or Connect
AVS › AVS Forum › Display Devices › Display Calibration › HCFR - Open source projector and display calibration software
New Posts  All Forums:Forum Nav:

HCFR - Open source projector and display calibration software - Page 87

post #2581 of 3436
So Zoyd, now I can choose refresh and it will automatically set my intergration to 1 second and I can chose my ccss. or is it still the same as 3.1.0.0?

Also i remember you saying over at HDJ that you owned a GT30 for a brief period,so we know there is something funky going on gamma wise in custom mid.

I would like you to take a look at this calibration I did in Custom Mid.As you can see the Gamma Y matches perfectly with itu-1886 but the gamma graph chart says a whole different story.Just curious as to which to believe the Gamma Y reading or the gamma plot graph....Thanks....Little Buddy.smile.gif

Custom 2.2 4% apl hdmi off (2).zip 8k .zip file
post #2582 of 3436

I think i came across another issue. When trying to create a meter correction file from a reference measure the programm just quits.

It's like this:

 

1. I have a sheet opened with primaries and secondaries measured and the "reference measure" checkbox checked

2. i open a new sheet with my other probe and click Advanced -> meter correction file -> create using existing reference measures

3. back to desktop

 

if i measure the primaries in that new sheet everything works as expected.

post #2583 of 3436
Why doesn't HCFR 3.1.0.0 show 0-10% WB RGB IRE/gray scale? It is not important? Whenever I do the black level pattern test, I can tell that my blacks do not align properly, some bars are purplish, some greenish, some reddish... slightly, yet I can tell. I remember the original HCFR 3.0.4.0 did show 0-10% IRE/gray scale but each reading would provide different results. AFAIK you need a spectrometer to properly measure blacks. I do know that at some point I had my blacks almost perfect to my eye, but now any attempt to fix them results in screwed up RGB WB 10-90% IRE.

I use my HDTV as a PC monitor in PC 4:4:4/RGB Mode and I use 0-255 full range! Maybe 0-10% IRE affects only full 0-255 range, but not 16-235???
post #2584 of 3436
Quote:
Originally Posted by MonarchX View Post

Why doesn't HCFR 3.1.0.0 show 0-10% WB RGB IRE/gray scale? It is not important? Whenever I do the black level pattern test, I can tell that my blacks do not align properly, some bars are purplish, some greenish, some reddish... slightly, yet I can tell. I remember the original HCFR 3.0.4.0 did show 0-10% IRE/gray scale but each reading would provide different results. AFAIK you need a spectrometer to properly measure blacks. I do know that at some point I had my blacks almost perfect to my eye, but now any attempt to fix them results in screwed up RGB WB 10-90% IRE.
It does show 10%. HCFR probably doesn't display errors at 0% video levels because most meters can't read color accurately that low anyway, spectros included (spectros generally have worse low-light ability). You can do a 20pt run and see how 5% video levels are measuring by going to Measures -> Parameters -> and setting the number of grayscale levels to 20; then use the 5% step grayscale patterns.

But you're still obviously limited by the controls on your display and how the display performs. Remember that the software only sees the points that you're measuring... it doesn't see the in-between points. If you have a 10pt white balance, doing a 2pt calibration first might help with banding. Other than that, there's not much else you can do about that. Some displays with 10pt systems are just a pain in the ass to get smooth gradients, while others with even just a 2pt system might have a smoother looking grayscale, albeit probably with higher errors overall.
Quote:
I use my HDTV as a PC monitor in PC 4:4:4/RGB Mode and I use 0-255 full range! Maybe 0-10% IRE affects only full 0-255 range, but not 16-235???
If your display is expecting the RGB range that your PC is outputting then it shouldn't matter if you use full range or limited range. Although I recommend full range whenever possible when it comes to PCs, because outputting limited can sometimes cause banding.
Edited by rahzel - 11/10/13 at 1:12pm
post #2585 of 3436
Thread Starter 
Quote:
Originally Posted by PeterLewis View Post

So Zoyd, now I can choose refresh and it will automatically set my intergration to 1 second and I can chose my ccss. or is it still the same as 3.1.0.0?
The behavior of the integration time setting was described several posts ago, it has not changed.
Quote:
Also i remember you saying over at HDJ that you owned a GT30 for a brief period,so we know there is something funky going on gamma wise in custom mid.

I would like you to take a look at this calibration I did in Custom Mid.As you can see the Gamma Y matches perfectly with itu-1886 but the gamma graph chart says a whole different story.

That's because Gamma Y is your target, the Y row are your measurements and you can see that the measurements are significantly higher than the targets at the high end, hence the drop in gamma. This is typical of the GT30 in mid-panel brightness.
post #2586 of 3436
Thread Starter 
Quote:
Originally Posted by unknown577 View Post

I think i came across another issue. When trying to create a meter correction file from a reference measure the programm just quits.
It's like this:

1. I have a sheet opened with primaries and secondaries measured and the "reference measure" checkbox checked
2. i open a new sheet with my other probe and click Advanced -> meter correction file -> create using existing reference measures
3. back to desktop

if i measure the primaries in that new sheet everything works as expected.

Thanks for finding that, I knew something was up but couldn't reproduce it. I've made test build that fixes that and added an adaptive averaging mode that might be useful for older probes. When engaged it will average a number of samples that varies with the light level. Of course it will slow down your probe but give it a try and see if it's worth it.

test it
post #2587 of 3436

Yes, it doesn't crash anymore. But now i cannot load chc calibration files anymore that i created with 3.1.0.0, only one that i created with an older version. Weirdly there is not always an error message, for some files nothing happens at all except for a short flickering of the cursor.

post #2588 of 3436
This is probably a stupid question but I'm new to measuring. The new version of ColorHCFR when measuring Gray, Primary, and Secondard asks for a 100% Gray and 100% White now but my AVS 709 patterns only have 100% White. So I basically end up measuring the same pattern twice or should I use 99% Grey for 100 Gray?
post #2589 of 3436
Thread Starter 
100% gray = 100% White
post #2590 of 3436
100% grey = 100% white

Edit : grilled by zoyd biggrin.gif
post #2591 of 3436
Thread Starter 
Quote:
Originally Posted by unknown577 View Post

Yes, it doesn't crash anymore. But now i cannot load chc calibration files anymore that i created with 3.1.0.0, only one that i created with an older version. Weirdly there is not always an error message, for some files nothing happens at all except for a short flickering of the cursor.

I might have to sort out some save file compatibility, test version 2 is in the testbuilds folder.
post #2592 of 3436
Thread Starter 
Quote:
Originally Posted by rahzel View Post

Didn't more recent versions of HCFR have an option to not include luminance in error reporting for grayscale? I thought I had mine setup that way. By 'old versions of HCFR', I simply mean <3.1.0.0 (not including the betas). I'm just kind of surprised my errors are so low on my ST50 with just a 2pt system. I know that CalMAN also defaults to dE2000 with or without luminance. If you don't mind and have time, can you compare these two like you have before? The differences in the errors reported by the two softwares that is.

All of the color difference calculations have been verified against the Bruce Lindbloom calculator. dE2000 just has a compressed scale compared to what you are used to seeing for grayscale errors. Apples and oranges.
post #2593 of 3436
Thanks.

FWIW, I wasn't doubting the numbers, was just a bit surprised. I'll trust your word. smile.gif
Edited by rahzel - 11/10/13 at 8:48pm
post #2594 of 3436
Quote:
Originally Posted by unknown577 View Post

Yes, it doesn't crash anymore. But now i cannot load chc calibration files anymore that i created with 3.1.0.0, only one that i created with an older version. Weirdly there is not always an error message, for some files nothing happens at all except for a short flickering of the cursor.

I noticed from v3.1.x.x you have to uninstall completely as it was going crashy crashy upon a new calibration project for the DVD Images selection.
post #2595 of 3436
Quote:
Originally Posted by zoyd View Post

100% gray = 100% White
Quote:
Originally Posted by realzven View Post

100% grey = 100% white

Edit : grilled by zoyd biggrin.gif

Thanks! smile.gif That's what I thought I just wanted to be sure.
post #2596 of 3436
Quote:
Originally Posted by rahzel View Post

Focus more on lower saturation levels. Actual content rarely has 100% saturation. I recommend using the 75% luminance / 75% saturation patterns. Also make sure you set the color standard in HCFR to rec.709 75%.
Answered in the HCFR thread.

Hmm, would it matter if 100% is outside of the Rec.709 triangle? I have a screwy blue for this.

jiq1.jpg
post #2597 of 3436
@Zoyd

Would it much of a hassle to include a 50% Rec 709 color space to the available options? I have a bit of trouble getting colors right on my european VT30 so I thought of trying calibrating at lower saturations. The GCD disk has patterns down to 25% saturation actually so if you are at it maybe you should add 25% Rec 709 as well to cover all bases.
post #2598 of 3436
Thread Starter 
Quote:
Originally Posted by xvfx View Post

Hmm, would it matter if 100% is outside of the Rec.709 triangle? I have a screwy blue for this.

not if you can hit the 75% saturated target, you'll just have to see if it's possible.
post #2599 of 3436
Thread Starter 
Quote:
Originally Posted by Barsk View Post

@Zoyd

Would it much of a hassle to include a 50% Rec 709 color space to the available options? I have a bit of trouble getting colors right on my european VT30 so I thought of trying calibrating at lower saturations. The GCD disk has patterns down to 25% saturation actually so if you are at it maybe you should add 25% Rec 709 as well to cover all bases.

yes, those extra color space modes are a pain. Try the CC6 mode, that includes 4 lower saturation points so it might work better for you than 75%.



edit:

Current version is now 3.1.0.2 - no new features, just a fix for calculating a matrix correction and a (very small) error in the dE94 formula. I left the adaptive averaging feature out but it's in the test build folder if you still want to test it.
Edited by zoyd - 11/11/13 at 5:13am
post #2600 of 3436
Quote:
Originally Posted by zoyd View Post

yes, those extra color space modes are a pain. Try the CC6 mode, that includes 4 lower saturation points so it might work better for you than 75%.
.

Ok. that was unfortunate.

What is the CC6 mode and how do I use it? It tried google it and found virtually nothing...
post #2601 of 3436
Thread Starter 
Won't find it on google, it's my own creation. smile.gif It's one of the reference color spaces, look for CC6(GCD) or CC6(MCD). It will set the primaries and secondaries references to six color check patterns found on both those disks. When you run a primaries/secondaries sweep it will ask for a color from that 24 pattern sequence and then you can calibrate to each of the colors just like regular primaries.
post #2602 of 3436
Thanks!

I'll give it a try!
post #2603 of 3436

If I'm only adjusting the gamma of, say the 70% point, what I need to measure is the 0% Y value and the 100% Y value, and the target Y for my 70% point should be calculated and show up on the screen for me to shoot for that value

1bHk

post #2604 of 3436
Thread Starter 
yes, you can do that although you need at least 2 points other than 0 and 100 for anything to show up on the gamma graph.
post #2605 of 3436
Thread Starter 
Quote:
Originally Posted by Barsk View Post


I also suggest that the combining of the refresh/no-refresh inside the ccss select list violates standard GUI "guidelines" as how GUI's should behave. Or at lest how people expects them to. A select list is for selecting ONE thing and one thing only. If you need to select two different things - make two select lists.

This is sorted out now (re-download 3.1.0.2 if you already have it). There is still only 1 selection box containing display type but if any of the selections require a refresh mode calibration (e.g. plasma or crt edrs) it will be requested automatically, you do not have to choose refresh mode.



edit: bonus tip - If you drop a .ccmx file created via ArgyllCMS or dispCalGui into your user\APPDATA\roaming\color folder (where the ccss files live), it will become a display type choice and apply that correction automatically. Other places that ArgyllCMS looks for both .ccss and .ccmx for autoloading are user\APPDATA\roaming\ArgyllCMS and C:\PROGRAMDATA\ArgyllCMS
Edited by zoyd - 11/11/13 at 9:31am
post #2606 of 3436
Quote:
Originally Posted by zoyd View Post

Current version is now 3.1.0.2 - no new features, just a fix for calculating a matrix correction and a (very small) error in the dE94 formula. I left the adaptive averaging feature out but it's in the test build folder if you still want to test it.

 

Looks good. I can load my previously saved files and it won't crash calculatin a matrix correction for an empty sheet.

What about this averaging feature? Will it make a big difference for dark readings?

post #2607 of 3436
Thread Starter 
Quote:
Originally Posted by unknown577 View Post

Looks good. I can load my previously saved files and it won't crash calculatin a matrix correction for an empty sheet.
What about this averaging feature? Will it make a big difference for dark readings?

Well I was hoping you'd tell me with some test results from the v2 test version. Run your dtp94 with and without the switch on for some low level patterns and see if it improves stability.
post #2608 of 3436
Zoyd baby,

I rechecked my THX calibration with 3.1.0.2 and it matches perfectly with 3.0.5.2

Also the 4% apl had me set my color to 48 +1 tint and soI decided to check my calibration with the GCD 10% patterns and lo and behold the greyscale actually measured better but the gamma was lower and it also made me lower color to 45 +1 tint.

So Zoyd looking at both calibration which color calibration to believe...48 or 45?

THX 4% apl 76 contrast 3.zip 1k .zip file

THX 10% GCD.zip 2k .zip file
post #2609 of 3436

Oh alright, i will tell you the results tomorrow when i have time for some tests.

post #2610 of 3436
Thread Starter 
Quote:
Originally Posted by PeterLewis View Post

Zoyd baby,

The next time you call me that or "little buddy" you're on my blocklist. wink.gif
Quote:
I rechecked my THX calibration with 3.1.0.2 and it matches perfectly with 3.0.5.2

Also the 4% apl had me set my color to 48 +1 tint and soI decided to check my calibration with the GCD 10% patterns and lo and behold the greyscale actually measured better but the gamma was lower and it also made me lower color to 45 +1 tint.

So Zoyd looking at both calibration which color calibration to believe...48 or 45?

Exactly which patterns are these? By GCD 10% do you mean on black background or 22% background. By 4% APL do you mean Mascior's 4% APL?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › HCFR - Open source projector and display calibration software