HCFR - Open source projector and display calibration software - Page 263 - 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: 
 746Likes
Reply
 
Thread Tools
post #7861 of 7871 Old Yesterday, 05:08 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 886
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 219 Post(s)
Liked: 173
Quote:
Originally Posted by ck1223 View Post
I'm not using Argyll drivers. I'm using the drivers from the Colormunki Display CD. If I were already using Argyll Drivers of course I'd do the "work".
If you are using a not-ancient version of HCFR, you are using the ArgyllCMS drivers. But in any case, running "dispwin" doesn't involve any of that, since there is no need to access an instrument to test ChromeCast.
gwgill is offline  
Sponsored Links
Advertisement
 
post #7862 of 7871 Old Yesterday, 05:22 PM
Member
 
ck1223's Avatar
 
Join Date: Jun 2009
Posts: 20
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 10
Quote:
Originally Posted by gwgill View Post
If you are using a not-ancient version of HCFR, you are using the ArgyllCMS drivers. But in any case, running "dispwin" doesn't involve any of that, since there is no need to access an instrument to test ChromeCast.
So in an effort to solve this:

Running latest HCFR (v 3.4.2)
1. Tested static IP... didn't work. (note that no Argyll drivers were manually downloaded)
2. Uninstalled Colormunki Display and X-rite Service Manager (drivers). Downloaded Argyll driver package v1.9.1 and pointed Windows to use the 1.9.1 Argyll drivers.
3. With static IP and 1.9.1 Argyll Drivers, Chromecast worked without issue.
4. As a test, disabled static IP with latest Argyll drivers and no longer worked.

So it looks like the best case scenario to get this to work is to have your computer as a static IP with the latest Argyll drivers.

Thank you
ck1223 is online now  
post #7863 of 7871 Old Yesterday, 05:59 PM
AVS Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 1,443
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 826 Post(s)
Liked: 151
Quote:
Originally Posted by ck1223 View Post
Any thoughts on the pattern type I should be using considering the TV technology?
I'm not familiar with your TV model. I calibrate mostly projectors and LCD TVs without auto-dimming; for those the choice of pattern type does not affect the calibration results.
Dominic Chan is online now  
post #7864 of 7871 Old Yesterday, 07:28 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 886
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 219 Post(s)
Liked: 173
Quote:
Originally Posted by ck1223 View Post
3. With static IP and 1.9.1 Argyll Drivers, Chromecast worked without issue.
4. As a test, disabled static IP with latest Argyll drivers and no longer worked.
Thanks for testing - although I'm not sure if you actually used V1.9.1 dispwin or not, since it's nothing to do with drivers (i.e. nothing to do with .sys files).

Unfortunately I'm unable to reproduce the problem - in my local network I can find my ChromeCast whether my computer has a fixed IPV4 address, or not.
gwgill is offline  
post #7865 of 7871 Old Yesterday, 09:14 PM
Member
 
ck1223's Avatar
 
Join Date: Jun 2009
Posts: 20
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 10
Quote:
Originally Posted by Dominic Chan View Post
Before doing the 2-pt calibration, I would run the grey scale sweep to see what it looks like "as is". This will help decide whether a different preset should be used as the starting point.
Making drastic adjustments to the Bias controls based on reading 30 IRE can have detrimental effects on lower levels; i.e., the lower levels may get over-compensated. With many TVs, I get better results by leaving the Bias controls at default settings and use the 10-pt to get the correct grey scale.


The Colormunki Display should be similar to the i1D3, other than speed and software compatibility. Colorimeters are actually better than spectros at low levels, but in any case 30 IRE should not be a challenge for any meter.

You can run the HCFR Near Black grey scale to confirm that blacks are not crushed.


In general you should reduce the R/G/B gain to get the proper colour temperature. Raising the gain to Max can cause clipping even when the contrast is reduced.
I've added pre calibration images. I used my Chromecast to display the patterns.
Attached Thumbnails
Click image for larger version

Name:	Pre Cal Pro RGB.PNG
Views:	1
Size:	587.6 KB
ID:	1691233   Click image for larger version

Name:	Pre Cal Pro nEAR bLACK.PNG
Views:	0
Size:	76.5 KB
ID:	1691241   Click image for larger version

Name:	Pre Cal Pro n EAR wHITE.PNG
Views:	0
Size:	375.2 KB
ID:	1691249   Click image for larger version

Name:	Pre Cal Pro Luminance.PNG
Views:	0
Size:	606.3 KB
ID:	1691257   Click image for larger version

Name:	Pre Cal Pro Grayscale.PNG
Views:	0
Size:	108.9 KB
ID:	1691265  

Click image for larger version

Name:	Pre Cal Pro Gamma.PNG
Views:	0
Size:	514.7 KB
ID:	1691273   Click image for larger version

Name:	Pre Cal Pro Colourtemp.PNG
Views:	0
Size:	592.9 KB
ID:	1691281   Click image for larger version

Name:	Pre Cal Pro CIE.PNG
Views:	0
Size:	574.2 KB
ID:	1691289  

Last edited by ck1223; Today at 06:30 AM.
ck1223 is online now  
post #7866 of 7871 Old Yesterday, 10:03 PM
AVS Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 1,443
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 826 Post(s)
Liked: 151
Quote:
Originally Posted by ck1223 View Post
I've added pre calibration images. I used my Chromecast to display the patterns.
I'm not sure how to correlate these pre-calibration images to your earlier results. You previously said that "the 30% Blue Bias is nearly maxed to achieve a deltaE < 1", and yet the pre-calibration measurement shows a deltaE of 0.5 without raising the Blue Bias.

In any case, the RGB levels graph and the Near Black graph support my earlier suggestion that the Blue bias should not be raised; if anything, it should be lowered. Note that the bias controls have a much bigger effect at 0-10 IRE than at 30 IRE.

For the high end, I would lower the Green gain rather than raising the Red gain.

The gamma is at almost 2.4 rather than 2.2. That will make the shadows dark.

Last edited by Dominic Chan; Yesterday at 10:13 PM.
Dominic Chan is online now  
post #7867 of 7871 Old Today, 05:34 AM
Member
 
ck1223's Avatar
 
Join Date: Jun 2009
Posts: 20
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 10
Quote:
Originally Posted by Dominic Chan View Post
I'm not sure how to correlate these pre-calibration images to your earlier results.
The images I uploaded were an untouched setting from the TV. I did not make any changes to it. (You can see how bright the luminance is at 100 IRE in lieu of the standard 100cd/m2 target). This was using the Chromecast and builtin patterns of HCFR. The color settings (default) are as follows:

Rg MAX
Gg MAX
Bg MAX

Rb 0
Gb 0
Bb 0

All 10 point RGB values are at 0.



If you wanted to see the graphs from what I was describing before, I've attached them. Note that scales are different from the uncalibrated images. You can see how the Red takes a dive the brighter the image, and the Blue is at 120% at 10% IRE. This was using the 5% patterns on GCD disc so we're not exactly comparing apples to apples. Unfortunately I did not take Near Black and Near Whites.

In any case, your recommendation of lowering Green Gain at 80IRE is what I do, as well as blue. When I am satisfied with the 80% deltaE and start to adjust the 30% IRE, the 30% deltaE is thrown off because the adjustment of the 80% RGB Gains I just made. So I tweak 30%, check 80% and tweak that etc. etc. until I dial in 80 and 30 where I want it. Unfortunately that ended up with 30%IRE Blue at +18 (+20 is max).

Basically I think I should just start over lol
Attached Thumbnails
Click image for larger version

Name:	Post Cal Pro RGB.PNG
Views:	5
Size:	591.0 KB
ID:	1690985   Click image for larger version

Name:	Post Cal Pro Luminance.PNG
Views:	4
Size:	598.3 KB
ID:	1690993   Click image for larger version

Name:	Post Cal Pro Grayscale.PNG
Views:	5
Size:	102.2 KB
ID:	1691001   Click image for larger version

Name:	Post Cal Pro Gamma.PNG
Views:	5
Size:	517.9 KB
ID:	1691009   Click image for larger version

Name:	Post Cal Pro Colourtemp.PNG
Views:	5
Size:	592.5 KB
ID:	1691017  

Click image for larger version

Name:	Post Cal Pro CIE.PNG
Views:	4
Size:	570.0 KB
ID:	1691025  
ck1223 is online now  
post #7868 of 7871 Old Today, 06:18 AM
Member
 
ck1223's Avatar
 
Join Date: Jun 2009
Posts: 20
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 10
Quote:
Originally Posted by Dominic Chan View Post
I'm not sure how to correlate these pre-calibration images to your earlier results. You previously said that "the 30% Blue Bias is nearly maxed to achieve a deltaE < 1", and yet the pre-calibration measurement shows a deltaE of 0.5 without raising the Blue Bias.

In any case, the RGB levels graph and the Near Black graph support my earlier suggestion that the Blue bias should not be raised; if anything, it should be lowered. Note that the bias controls have a much bigger effect at 0-10 IRE than at 30 IRE.

For the high end, I would lower the Green gain rather than raising the Red gain.

The gamma is at almost 2.4 rather than 2.2. That will make the shadows dark.
Here are some images, when I target proper black level and 100cd/m2. I target a bit over 100 since peak luminance normally drops when adjusting RGB later.

The changes I've made compared to the uncalibrated Sony Default Setting...

Changed the Global Gamma from a "-2" to "0"
Dropped the backlight from 35 to 9. This gave me my target luminance.
Contrast stayed at 90 out of 100.

No changes have been made to any RGB values.
Attached Thumbnails
Click image for larger version

Name:	BCG Cal Pro RGB.PNG
Views:	5
Size:	589.7 KB
ID:	1691385   Click image for larger version

Name:	BCG Cal Pro Near White.PNG
Views:	4
Size:	374.1 KB
ID:	1691393   Click image for larger version

Name:	BCG Cal Pro Near Black.PNG
Views:	4
Size:	77.6 KB
ID:	1691401   Click image for larger version

Name:	BCG Cal Pro Luminance.PNG
Views:	4
Size:	599.7 KB
ID:	1691409   Click image for larger version

Name:	BCG Cal Pro Grayscale.PNG
Views:	4
Size:	120.6 KB
ID:	1691417  

Click image for larger version

Name:	BCG Cal Pro Gamma.PNG
Views:	3
Size:	516.3 KB
ID:	1691425   Click image for larger version

Name:	BCG Cal Pro Colourtemp.PNG
Views:	2
Size:	594.0 KB
ID:	1691433   Click image for larger version

Name:	BCG Cal Pro CIE.PNG
Views:	2
Size:	572.9 KB
ID:	1691441  

Last edited by ck1223; Today at 06:39 AM.
ck1223 is online now  
post #7869 of 7871 Old Today, 07:09 AM
AVS Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 1,443
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 826 Post(s)
Liked: 151
Quote:
Originally Posted by ck1223 View Post
If you wanted to see the graphs from what I was describing before, I've attached them.
I didn't know that for your TV the default Gain setting is MAX. That actually makes sense, since as I mentioned previously, the gains should normallt not be increased beyond their default values.

Quote:
So I tweak 30%, check 80% and tweak that etc. etc. until I dial in 80 and 30 where I want it. Unfortunately that ended up with 30%IRE Blue at +18 (+20 is max).
That is the pitfall of the 2-pt calibration process as described in most tutorials. You boosted Blue Bias to get a good reading a 30 IRE, but that actually made the 10 IRE worse.
I would adjust the RGB gains, then run a greyscale sweep to see if bias settings need to be changed by looking at the overall low end (not just one point). After that, use the 10-point controls to fine-tune the greyscale.
Also, for TVs with 10-pt control, I find it better to adjust the gains based 100 IRE rather than 70 or 80 IRE. I believe ConnecTEDD (of Ted's Calibration Disc fame) said the same thing in a previous post.
ck1223 likes this.

Last edited by Dominic Chan; Today at 07:16 AM.
Dominic Chan is online now  
post #7870 of 7871 Old Today, 08:50 AM
Member
 
ck1223's Avatar
 
Join Date: Jun 2009
Posts: 20
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 10
Quote:
Originally Posted by Dominic Chan View Post
I didn't know that for your TV the default Gain setting is MAX. That actually makes sense, since as I mentioned previously, the gains should normallt not be increased beyond their default values.


That is the pitfall of the 2-pt calibration process as described in most tutorials. You boosted Blue Bias to get a good reading a 30 IRE, but that actually made the 10 IRE worse.
I would adjust the RGB gains, then run a greyscale sweep to see if bias settings need to be changed by looking at the overall low end (not just one point). After that, use the 10-point controls to fine-tune the greyscale.
Also, for TVs with 10-pt control, I find it better to adjust the gains based 100 IRE rather than 70 or 80 IRE. I believe ConnecTEDD (of Ted's Calibration Disc fame) said the same thing in a previous post.
Thanks! I'll give it a shot. So basically don't use RGB Bias? Just RGB Gain @ 100IRE and 10 point RGB?

Sent from my Nexus 6P using Tapatalk
ck1223 is online now  
post #7871 of 7871 Old Today, 09:16 AM
AVS Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 1,443
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 826 Post(s)
Liked: 151
Quote:
Originally Posted by ck1223 View Post
Thanks! I'll give it a shot. So basically don't use RGB Bias? Just RGB Gain @ 100IRE and 10 point RGB?
In your case the Near Black Grey Scale graph shows that a slight reduction in Blue Bias may help. Note that this is opposite to what you were doing.

The 10-point controls may not be able to affect the range below 10 IRE.

As a side note, posting the HCFR data file is more useful than the screen shots, at least for people who use HCFR.
ck1223 likes this.
Dominic Chan is online now  
Sponsored Links
Advertisement
 
Reply Display Calibration



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