Basic Guide to Color Calibration using a CMS (updated and enhanced) - Page 25 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #721 of 1936 Old 03-13-2008, 06:26 PM
Advanced Member
 
thomasl's Avatar
 
Join Date: Apr 2004
Posts: 592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Eric,

what variation are you seeing?

I've been using a Eye-One Display sensor and HCFR to calibrate both some older CRTs and a new LCD tv. I've noticed that if I do successive measurements that I can get up to .003 variance in x,y measurements for RGBCYM - typically though it is .002 and I've noticed that it is mainly red and green that show the most variance. Blue rarely varies but if it does, it is only by .001. What percent brightness patterns are you using - 100 or 75 and do you see the same level of variance with both sets of patterns?

In general I would agree that once you get down into the extremely fine tweaking to get dE as low as possible that you run up against variances in all the parts involved. One thing to remember is that you can see the full 6 digit precision value for x,y within HCFR - this will show exactly how much difference there is. For example, .061 and .064 may actually be .061499 and .063500 - thus instead of .003 difference, it is ~.002.

cheers,


--tom
thomasl is offline  
Sponsored Links
Advertisement
 
post #722 of 1936 Old 03-13-2008, 07:43 PM
AVS Special Member
 
Eric Carroll's Avatar
 
Join Date: Sep 2007
Location: Toronto, Canada
Posts: 1,069
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I am using 100 IRE windows and getting (from memory) +- .002 on x,y easily when I am not touching anything. Luminance as I recall is even more. I will have to try to characterize it a bit more next time around.

How do you see full six digits in HCFR? Does it mean anything?

Would CalMan be better at this? I started with HCFR just to get a feel on the process but I am not adverse to upgrading.

Is the sensor really got six decimal places of precision? 3 more digits if its +- .002 is not really much help.

Interestingly, I found the i1 Display 2 spec sheet. It states a "repeatability" (which I take to me precision) of x,y of +- .001 and accuracy of +- .004. That's a lot in terms of dE.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
Eric Carroll is offline  
post #723 of 1936 Old 03-13-2008, 09:06 PM
Advanced Member
 
thomasl's Avatar
 
Join Date: Apr 2004
Posts: 592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Eric, I found if you click on Editable Data in the Measures tab, you can then click on the x,y values and it'll expand to the 6 digit reading. Also, I believe you can export them to an Excel spreadsheet.

I have never used CalMan - only HCFR. I suspect that if you want more precise repeatable readings you'd need to upgrade to the i1Pro but it seems to comes with a big jump in price.

I did some more testing this evening and I was getting repeatable variation of .002 for x,y readings on the colors with red being the one that seems to vary the most. Blue, green and the secondaries not so much. I have no idea if that means anything at all. I haven't played around with the affect on the various dE calculations yet but can take a look in the morning.

Btw, what dE are you using? I assume that you're using Tom's CIELAB dE1976 and dE 1994 spreadsheet calculations?

cheers,


--tom
thomasl is offline  
post #724 of 1936 Old 03-13-2008, 09:43 PM
AVS Special Member
 
Eric Carroll's Avatar
 
Join Date: Sep 2007
Location: Toronto, Canada
Posts: 1,069
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
HCFR gives you dE. I am also using a spreadsheet I found written by Gary Fritz and Shawn Rader with my modifications so I can enter luminance directly, have it normalize and compare to targets.

Thanks for the HCFR tip.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
Eric Carroll is offline  
post #725 of 1936 Old 03-14-2008, 04:12 AM
Advanced Member
 
thomasl's Avatar
 
Join Date: Apr 2004
Posts: 592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Eric, the RGBYCM dE (< 2) values that Tom suggests are for CIELAB 1994 dE color error calculations. This formula is different than the one HCFR uses. HCFR is using dELuv. The Luv values tend to be higher than the ones produced by dE1994 when looking at color error. Here's an example of the type of dE numbers returned by the various formulas:

Luv 1976 1994
yellow 6.34 3.98 1.79

I think if you're shooting for Luv < 1 that that will be within the variance of the sensor. Although I haven't found an official target for dELuv, I'd shoot for a Luv/HCFR value < 10 and a dE1994 value of < 2 as Tom suggests and then I'd half those numbers if you're really looking to fine tune things.

If you search this thread, you'll find some spreadsheets that Tom posted that will calculate dE1976 and dE1994 values.

hope this helps,


--tom
thomasl is offline  
post #726 of 1936 Old 03-14-2008, 08:06 AM
AVS Special Member
 
Eric Carroll's Avatar
 
Join Date: Sep 2007
Location: Toronto, Canada
Posts: 1,069
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
yes, it helps alot. I suspected it was a beginner's error. Thanks very much.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
Eric Carroll is offline  
post #727 of 1936 Old 03-14-2008, 09:27 AM
Advanced Member
 
JoeProcopio's Avatar
 
Join Date: Dec 2006
Location: New Jersey
Posts: 655
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Tom...first I want to thank you for this great introduction to calibrating. I have a Toshiba 42HL167 and bought an i1 Display 2...following your post, and playing with the HCFR software, I was able to bring this beast into a pretty good place. I realize I won't get this set perfect, but it is close to some published results in magazines for the HL167 series.

I do have 2 questions for you...
1) if you set the TV up for HD 709 colors, how does this affect SD colors...meaning..setup for HD, watching SD broadcasts
2) you mentioned in another post about setting the Blue up with the color control, and then Cyan with tint..but in the first post, you have Red being setup...why 2 different references


again, thanks so much...I have been able to help other HL167 owners get their set better, and a lot of that is because of your descriptions in the first post.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
JoeProcopio is offline  
post #728 of 1936 Old 03-14-2008, 11:06 AM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,418
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 216
I have added a section to the tutorial on calculating dE and included a spreadsheet that allows users to calculate dE from the SMPTE-C or Rec. 709 standards against your own test data. The spreadsheet includes the CIELUV, CIELAB, and CIE94 calculations.

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

TomHuffman is offline  
post #729 of 1936 Old 03-14-2008, 11:10 AM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,418
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 216
Quote:
Originally Posted by JoeProcopio View Post

1) if you set the TV up for HD 709 colors, how does this affect SD colors...meaning..setup for HD, watching SD broadcasts
2) you mentioned in another post about setting the Blue up with the color control, and then Cyan with tint..but in the first post, you have Red being setup...why 2 different references.

SD programming will be slightly oversaturated and a little purplish when viewed with a set calibrated to the Rec. 709, but it is a subtle difference.

It depends upon whether you have color decoding controls. If you do, then adjust Color against blue, then use the color decoding controls to adjust red and green. If you don't have color decoding controls (most TVs don't), I prefer to adjust the Color control against red, because the eye is more sensitive to errors in red.

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

TomHuffman is offline  
post #730 of 1936 Old 03-14-2008, 12:56 PM
Advanced Member
 
thomasl's Avatar
 
Join Date: Apr 2004
Posts: 592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Tom, what is your opinion (other than get a better TV ) on sets with no dedicated color decoder controls yet show the classic over-brightness of red and under-brightness of blue. I have an older (7 years old) 27 inch Toshiba CRT that calibrated nicely to D65 as far as the grayscale goes but after tweaking color and tint, I ended up with red at +35-40 percent and blue at -20-25 percent - these numbers were derived from the Accupel Color Gamut calculator's Y numbers given my primary color points for the set. Green is < -5 percent. I suspect many folks suffer from this. I essentially set the color control using green as the middle of the seesaw knowing that going in either direction would make either red better at the expense of green/blue or blue better at the expense of green/red.. In these situations, is it typical to get red down under +20 even knowing that it will end up putting blue at -40-50 or just balance it as I did and tweak it if red bothers one during actual viewing.

thanks,


--tom
thomasl is offline  
post #731 of 1936 Old 03-14-2008, 01:28 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,418
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 216
Quote:
Originally Posted by thomasl View Post

Tom, what is your opinion (other than get a better TV) on sets with no dedicated color decoder controls yet show the classic over-brightness of red and under-brightness of blue.

Adjust it to personal taste or to what gives the lowest average error. The only other solution short of getting a better display is to add an external processor that adds color decoding controls.

The ability of consumers to measure display performance has dramatically outstripped the manufacturers' willingness to market accurate displays at an attractive price.

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

TomHuffman is offline  
post #732 of 1936 Old 03-14-2008, 03:02 PM
Advanced Member
 
JoeProcopio's Avatar
 
Join Date: Dec 2006
Location: New Jersey
Posts: 655
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by TomHuffman View Post

SD programming will be slightly oversaturated and a little purplish when viewed with a set calibrated to the Rec. 709, but it is a subtle difference.

It depends upon whether you have color decoding controls. If you do, then adjust Color against blue, then use the color decoding controls to adjust red and green. If you don't have color decoding controls (most TVs don't), I prefer to adjust the Color control against red, because the eye is more sensitive to errors in red.

ahhh...so i will go back and set blue...i have a CMS in my Tosh HL167. is blue the same percentage of white as red is? and that would be what I am seeing on SD programming, the slight oversaturation and purplish look, that explains it!!


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
JoeProcopio is offline  
post #733 of 1936 Old 03-15-2008, 08:12 AM
Advanced Member
 
thomasl's Avatar
 
Join Date: Apr 2004
Posts: 592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks Tom.

I agree with manufacturer's willingness to market accurate displays. As best I can tell, the correlation between price and accuracy isn't always true either. A $400 CRT tv that I own has better color performance than the 3.5x more expensive new Samsung LCD that I also own. For me, I don't expect accuracy out of the box with consumer electronics but I'd like the ability to configure things. The more configurable the better since it allows everyone to get the maximum out of their purchase. I hope in the future that standard RGBYCM color management becomes common place with displays.

I've noticed another interesting behavior on the Samsung LNT4061F. I did all of my calibrating using 75 percent window color patterns - I then went in to see what it looked like with 100 percent window patterns. I wasn't shocked to see some differences in color decoding performance but nothing that changed the actual color control setting - but what was interesting is that the color points moved themselves. This was repeatable with numerous measurements done in sequence and didn't seem to be sensor variance since it was a bit more of a change than that. Green and red both moved ~+.004 on the x axis. Green is somewhat undersaturated on this set and a bit leaning toward blue with the 75 percent patterns but with the 100 percent patterns, it's still undersaturated but it loses it's leaning toward blue. Has anyone seen this behavior with color points moving depending on the brightness of the window color pattern being measured? What this affects, subtly, is the position of the secondaries with tint/hue control moving about two notches to get the optimal setting.

cheers,


--tom
thomasl is offline  
post #734 of 1936 Old 03-16-2008, 02:55 AM
Member
 
d6500's Avatar
 
Join Date: Mar 2008
Posts: 16
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi tom,

Would you say that the 21%/100% for color luminance would be accurate for all display types?.

I used this method for setting color luminance on a pioneer panel, for which the blue filter method indeed yields inaccurate results(color way too high).

This way, the color seems to be a bit on the high side also....

Compliments on the topic

Dimitri

ISF certified
Imaging Science Solutions

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
d6500 is offline  
post #735 of 1936 Old 03-16-2008, 07:44 AM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,418
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 216
Yes, the display type doesn't matter. The only caveat is that the measured primaries may make a difference. The 21%, 71%, 8% targets are only approximations. To get the absolutely precise figure, you have to calculate the targets from the measured primaries. I posted a spreadsheet some time ago for this.

http://www.avsforum.com/avs-vb/showt...0#post13172540

Greg Rogers has also posted a small application that does the same thing and adds secondary color and dE calculations.

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

TomHuffman is offline  
post #736 of 1936 Old 03-18-2008, 03:32 PM
Member
 
d6500's Avatar
 
Join Date: Mar 2008
Posts: 16
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi tom,

I've used the excell file.

Does it look strange to you?..

Red is at 29%, if I'm not mistaken.

Dimitri

 

pioneer ymatrix..txt 0.2919921875k . file

ISF certified
Imaging Science Solutions

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
d6500 is offline  
post #737 of 1936 Old 03-18-2008, 04:18 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,418
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 216
It seems high, but what really looks strange are the green xy coordinates. That's about the most inaccurate green I've ever seen and no doubt, along with an undersaturated red, the explanation for the high red lightness target.

Quote:
Originally Posted by d6500 View Post

Hi tom,

I've used the excell file.

Does it look strange to you?..

Red is at 29%, if I'm not mistaken.

Dimitri


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

TomHuffman is offline  
post #738 of 1936 Old 03-18-2008, 04:25 PM
Member
 
d6500's Avatar
 
Join Date: Mar 2008
Posts: 16
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Tom,

It's a gen6 pioneer plasma.

Which has pretty good red and blue primaries, but awfull green. The problem is that if I adjust the primaries, the grayscale is upset, so I only adjust secondaries for color accuracy.

Green is way oversaturated by default.

I guess this means the matrix calculation is not accurate because of this?...

Dimitri

ISF certified
Imaging Science Solutions

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
d6500 is offline  
post #739 of 1936 Old 03-18-2008, 05:04 PM
AVS Special Member
 
rickardl's Avatar
 
Join Date: Nov 2003
Location: Stockholm, Sweden
Posts: 1,963
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 43
Quote:
Originally Posted by d6500 View Post

Tom,

It's a gen6 pioneer plasma.

Which has pretty good red and blue primaries, but awfull green. The problem is that if I adjust the primaries, the grayscale is upset, so I only adjust secondaries for color accuracy.

Green is way oversaturated by default.

I guess this means the matrix calculation is not accurate because of this?...

Dimitri

My green is also off but not so much but I still end up with about 26% using
the spreadsheet. I have a Sony LCD and using EyeOne Display 2.
rickardl is offline  
post #740 of 1936 Old 03-18-2008, 09:53 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,418
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 216
Quote:
Originally Posted by d6500 View Post

Green is way oversaturated by default.

I guess this means the matrix calculation is not accurate because of this?...

Not at all. The whole point of the matrix calculation is to indicate the proper lightness targets relative to the measured primary coordinates. It is just that the further those coordinates are from their intended points, the further the lightness targets may be from theirs.

It seems to me that deviations in hue have a bigger effect on the lightness targets than over or under saturation. Your green primary is not only oversaturated. It is also shifted considerably towards cyan.

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

TomHuffman is offline  
post #741 of 1936 Old 03-24-2008, 11:20 AM
Member
 
ragingd's Avatar
 
Join Date: Sep 2003
Posts: 144
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
How exactly do you measure brightness with the meter? I know you have to measure 100% window and then get the 10% window to .65% of the white window. So if I measure 30 ftl @ 100% white then 10% should be .195 ftl is this correct? Thanks for any help.
ragingd is offline  
post #742 of 1936 Old 03-24-2008, 12:06 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,621
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 28 Post(s)
Liked: 167
Quote:
Originally Posted by ragingd View Post

How exactly do you measure brightness with the meter? I know you have to measure 100% window and then get the 10% window to .65% of the white window. So if I measure 30 ftl @ 100% white then 10% should be .195 ftl is this correct? Thanks for any help.

That's about right.

The way you calculate it is (I'm making the variable names up here because I can't remember the real ones off the top of my head)

W = max white luamance
B = black level lumanance
p = percent white you are calculating for (10%) expressed as decimal (.1)
G = desired Gamma
Y = luma output for your grey feild.

y = ((W-B) * p^g) + B
So for you
y = ((30-B) * (.1)^(2.2)) + B

The end results is something close to 1.9ftl + your black level.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is online now  
post #743 of 1936 Old 03-25-2008, 12:43 PM
Advanced Member
 
JoeProcopio's Avatar
 
Join Date: Dec 2006
Location: New Jersey
Posts: 655
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
did we ever get an answer to the 100% vs 75% colors question?

Quote:
Originally Posted by thomasl View Post

I've noticed another interesting behavior on the Samsung LNT4061F. I did all of my calibrating using 75 percent window color patterns - I then went in to see what it looked like with 100 percent window patterns... ...Has anyone seen this behavior with color points moving depending on the brightness of the window color pattern being measured?



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
JoeProcopio is offline  
post #744 of 1936 Old 03-25-2008, 07:02 PM
Advanced Member
 
thomasl's Avatar
 
Join Date: Apr 2004
Posts: 592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Joe,

This weekend I'll probably double check those results but the Eye-One Display was giving consistent readings where both green and red were shifted mostly on the x axis (about .007) between a 75 and 100 percent window pattern. I chalk it up as consumer display variance. Without a full set of test patterns that cover both saturation and brightness (i.e. 16 patterns that cover 25/50/75/100 percent brightness for each saturation level (25/50/75/100)), it's tough to get a complete idea of what the display is doing. I doubt though that the differing results would result in different user menu level settings. i.e. color/tint would be the same as I set them for the 75 percent brightness windows. My set does have not have a CMS.

cheers,


--tom
thomasl is offline  
post #745 of 1936 Old 03-25-2008, 07:46 PM
Advanced Member
 
JoeProcopio's Avatar
 
Join Date: Dec 2006
Location: New Jersey
Posts: 655
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
great let me know. I used 75% patterns to set my CMS, the Tosh HL167 has one, although somewhat limited in how affective it is...but i used 100% to set color level (not sure what i used for tint)


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
JoeProcopio is offline  
post #746 of 1936 Old 03-25-2008, 08:02 PM
Advanced Member
 
Gregg Loewen's Avatar
 
Join Date: Feb 2002
Location: New England
Posts: 771
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 39
hi guys

I just did one of the new Epson cinema 1080 UBs. Gawd this bb is nice...and the CMS works very well, with xyY adjustments for all primaries and secondaries.

President, Lion Audio Video Consultants Inc.
Lead THX Video Standards Instructor
Gregg Loewen is offline  
post #747 of 1936 Old 03-26-2008, 03:35 PM
Member
 
Orwellflash's Avatar
 
Join Date: Jun 2006
Location: New Mexico
Posts: 196
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Gregg, the Epson caught my attention when I saw it is one of the few proj. with a full CMS. Were you able to get all of the primaries and secondaries, grey scale properly calibrated? No limitations due to the bulb spectrum, etc?

That's a picture of Leadbelly's grave in Shiloh Baptist Church Cemetery Mooringsport Caddo Parish Louisiana (with Arlo Gutherie's guitar picks placed in memoriam). Arlo put them on his grave in 2005 and they were still there in 2012.
Orwellflash is offline  
post #748 of 1936 Old 03-27-2008, 04:28 PM
Member
 
jouko342's Avatar
 
Join Date: Jan 2008
Posts: 172
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
Quick q:

"Adjust the Brightness setting so that this test pattern measures as close as possible to .65% of the 100% white window."

What measure is he referring to? Is it Y or some other metric? So, if Y is 25, .65% of that is 0.1625, correct?
jouko342 is offline  
post #749 of 1936 Old 03-27-2008, 04:43 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,621
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 28 Post(s)
Liked: 167
Quote:
Originally Posted by jouko342 View Post

Quick q:

"Adjust the Brightness setting so that this test pattern measures as close as possible to .65% of the 100% white window."

What measure is he referring to? Is it Y or some other metric? So, if Y is 25, .65% of that is 0.1625, correct?

Yes that's almsot exactly right. Although it should techincally be (.1)^(desired gamma, 2.2) + black level. so if you black level mesaures .1 then your real target for a 2.2 gamma is (24.9 * (.1^2.2)) + .1 = .257


Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is online now  
post #750 of 1936 Old 03-27-2008, 05:05 PM
Member
 
jouko342's Avatar
 
Join Date: Jan 2008
Posts: 172
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
Quote:
Originally Posted by sotti View Post

Yes that's almsot exactly right. Although it should techincally be (.1)^(desired gamma, 2.2) + black level. so if you black level mesaures .1 then your real target for a 2.2 gamma is (24.9 * (.1^2.2)) + .1 = .257


Ah, thanks. But aren't you implying that I should set the white point based on black point (well, 10% level), instead of the other way around?

The guide measures white point first, then the black point based on that. I get 21.2Y for my 100% white point, but can't get even close to ~.14Y for 10% black the calcs call for - more like .5. But I imagine if I start from that .5, and set the white point based on your formula, it could be doable.

Is there a wrong way to do this, and does the order matter?
jouko342 is offline  
Reply Display Calibration

Tags
X Rite Eodis3 I1display Pro , X Rite Cmundis Colormunki Display , X Rite Eye One Display Lt , Calman 4 0 Calibration Software Home License , Aemc Ca813 Peak Function Lightmeter 200 000 Lux Range 0 01 Lux Resolution 3 Accuracy
Gear in this thread - Ca813 by PriceGrabber.com

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