Official JVC RS20 / HD750 Calibration and CMS thread - Page 16 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #451 of 1782 Old 12-30-2008, 03:45 AM
AVS Special Member
 
JeffY's Avatar
 
Join Date: Mar 2001
Location: England
Posts: 3,667
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by lovingdvd View Post

Thanks. No, pixel shift was not any help. It mainly just shifted the issue from red to green. Also didn't help because of its unevenness.

For those of you with the RS20, what does your convergence look like up close on a single pixel pattern? Can you spot any fringing in text? Does small text look very sharp? I sure hope mine is an exception, along with a few others that posted about similar issues. If so that would greatly increase my changes of getting a much better unit. I anticipate having a new one later in the week.


Some fringing is to be expected on any 3 chip projector, you will even get it on a single chipper due to CA. I consider half a pixel or less across the whole of the screen to be good. Lens shift will also most likely affect the results you get. Some issues are actually in the source material so don't get fooled by that, use a test screen you can trust.
JeffY is offline  
Sponsored Links
Advertisement
 
post #452 of 1782 Old 12-30-2008, 07:02 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,818
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 23
Quote:
Originally Posted by JeffY View Post

Some fringing is to be expected on any 3 chip projector, you will even get it on a single chipper due to CA. I consider half a pixel or less across the whole of the screen to be good. Lens shift will also most likely affect the results you get. Some issues are actually in the source material so don't get fooled by that, use a test screen you can trust.

Unfortunately my MC was more than that. Have a look at the pics I posted in the RS20 owners thread and let me know how yours compares to that. Thanks.
lovingdvd is offline  
post #453 of 1782 Old 12-31-2008, 01:34 AM
Senior Member
 
Rob Dingen's Avatar
 
Join Date: Aug 2000
Location: Netherlands
Posts: 460
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi

I spent 2 days with the RS20 and tweaked the cms.
The calibrations where done with color control on -20 -10 and 0 and on 100% and 75%.
If you lower the color control the linearity gets worse from 0% 25% 50% 75% 100% especially on the 50% and 75% so leave the color control to 0.
Then if you calibrate to get it right on 75% patterns the gamut is to large en red and green are over saturated.
So in a shell CMS calibration on 100% and color control on 0 works best for me.

Rob
Rob Dingen is offline  
post #454 of 1782 Old 12-31-2008, 01:55 AM - Thread Starter
AVS Club Gold
 
Mark Petersen's Avatar
 
Join Date: Dec 2000
Posts: 4,671
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
I finally have some time to spend calibrating my RS20. I started off measuring the greyscale and saw that it was off considerably as others have reported. As the first step in getting RGB to track properly I picked an IRE where the RGB gain/offset adjustment would be most beneficial (70% in this case) and then went back and spent hours tweaking the RGB gamma settings to get the greyscale to track along the whole IRE range.

When I was finished I found that my biggest error was at 100% IRE and there isn't a corresponding gamma adjustment at this level to improve the values. So now I'm thinking that the best place to use the gain/offset adjustment is at 100% IRE and then use the gamma adjustments to get the greyscale right elsewhere down the curve. Have others followed this procedure too?
Mark Petersen is offline  
post #455 of 1782 Old 12-31-2008, 02:00 AM - Thread Starter
AVS Club Gold
 
Mark Petersen's Avatar
 
Join Date: Dec 2000
Posts: 4,671
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Rob Dingen View Post

Hi

I spent 2 days with the RS20 and tweaked the cms.
The calibrations where done with color control on -20 -10 and 0 and on 100% and 75%.
If you lower the color control the linearity gets worse from 0% 25% 50% 75% 100% especially on the 50% and 75% so leave the color control to 0.
Then if you calibrate to get it right on 75% patterns the gamut is to large en red and green are over saturated.
So in a shell CMS calibration on 100% and color control on 0 works best for me.

Rob

Interesting info, thanks for sharing Rob. So at 100% and color at 0 the tracking is the best, but how far off are the primaries? Do you mind posting your CMS settings and also measurements?
Mark Petersen is offline  
post #456 of 1782 Old 12-31-2008, 02:51 AM
AVS Special Member
 
JeffY's Avatar
 
Join Date: Mar 2001
Location: England
Posts: 3,667
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Rob Dingen View Post

Hi

I spent 2 days with the RS20 and tweaked the cms.
The calibrations where done with color control on -20 -10 and 0 and on 100% and 75%.
If you lower the color control the linearity gets worse from 0% 25% 50% 75% 100% especially on the 50% and 75% so leave the color control to 0.
Then if you calibrate to get it right on 75% patterns the gamut is to large en red and green are over saturated.
So in a shell CMS calibration on 100% and color control on 0 works best for me.

Rob

Very interesting Rob, I'll have to give this a try to see if I can get a better result.
JeffY is offline  
post #457 of 1782 Old 12-31-2008, 04:50 AM
Senior Member
 
Rob Dingen's Avatar
 
Join Date: Aug 2000
Location: Netherlands
Posts: 460
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi

Ok here are my measurements maybe they could be better with more time but this was a projector that I calibrate for a customer so it has to go back.
I made some compromises between delta xy and delta luma because both errors are visible.
Now it looks a little more saturated as the THX setting but by a Small margin.
So with a better gray scale tracking and acceptable gamut I prefer this over the THX setting.

Red Green Blue Yellow Cyan Magenta White
0.640 0.300 0.149 0.418 0.219 0.318 0.310 x
0.330 0.611 0.060 0.508 0.331 0.152 0.327 y
10.243 34.234 3.525 39.567 38.040 13.874 48.391 Y
0.2 3.0 0.7 6.8 5.0 3.2 delta E
0.000 0.011 0.001 0.003 0.006 0.004 delta xy
-0.5 % -1.1 % +0.9 % -11.9 % -0.2 % +0.7 % delta luma

Col Hue Sat Bright
R 3 -16 11
Y 3 -19 30
G 2 -7 30
C 1 -4 30
B 7 -13 6
M 1 -10 17

Rob
Rob Dingen is offline  
post #458 of 1782 Old 12-31-2008, 04:57 AM
AVS Special Member
 
Lawguy's Avatar
 
Join Date: Sep 2004
Posts: 5,705
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Rob Dingen View Post

Hi

Ok here are my measurments maybe they could be better with more time but this was a projector that I calibrate for a customer so it has to go back.
I made some compomises between delta xy and delta luma becouse both errors are visable.
Now it looks a little more saturated as the THX setting but by a smal margin.
So with a better grayscale tracking and acceptable gamut I prefer this over the THX setting.

Red Green Blue Yellow Cyan Magenta White
0.640 0.300 0.149 0.418 0.219 0.318 0.310 x
0.330 0.611 0.060 0.508 0.331 0.152 0.327 y
10.243 34.234 3.525 39.567 38.040 13.874 48.391 Y
0.2 3.0 0.7 6.8 5.0 3.2 delta E
0.000 0.011 0.001 0.003 0.006 0.004 delta xy
-0.5 % -1.1 % +0.9 % -11.9 % -0.2 % +0.7 % delta luma

Col Hue Sat Bright
R 3 -16 11
Y 3 -19 30
G 2 -7 30
C 1 -4 30
B 7 -13 6
M 1 -10 17

Rob

What did your 75% numbers look like?

Thanks.

Affable Nitwit
Lawguy is offline  
post #459 of 1782 Old 12-31-2008, 04:58 AM
Advanced Member
 
tryingtimes's Avatar
 
Join Date: Oct 2001
Location: UK
Posts: 991
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Rob Dingen View Post

Hi

Ok here are my measurements maybe they could be better with more time but this was a projector that I calibrate for a customer so it has to go back.
I made some compromises between delta xy and delta luma because both errors are visible.
Now it looks a little more saturated as the THX setting but by a Small margin.
So with a better gray scale tracking and acceptable gamut I prefer this over the THX setting.

Code:
Red             Green           Blue            Yellow          Cyan            Magenta         White
0.640           0.300           0.149           0.418           0.219           0.318           0.310           x
0.330           0.611           0.060           0.508           0.331           0.152           0.327           y
10.243          34.234          3.525           39.567          38.040          13.874          48.391          Y
0.2             3.0             0.7             6.8             5.0             3.2                             delta E
0.000           0.011           0.001           0.003           0.006           0.004                           delta xy
-0.5 %          -1.1 %          +0.9 %          -11.9 %         -0.2 %          +0.7 %                          delta luma

Col   Hue    Sat    Bright
R     3      -16    11
Y     3      -19    30
G     2      -7     30
C     1      -4     30
B     7      -13    6
M     1      -10    17
Rob

I've put the table into code, so that it displays correctly (although it still looks a bit broken)
It's a bit easier to read though.
tryingtimes is offline  
post #460 of 1782 Old 12-31-2008, 06:07 AM
Senior Member
 
Rob Dingen's Avatar
 
Join Date: Aug 2000
Location: Netherlands
Posts: 460
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi

Here are the 75% settings and normal 100% measurements:

red green blue yellow cyan magenta white
x 0.669 0.297 0.150 0.436 0.214 0.322 0.310
y 0.326 0.638 0.062 0.534 0.321 0.151 0.330
Y 10.880 35.311 3.756 48.144 36.602 10.408 51.980
dE 19.4 9.5 1.4 12.7 7.7 17.7
dxy 0.029 0.038 0.002 0.033 0.013 0.004
dluma -1.6 % -5.0 % +0.1 % -0.2 % -10.6 % -29.7 %

The settings:
Col Hue Sat Bright
R 2 -5 5
Y 0 0 23
G 3 -8 21
C 4 -6 25
B 7 -13 6
M 1 -30 -3

Don't know how to get it nice onscreen guys hope tryingtimes does it again.

Rob
Rob Dingen is offline  
post #461 of 1782 Old 12-31-2008, 06:16 AM
Advanced Member
 
tryingtimes's Avatar
 
Join Date: Oct 2001
Location: UK
Posts: 991
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by rob dingen View Post

hi

here are the 75% settings and normal 100% measurements:

Code:
          red       green      blue      yellow     cyan      magenta    white
x         0.669     0.297      0.150     0.436      0.214     0.322      0.310
y         0.326     0.638      0.062     0.534      0.321     0.151      0.330
y        10.880    35.311      3.756     48.144    36.602    10.408     51.980
de       19.4       9.5        1.4       12.7       7.7      17.7       
dxy       0.029     0.038      0.002     0.033      0.013     0.004     
dluma    -1.6 %    -5.0 %     +0.1 %    -0.2 %    -10.6 %   -29.7 %

the settings:
Col  hue    sat   bright
r    2      -5      5
y    0       0      23
g    3      -8      21
c    4      -6      25
b    7      -13     6
m    1      -30    -3
don't know how to get it nice onscreen guys hope tryingtimes does it again.

Rob

Here you go!
Edit: The way to do it is make it look nice in notepad, then when you paste it into the forums put {code} {/code} around it (replace curly brackets with square ones).

Hope that helps
tt
tryingtimes is offline  
post #462 of 1782 Old 12-31-2008, 08:03 AM
AVS Special Member
 
mandarax's Avatar
 
Join Date: Mar 2001
Location: Ontario, CA
Posts: 2,893
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
My eyeballs thank you Tryin ... and thanks for the info Rob and Mark. Looking at those numbers on an iphone while waiting for a firmware update on a bluray player is tough. Rob what were you calibrating with?

I reported before that Sencore was going to show their new software calibration at CES. I guess it isnt for prime time yet and they won't be showing it afterall.

Can you guys in Europe get the hubble which is equivilent to the OTC 1000. Don't seem to be able to get just the hardware here in North America?

Cheers and Happy New Year guys.
mandarax is offline  
post #463 of 1782 Old 12-31-2008, 09:13 AM
AVS Special Member
 
JeffY's Avatar
 
Join Date: Mar 2001
Location: England
Posts: 3,667
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by JeffY View Post

Very interesting Rob, I'll have to give this a try to see if I can get a better result.


I gave it a go but I can't say the results where any better. I always check with ramps to see nothing screwey is going on, I see no point in getting one bit accurate if everything else is messed up.
JeffY is offline  
post #464 of 1782 Old 12-31-2008, 01:56 PM - Thread Starter
AVS Club Gold
 
Mark Petersen's Avatar
 
Join Date: Dec 2000
Posts: 4,671
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Tom, Ric or others,

What is the difference between delta E and delta C measures for color differences? I'm using progressivelabs software which only uses the delta C convention and the documentation suggests that delta C values of less than 3 should not be perceptible. The software is also capable of using xyY and Luv conventions (and others), I'm sticking with xyY if only because that's the convention that others seem to be following which helps in comparing results. But is xyY truly better than other conventions like Luv?
Mark Petersen is offline  
post #465 of 1782 Old 12-31-2008, 02:01 PM
umr
AVS Club Gold
 
umr's Avatar
 
Join Date: Oct 2002
Location: USA
Posts: 10,134
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 29
Quote:
Originally Posted by Mark Petersen View Post

Tom, Ric or others,

What is the difference between delta E and delta C measures for color differences? I'm using progressivelabs software which only uses the delta C convention and the documentation suggests that delta C values of less than 3 should not be perceptible. The software is also capable of using xyY and Luv conventions (and others), I'm sticking with xyY if only because that's the convention that others seem to be following which helps in comparing results. But is xyY truly better than other conventions like Luv?

For our use it does not really matter much which measurement system is used. You just need to know the target values for the system you are using.

I never found the progressive labs delta C values to be very useful. They are never large enough to me for real errors. I am not sure how Cliff is calculating it. Delta E can be calculated many different ways.
umr is offline  
post #466 of 1782 Old 12-31-2008, 02:08 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,818
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 23
Quote:
Originally Posted by Rob Dingen View Post

Hi

Ok here are my measurements maybe they could be better with more time but this was a projector that I calibrate for a customer so it has to go back.
I made some compromises between delta xy and delta luma because both errors are visible.
Now it looks a little more saturated as the THX setting but by a Small margin.
So with a better gray scale tracking and acceptable gamut I prefer this over the THX setting.

Red Green Blue Yellow Cyan Magenta White
0.640 0.300 0.149 0.418 0.219 0.318 0.310 x
0.330 0.611 0.060 0.508 0.331 0.152 0.327 y
10.243 34.234 3.525 39.567 38.040 13.874 48.391 Y
0.2 3.0 0.7 6.8 5.0 3.2 delta E
0.000 0.011 0.001 0.003 0.006 0.004 delta xy
-0.5 % -1.1 % +0.9 % -11.9 % -0.2 % +0.7 % delta luma

Col Hue Sat Bright
R 3 -16 11
Y 3 -19 30
G 2 -7 30
C 1 -4 30
B 7 -13 6
M 1 -10 17

Rob

Attached is what the gamut for the above posted numbers would look like. Note that I did not have a white point to go by so it is estimated.

I was confused about your dE. Is this CIELUV or CIE94? I could not match up your dE based on the values. Thanks for the tips and posting this!
LL
lovingdvd is offline  
post #467 of 1782 Old 12-31-2008, 02:15 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 8,818
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 23
Quote:
Originally Posted by Rob Dingen View Post

Hi

Here are the 75% settings and normal 100% measurements:

red green blue yellow cyan magenta white
x 0.669 0.297 0.150 0.436 0.214 0.322 0.310
y 0.326 0.638 0.062 0.534 0.321 0.151 0.330
Y 10.880 35.311 3.756 48.144 36.602 10.408 51.980
dE 19.4 9.5 1.4 12.7 7.7 17.7
dxy 0.029 0.038 0.002 0.033 0.013 0.004
dluma -1.6 % -5.0 % +0.1 % -0.2 % -10.6 % -29.7 %

The settings:
Col Hue Sat Bright
R 2 -5 5
Y 0 0 23
G 3 -8 21
C 4 -6 25
B 7 -13 6
M 1 -30 -3

Don't know how to get it nice onscreen guys hope tryingtimes does it again.

Rob

Rob - do I understand correctly that this is your calibration using 75% stimulus patterns, whereas the #s posted just before this were for using 100% stimulus patterns? And in both cases it was with 100% saturation patterns?

Also do I understand correctly that you are saying that after much experimentation you think it is better to use 100% stimulus patterns and calibrate for that, rather than using 75%, before the former provides for an overall more balanced tracking at various stimulus levels?

And also do I understand that in both sets of measurements, you had the color control at 0? I'm not sure how that would be possible, as I and many others found that one cannot get green lowered enough simply by using the CMS and therefore required the color control to be lowered. Did you find otherwise?

For instance in my case I measured with 75% stimulus patterns and with Green saturation in the CMS reduced to the max of -30, green only moved down toward the white point a tad (barely moved). It wasn't until I moved the color control down quite a bit that green starting to really respond to its CMS saturation control. You found otherwise?
lovingdvd is offline  
post #468 of 1782 Old 12-31-2008, 02:44 PM
AVS Special Member
 
Frank Derks's Avatar
 
Join Date: Jul 2002
Location: Region A,B,C
Posts: 1,890
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 97 Post(s)
Liked: 88
Perhaps this lack of green control is because one or more colors are maxed out in the RGB gain settings together with the contrast control?

A solution might be to adjust grayscale and contrast for a slightly lower light output to give more headroom for the color adjustment controls.
Frank Derks is offline  
post #469 of 1782 Old 12-31-2008, 02:54 PM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by umr View Post

For our use it does not really matter much which measurement system is used. You just need to know the target values for the system you are using.

I never found the progressive labs delta C values to be very useful. They are never large enough to me for real errors. I am not sure how Cliff is calculating it. Delta E can be calculated many different ways.

Delta E is really for Print, not video......

Here is some of the info on calculations
http://www.gamutvision.com/docs/gamu...equations.html

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #470 of 1782 Old 12-31-2008, 02:57 PM
umr
AVS Club Gold
 
umr's Avatar
 
Join Date: Oct 2002
Location: USA
Posts: 10,134
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 29
Quote:
Originally Posted by GlenC View Post

Delta E is really for Print, not video......

Here is some of the info on calculations
http://www.gamutvision.com/docs/gamu...equations.html

Most people here use delta E* which drops the luminance like the delta C. I am familiar with these since I have programmed them all in the past. Just did not remember delta C off the top of my head since I found it horrible. I had no reason to remember it.

As I said there are many ways to calculate delta E.
umr is offline  
post #471 of 1782 Old 12-31-2008, 03:01 PM
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,360
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 204
Quote:
Originally Posted by Mark Petersen View Post

What is the difference between delta E and delta C measures for color differences? I'm using progressivelabs software which only uses the delta C convention and the documentation suggests that delta C values of less than 3 should not be perceptible. The software is also capable of using xyY and Luv conventions (and others), I'm sticking with xyY if only because that's the convention that others seem to be following which helps in comparing results. But is xyY truly better than other conventions like Luv?

Mark, this is a complicated question.

First, all of the major dE calculations allow you to break out LCH (Lightness, Chroma, Hue) components. This is what I have been using to provide % error numbers, except I use LSH. The saturation value can be derived from L and C.

Second, if you look at errors in the white point, you will see that
1) dL is ignored (except in a perverse option in CalMan that Bill Blackwell and I have already argued about).
2) The dH value is negligible.

This leaves dC, which forms the bulk of the dE value for white. Thus, at least for gray scale measurments, this is a perfectly reasonable option. It would NOT be proper to use this for pri/sec color grading.

Third, ALL dE calculations should use either Luv or Lab values. xyY is simply not sufficiently percpetually uniform to provide accurate dE results. The 1976 dE formulas allow Luv or Lab values. All subsequent dE formulas use Lab exclusively.

Tom Huffman
ChromaPure Software/AccuPel Video Signal Generators
ISF/THX Calibrations
Springfield, MO

TomHuffman is offline  
post #472 of 1782 Old 12-31-2008, 03:02 PM
AVS Special Member
 
JeffY's Avatar
 
Join Date: Mar 2001
Location: England
Posts: 3,667
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Do people still check with colour filters or does that go out of the window when using a probe?
JeffY is offline  
post #473 of 1782 Old 12-31-2008, 03:10 PM
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,360
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 204
Quote:
Originally Posted by GlenC View Post

Delta E is really for Print, not video......

No. dE is a device independent measurement of color error. Some industries have preferred one measurement over another, but this indicates no technical limitations to using it elsewhere.

For example, the video industry has traditionally used Luv-based calculations because Luv provides a linear chromaticity diagram that Lab does not. This has led some to incorrectly assume that Lab-based calculations are somehow inaccurate for video.

Finally, as I mentioned to Mark, you MUST use dE for pri/sec. color grading. dC would be a grossly inadequate tool in this context, though it is perfectly fine for gray scale applications, which is how Progressive Labs uses it.

Tom Huffman
ChromaPure Software/AccuPel Video Signal Generators
ISF/THX Calibrations
Springfield, MO

TomHuffman is offline  
post #474 of 1782 Old 12-31-2008, 05:14 PM - Thread Starter
AVS Club Gold
 
Mark Petersen's Avatar
 
Join Date: Dec 2000
Posts: 4,671
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by umr View Post

For our use it does not really matter much which measurement system is used. You just need to know the target values for the system you are using.

I never found the progressive labs delta C values to be very useful. They are never large enough to me for real errors. I am not sure how Cliff is calculating it. Delta E can be calculated many different ways.

Thanks Jeff. Yes, it would be nice if the delta C calculation was specified in the doc. It would also be interesting if the other delta calculations were done and displayed so a person could get a feel for the magnitude of each number in different scenarios (too much or too luminance, or saturation or hue, etc.).

Quote:
Originally Posted by Frank Derks View Post

Perhaps this lack of green control is because one or more colors are maxed out in the RGB gain settings together with the contrast control?

A solution might be to adjust grayscale and contrast for a slightly lower light output to give more headroom for the color adjustment controls.

You read my mind The gain setting is the only other adjustment that I can think of that hasn't yet been tried so I was going to try this as an experiment tonight. I'll let you know how it pans out.

Quote:
Originally Posted by GlenC View Post

Here is some of the info on calculations
http://www.gamutvision.com/docs/gamu...equations.html

Thanks for the link. Lots of good stuff in there.

Quote:
Originally Posted by TomHuffman View Post

Mark, this is a complicated question.

First, all of the major dE calculations allow you to break out LCH (Lightness, Chroma, Hue) components. This is what I have been using to provide % error numbers, except I use LSH. The saturation value can be derived from L and C.

Second, if you look at errors in the white point, you will see that
1) dL is ignored (except in a perverse option in CalMan that Bill Blackwell and I have already argued about).
2) The dH value is negligible.

This leaves dC, which forms the bulk of the dE value for white. Thus, at least for gray scale measurments, this is a perfectly reasonable option. It would NOT be proper to use this for pri/sec color grading.

Third, ALL dE calculations should use either Luv or Lab values. xyY is simply not sufficiently percpetually uniform to provide accurate dE results. The 1976 dE formulas allow Luv or Lab values. All subsequent dE formulas use Lab exclusively.

Thanks for the info Tom. So you're saying that not only should I be using some form of dE for primary/secondary measurements but that I should also switch to Luv or Lab conventions then too correct?
Mark Petersen is offline  
post #475 of 1782 Old 12-31-2008, 05:16 PM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by TomHuffman View Post

No. dE is a device independent measurement of color error. Some industries have preferred one measurement over another, but this indicates no technical limitations to using it elsewhere.

For example, the video industry has traditionally used Luv-based calculations because Luv provides a linear chromaticity diagram that Lab does not. This has led some to incorrectly assume that Lab-based calculations are somehow inaccurate for video.

Finally, as I mentioned to Mark, you MUST use dE for pri/sec. color grading. dC would be a grossly inadequate tool in this context, though it is perfectly fine for gray scale applications, which is how Progressive Labs uses it.

Correction - ΔE is a measurement of the combined color and luminance error.

Only when Luminance is factored into the equation is true ΔE. IIRC, no calibration program brings luminance into the equation. I agree that ΔE could be used, however you need a reference luminance value to start with. In print, it is simple because you are measuring the color under a reference light source. This could get extremely complex to do with any level of accuracy due to the "reference" luminance changes with nearly every change in RGB levels. I agree that it could be useful as long as you can get good adjustments on grayscale and gamma and color space and color gamma. For the most part we need to focus on the x,y values, especially grayscale and hope gamma can be close. We are more sensitive to wrong colors than luminance to a degree.

ΔE*xy = ( (Y*2-Y*1)2 + (x*2-x*1)2 + (y*2-y*1)2 )1/2 = (Δy*2 + Δx*2 + Δy*2 )1/2

ΔC*xy = ((x*2-x*1)2 + (y*2-y*1)2 )1/2 = (Δx*2 + Δy*2 )1/2

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #476 of 1782 Old 12-31-2008, 07:12 PM
AVS Club Gold
 
JimmyR's Avatar
 
Join Date: May 2002
Location: Runco Island CA.
Posts: 4,605
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:


Mark Petersen:You read my mind The gain setting is the only other adjustment that I can think of that hasn't yet been tried so I was going to try this as an experiment tonight. I'll let you know how it pans out.

Mark, have you tried calibrating using your DVDO VP50 non-pro "Saturation" control instead of the main RS20 Color ?
I haven't been able to do any correction because my replacement has not landed (non working H/V shift).
JimmyR is offline  
post #477 of 1782 Old 12-31-2008, 08:04 PM
Advanced Member
 
RickS's Avatar
 
Join Date: Aug 1999
Location: AB,Canada
Posts: 789
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 20
Quote:


You read my mind The gain setting is the only other adjustment that I can think of that hasn't yet been tried so I was going to try this as an experiment tonight. I'll let you know how it pans out.

Mark,

When I had my 750 calibrated the gain settings were used to get the best picture possible.

Rick
RickS is offline  
post #478 of 1782 Old 12-31-2008, 08:19 PM
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,360
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 204
Quote:
Originally Posted by GlenC View Post

Correction - ΔE is a measurement of the combined color and luminance error.

Only when Luminance is factored into the equation is true ΔE. IIRC, no calibration program brings luminance into the equation.

ColorFacts and Progressive Labs do not because neither uses dE for pri/sec. color grading. However, CalMan does and HCFR has a Preferences toggle that allows you to turn it on and off. Greg Rogers' Display Calibration Calculator uses L for pri/sec. grading and does NOT use it for grayscale, which is the correct implementation.

Tom Huffman
ChromaPure Software/AccuPel Video Signal Generators
ISF/THX Calibrations
Springfield, MO

TomHuffman is offline  
post #479 of 1782 Old 12-31-2008, 08:22 PM
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,360
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 204
Quote:
Originally Posted by Mark Petersen View Post

Thanks for the info Tom. So you're saying that not only should I be using some form of dE for primary/secondary measurements but that I should also switch to Luv or Lab conventions then too correct?

Yes. The 1976 and all subsequent dE formulas use Luv or Lab data, so xyY must first be converted into Lab or Luv before the dE calculation can occur.

Tom Huffman
ChromaPure Software/AccuPel Video Signal Generators
ISF/THX Calibrations
Springfield, MO

TomHuffman is offline  
post #480 of 1782 Old 12-31-2008, 08:31 PM
AVS Addicted Member
 
millerwill's Avatar
 
Join Date: Apr 2004
Location: Berkeley, CA
Posts: 11,358
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 27
Question re the 'Color Space' setting, under 'HDMI' in the 'Input Signal' part of the Menu: do people find that 'Auto' is appropriate? I recall that several have said that 'Auto' does NOT work correctly for the 'Input' selection under 'HDMI'.

If not, what is recommended: RGB, YCbCr(4:4:4), or YCbCr(4:2:2)?
millerwill is offline  
Reply Digital Hi-End Projectors - $3,000+ USD MSRP

User Tag List

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