HCFR - Open source projector and display calibration software - Page 132 - AVS Forum
Forum Jump: 
 84Likes
Reply
 
Thread Tools
post #3931 of 4913 Old 07-08-2014, 01:13 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
excellent CCSG numbers

2.2 power law
avg. dE2000 1.02, max 3.42, worst 10% 2.64

2.35 power law
avg. dE2000 0.98, max 2.48, worst 10% 1.88

I wouldn't bother with a 3dLUT with those numbers unless you just want to learn how to do it. It will pull those outliers in somewhat but you're not going to notice it in real material.
zoyd is online now  
Sponsored Links
Advertisement
 
post #3932 of 4913 Old 07-08-2014, 01:28 PM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
I would have to say I threw in some hd movie trailors and switched back and forth from 2.2 and 2.35 and liked the 2.35. It seemed to make the blacks darker.

Anyhow I finally learned this somewhat. I appreciate all the help and patience from all of you

I am first going to tackle my bluray player and ps4 first.

Then going to do my sharp70inch tv and maybe my 42 plasma.

After that I will for sure try a 3dlut. The more I learn the better off I feel. I mean now I can fly through and level out a greyscale and gamma.
Holiday121 is online now  
post #3933 of 4913 Old 07-08-2014, 01:39 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
If you want to try a BT.1886 type gamma calibration tick the "editable data" box on the gray scale page and enter 0.005 for the black reading. Also tick ITU-R BT.1886 on the gamma set-up page with 0 in the Relative/Absolute box. This will target a curve which has 2.2 gamma at 10% and 2.35 gamma above 40%.
zoyd is online now  
post #3934 of 4913 Old 07-08-2014, 01:50 PM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
Ahhh so it's like 2 gammas spread across the 0-100 mark?
Holiday121 is online now  
post #3935 of 4913 Old 07-08-2014, 02:00 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
sort-of, it depends on the absolute black and white levels as to what the function resembles. so in your case if you set black = 0.005 cd/m^2 and white is 54.6 cd/m^2 the curve will resemble a power law curve with gamma = 2.2 at 10% and a power law curve with gamma = 2.35 at 100%
zoyd is online now  
post #3936 of 4913 Old 07-08-2014, 02:03 PM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
Interesting I will
Give that a try maybe tommorow. I spent 8 hours today on it.

Also is there a way to calibrate like direct tv feed to my tv?
Holiday121 is online now  
post #3937 of 4913 Old 07-08-2014, 02:27 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
Quote:
Originally Posted by Holiday121 View Post
Also is there a way to calibrate like direct tv feed to my tv?
no.
zoyd is online now  
post #3938 of 4913 Old 07-08-2014, 02:32 PM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
Quote:
Originally Posted by zoyd View Post
no.
So basically you just calibrate the tv and just feed it the direct tv signal n hope for the best.

I can use the avs disk in the ps4 atleast.
Holiday121 is online now  
post #3939 of 4913 Old 07-08-2014, 02:42 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
Quote:
Originally Posted by Holiday121 View Post
So basically you just calibrate the tv and just feed it the direct tv signal n hope for the best.

I can use the avs disk in the ps4 atleast.
You've calibrated it to industry standards and hope that the cable/sat provider provides the appropriate signal, many don't but there is nothing you can do about it.

With the ps4 you can use a test disk but use one with color checker patterns (GCD, MCD, or Ted's disk) and if you find a problem you could devote a different calibration memory slot for it. The ps4 will probably test fine since ps3s do.
zoyd is online now  
post #3940 of 4913 Old 07-08-2014, 02:49 PM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
I have 5 different user presets so I can easily dedicate one for the ps4 and one for a standalone bluray player.

I'll let you know if ps4 looks any different in the readings
Holiday121 is online now  
post #3941 of 4913 Old 07-08-2014, 04:38 PM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
Quote:
Originally Posted by zoyd View Post
There are three main considerations when deciding on what type of transfer function to calibrate to (gamma is the exponent of the transfer function). There is the source display, the target display, and lighting conditions. [...] So given this situation it's best to keep 2 or 3 different presets around and after awhile you'll find one that works most of the time.
Here is my unbiased (not much riding on either option.. ) look at it. I found myself deviating to bt1886 at times, but it depended entirely on the source material. If it has the image detail and isnt overall mastered to be "pop artsy (vibrant colors)" I can swallow bt1886 - because the volume the image gets "makes sense" (is supported by lighting characteristics and visual information). In all other cases the picture for me just looks off (midtones darken without becoming any more lifelike) and I much prefer it at power 2.2. Even in dimly lit viewing environments.

In regards to the color of white curve - when optimizing mine for a time deviated into "warmer" territory at IRE 30,40 and in direct comparison I noticed deviations between skintones in critical viewing. Also when choosing to optimize the white curve a bit at the detriment (all sub dE 1,5) of greyscale - in my case I got a bit better color checker values.

Thats why I attributed it to adhering to the white curve graph.

I guess at this point we are arguing about how "strictly" the dE formula judges regarding certain color shifts. And this cant ever end well. So I give in...

Last edited by harlekin; 07-08-2014 at 04:41 PM.
harlekin is offline  
post #3942 of 4913 Old 07-08-2014, 04:47 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
Quote:
Originally Posted by harlekin View Post
I guess at this point we are arguing about how "strictly" the dE formula judges regarding certain color shifts. And this cant ever end well. So I give in...
The problem is that if someone were to find the perfect settings satisfying the perfect calibration metric they would probably vanish in a puff of smoke along with their perfect display, and not be able to tell us about it.
zoyd is online now  
post #3943 of 4913 Old 07-08-2014, 04:56 PM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
Aha! No - not giving in so fast..

I had one other greyscale configuration where the blue line took off at ire 10 and 20 - which showed almost no dE increase in greyscale (it was just very noticeable on the graph) - but moved the white point line significantly. I have both configurations as presets on my TV right now - and the difference is very apparent (with greyscale being "hotter" (blue) at IRE 10 and 20, especially the picture background "pops" visibly more, although Greyscale dE's dont differ significantly).
harlekin is offline  
post #3944 of 4913 Old 07-08-2014, 04:58 PM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
Quote:
Originally Posted by zoyd View Post
The problem is that if someone were to find the perfect settings satisfying the perfect calibration metric they would probably vanish in a puff of smoke along with their perfect display, and not be able to tell us about it.
I hate the red tint I get when feeding certain SD material. Satisfactory?

Have a nice one, I'm off for today..
harlekin is offline  
post #3945 of 4913 Old 07-08-2014, 05:47 PM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
Ya I'm in the situation of be even knowing what it's supposed to look like and what it ain't.

So basically I'm going by graphs which ain't the best thing to do but ya know.

Also I haven't really sat down and watched some material since the 2 1/2 month old doesn't seem to sleep haha
Holiday121 is online now  
post #3946 of 4913 Old 07-09-2014, 02:00 AM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
Its the _only_ thing to go by. Sadly. All other judgements could only be made in direct comparison between two settings options - where both could be switched instantaneously. And only against material you _know_ has gone through post processing and has a bias for color representation. F.e. some transfers even in the 2000er years bring the characteristics of the film they were shot on with them and its a certain "flair" the DP was going for.

Also I have certain reference material for vibrant colors, washed out (undersaturated) colors, daylight, warm ambient light, ... all with enough faces mixed in that you can go by skin color as reference. At a certain point you also can see when the makeup department has messed up (not useable as reference material), daylight scenes are especially valuable when the white balance was done right with minimal post processing - because you can just tell what daylight looks like reflecting of skin and when its "off". But you cant correct nuances "by eye".

And as a joker I also have a black and white (greytones) scene with a predominant white background, and much detail (fur coat, skin..) which I love to use to show of how your visual system betrays you - because I can show the scene for 20 seconds on D75 and then switch to D65 - and you will scream, that the picture is too warm - until you wait another 20 seconds and then switch back to D75..

The greyscale calibration we have done is entirely dependent on the white point (or the line.. ), so not even being able to judge that correctly just off your eyes (it becomes apparent with material you know, 7000k (custom whitepoint ) and upwards pop too much, feel sterile, but only because of "feel" and not because you could pinpoint something that is wrong.

So why are we going with a certain whitepoint at all? We are trying to keep the visual chain according to the standards used in production, post processeing. D65 is chosen as a whitepoint, because "it is the color of white that a white sheet of paper has under "perfect" sunlight". Why do we do that? Because camera men (at least used to) do white balance against that reference.

Last edited by harlekin; 07-09-2014 at 02:13 AM.
harlekin is offline  
post #3947 of 4913 Old 07-09-2014, 06:32 AM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
What are some of the movies and scenes you test with
Holiday121 is online now  
post #3948 of 4913 Old 07-09-2014, 08:12 AM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
@zoyd :

Here is the data from the Color Checker SG comparison (Calman and HCFR) I promised. All real measurements (urgh...).

Values are listed as avg/max dEs.
All measured at power 2.2 gamma (to make sure its not the bt1886 formula).
Measured against two different greyscale settings and two different color slider settings.
Measured both with the white LED correction active for the i1d3.
Both measured on a RGB full device chain (PC and TV).
Made sure 100% saturated colors were not part of Calmans CC SG run.
edit: All measured against CIE2000.



- As you can see the values I get through HCFR are significantly different from those I get through Calman.
- Although their tendency ("the story they tell") remains about the same.
- the full/SMPTE setting in Calman does significantly impact my measurements.
(most noticeably at 49 color (TV slider) max dE, but interestingly noticeably less at 50 color (TV slider) max dE.

Between Calman and HCFR I get significant changes in breakout values.

Between full/SMPTE (in Calman) I get significant changes in breakout (max) values.

Care to take a guess on what it means? (To take the integrated pattern generators out of the occasion I could measure against Teds pattern disk - which I probably will do - but first I have to seriously search for my masochistic streak again.. )

Id very much appreciate you taking a look at it.

Last edited by harlekin; 07-09-2014 at 10:12 AM.
harlekin is offline  
post #3949 of 4913 Old 07-09-2014, 09:25 AM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
Quote:
Originally Posted by zoyd View Post
If you want to try a BT.1886 type gamma calibration tick the "editable data" box on the gray scale page and enter 0.005 for the black reading. Also tick ITU-R BT.1886 on the gamma set-up page with 0 in the Relative/Absolute box. This will target a curve which has 2.2 gamma at 10% and 2.35 gamma above 40%.


When I do this and run a sweep when I click the gamma box will there be a wave line going up from 40?
Holiday121 is online now  
post #3950 of 4913 Old 07-09-2014, 09:54 AM
AVS Special Member
 
HDTVChallenged's Avatar
 
Join Date: Jan 2002
Posts: 8,537
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 142
Quote:
Originally Posted by harlekin View Post
I guess at this point we are arguing about how "strictly" the dE formula judges regarding certain color shifts. And this cant ever end well. So I give in...
LOL ... to be honest, I think dE formulae ... especially the ones that have various "fudge factors" ... are as much about feeling "good" about the "errors" you can't fix as they are about quantifying the error itself.
HDTVChallenged is offline  
post #3951 of 4913 Old 07-09-2014, 10:04 AM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
@Holiday121 : By now you ought to have figured out which line is the so called "Gamma Target" in the graph - and even if you didnt - google picture search for bt1886 gamma HCFR should have answerd your question as well... The exact "body" of the curve is calculated for the individual black and white points measured(/substituted), so again dont look at what gammas the curve actually touches - just go with the target gamma line.

As for what I use for critical viewing - thats a very personal question. And you are especially lucky, that I am manly enough to share my choices and am not shattered in my confidence as human being, and as a guy - whatsoever - by making it public. As you might have guessed, as I had to watch them over and over, and over again -I chose clips were it would take some time for me to get bored by. Not a fan of this style of music (music? you'll see. ) in particular, just appreciating the "art". And that they are short so you minimize the time searching for scenes.

Code:
h**p://www.youtube.com/watch?v=zdKTgwffmdo
- Highly saturated colors, greyscale - last shot is the greyscale/white shot I spoke about when talking about the adaptability of our eyes to a whitepoint (D65 vs D75).

Code:
h**p://www.youtube.com/watch?v=didptMJxjpE
- Undersaturated colors. Highly postprocessed, except very few shots outdoors (most are color corrected as well).

Code:
h**p://www.youtube.com/watch?v=TpC3TBIvRSk
- used ONLY for the on set shots, with warm lighting and a heavy emphasis on skin colors (dont use the alleyway shots, or the classroom shots (heavily post processed, lighting isnt uniform, ...).

Then I use three more videos in this set, which are especially challenging over a broad palette of colors (all kinds of skin tones, blue skies, greens, lighting conditions..), just to notice "if something feels off":
Code:
h**p://www.youtube.com/watch?v=zEVd9pSG85Q
Code:
h**p://www.youtube.com/watch?v=AG0jlKdB1s0
Code:
h**p://www.youtube.com/watch?v=j7_lSP8Vc3o
-

_Download_ HD versions of the videos (JDownloader f.e.) and play them only in a player (software) that you have verified as "accurate" by doing a pattern measuring run through the player (GCD Calibration Patterns) - trying to eliminate as much uncertainty potential as possible, in each step.
_

Also when I did a major change in configurations, I always checked against daylight scenes in "Quantum of Solace" - but you are invited to find different and/or better material for critical viewing.

(The New PQ Tier Thread for Blu-Ray - Rankings might be a good start.. )

Last edited by harlekin; 07-09-2014 at 10:39 AM.
harlekin is offline  
post #3952 of 4913 Old 07-09-2014, 10:11 AM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
Quote:
Originally Posted by HDTVChallenged View Post
LOL ... to be honest, I think dE formulae ... especially the ones that have various "fudge factors" ... are as much about feeling "good" about the "errors" you can't fix as they are about quantifying the error itself.
Truer words probably were never spoken..

This is the diagram that says it all:
http://en.wikipedia.org/wiki/Color_d...31_MacAdam.png

And here is the context:
http://en.wikipedia.org/wiki/Color_difference#Tolerance
harlekin is offline  
post #3953 of 4913 Old 07-09-2014, 10:18 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
Quote:
Originally Posted by harlekin View Post
Id very much appreciate you taking a look at it.
The only thing of significance I see in that data is that the CM average is systematically between 0.1 and 0.2 dE higher than HCFR. Without having all the data I wouldn't be able track that kind of offset down (no, don't send it to me). One thing I would check though is whether the ccss correction is being applied identically by both programs. So run the comparison without any correction, this is called rawXYZ in CM. btw, exactly what setting in CM are you changing that corresponds to your FULL and SMPTE lines?

ps: If you are using the power law formula in HCFR make sure you have unticked "use measured gamma"

Last edited by zoyd; 07-09-2014 at 10:22 AM.
zoyd is online now  
post #3954 of 4913 Old 07-09-2014, 10:21 AM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
The exact "body" of the curve is calculated for the individual black and white points measured(/substituted), so again dont look at that gammas the curve actually touches - just go with the target gamma line.

The target line hcfr did not change when I selected 1886 and and the target. It stayed a straight line

Making the adjustments of the rgb gamma from 0-30 isn't a problem for 2.2 there and 2.4 from 40-100 but won't that mess up the greyscale?
Holiday121 is online now  
post #3955 of 4913 Old 07-09-2014, 10:29 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
Quote:
Originally Posted by Holiday121 View Post
The target line hcfr did not change when I selected 1886 and and the target. It stayed a straight line

Making the adjustments of the rgb gamma from 0-30 isn't a problem for 2.2 there and 2.4 from 40-100 but won't that mess up the greyscale?
You need to read posts more carefully, I said to edit your 0% and change it to 0.005 first.
zoyd is online now  
post #3956 of 4913 Old 07-09-2014, 10:31 AM
AVS Special Member
 
Holiday121's Avatar
 
Join Date: Feb 2009
Posts: 1,708
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 38
I did edit it. but do I edit all 3 or just one line with it? just the Y?
Holiday121 is online now  
post #3957 of 4913 Old 07-09-2014, 10:35 AM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
@zoyd : Thanks for taking a look. I have access to a full version of Calman - but personally only have the (free) Color Checker version of it. Otherwise I already would have drilled down on the xyY values and posted a cross section.. (*ofcourse* ).

The increase of dE seems to be uniform, but more "exaggerated" at higher values (from what I saw looking at the bar graphs in Calman).

Here are the three hard questions for the data I posted:

- Why on earth - is there this this gap between measuring at full>SMPTE in Calman. According to you prior comment this is purely the math the data seeps through? (No difference in patterns Calman produces, nor in the way it drives the i1d3.)

- Which of the configurations would you choose (some of "the story" gets counteracted by the SMPTE readings).
(I gave up and chose the one with better greyscale tracking.. )

- What is your guess on why there is a difference at all. I know that you cant look into what sorcery Calman does on behalf of its dE formulas - but lets say I already at least looked at the differences between i1d3 correction tables (LED, CCFL IPS, non refresh) in HCFR (not in Calman), and the difference wasnt big enough to explain the higher dE deviations... Also, I will accept "we will never know" as an answer.

The point is that the "picture" all those readings paint in relation to other potential configurations I tested is "uniform", so regardless of which numbers I choose to go by I end up at the same two configurations.

Last edited by harlekin; 07-09-2014 at 10:39 AM.
harlekin is offline  
post #3958 of 4913 Old 07-09-2014, 10:40 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
Quote:
Originally Posted by Holiday121 View Post
I did edit it. but do I edit all 3 or just one line with it? just the Y?
Take one of your already filled-in gray scale runs and place 0.005 into the 0% Y box, the other two boxes should fill in automatically (you may have to enter it twice).
zoyd is online now  
post #3959 of 4913 Old 07-09-2014, 10:44 AM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 267
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
Quote:
Originally Posted by Holiday121 View Post
The target line hcfr did not change when I selected 1886 and and the target. It stayed a straight line
You have edited your black point? zoyd told you which value you should substitute for your unmeasurable black.

The line (target) shouldnt be straight any more.
harlekin is offline  
post #3960 of 4913 Old 07-09-2014, 10:50 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,910
Mentioned: 37 Post(s)
Tagged: 0 Thread(s)
Quoted: 375 Post(s)
Liked: 409
Quote:
Originally Posted by harlekin View Post
@zoyd : Thanks for taking a look. I have access to a full version of Calman - but personally only have the (free) Color Checker version of it. Otherwise I already would have drilled down on the xyY values and posted a cross section.. (*ofcourse* ).

The increase of dE seems to be uniform, but more "exaggerated" at higher values (from what I saw looking at the bar graphs in Calman).

Here are the three hard questions for the data I posted:

- Why on earth - is there this this gap between measuring at full>SMPTE in Calman. According to you prior comment this is purely the math the data seeps through? (No difference in patterns Calman produces, nor in the way it drives the i1d3.)
I don't know.

Quote:
Originally Posted by harlekin
- Which of the configurations would you choose (some of "the story" gets counteracted by the SMPTE readings).
(I gave up and chose the one with better greyscale tracking.. )
or flip a coin, they are all identical with respect to human visible color differences.

Quote:
Originally Posted by harlekin
- What is your guess on why there is a difference at all. I know that you cant look into what sorcery Calman does on behalf of its dE formulas - but lets say I already at least looked at the differences between i1d3 correction tables (LED, CCFL IPS, non refresh) in HCFR (not in Calman), and the difference wasnt big enough to explain the higher dE deviations... Also, I will accept "we will never know" as an answer.
I would still check without a correction, as you saw when I did the test there was no measurable difference between the two programs. (I used a fixed offset matrix entered into both programs). Also if you didn't catch my ps. make sure that the HCFR power law gamma "use measured gamma" box is unchecked.
harlekin likes this.
zoyd is online now  
Reply Display Calibration

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off