HCFR - Open source projector and display calibration software - Page 205 - AVS Forum | Home Theater Discussions And Reviews
Baselworld is only a few weeks away. Getting the latest news is easy, Click Here for info on how to join the Watchuseek.com newsletter list. Follow our team for updates featuring event coverage, new product unveilings, watch industry news & more!


Forum Jump: 
 567Likes
 
Thread Tools
Old 06-14-2015, 03:43 PM
Member
 
Sithuk's Avatar
 
Join Date: Mar 2011
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 10
I run HCFR on my laptop with my projector as an extended desktop.

Question 1 (Madvr related crash)
I have to manually close madvr after each HCFR reading run, otherwise when I start a new reading the test patterns do not initialise. When I start a new set of readings when madvr is open on the extended desktop, madvr goes full screen (black) as it should do, but the patterns don't start afterward. When I manually close madvr after each set of readings the issue goes away, as the test patterns start correctly the next time HCFR starts madvr. Does anyone know of a fix?

Question 2 (Video signal selection RGB Full)
My video chain is laptop to AV receiver to Projector. I also have a blu-ray player plugged into the AV receiver. I have my Ati catalyst settings for the laptop set to RGB Full. I have set HFCR to create test patterns as 0-255, MadVR is set to output 16-255 to my projector via the AV receiver. The projector is on auto signal recognition.
I use 16-235 so that I can calibrate the projector for both the blu-ray player and the HTPC (my laptop).
Is there any reason why this setup should be less than ideal? Is there a significant benefit in trying to get the bluray player to output in PC mode (0-255) and matching the projector to receive same? I understood that the conversion would happen anyway, and that I was better to have the conversion happen in MadVR. I am wondering if the projector detects when the signal changes from RGB Full (0-255) of the windows desktop to the 16-235 of the MadVR output. The HCFR calibration appears to be working ok so far.
Sithuk is offline  
Sponsored Links
Advertisement
 
Old 06-14-2015, 04:34 PM
AVS Special Member
 
Join Date: Sep 2014
Posts: 2,332
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 1048 Post(s)
Liked: 545
Quote:
Originally Posted by Sithuk View Post
I run HCFR on my laptop with my projector as an extended desktop.

Question 1 (Madvr related crash)
I have to manually close madvr after each HCFR reading run, otherwise when I start a new reading the test patterns do not initialise. When I start a new set of readings when madvr is open on the extended desktop, madvr goes full screen (black) as it should do, but the patterns don't start afterward. When I manually close madvr after each set of readings the issue goes away, as the test patterns start correctly the next time HCFR starts madvr. Does anyone know of a fix?

Question 2 (Video signal selection RGB Full)
My video chain is laptop to AV receiver to Projector. I also have a blu-ray player plugged into the AV receiver. I have my Ati catalyst settings for the laptop set to RGB Full. I have set HFCR to create test patterns as 0-255, MadVR is set to output 16-255 to my projector via the AV receiver. The projector is on auto signal recognition.
I use 16-235 so that I can calibrate the projector for both the blu-ray player and the HTPC (my laptop).
Is there any reason why this setup should be less than ideal? Is there a significant benefit in trying to get the bluray player to output in PC mode (0-255) and matching the projector to receive same? I understood that the conversion would happen anyway, and that I was better to have the conversion happen in MadVR. I am wondering if the projector detects when the signal changes from RGB Full (0-255) of the windows desktop to the 16-235 of the MadVR output. The HCFR calibration appears to be working ok so far.

Won't attempt to wrestle with your configuration, here are the settings that work with my HTPC, no BD hardware setup:

HCFR, generator 16-235
Windows Media Center, 16-235
Nvidia driver, RGB Full
AVR, movie mode 16-235
Samsung F8500, low 16-235

Last edited by bmcn; 06-14-2015 at 04:46 PM.
bmcn is offline  
Old 06-14-2015, 05:05 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,049
Mentioned: 73 Post(s)
Tagged: 0 Thread(s)
Quoted: 1131 Post(s)
Liked: 886
Quote:
Originally Posted by Sithuk View Post
I run HCFR on my laptop with my projector as an extended desktop.

Question 1 (Madvr related crash)
I have to manually close madvr after each HCFR reading run, otherwise when I start a new reading the test patterns do not initialise. When I start a new set of readings when madvr is open on the extended desktop, madvr goes full screen (black) as it should do, but the patterns don't start afterward. When I manually close madvr after each set of readings the issue goes away, as the test patterns start correctly the next time HCFR starts madvr. Does anyone know of a fix?
That's not something I can reproduce, madTPG works reliably on my extended screen in either windowed or fullscreen mode and whether I start it manually or have HCFR autostart it.

Quote:
Question 2 (Video signal selection RGB Full)
My video chain is laptop to AV receiver to Projector. I also have a blu-ray player plugged into the AV receiver. I have my Ati catalyst settings for the laptop set to RGB Full. I have set HFCR to create test patterns as 0-255, MadVR is set to output 16-255 to my projector via the AV receiver. The projector is on auto signal recognition.
I use 16-235 so that I can calibrate the projector for both the blu-ray player and the HTPC (my laptop).
Is there any reason why this setup should be less than ideal?
No, that sounds right. And it doesn't matter what you set HCFR levels to since you are using madTPG to set the levels.

Quote:
Is there a significant benefit in trying to get the bluray player to output in PC mode (0-255) and matching the projector to receive same?
no

Quote:
I understood that the conversion would happen anyway, and that I was better to have the conversion happen in MadVR. I am wondering if the projector detects when the signal changes from RGB Full (0-255) of the windows desktop to the 16-235 of the MadVR output. The HCFR calibration appears to be working ok so far.
As long as the projector is matched to the madTPG output you are fine.
zoyd is online now  
Old 06-14-2015, 11:54 PM
AVS Special Member
 
jhughy2010's Avatar
 
Join Date: Jan 2014
Location: California
Posts: 1,759
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 816 Post(s)
Liked: 256
Quote:
Originally Posted by LastButNotLeast View Post
Please define "really screwed up." If you calibrated to 75%, then the 75% yellow should have a very low dE. 100% may be way off, but that is less important since very little actual viewing contains 100% yellow (or anything else, for that matter, which is why 75% has become popular). What are the other dEs for your yellow saturation sweep?
Quote:
Originally Posted by zoyd View Post
The x,y values matching (the target arrow on mark) does not always correspond to minimum dE because that does not include luminance. You should always try to get the 3 RGB bars (or HSV bars) at 100% for minimum dE which corresponds to minimum perceptual error. Upload your .chc file so I can see what you've done if you have further questions.
I finally had an opportunity to run through some tests tonight and make some adjustments. I first adjusted my gray scale to compromise a little for the high end. The result was a little higher than dE of 2.0 on 70-100%.

I adjusted CMS using 75% saturation continuous measures (what I did not do though is change the standard to 75%/75% under preferences as I didn't think this was necessary for adjusting with a continuous measure and only focusing on dE and HSV bars... correct me if I am wrong please).

What I found frustrating is that adjusted/calibrating to 75% saturation didn't get me a low dE for both yellow and green in particular. Their dE measures were rather high for the 75% saturation as compared to their 25% and 50% saturation.

I felt like I was able to get the lowest dE for the primary/secondary colors when on the color checker and running the continuous measures there instead. Which saturation level is the primary/secondary colors under color checker? None of the results in color checker correspond to the results in the associated saturation scale (under view). My guess would be the color checker has saturation levels between 50-75% for the primaries/secondaries.

Any help would be greatly appreciated.

http://www.avsforum.com/forum/attach...1&d=1434347646
Attached Files
File Type: zip 6-14-2015.zip (46.9 KB, 25 views)
jhughy2010 is online now  
Old 06-15-2015, 05:51 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,049
Mentioned: 73 Post(s)
Tagged: 0 Thread(s)
Quoted: 1131 Post(s)
Liked: 886
Quote:
Originally Posted by jhughy2010 View Post
I finally had an opportunity to run through some tests tonight and make some adjustments. I first adjusted my gray scale to compromise a little for the high end. The result was a little higher than dE of 2.0 on 70-100%.
In the data file you sent the upper end of the grayscale is pretty far off (dE00=11 @100% ). I'm not familiar with your display or it's controls but you might try calibrating 100%/30% stimulus with the 2pt control to see if you can get it more even across the levels. I wouldn't do the CMS until the grayscale was more linear because that will affect the results of CMS changes. Do you have a 10 point grayscale control?

Quote:
I adjusted CMS using 75% saturation continuous measures (what I did not do though is change the standard to 75%/75% under preferences as I didn't think this was necessary for adjusting with a continuous measure and only focusing on dE and HSV bars... correct me if I am wrong please).
That's fine.

Quote:
What I found frustrating is that adjusted/calibrating to 75% saturation didn't get me a low dE for both yellow and green in particular. Their dE measures were rather high for the 75% saturation as compared to their 25% and 50% saturation.

I felt like I was able to get the lowest dE for the primary/secondary colors when on the color checker and running the continuous measures there instead. Which saturation level is the primary/secondary colors under color checker? None of the results in color checker correspond to the results in the associated saturation scale (under view). My guess would be the color checker has saturation levels between 50-75% for the primaries/secondaries.
The Red, Green, Blue... colors in the color checker are generic names and do not correspond to "Primary" colors since they may not even lie on the color axis. So how well they calibrate compared to the 75%/75% colors will depend on the linearity of your display after grayscale and CMS adjustments. Part of your frustration may be that (from the looks of your measurements) this display may be difficult to calibrate.
jhughy2010 likes this.
zoyd is online now  
Old 06-15-2015, 07:44 AM
AVS Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 1,063
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 608 Post(s)
Liked: 112
Hi zoyd,

Couple of minor feature requests for your consideration in future versions:

1. Make the Preference settings "sticky". For example, I use 20-point grey scale as default as 10-point is not fine enough to BT.1886, at the low end (need to set 5 IRE). The current version reverts to 10-point when starting a new set, even if I set the Preference to 20.

2. Related to 1. above, when switching from 10-point to 20-point, preserve the measured values. My JVC projector has 5% increments only for 5, 15, and 95, so it would be easier to do a 10-pt run, then switch to 20-pt and measure those 3 additional points.

Thanks.
Kerlucun likes this.
Dominic Chan is offline  
Old 06-15-2015, 08:32 AM
AVS Special Member
 
jhughy2010's Avatar
 
Join Date: Jan 2014
Location: California
Posts: 1,759
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 816 Post(s)
Liked: 256
Quote:
Originally Posted by zoyd View Post
In the data file you sent the upper end of the grayscale is pretty far off (dE00=11 @100% ). I'm not familiar with your display or it's controls but you might try calibrating 100%/30% stimulus with the 2pt control to see if you can get it more even across the levels. I wouldn't do the CMS until the grayscale was more linear because that will affect the results of CMS changes. Do you have a 10 point grayscale control?
My chart says that at 100% the dE is 7.0, is that what you mean? I will definitely try 100% then 30% again next time. One minor adjustment downward at 100% drastically increases dE for 70%, 80% and 90%. My display (Sharp SQ) has garnered a reputation as being very difficult to calibrate. The 10 point grayscale control is present but hasn't been used. Other members that have calibrated this display say that there is something wrong with the 10 point (doesn't work very well) and that it cancels out the 2 point. Just like CMS -Value. It has absolutely no affect so it isn't used either. Pretty useless.

Quote:
Originally Posted by zoyd View Post
The Red, Green, Blue... colors in the color checker are generic names and do not correspond to "Primary" colors since they may not even lie on the color axis. So how well they calibrate compared to the 75%/75% colors will depend on the linearity of your display after grayscale and CMS adjustments. Part of your frustration may be that (from the looks of your measurements) this display may be difficult to calibrate.
That is good to know. I will stick with calibrating to 75% saturation and not the colorchecker.

I wonder if I can get grayscale adjusted more properly by adjusting gamma control. I have it at -1 right now I'll experiment with -2. Brightness, contrast and color were all adjusted using the guide . I was unable to discern anything with the eye attempting to calibrate contrast and brightness. For contrast I tried to keep it in the 35-40 ftL range with the 100% white test. I did the calculation for brightness then color. These weren't drastic adjustments if any from the displays default.
jhughy2010 is online now  
Old 06-15-2015, 08:55 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,049
Mentioned: 73 Post(s)
Tagged: 0 Thread(s)
Quoted: 1131 Post(s)
Liked: 886
Quote:
Originally Posted by jhughy2010 View Post
My chart says that at 100% the dE is 7.0, is that what you mean?
ok sorry, I had the program set to a different grayscale Y weighting value so your measurements actually look fine except for the 100% point at dE=6.7. Try to get that closer to D65 and redo the colors. I'm not sure if you need to use the 75%/75% setting for that display or not.
zoyd is online now  
Old 06-15-2015, 12:44 PM
Member
 
Basstrix's Avatar
 
Join Date: Dec 2012
Location: AZ
Posts: 115
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 27
Hi Jhughly,

I took a look at your .chc file and see some familiar problems. I see that you have the same saturation tracking issues with cyan, blue, green and magenta along with an inability to dial in blue at 100%. Go to Measures/Parameters/Saturation Color Levels and set it to 10 and you'll see what I mean.

I can get a good grayscale with 2 point but can get excellent (dE < 1) using the 10 point. It's nowhere near as easy as my 2 Vizio panels, but it can be done.

My early attempts were to calibrate at 100% color/100% sat. I could get decent dE's on the primaries/secondaries, but the picture didn't look good. I didn't have the internal pattern generator set up at that time, so I had no idea how badly the color checker was, but it was obvious the picture wasn't good.

Next, I learned how to cal to the 75% saturation. I wasn't aware of (or perhaps it wasn't present) the color space Zoyd had set up, so I used a spreadsheet developed by stereomandan to come up with x, y, Y targets. The picture improved, but could still tell is wasn't great.

Later still, I learned that I could obtain the best looking picture by using a combination of color check patterns as the final check and to use 8-10 divisions on the saturation scales and adjusting hue, saturation, and value (in that order) to get the best color check dE.

Here's what I've learned so far:

Start with Movie mode. I have the control cal software that allows one to add ISF Day and ISF Night modes and has a very different basis for the calibration. I've found I get better results from the built in Movie mode as the basis.

Set contrast using a full color clip pattern. GCD has a good one and AVSHD is good too. Make sure you're not clipping anything above 237 in the least. Setting contrast in this way will allow the "value" adjustment to have more influence when setting primary/secondaries.

Set brightness using the black clip, as usual. Use backlight to get the output you're looking for. Leave color, tint, and sharpness at zero. I've found that I get the most control from the CMS when the Gamut is set to Standard....similar to setting the contrast using color clip pattern. The gamut and contrast are heavily tied together. If you use expanded gamut, you'll have to substantially reduce contrast to prevent clipping.

When setting grayscale, you may find that bumping Gamma to -1 gets you a better overall dE. I use -1 when using the control cal program, but 0 works best when I use Movie mode.

Turn the auto controls off.

Don't be afraid to use green to raise Y to get your grayscale dE's dialed in. I do as much as I can with red & blue and then bring green in as needed. If you think you're using too much green, try changing the TV's gamma setting.

Once you get grayscale dialed in, move on to the CIE plot. Set your saturation to 8-10 divisions and run the full sweep including CC to see where you're at. Adjust hue, saturation, and value (in that order) to get the best color check dE.

Think of the zero sat point for each color as the center of a circle....when observing the location of a given colorcheck (CC) relative to it's target, use hue for angular adjustment, saturation for gross radial, and value to fine tune. You're not looking to get the CC's perfect, more to get them all as low as possible while keeping the saturation sweeps fairly accurate. I place priority on the "memory" colors.....skin tones, greens, sky blue tones....colors that we easily recognize as being off.

I can actually get my Vizio's to have great dE's across sat sweeps and good CC values overall....very hard with the sharp....in fact, I'm running out of ideas and plan to write a post asking for help on how to improve the saturation tracking. I have seen the tracking-linearity improve by making drastic changes to contrast, standard Gamut shows more linear tracking than Expanded, but beyond that, I'm not sure where to look. If I find anything, I'll be sure to let you know.

I hope all this is helpful.
BT
JGrana and jhughy2010 like this.

Vizio M652i-B2....................HCFR 3.3.1
Vizio E600i-B3.....................i1 Display Pro
Sharp LC-70C8470U............ControlCAL Elite X5 Edition v2
Onkyo TX-SR606.................AVSHD-709
PS-3
Homegrown HTPC
Basstrix is offline  
Old 06-15-2015, 12:55 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,049
Mentioned: 73 Post(s)
Tagged: 0 Thread(s)
Quoted: 1131 Post(s)
Liked: 886
Quote:
Originally Posted by Basstrix View Post
Set contrast using a full color clip pattern. GCD has a good one and AVSHD is good too. Make sure you're not clipping anything above 237 in the least. Setting contrast in this way will allow the "value" adjustment to have more influence when setting primary/secondaries.

Thanks for the tips - clipping high/low are also available now from the internal pattern generator (Ctrl-2 and Ctrl-3).
zoyd is online now  
Old 06-15-2015, 01:11 PM
Member
 
Basstrix's Avatar
 
Join Date: Dec 2012
Location: AZ
Posts: 115
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 27
Quote:
Originally Posted by zoyd View Post
Thanks for the tips - clipping high/low are also available now from the internal pattern generator (Ctrl-2 and Ctrl-3).
Zoyd, I've tried these, but for some reason, they don't flash on my setup. Are they supposed to flash? I find the flashing makes it easier, particularly for blue, to detect clipping.

Vizio M652i-B2....................HCFR 3.3.1
Vizio E600i-B3.....................i1 Display Pro
Sharp LC-70C8470U............ControlCAL Elite X5 Edition v2
Onkyo TX-SR606.................AVSHD-709
PS-3
Homegrown HTPC
Basstrix is offline  
Old 06-15-2015, 02:04 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,049
Mentioned: 73 Post(s)
Tagged: 0 Thread(s)
Quoted: 1131 Post(s)
Liked: 886
Quote:
Originally Posted by Basstrix View Post
Zoyd, I've tried these, but for some reason, they don't flash on my setup. Are they supposed to flash?
No, they aren't animated. I always pause the flashing ones because they are distracting - maybe I'm in the minority.
zoyd is online now  
Old 06-15-2015, 02:28 PM
Member
 
Basstrix's Avatar
 
Join Date: Dec 2012
Location: AZ
Posts: 115
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 27
Quote:
Originally Posted by zoyd View Post
No, they aren't animated. I always pause the flashing ones because they are distracting - maybe I'm in the minority.
Blue is really the only color that animation helps me. I think they are great additions and appreciate all the work you put into this program.

Vizio M652i-B2....................HCFR 3.3.1
Vizio E600i-B3.....................i1 Display Pro
Sharp LC-70C8470U............ControlCAL Elite X5 Edition v2
Onkyo TX-SR606.................AVSHD-709
PS-3
Homegrown HTPC
Basstrix is offline  
Old 06-15-2015, 02:52 PM
AVS Special Member
 
Join Date: Sep 2014
Posts: 2,332
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 1048 Post(s)
Liked: 545
Quote:
Originally Posted by zoyd View Post
hey that's looking real nice. Up the red component of green in the CMS to align it with the reference in the Plasma-OPT space, this will help with skin tones. After that tweak the flesh tone control (try +4) until the pantone skin sweep averages under 1 dE.
Zoyd, increased Red component of green to 18 which lowered green's dE to 0.6.

Then adjusted flesh tone to +4 and skin dE's increased. Lowest Skin dE's were obtained with flesh tone set to 0. Was expecting different results.
Attached Files
File Type: zip 8500 Settings 061515.zip (7.5 KB, 2 views)
File Type: zip 06152015 Complete.zip (41.3 KB, 6 views)
File Type: zip 06152015 Skin.zip (41.3 KB, 8 views)
bmcn is offline  
Old 06-15-2015, 03:48 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,049
Mentioned: 73 Post(s)
Tagged: 0 Thread(s)
Quoted: 1131 Post(s)
Liked: 886
Quote:
Originally Posted by bmcn View Post
Zoyd, increased Red component of green to 18 which lowered green's dE to 0.6.

Then adjusted flesh tone to +4 and skin dE's increased. Lowest Skin dE's were obtained with flesh tone set to 0. Was expecting different results.
Well you're really at the point of diminishing returns now, you'd have to play with a slightly higher color value (get red 50% sat closer to target) and clip a little in the process. The highest error in the skin tones is 1.9 so maybe you should stop now, have a beer and watch some teevee.
jhughy2010 and bmcn like this.
zoyd is online now  
Old 06-15-2015, 04:38 PM
AVS Special Member
 
jhughy2010's Avatar
 
Join Date: Jan 2014
Location: California
Posts: 1,759
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 816 Post(s)
Liked: 256
Quote:
Originally Posted by Basstrix View Post
Go to Measures/Parameters/Saturation Color Levels and set it to 10 and you'll see what I mean.
This is very helpful thank you! It's actually nice to see that I'm not the only one struggling with the Sharp. Honestly, I have no idea how to use the 10 point. I'll check out

Quote:
Originally Posted by Basstrix View Post
I can get a good grayscale with 2 point but can get excellent (dE < 1) using the 10 point. It's nowhere near as easy as my 2 Vizio panels, but it can be done.
Honestly, I have no idea how to use the 10 point. I'll check out some "how to's" later. For now I will concentrate on getting the 2 point dialed in.

Quote:
Originally Posted by Basstrix View Post
Later still, I learned that I could obtain the best looking picture by using a combination of color check patterns as the final check and to use 8-10 divisions on the saturation scales and adjusting hue, saturation, and value (in that order) to get the best color check dE.
So for this you are calibrating 80, 90 and 100% saturation? Are you focusing on dE only when making adjustments to CMS?

Quote:
Originally Posted by Basstrix View Post
Set contrast using a full color clip pattern. GCD has a good one and AVSHD is good too. Make sure you're not clipping anything above 237 in the least. Setting contrast in this way will allow the "value" adjustment to have more influence when setting primary/secondaries.
Do you remember what the file name is for this test on AVSHD?

Quote:
Originally Posted by Basstrix View Post
Set brightness using the black clip, as usual. Use backlight to get the output you're looking for. Leave color, tint, and sharpness at zero. I've found that I get the most control from the CMS when the Gamut is set to Standard....similar to setting the contrast using color clip pattern. The gamut and contrast are heavily tied together. If you use expanded gamut, you'll have to substantially reduce contrast to prevent clipping.

When setting grayscale, you may find that bumping Gamma to -1 gets you a better overall dE. I use -1 when using the control cal program, but 0 works best when I use Movie mode.

Don't be afraid to use green to raise Y to get your grayscale dE's dialed in. I do as much as I can with red & blue and then bring green in as needed. If you think you're using too much green, try changing the TV's gamma setting.

Once you get grayscale dialed in, move on to the CIE plot. Set your saturation to 8-10 divisions and run the full sweep including CC to see where you're at. Adjust hue, saturation, and value (in that order) to get the best color check dE.

Think of the zero sat point for each color as the center of a circle....when observing the location of a given colorcheck (CC) relative to it's target, use hue for angular adjustment, saturation for gross radial, and value to fine tune. You're not looking to get the CC's perfect, more to get them all as low as possible while keeping the saturation sweeps fairly accurate. I place priority on the "memory" colors.....skin tones, greens, sky blue tones....colors that we easily recognize as being off.

I can actually get my Vizio's to have great dE's across sat sweeps and good CC values overall....very hard with the sharp....in fact, I'm running out of ideas and plan to write a post asking for help on how to improve the saturation tracking. I have seen the tracking-linearity improve by making drastic changes to contrast, standard Gamut shows more linear tracking than Expanded, but beyond that, I'm not sure where to look. If I find anything, I'll be sure to let you know.

I hope all this is helpful.
BT
Again, great overall advice and I seriously appreciate your help. It seems there are a ton on here that have sets that are extraordinarily easier to calibrate. I like the Sharp but will be hesitant in purchasing another due to this frustration.

I will make those adjustments you mentioned (already in movie mode and have everything extra turned off). I know that I was in standard for color versus advanced. I will bring gamma back to 0 from -1 and see what kind of results I get with it there.

Definitely was using a lot of green to adjust grayscale. I will bring it back to 0 and start over. Can you elaborate on the CIE plot? Should I be using the diagram to calibrate color? It seemed like the consensus here was just focus on dE.
jhughy2010 is online now  
Old 06-15-2015, 05:03 PM
AVS Special Member
 
Join Date: Sep 2014
Posts: 2,332
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 1048 Post(s)
Liked: 545
Quote:
Originally Posted by jhughy2010 View Post
Do you remember what the file name is for this test on AVSHD?

HCFR has a High Level Clipping pattern and on AVS the flashing color clipping pattern is found in the Misc Patterns, A-Additional folder.
bmcn is offline  
Old 06-15-2015, 06:13 PM
AVS Special Member
 
xvfx's Avatar
 
Join Date: Feb 2013
Location: UK
Posts: 1,280
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 380 Post(s)
Liked: 211
Quote:
Originally Posted by bmcn View Post
HCFR has a High Level Clipping pattern and on AVS the flashing color clipping pattern is found in the Misc Patterns, A-Additional folder.
This is what I don't like about AVSHD disc and only run with Ted's. AVS makes you think you aren't clipping but put up Ted's 21 saturation steps and you've clipped a lot near 100%.
xvfx is offline  
Old 06-15-2015, 06:34 PM
Member
 
Basstrix's Avatar
 
Join Date: Dec 2012
Location: AZ
Posts: 115
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 27
Quote:
Originally Posted by jhughy2010 View Post

So for this you are calibrating 80, 90 and 100% saturation? Are you focusing on dE only when making adjustments to CMS?
No, you're adjusting Hue, Saturation, and Value for a given color and observing a saturation plot of 8-10 points on said color.

For example, when tuning red:
In the upper window (data window), select red saturation scale. In the lower window, show the CIE diagram. Make sure that saturations, sat targets, cc, and cc targets are checked to show. Watch the Average dE for red saturation along with delta luminance for each level (% saturation) as you make adjustments to hue, sat, and value. Once you get the overall dE minimized and delta luminance close to zero, run color check and tweak red to get the color check colors most influenced by red closer to the targets. At this point, you're looking at the dE for the group of colors most influenced by red (dark skin, light skin, moderate red, red, orange, etc) It's tedious, but yields a very nice picture for panels that are inherently difficult to calibrate. Once you get some improvement in red, move on to green, then blue, and so on. Don't try to get anything perfect on the first pass. It's an iterative process and each color will affect the next, so it does no good to spend a bunch of time on one color only to return to it and have lost all the accuracy you spent so much time to get.

That being said, you may be better off getting comfortable calibrating to a 75/75 color space and then working up to the one described above.

The 75/75 space is basically a compromise for panels that don't have perfectly linear tracking and/or cannot achieve adequate calibration results at 100%. Using the entire saturation sweep in combination with the Color Check is just another type of compromise for a difficult panel. Personally, I think most panels can benefit from this method but those like our's (sharp with tracking issues) benefit the most.....this is only my opinion and I'm not an expert.

Quote:
Originally Posted by jhughy2010 View Post
Do you remember what the file name is for this test on AVSHD?
Someone responded with the correct folder for where it's located.

Quote:
Originally Posted by jhughy2010 View Post

I will make those adjustments you mentioned (already in movie mode and have everything extra turned off). I know that I was in standard for color versus advanced. I will bring gamma back to 0 from -1 and see what kind of results I get with it there.

Definitely was using a lot of green to adjust grayscale. I will bring it back to 0 and start over. Can you elaborate on the CIE plot? Should I be using the diagram to calibrate color? It seemed like the consensus here was just focus on dE.
I attached a copy of the spreadsheet I use to keep track of my settings (in case you don't already have one). It should be pretty similar to your set....and you can always edit to add/remove anything you can't use.
Attached Files
File Type: zip Sharp.zip (19.6 KB, 6 views)
jhughy2010 likes this.

Vizio M652i-B2....................HCFR 3.3.1
Vizio E600i-B3.....................i1 Display Pro
Sharp LC-70C8470U............ControlCAL Elite X5 Edition v2
Onkyo TX-SR606.................AVSHD-709
PS-3
Homegrown HTPC
Basstrix is offline  
Old 06-15-2015, 06:52 PM
Member
 
Basstrix's Avatar
 
Join Date: Dec 2012
Location: AZ
Posts: 115
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 27
Here's a slide which attempts to illustrate with a picture what I'm trying to describe with words.
Attached Thumbnails
Click image for larger version

Name:	sat sweep tuning example.JPG
Views:	124
Size:	180.8 KB
ID:	778329  
jhughy2010 likes this.

Vizio M652i-B2....................HCFR 3.3.1
Vizio E600i-B3.....................i1 Display Pro
Sharp LC-70C8470U............ControlCAL Elite X5 Edition v2
Onkyo TX-SR606.................AVSHD-709
PS-3
Homegrown HTPC
Basstrix is offline  
Old 06-15-2015, 10:30 PM
AVS Special Member
 
jhughy2010's Avatar
 
Join Date: Jan 2014
Location: California
Posts: 1,759
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 816 Post(s)
Liked: 256
^^Incredibly helpful thanks much Basstrix. I am going to set up now and give it a fresh start. Will post results when I'm done.
jhughy2010 is online now  
Old 06-16-2015, 12:08 AM
AVS Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 2,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 75
Quote:
Originally Posted by Dominic Chan View Post
1. Make the Preference settings "sticky". For example, I use 20-point grey scale as default as 10-point is not fine enough to BT.1886, at the low end (need to set 5 IRE). The current version reverts to 10-point when starting a new set, even if I set the Preference to 20.
Another feature request: have the zoom settings stick, I always change the RGB levels to 80% and dE to 0-5. Or is this already possible?
rickardl is offline  
Old 06-16-2015, 10:14 PM
AVS Special Member
 
jhughy2010's Avatar
 
Join Date: Jan 2014
Location: California
Posts: 1,759
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 816 Post(s)
Liked: 256
Overall I'm really happy with the results here. Thank you guys for all your help! Basstrix I think I got some pretty good dE values for the 10 saturation measurements you had me take. I had to set the standard to 75%/75% for the dE to reflect properly (I think this was just the scale though as adjusting CMS had same affect regardless of standard).

A couple things to note... no matter what I did 100% grayscale was botched. I opted to compromise significantly here so that the rest of the values had lower dE. Even if I attempted to get 100% dE down to say in the 5-6dE range then 80% would be in the 5dE range as well. I think the compromise was worth it.

100% color saturation was impossible to achieve. Focusing on 60, 70, 80 and 90% saturation levels was much simpler and yielded pretty accurate results (the CMS adjustments weren't extreme as they would have been for 100%).
Attached Files
File Type: zip 6-16-2015 Movie Mode.zip (47.6 KB, 42 views)
jhughy2010 is online now  
Old 06-17-2015, 06:35 AM
Member
 
Sithuk's Avatar
 
Join Date: Mar 2011
Posts: 30
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 10
Quote:
Originally Posted by zoyd View Post
No, that sounds right. And it doesn't matter what you set HCFR levels to since you are using madTPG to set the levels.
Question 1: test pattern options
The HCFR test pattern settings options for 1-255 and 16-235. Does that button have no effect as the madTPG settings will override whatever is selected?

Quote:
Is there a significant benefit in trying to get the bluray player to output in PC mode (0-255) and matching the projector to receive same?

>>>no
Question 2: video chain 1-255 16-235 settings
I discovered my projector is set to 1-255. My video chain is therefore:
Gfx card: 1-255
HCFR test pattern generator: 1-255
MadVR settings for projector: 16-235
Projector: 1-255

My blu-ray player has been displaying BTB and WTW when using the AVS709 test disc so it must be in 1-255 mode too.

I finished a calibration last night and the picture looks good. I don't understand how it can if MadVR is outputting at 16-235 and the projector is in 1-255. Should I switch MadVR to 1-255 and keep my entire video chain 1-255, or do something else? My primary focus is movie quality, I don't mind about desktop clipping.

Quote:
I understood that the conversion would happen anyway, and that I was better to have the conversion happen in MadVR. I am wondering if the projector detects when the signal changes from RGB Full (0-255) of the windows desktop to the 16-235 of the MadVR output. The HCFR calibration appears to be working ok so far.

>>>As long as the projector is matched to the madTPG output you are fine.
Question 3: minimising video chain conversions
Is there really no significant video quality difference between setting the video chain to 1-255 or to 16-235, as long as the projector setting is matched to the MadVR output setting? I had read it was preferable to minimise the number of conversions and that projectors tend to be designed to operate at 16-235, so sending a 1-255 signal would result in a conversion in the projector?

Last edited by Sithuk; 06-17-2015 at 06:41 AM.
Sithuk is offline  
Old 06-17-2015, 08:00 AM
AVS Special Member
 
Join Date: Sep 2014
Posts: 2,332
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 1048 Post(s)
Liked: 545
Quote:
Originally Posted by zoyd View Post
Well you're really at the point of diminishing returns now, you'd have to play with a slightly higher color value (get red 50% sat closer to target) and clip a little in the process. The highest error in the skin tones is 1.9 so maybe you should stop now, have a beer and watch some teevee.

The results are worth the effort and very satisfying. Hockey and basketball final games looked spectacular and the beer was good. Might take one more run to adjust saturations, if only to understand the concept. Thanks for your tireless work on HCFR.
bmcn is offline  
Old 06-17-2015, 03:09 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,049
Mentioned: 73 Post(s)
Tagged: 0 Thread(s)
Quoted: 1131 Post(s)
Liked: 886
3.3.2 is a bit of a usability update taken from some recent suggestions:
  • Added worst 10% dE filter to CIE diagram.
  • Made user choice of number of gray, near black, near white, and saturation steps sticky.
  • Made user adjusted plot scales sticky.
  • Added Belle Nuit test image.
  • Smoother updates during real-time measurements.
  • Added automatic switching to active grid during real-time measurements.
zoyd is online now  
Old 06-17-2015, 05:16 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,049
Mentioned: 73 Post(s)
Tagged: 0 Thread(s)
Quoted: 1131 Post(s)
Liked: 886
Quote:
Originally Posted by Sithuk View Post
Question 1: test pattern options
The HCFR test pattern settings options for 1-255 and 16-235. Does that button have no effect as the madTPG settings will override whatever is selected?
That's correct, madTPG overrides this option.

Quote:
Question 2: video chain 1-255 16-235 settings
I discovered my projector is set to 1-255. My video chain is therefore:
Gfx card: 1-255
HCFR test pattern generator: 1-255
MadVR settings for projector: 16-235
Projector: 1-255

My blu-ray player has been displaying BTB and WTW when using the AVS709 test disc so it must be in 1-255 mode too.

I finished a calibration last night and the picture looks good. I don't understand how it can if MadVR is outputting at 16-235 and the projector is in 1-255. Should I switch MadVR to 1-255 and keep my entire video chain 1-255, or do something else? My primary focus is movie quality, I don't mind about desktop clipping.
Different devices use different terminology. When madVR is set to 16-235 it still outputs 0-255, but it sends black as 16 and white as 235 so you are fine with this setting.


Quote:
Question 3: minimising video chain conversions
Is there really no significant video quality difference between setting the video chain to 1-255 or to 16-235, as long as the projector setting is matched to the MadVR output setting? I had read it was preferable to minimise the number of conversions and that projectors tend to be designed to operate at 16-235, so sending a 1-255 signal would result in a conversion in the projector?
It is true that minimizing conversions is the safest approach although it tends to get exaggerated like everything else in this business. It looks like you have a good set-up where madVR is sending video levels and allows WTW and your projector is set for 16=black and 235=white, 255=above white, that's all you need to worry about really.
zoyd is online now  
Old 06-17-2015, 06:37 PM
Member
 
Basstrix's Avatar
 
Join Date: Dec 2012
Location: AZ
Posts: 115
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 27
Quote:
Originally Posted by jhughy2010 View Post
Overall I'm really happy with the results here. Thank you guys for all your help! Basstrix I think I got some pretty good dE values for the 10 saturation measurements you had me take. I had to set the standard to 75%/75% for the dE to reflect properly (I think this was just the scale though as adjusting CMS had same affect regardless of standard).

A couple things to note... no matter what I did 100% grayscale was botched. I opted to compromise significantly here so that the rest of the values had lower dE. Even if I attempted to get 100% dE down to say in the 5-6dE range then 80% would be in the 5dE range as well. I think the compromise was worth it.

100% color saturation was impossible to achieve. Focusing on 60, 70, 80 and 90% saturation levels was much simpler and yielded pretty accurate results (the CMS adjustments weren't extreme as they would have been for 100%).
The saturation sweeps do look good. I suggest you switch back to the full color space and run them again. It doesn't really affect the net result, but I have no way of judging how close you are on luminance. From what I see of

Your average dE on color check is 3.18 with a max of 7.35. I think you can improve that and get an average below 2.

Your green and yellow are both oversaturated. I think hue can improve on yellow as well.

In color check, your blues look good, but greens are not so good.

I think you're on the right track and getting close!
jhughy2010 likes this.

Vizio M652i-B2....................HCFR 3.3.1
Vizio E600i-B3.....................i1 Display Pro
Sharp LC-70C8470U............ControlCAL Elite X5 Edition v2
Onkyo TX-SR606.................AVSHD-709
PS-3
Homegrown HTPC
Basstrix is offline  
Old 06-17-2015, 07:15 PM
AVS Special Member
 
jhughy2010's Avatar
 
Join Date: Jan 2014
Location: California
Posts: 1,759
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 816 Post(s)
Liked: 256
Quote:
Originally Posted by Basstrix View Post
The saturation sweeps do look good. I suggest you switch back to the full color space and run them again. It doesn't really affect the net result, but I have no way of judging how close you are on luminance. From what I see of

Your average dE on color check is 3.18 with a max of 7.35. I think you can improve that and get an average below 2.

Your green and yellow are both oversaturated. I think hue can improve on yellow as well.

In color check, your blues look good, but greens are not so good.

I think you're on the right track and getting close!
Thanks for the suggestion I will try to fix yellow and green the next time I have a moment.

Can you clarify how I switch back to the full color space?
jhughy2010 is online now  
Old 06-17-2015, 07:30 PM
Member
 
Basstrix's Avatar
 
Join Date: Dec 2012
Location: AZ
Posts: 115
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 27
Quote:
Originally Posted by jhughy2010 View Post
Thanks for the suggestion I will try to fix yellow and green the next time I have a moment.

Can you clarify how I switch back to the full color space?
In HCFR go to Advanced/Preferences and on the References tab at the top is Color Space and Standard. Select HDTV-Rec 709 instead of HDTV-Rec 709 (75%/75%).
jhughy2010 likes this.

Vizio M652i-B2....................HCFR 3.3.1
Vizio E600i-B3.....................i1 Display Pro
Sharp LC-70C8470U............ControlCAL Elite X5 Edition v2
Onkyo TX-SR606.................AVSHD-709
PS-3
Homegrown HTPC
Basstrix is offline  
Sponsored Links
Advertisement
 
 
Thread Tools


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