HCFR - Open source projector and display calibration software - Page 181 - AVS | 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: 
 276Likes
Reply
 
Thread Tools
post #5401 of 5886 Old 02-10-2015, 03:28 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
@cfraser

Good suggestions, thanks. You can also copy the Y value out of the measurement grid and when you paste it in Excel for example you'll get 6 decimal precision.

Quote:
Originally Posted by flyt100 View Post
I've been using HCFR(now using 3.1.6) for a few months with my i1d3, but just switched to using the internal generator. It sure is faster than using a DVD! Everything seems to work OK, and I get similar results to using a DVD, however I have one issue I can't figure out. I am using the 75%/75% reference, and when I measure primary and secondary colors using the automated method (ex. F4), things look fine. When I want to adjust a single color and use the view->test colors, I have a problem. I click on the reference square in the CIE diagram (ex. for red saturated at 75%), and the test color seems to be almost fully saturated looking at the numbers, with the eye, and the measurement point. The automated measurements look and test fine, but the RGB numbers that pop up there, aren't the ones I get when clicking on the CIE and using the test color window. Any ideas?
Working with an automatic pattern generator is the way to go. Unfortunately the CIE chart color picker does not work properly when using the special color 75%/75% color space, you have to be in one of the normal color space settings.
zoyd is online now  
Sponsored Links
Advertisement
 
post #5402 of 5886 Old 02-10-2015, 04:04 AM
Member
 
flyt100's Avatar
 
Join Date: Apr 2003
Posts: 84
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 10
Quote:
Originally Posted by zoyd View Post
Working with an automatic pattern generator is the way to go. Unfortunately the CIE chart color picker does not work properly when using the special color 75%/75% color space, you have to be in one of the normal color space settings.
Thanks, I thought I was going crazy!

This may be more of a process question then. When using the DVD, to adjust the CMS for a color, I'd put the color on the display (ex. 75%/75%), then adjust the CMS while using continuous measurements. Is there a practical way to do something similar using the internal generator? The automatic generator is great to check them all, but instant feedback (continuous measurements) seems more useful while adjusting. Any guidance on the appropriate process would be appreciated...

Thanks for a great piece of software, it's been extremely useful!
flyt100 is offline  
post #5403 of 5886 Old 02-10-2015, 04:38 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
Yes, select the color from the primaries/secondaries view and it will display that color for real-time adjustments
zoyd is online now  
post #5404 of 5886 Old 02-10-2015, 05:58 AM
Member
 
flyt100's Avatar
 
Join Date: Apr 2003
Posts: 84
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 10
Quote:
Originally Posted by zoyd View Post
Yes, select the color from the primaries/secondaries view and it will display that color for real-time adjustments
Thanks for the help! I think this is where I must be looking past the solution over and over. I think I have been trying that, but must be doing something incorrectly. In the measures tab, I have primaries and secondaries selected under view, but selecting one of the color columns doesn't change the color in the test color window. I must not be looking in the right place?
flyt100 is offline  
post #5405 of 5886 Old 02-10-2015, 06:21 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
yes that's right, also Preferences->General->"Detect primary colors during user measures" and "Display test color during measures" need to be ticked (which is the default), and you need to run a primaries/secondaries sweep prior to automatic color detection.
zoyd is online now  
post #5406 of 5886 Old 02-10-2015, 06:36 AM
Member
 
flyt100's Avatar
 
Join Date: Apr 2003
Posts: 84
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 10
Quote:
Originally Posted by zoyd View Post
yes that's right, also Preferences->General->"Detect primary colors during user measures" and "Display test color during measures" need to be ticked (which is the default), and you need to run a primaries/secondaries sweep prior to automatic color detection.
"Display test color during measures" was the culprit. Thanks a million!!!
flyt100 is offline  
post #5407 of 5886 Old 02-10-2015, 01:29 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
Thanks zoyd for not laughing at my little suggestions. They've been in the back of my mind for a while and I felt ashamed/ungrateful/etc. to even mention them... I started using HCFR about 2 years ago.

Is there some kind of (written) guidance about using the built-in pattern generator? I have played with it, but I still don't use it. I would like to as it makes obvious sense and is convenient. I use a pretty basic laptop with some type of built-in Intel video shenanigans->HDMI. Here are my "concerns", that may have nothing to do with HCFR, IDK really. Mostly they have to do with the lack of settings on my laptop, I think, or really the lack of specifics. The Intel app and laptop doc are just too dumbed down re video, and I like to be "exact" i.e. know what format signals I'm sending etc.

How do I use my laptop display as well as the display under test (DUT)? My laptop doesn't do 1920x1080, so if I want to output in that format to the DUT, I can only use the DUT to also see the HCFR program. Do other people who use the internal generator just use the DUT for HCFR and not their laptop display? It's been a few days since I've tried this so I forget exactly, but there seemed to be some display "interference" from HCFR when doing this (a displayed program window or something IIRC). Is there some way to configure HCFR so this doesn't happen? (i.e just send the test signal to the "second" display/DUT when both laptop and DUT are configured) Do I need a laptop with certain special video capabilities? I guess the gist of this is: how am I supposed to configure HCFR to use the internal generator as intended/properly? Thanks.

Edit: it just came back to me what the main issue was: I need to use ControlCal to make the actual adjustments to my main display. CC doesn't "disappear" the way HCFR sort of does when using it to generate the test sigs. I tried using the DUT as an "extended" display, with HCFR dragged so it's displayed on that, and CC remains on the laptop's display. That didn't work well at all, besides that you keep having to do crazy mouse movements to get the pointer on the correct display depending on app (you do tend to go back and forth...).

And just try doing continuous-readings type of adjustments with this arrangement, moving HCFR from display to display, with your sensor in the center of the DUT too. I guess you could use two computers (a separate one for ControlCal) - not gonna happen. If HCFR could "force" the test signals only to the "second"/extended display/DUT would be a better way. (Maybe that can already be done and I'm not using it right.) Besides using the BDP for the test signal, which I eventually went back to using after giving up on the neater way. Will try again tonight.

Last edited by cfraser; 02-10-2015 at 03:42 PM.
cfraser is offline  
post #5408 of 5886 Old 02-10-2015, 03:47 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
Quote:
Originally Posted by cfraser View Post
Thanks zoyd for not laughing at my little suggestions. They've been in the back of my mind for a while and I felt ashamed/ungrateful/etc. to even mention them... I started using HCFR about 2 years ago.
no problem, I like easy fix-its like that.

Quote:
Is there some kind of (written) guidance about using the built-in pattern generator? I have played with it, but I still don't use it. I would like to as it makes obvious sense and is convenient. I use a pretty basic laptop with some type of built-in Intel video shenanigans->HDMI. Here are my "concerns", that may have nothing to do with HCFR, IDK really. Mostly they have to do with the lack of settings on my laptop, I think, or really the lack of specifics. The Intel app and laptop doc are just too dumbed down re video, and I like to be "exact" i.e. know what format signals I'm sending etc.
HCFR has no control over your graphics card so there really isn't anything associated with the program to write about other than to match the output levels (16-235 or 0-255) to what your display is expecting. Most people start out by comparing DVD output to computer output to ensure that this level selection is correct and that grayscale and basic color calibration number are the same. You'll have to seek help elsewhere if there are discrepancies. There is a switch in the program to move some of the dialogs off-center if you have to run HCFR on the display you are calibrating but if you have an HDMI port you should be able to use that. Resolution doesn't matter if you are sending RGB over the HDMI line which is recommended.
zoyd is online now  
post #5409 of 5886 Old 02-10-2015, 07:08 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
When you set the laptop for extended display, HCFR will have two displays available in the dropdown under the generator configuration->display image parameters. Choose the 2nd display and all patterns will go there while the program stays on display 1.
zoyd is online now  
post #5410 of 5886 Old 02-10-2015, 07:25 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
^ Thanks. As you can see, I must have deleted my explanatory post just before you posted *exactly* what I had just discovered! The thing is, even though Windows calls my displays by different names (Mobile PC Display and Pioneer-TV in this case), the drop-down box calls them both the same name (Generic PnP Monitor). I didn't realize (or notice) that though the same display name was in there twice, that they actually represented the two different displays. D'oh.

Edit: now we're cooking. Working great. But for some reason I need to configure the HCFR RGB as 0-255, even though both the graphics card and display are configured for 16-235. I'll probably figure that out too. I am a bit confused as to how to leave the graphics card settings, as there are gamma/brightness/contrast/RGB range/etc. controls for that, and they definitely affect the HCFR HDMI output, as expected. <- default settings seem good, looks like I had changed those a long time ago (the odd one is I had set the graphics board gamma to 2.2, my display was previously calibrated for a gamma of 2.2, and the result of that is a measured display gamma of 1.0; that actually makes sense now that I put all the settings together)

Last edited by cfraser; 02-10-2015 at 09:16 PM. Reason: more testing revealed irrelevant info: removed
cfraser is offline  
post #5411 of 5886 Old 02-11-2015, 03:52 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
Quote:
Originally Posted by cfraser View Post
Edit: now we're cooking. Working great. But for some reason I need to configure the HCFR RGB as 0-255, even though both the graphics card and display are configured for 16-235. I'll probably figure that out too.
This often trips people up since it's non-intuitive but it will get you the right levels for everything sent over HDMI, including desktop applications. Set the graphics card to 0-255 if you are able for calibration, sometimes this needs a registry hack. 0-255 is the natural state of the graphics card and it means that levels sent to it will not be scaled, when set to 16-235 it is scaling to video levels which is why you are having to set HCFR to 0-255 to get it correct.
zoyd is online now  
post #5412 of 5886 Old 02-11-2015, 06:35 AM
Member
 
flyt100's Avatar
 
Join Date: Apr 2003
Posts: 84
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 10
Thanks for the extended display tip a couple messages above. I was battling the issues cfraser was, not knowing that you could force the patterns to one display or the other.

One thing I ran into last night was when looking at the primary/secondary measurements, it seems that when I do a "measure greyscale, primary, secondary" the white level seems to be measured at @100% , even though I'm doing 75/75, throwing off all of the other calculations. It seems OK when I just do a "measure primary and secondary". I could do some more experimenting tonight, but it seemed to be consistent.
flyt100 is offline  
post #5413 of 5886 Old 02-11-2015, 09:21 AM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
Note: This post was originally just after post #5408 . Re-posted for continuity, to put zoyd's responses in context, and in case flyt100 and I aren't the only ones wondering about this. [I deleted it because it seemed useless, since it was before I figgered more stuff out and I thought it was irrelevant and possibly misleading.]

^ I do in fact always use RGB as it is recommended (less internal conversions) for my main display (the one I'm working on right now, 9G Kuro).

I did notice that in the (I assume) original documentation for the Measures/Generator/Configure Help doc an option to do exactly what I need to do is shown: select a monitor out of those connected to send the test patterns to. It says I need to be in the "wide office" mode of Windows to do this. I have no idea what that mode is, I'm running W7 on the laptop, I guessed they might have meant what is now called the "extended" mode. That really doesn't work any differently than just "running" HCFR on the display under calibration, it seems. I may be able to use this after all, will play some more...

In the same Measures/Generator/Configure window I also don't get an option to select a Target pattern output monitor, even though Windows knows both displays are there, no matter how I configure the two displays (many ways actually, I've tried them all). Again, this seems to be the feature I'm looking for.

[That ControlCal thing was a red herring, my memory wasn't correct really, it is not the "root" impediment.]
cfraser is offline  
post #5414 of 5886 Old 02-11-2015, 09:32 AM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
Quote:
Originally Posted by zoyd View Post
0-255 is the natural state of the graphics card and it means that levels sent to it will not be scaled, when set to 16-235 it is scaling to video levels which is why you are having to set HCFR to 0-255 to get it correct.
That is the logic that nailed it for me: the Intel app that configures the graphics card talks about scaling, and we don't want any scaling, we'll do it in HCFR if required. There really isn't much technical info in the app though, like for instance a contrast of 40 and a brightness of 0 (both defaults here) is in fact no molesting of the HDMI video signal, but you can only find that out by trial and error.

As you said, it's best to figure settings out in comparison to a known video source. It's actually quite simple to configure the laptop and graphics card for HCFR in my case, but I had no way to know that. You do learn a lot more when you make things difficult, always new mistakes to learn from...
cfraser is offline  
post #5415 of 5886 Old 02-11-2015, 05:13 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
I've uploaded a new version of HCFR, there hasn't been much testing with the new color checker patterns so let me know if you run into any oddities. Note that if you use the madVR TPG you need to make sure you have the latest version of madVR (0.87.13)

3.1.7
-------
  • Added additional color checker options for CalMAN Classic, CalMAN SG skin tones, and ChromaPure skin tones. Pattern sets enabled with the internal generator, madVR, and manual DVD. Use Ted's disk for these patterns.
  • Added special color space - "OPT", optimized for generic plasma calibration.
  • Changed selected color display precision to 4 decimals for ftL field.
  • Fixed some discrepancies between delta luminance and HSV bar indicator in special color spaces.
  • Fixed dE bar indicator obscuration of dE value when dE > 4.
  • Updated madVRTPG related code to current API which allows window area, APL value, and OSD state to be set from within HCFR. madVR version 0.87.13 must be installed. madTPG will start automatically if not already running.
  • Added warnings if program is configured incorrectly prior to doing a saturation run. Normal color space 100% luminance level primaries/secondaries is required.

Last edited by zoyd; 02-12-2015 at 03:48 AM.
zoyd is online now  
post #5416 of 5886 Old 02-11-2015, 06:24 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
^ Thanks!

I'll give you a color checker response. By coincidence I just ran the one in the previous HCFR version (3.1.6) for the first time. (You know that tonight is the first time I've ever run HCFR with the internal generator "properly", based on my recent failings.) I have not adjusted anything at all on the Pio 151 display since "calibrating" it a few days ago with HCFR and a BDP/test disc.

I don't know what to say except that the worst dE is 1.6 (Purplish blue) and the vast majority are <1. I have never run any sort of color checker on this display before. And there are slight differences in basic cal results from my BDP method, but nothing to remark on. I'm just saying this so that it's understood I did not cal this display to this HCFR internal source (yet), and the patterns are slightly different from those on my test BD, and it is a plasma so test pattern variation will cause slight discrepancies, etc. (enough disclaimers). I guess the best I can say with zero experience with them is the color checker pattern results are very credible to me.

Edit: also, very specific to me, is I have a Pio panel that has the Blue at (0.150, 0.068) instead of the ideal (0.150,0.060), which is about as bad as this batch of panels came re blue, so anything involving blue will always be a tad off and always gives significantly larger dEs than everything else

Last edited by cfraser; 02-11-2015 at 06:41 PM.
cfraser is offline  
post #5417 of 5886 Old 02-11-2015, 07:45 PM
Member
 
flyt100's Avatar
 
Join Date: Apr 2003
Posts: 84
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 10
I just tried 3.1.7. I was able to direct the patterns to the extended display as you stated, and it worked much better than the way I was doing it. I understand that isn't a new feature, but I'm glad you pointed it out!

I did notice one issue. When I click on the colors in the primary/secondary measurements to show that color, they all work, except for red (that had worked on the previous version.).


The other thing I had mentioned previously is the same (I didn't expect that you were able to look at it yet, so it's here for completeness):

When looking at the primary/secondary measurements, when I do a "measure greyscale, primary, secondary" the white level value is shown as the 100% value, even though I'm doing 75/75, throwing off all of the other calculations. It is OK when I just do a "measure primary and secondary"

Last edited by flyt100; 02-11-2015 at 08:01 PM.
flyt100 is offline  
post #5418 of 5886 Old 02-11-2015, 08:37 PM
AVS Special Member
 
LarryInRI's Avatar
 
Join Date: Jun 2007
Location: in RI
Posts: 5,306
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 168
Hi, zoyd. Great job. Now please explain this item : "Added special color space - "OPT", optimized for generic plasma calibration."

Larry
LarryInRI is offline  
post #5419 of 5886 Old 02-12-2015, 01:59 AM
Newbie
 
birdiebirdies's Avatar
 
Join Date: Mar 2009
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 0
Probably a stupid question... what are the two CSV files in the doanload folder used for?
birdiebirdies is offline  
post #5420 of 5886 Old 02-12-2015, 03:46 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
Quote:
Originally Posted by flyt100 View Post

I did notice one issue. When I click on the colors in the primary/secondary measurements to show that color, they all work, except for red (that had worked on the previous version.).


The other thing I had mentioned previously is the same (I didn't expect that you were able to look at it yet, so it's here for completeness):
Thanks, I'll take a look.


Quote:
Originally Posted by LarryInRI View Post
Hi, zoyd. Great job. Now please explain this item : "Added special color space - "OPT", optimized for generic plasma calibration."

Larry
Plasmas have similar spectral distributions and cross-talk between the primaries which tend to create a non-linear response in the same portions of the gamut. This color space puts the control points as close as possible to those areas along each of the color axes. I haven't played around with it much so I really don't know if it will do much better than the 75%/75% scheme but in theory it should.

Quote:
Originally Posted by birdiebirdies View Post
Probably a stupid question... what are the two CSV files in the doanload folder used for?
You can use those with the internal pattern generator to randomly test your calibration, rename either one to colors.csv and it will populate the "User defined" selection of the color checker.
zoyd is online now  
post #5421 of 5886 Old 02-12-2015, 04:03 AM
Newbie
 
birdiebirdies's Avatar
 
Join Date: Mar 2009
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 0
not sure I understnad it.
Where exactly I'd have to put the Colors.csv and where in hcfr would I be able to see/use this?
birdiebirdies is offline  
post #5422 of 5886 Old 02-12-2015, 10:24 AM
Member
 
mot67's Avatar
 
Join Date: Nov 2005
Posts: 83
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
First of all thank you all for a very informative thread and great, free software!
I've been trying to calibrate my over 3 years old Samsung PN59D7000 plasma using a brand new Colormunki Display and HCFR 3.1.6.
This is my first time ever calibrating and I'd appreciate any comments if there is something obviously wrong in my results.

Before calibrating I resetted to defaults and used Movie mode, made basic brightness/contrast adjustments, Movie mode and Warm2 Colour tone.
HCFR was set to Plasma display type, HDTV-REC 709 color space, ITU-R BT 1886 0=Absolute/100%, 2.2 Ref./avg. Gamma.
For patterns I used AVS HD 709 windowed grayscale and 100% color patterns.

Before calibration the grayscale had way too much red and too low green. I had to lower red gain to 0, raise green to almost max 47 and blue 40 to get it near 100%. Tweaking the 30% offsets and 80% gains I already got a pretty good grayscale with a peak less than 10% for red and blue in 70% which I was able to flatten somewhat by adjusting 10 point (-1 red and blue for 6 and -2 r/b for 7).
I was not able to get the gamma very good, at least according to the curve.

Overall I'm quite happy with the image now. Funny how my eyes got used to such reddish image I had before that I didn't realize how off it was
However, there is a light greenish tint to black and white images. For example in Casino Royales beginning it shows clearly and sometimes in more colourful scenes if there's some light gray mist, clouds etc..
It was there before calibration too, but then it was a pinkish/purpleish/greenish tint. This slight light green tint is better anyway
Attached Thumbnails
Click image for larger version

Name:	Default Grayscale.jpg
Views:	49
Size:	80.3 KB
ID:	541265   Click image for larger version

Name:	Calibrated Grayscale.jpg
Views:	46
Size:	97.5 KB
ID:	541273   Click image for larger version

Name:	Default Color.jpg
Views:	44
Size:	65.4 KB
ID:	541281   Click image for larger version

Name:	Calibrated Color.jpg
Views:	49
Size:	80.7 KB
ID:	541289   Click image for larger version

Name:	Default Luminance.jpg
Views:	43
Size:	132.5 KB
ID:	541297  

Click image for larger version

Name:	Calibrated Luminance.jpg
Views:	47
Size:	123.5 KB
ID:	541305   Click image for larger version

Name:	Default Gamma.jpg
Views:	53
Size:	162.8 KB
ID:	541313   Click image for larger version

Name:	Calibrated Gamma.jpg
Views:	50
Size:	102.9 KB
ID:	541321   Click image for larger version

Name:	Default RGB.jpg
Views:	56
Size:	180.7 KB
ID:	541329   Click image for larger version

Name:	Calibrated RGB.jpg
Views:	52
Size:	162.7 KB
ID:	541337  

Click image for larger version

Name:	Default CIE.jpg
Views:	60
Size:	105.0 KB
ID:	541345   Click image for larger version

Name:	Calibrated CIE.jpg
Views:	53
Size:	100.7 KB
ID:	541353  
mot67 is offline  
post #5423 of 5886 Old 02-12-2015, 11:20 AM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
Is it possible to "single-step" through the test patterns? e.g. suppose I want to slowly go through the grayscale test patterns, and stop at 80% for a while to adjust something

I suppose I could make my own pattern (don't know how yet), but it would be best if I could use the same patterns/format as I'm taking the "auto" readings with.

Thanks.

[I get the distinct impression from reading this thread that HCFR does a lot of "hidden"/non-obvious things that I haven't figured out yet.]
cfraser is offline  
post #5424 of 5886 Old 02-12-2015, 11:28 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
You can't "play/pause" a sweep. The only two options are the full sweep or single pattern selection of either grayscale or primaries/secondaries.
zoyd is online now  
post #5425 of 5886 Old 02-12-2015, 12:30 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
^ The "single pattern selection" would do the job just fine, better and more conveniently than what I said. I figured HCFR could do that somehow from what I read, but I couldn't figure out how to get it to work. I did manage to get it to work, some way, once, but then I couldn't make it work again. I think it filled up the whole screen with the pattern, it wasn't the same as the regular full-sweep pattern anyway. Sorry for asking about this feature that once again will probably be obvious once I know how.

I also got the impression from things I read here that after you had done a full sweep, you could select one of the "measure columns" and somehow get that single test pattern output. True?

Other subjects:

I can't get proper RGB 16-235 patterns output unless I select 0-255 in the generator. IOW I can't get full black unless I do that. I know they are really 16-235 because if I put my display in a 0-255 mode, the "black" output then appears gray etc., yet it looks correct when the display is in the 16-235 mode. [I can do the same test with a BDP/test disc changing from 16-235 to 0-255 outputs. It is truly varying the output RGB range.] This is not a problem for me, but would be if I really wanted 0-255. The graphics chipset is set up for the "full" RGB range.

How "accurate" are the target squares for saturation and color checker in the CIE diagram? I ask because I have really small errors for most of them (<1), yet the measured points/circles seem to fall noticeably outside the target squares, much more proportionately outside them than they do for similar dEs with primary/secondary measures. Maybe just my imagination...or do these apparently small errors for the saturation/checker readings somehow get "magnified" when plotted, IOW they're more significant errors?

Thanks.
cfraser is offline  
post #5426 of 5886 Old 02-12-2015, 12:48 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
Quote:
Originally Posted by cfraser View Post
I also got the impression from things I read here that after you had done a full sweep, you could select one of the "measure columns" and somehow get that single test pattern output. True?
Yes, select the column and that pattern will be displayed.

Quote:
I can't get proper RGB 16-235 patterns output unless I select 0-255 in the generator. IOW I can't get full black unless I do that. I know they are really 16-235 because if I put my display in a 0-255 mode, the "black" output then appears gray etc., yet it looks correct when the display is in the 16-235 mode. [I can do the same test with a BDP/test disc changing from 16-235 to 0-255 outputs. It is truly varying the output RGB range.] This is not a problem for me, but would be if I really wanted 0-255. The graphics chipset is set up for the "full" RGB range.
Sounds like your graphics card is scaling to 16-235 anyways, make sure it's not set to output Y'CC.

Quote:
How "accurate" are the target squares for saturation and color checker in the CIE diagram? I ask because I have really small errors for most of them (<1), yet the measured points/circles seem to fall noticeably outside the target squares, much more proportionately outside them than they do for similar dEs with primary/secondary measures. Maybe just my imagination...or do these apparently small errors for the saturation/checker readings somehow get "magnified" when plotted, IOW they're more significant errors?
They are exact. You can't compare distances on the CIE diagram with dE values very easily for two reasons:
  • dx and dy are non-linearly related to dE.
  • dE is 3-dimensional and the third component of dE related to luminance is not shown on the diagram.
Geoff D likes this.
zoyd is online now  
post #5427 of 5886 Old 02-12-2015, 02:15 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
^ Thanks. Yes, I did think dE was a 3D vector distance/measure. I did think your target squares were exact, but couldn't get straight in my mind why the measures appeared so "far" outside the targets, yet were so "close" in dE. Thanks for the correction, I'll pay better attention/look again when I get to my HCFR PC. I think it's time for me to re-read the basic documentation for calibration/CIE diagram...been a couple years and it's only something I do occasionally, like when somebody asks or I think one of my own displays "deserves" touching up.

So what are the target squares on the CIE diagram good for? I guess they're just for reference/looking at then, you can't practically actually use them to adjust anything. What can you use the color checker for? I know the automatic display calibration program for one of my PC monitors does a whole bunch of those, but IIRC all it says is good/bad for each of them because you can't really adjust for them AFAIK. To put another way: if the HCFR color checker gives a bad result somewhere and you want to make adjustments, how are you supposed to use the available HCFR tools to fix it? How do you know what to adjust? Is dE kind of just a "fancier" good/bad indicator?
cfraser is offline  
post #5428 of 5886 Old 02-12-2015, 03:32 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,499
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 743 Post(s)
Liked: 642
Quote:
Originally Posted by cfraser View Post
So what are the target squares on the CIE diagram good for? I guess they're just for reference/looking at then, you can't practically actually use them to adjust anything.
It's a quick check to see how bad (or good) things are, you can use it to make quick assessments of general display non-linearity after the gamut edge is aligned to your reference.

Quote:
What can you use the color checker for? I know the automatic display calibration program for one of my PC monitors does a whole bunch of those, but IIRC all it says is good/bad for each of them because you can't really adjust for them AFAIK. To put another way: if the HCFR color checker gives a bad result somewhere and you want to make adjustments, how are you supposed to use the available HCFR tools to fix it? How do you know what to adjust? Is dE kind of just a "fancier" good/bad indicator?
The color checker and saturation scans are proofing tools and not typically used for real-time adjustments because the colors are not aligned with any particular control in a display's CMS. You use the dE statistics from these scans to assess your final calibration's overall color performance. Sometimes calibrating the edge points perfectly is not the best solution for overall color performance and these tools will tell you that.
zoyd is online now  
post #5429 of 5886 Old 02-12-2015, 03:58 PM
AVS Special Member
 
LastButNotLeast's Avatar
 
Join Date: Feb 2007
Location: 08077
Posts: 5,558
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 574 Post(s)
Liked: 424
Quote:
Originally Posted by zoyd View Post
Sometimes calibrating the edge points perfectly is not the best solution for overall color performance and these tools will tell you that.
You might want to try a calibration using the 75%/75% setting (after recording your settings for this calibration) which, rather than aiming for 100%, which is not often seen in real-world viewing situations, aims for 75%, which is much more typical.
Running a color checker and comparing the results from the two calibrations will tell you which is "better."
Or just enjoy your set the way it is.

Did you really need to quote that entire post in your reply?
Downloadable FREE demo discs: Demonstration Blu-Ray Discs (Independently Authored)
Welcome to AVS - Get out while you still can!
LastButNotLeast is offline  
post #5430 of 5886 Old 02-12-2015, 05:37 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,428
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 104 Post(s)
Liked: 66
I like what you're saying about the edge points etc., I'm not a slave to them, and my display would say "I know, I know...". They are reasonable though, or at least they used to be. It's time to get into the service menu and give "the guns" the 10k tune-up, B can't be helped much IME, but R has drifted somewhat ("algorithmically aged lousily", plasma). Makes sense to adjust for best/correct look of real-world products rather than test signals, if one has to choose one or the other, but I'm nowhere close to that yet, just learning the ropes of the built-in generator.

I have always previously used the "75% APL" patterns on my test discs. Used similar with the generator, except for saturation and color checker. This 75/75 thing is new to me, I see it in the choices and have read a lot of recent comments about it in this thread, but haven't tried it yet.

What got me to look into touching up this display right now is I've been watching some B&W BDs. These are sourced from the early-mid 60s and are more taxing to the display's B&W capabilities than the relatively few B&W movie BDs I've watched were (the extreme majority of B&W content I have is on DVD). It could be the quantity though, with so many hours of getting used to it (it's a TV show), *I've* grown more critical with experience of the certain look. Sure doesn't look like a B&W show on an old B&W CRT, in every way...though I'm certain this display has a better grayscale by miles than any home TV I had ever was back in the day.

Edit:
Quote:
Originally Posted by zoyd View Post
Yes, select the column and that pattern will be displayed.
This only works reliably for me with the 10-point grayscale (likely the most important anyway). The primaries/secondaries sometimes works. All the rest never work. I'll keep trying...

And again: nope, not possible for me to get proper blacks with 16-235 generator setting and matching display settings. I verified by Oppo the display setting is definitely not the prob. So then I'll have to say it's my video "card" that must be doing it, though I can find no mention of this (potential?) RGB range problem by googling. [I might also say without being too arrogant that even the people who reviewed this laptop on so-called "professional" sites seemed clueless and almost oblivious to its HDMI capability. Having HDMI capability is a big deal to some laptop purchasers, you'd think a lot of them actually.] This video card gives no video type option and all references to video are to RGB, so I guess that's the only "option" (not totally unreasonably, since it is a PC). It appears the Intel video card app doesn't actually change the RGB output range, Intel offers no updated drivers, the chipset itself is fully capable of 0-255 video.

Last edited by cfraser; 02-12-2015 at 07:53 PM.
cfraser 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