.
never heard this before. and a 12 bit panel is already very unrealistic in the first place when the BD spec was planned.
There is no flaw in nVidia driver that you speak of. I had an AMD card and there was just as much banding with 1D LUT's as there was with nVidia cards when 8bit modes were selected for both AMD and nVidia cards. 12bit mode eradicates those bands and it is obvious on my TV when I toggle between 8bit and 12bit. madVR has nothing to do with this since I am talking about normal desktop gradients and 1D LUT. There is no dithering and it is not a "fake 12bit". I simply did not know about 12bit and High-Speed HDMI cables earlier, which is why I was curious if I needed a re-calibration. I ran measurements and know that I don't need a re-calibration.

Thanks for help!
Masharak is offline  
post #3713 of 4404 Old 06-04-2015, 02:12 AM
Senior Member
 
BlueChris's Avatar
 
Join Date: May 2014
Location: Athens, Greece
Posts: 354
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 191 Post(s)
Liked: 81
Guys can anyone help in the above post that i did with the images? my red went a bit wild after the 3Dlut in gamma scale... and it huppent only in my latest 3dlut creation with the above settings in dispcalguy...

thx..
BlueChris is offline  
post #3714 of 4404 Old 06-04-2015, 04:35 AM
Advanced Member
 
-Hitman-'s Avatar
 
Join Date: Nov 2004
Location: Manchester, UK
Posts: 816
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 134 Post(s)
Liked: 106
Hi Florian,


I've created a profile and 3dlut, however if I want to change the 3dlut settings, say to create a lut for madvr with an existing profile, I don't see an option to just re-create a 3dlut and the only way currently I can seem to be able to do this (get the option to create a 3dlut) is to run through a full profile creation from measurement data, which seems an unneeded additional process just to recreate the lut.


All I want to do is select a pre-made/saved profile, adjust the lut settings and re-create a new 3dlut.


Could i ask as suggestion if it could be possible to have 2 usage options that will set up dispcal settings for different users (full cal for PC users and Profile only - 3dlut creation for 3dlut only users options), maybe in the form of large selection buttons on the main "display and instrument" page to aid users calibration preference eg. 1 button/option for users who want to create a "full" calibration of embedded curves in a 3dlut for PC's and the other option 2 for profiling only with standard 3dlut creation (non embedded curves) for users who just want a profile and 3dlut created - with a locked down the calibration page (grey'ed out unneeded options) but with the interactive calibration option to set the display to d65 etc.. before profiling as this would make things a lot easier to run through the process and remove options 3dlut users don't need as currently it's seems to insist a calibration of curves take place and it's a bit cumbersome to find "as measured" setting to bypass this and have to re-enable when wanting to set white point again?


Thanks for a great upgrade BTW!

Last edited by -Hitman-; 06-04-2015 at 10:32 AM.
-Hitman- is offline  
post #3715 of 4404 Old 06-04-2015, 11:13 AM
Advanced Member
 
fhoech's Avatar
 
Join Date: Apr 2013
Location: Stuttgart, Germany
Posts: 823
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 404 Post(s)
Liked: 262
Quote:
Originally Posted by -Hitman- View Post
All I want to do is select a pre-made/saved profile, adjust the lut settings and re-create a new 3dlut.
You can do this, just go to the 3D LUT tab and uncheck "Create 3D LUT after profiling". The button at the bottom will then change to "Create 3D LUT...".

Quote:
Originally Posted by -Hitman- View Post
[...]1 button/option for users who want to create a "full" calibration of embedded curves in a 3dlut for PC's and the other option 2 for profiling only with standard 3dlut creation (non embedded curves) for users who just want a profile and 3dlut created - with a locked down the calibration page (grey'ed out unneeded options) but with the interactive calibration option to set the display to d65 etc.. before profiling as this would make things a lot easier to run through the process and remove options 3dlut users don't need as currently it's seems to insist a calibration of curves take place and it's a bit cumbersome to find "as measured" setting to bypass this and have to re-enable when wanting to set white point again?
The presets already do this and are set up in the way you describe. All 3D LUT presets by default don't use 1D LUT calibration.
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 #3716 of 4404 Old 06-04-2015, 02:40 PM
Advanced Member
 
-Hitman-'s Avatar
 
Join Date: Nov 2004
Location: Manchester, UK
Posts: 816
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 134 Post(s)
Liked: 106
Thanks Florian, makes sense now although it did seem to want to do a full calibration until I'd found the "as measured" setting then it all fell into place, also you mentioning toggling "3dlut creation after profile" is also what I was missing, was a late 1.30am calibration run/play last night :/


I'll play some more to get used to the slight changes here and there but all good - many thanks!
-Hitman- is offline  
post #3717 of 4404 Old 06-04-2015, 02:54 PM
Advanced Member
 
-Hitman-'s Avatar
 
Join Date: Nov 2004
Location: Manchester, UK
Posts: 816
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 134 Post(s)
Liked: 106
Do you need a sensor connected to get the "cal/profile/create 3dlut" buttons to appear at the bottom as I've just started Dispcal, loaded my profile/set madvr as display and de-selected create lut after profile but no buttons to do anything are there?




Edit: Yes I need a sensor connected, please could you add an override option to work without a sensor?


Thanks.

Last edited by -Hitman-; 06-04-2015 at 03:05 PM.
-Hitman- is offline  
post #3718 of 4404 Old 06-05-2015, 12:46 AM
Member
 
Sithuk's Avatar
 
Join Date: Mar 2011
Posts: 42
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 11
Can MadVR with a LUT correct the colour from a projector where the red component of the light from the bulb has dropped off? How does a LUT solve red drop off?
Sithuk is offline  
post #3719 of 4404 Old 06-05-2015, 10:08 AM
Member
 
VerGreeneyes's Avatar
 
Join Date: May 2013
Posts: 120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 13
Assuming you have the ability to calibrate your projector, sure. It'll basically tone back the green and blue channels so that they match the maximum intensity of the red channel, and compensate for the resulting gamma mismatch from only using part of the range. It can't do that for free of course - total brightness will be reduced, and it won't be able to use all 256 (or 1024 with 10-bit output) light intensity steps for the blue and green channels, so slightly different colors might end up looking the same. MadVR's dithering should help with that to some degree, but dithering won't do any good if the output values for adjacent colors are the same. In return, however, at least the color channels will be properly balanced.
VerGreeneyes is offline  
post #3720 of 4404 Old 06-06-2015, 10:18 AM
AVS Special Member
 
Luis Gabriel Gerena's Avatar
 
Join Date: Aug 2000
Location: Orem, Utah
Posts: 2,968
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 83 Post(s)
Liked: 61
Hello all!
First time I am able to get my Colormunkey Display to work with this so I did the calibration yesterday for my BenQ W1070 but I got some questions to make sure I did the right settings for my requirments.

Questions about the steps:

#1 Set the following settings in DispcalGUI

Select 'madVR' under Settings


I do NOT see a madVR option on the Settings drop down menu. I do see it under Display drop down menu so that is where I selected it. Is this a problem or just a difference on how the current version works?

#2 I did not see a Enable Low quality B2A tables but looks like by selecting MadVR this is done automatically....right?


#3 When creating the 3D Lut, I did not see the Enable 'Apply calibration (vcgt)' option. Problem here?

#4 Not sure but does this calibration affect the normal black and white level settings, thus I need to recalibrate after this? I am going to check tonight but just curious since I am at work right now.


Ok so now, this is my setup to see if I need any special options and to re-run the calibration.

Nvidia GTX 750
Windows 8.1
Using MPC-HC with latest madvr.
BenQ W1070 DLP projector. I was running Gamma 2.6 on the projector settings in a pitch black basement with partial ceiling and walls covered in black velvet fabric. Floor has a black carpet.

I watch exclusively ripped Blurays with MPC (except for 3D ones which I watch with PowerDVD).

Yesterday it finished after 1AM so I was too sleepy to test much but a quick Avengers check seem that it was a more redish look to the faces etc. Of course could be that my current one is too blue so I need to test more tonight but I want to make sure I did not miss anything since it seems the original Guide is a bit outdated.

I will appreciate any feedback.
Thanks!

Mario Lemieux is GOD!
Go Pens!!!
Aguante Boca!!!
If dolphins are so smart...why do they live in igloos?

Last edited by Luis Gabriel Gerena; 06-06-2015 at 10:28 AM.
Luis Gabriel Gerena 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