Basic Guide to Color Calibration using a CMS (updated and enhanced) - Page 31 - AVS Forum
Forum Jump: 
 
Thread Tools
Old 11-30-2008, 09:25 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,478
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 249
Quote:
Originally Posted by Gixxer1 View Post

What are your thoughts on the CMS in the new Pioneer 9G 111FD, still worthless?

I had Michael Chen calibrate my display but he used the filters first to adjust the color setting than some reference material which was "Flesh Tones" for the final setting of +9 for Pure mode and +6 for the ISF mode.

This was the only control he adjusted, the CMS is all 0's for RGBCMY and so far everyone here has told me that my color is set up completely wrong for my particular display, what is your opinion?

In the Pure mode very little CMS adjustment is required. Just a couple of ticks to get the secondaries right on the money.

Filters will NOT yield a correct result for Color and Tint on these displays. They usually recommend a much too high setting for Color.

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

TomHuffman is offline  
Sponsored Links
Advertisement
 
Old 12-01-2008, 07:08 AM
AVS Special Member
 
glaufman's Avatar
 
Join Date: Feb 2008
Posts: 1,011
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by TomHuffman View Post

Filters will NOT yield a correct result for Color and Tint on these displays. They usually recommend a much too high setting for Color.

Why is that?

-Greg
glaufman is offline  
Old 12-01-2008, 07:11 AM
umr
AVS Club Gold
 
umr's Avatar
 
Join Date: Oct 2002
Location: USA
Posts: 10,147
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 35
Quote:
Originally Posted by glaufman View Post

Why is that?

Because they use psuedo primaries that blend other colors to achieve the proper level of saturation. For example blue and red will be added to green to desaturate the native green primary. The blue and red will be removed by the green filter altering the actual level.
umr is offline  
Old 12-01-2008, 11:20 PM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
As it has been pointed out here, when setting Contrast, Plasmas should put out between 30 - 40 Footlamberts. THX documentation says something about 35. This is all well and good. So for those that have no actual Light Meter... Could you actually do this routine with a camera? They have built in meters right?

http://www.kodak.com/cluster/global/...am105kic.shtml

Just for an estimate? At least to make sure you are not putting out too much at a given setting?

If not this routine. Is there one that would work using a half decent camera and it's built in meter while in manual mode?

Or if I can put the camera in Auto mode and record the shutter speed, ISO and Fstop chosen by the camera when aimed at the screen with a 100% white window, there must be a calculation you can do to esitmate Candellas or some other measurement that can be converted to Footlamberts?

Sorry if this is a dumb idea.

NM I know it is now.

C.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-02-2008, 03:19 PM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by cvearl View Post

As it has been pointed out here, when setting Contrast, Plasmas should put out between 30 - 40 Footlamberts. THX documentation says something about 35. This is all well and good. So for those that have no actual Light Meter... Could you actually do this routine with a camera? They have built in meters right?

http://www.kodak.com/cluster/global/...am105kic.shtml

Just for an estimate? At least to make sure you are not putting out too much at a given setting?

If not this routine. Is there one that would work using a half decent camera and it's built in meter while in manual mode?

Or if I can put the camera in Auto mode and record the shutter speed, ISO and Fstop chosen by the camera when aimed at the screen with a 100% white window, there must be a calculation you can do to esitmate Candellas or some other measurement that can be converted to Footlamberts?

Sorry if this is a dumb idea.

C.

Bah... I found how to arrive at EV (set to 100 ISO and use an equasion to get EV from the F number and Shutter Speed) but once you arrive at EV and use a conversion chart somewhere you will find it only VERY VERY aproximate. Almost better off guessing.

My Display 1 LT can not get here fast enough.

DEC 8 - It has arrived and I cannot wait to get into this tonight! Thanks again Tom for this new addiction. Get's me off computers for a while. Sort of.

C.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-06-2008, 11:36 AM
Senior Member
 
dougmcbride's Avatar
 
Join Date: Nov 2002
Location: Auburn CA
Posts: 413
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi Tom - if you could be so kind could I ask you to have a glance at the attached files and give me your thoughts on a couple of questions below, plus any recommendations you may have? Sorry for the long post, but didn't want to leave holes in the data or methodology.

Background:

Mitsubishi WD-73736 display. Eye-1 Display 2 meter using both HCFR and CalMAN (they line up pretty well so I'm just attaching the HCFR files). Used the AVS HD 709 disk for test patterns (thanks for that!). Zip contains three files:

Baseline - all user controls centered and SM entries default
12-3-2008 - Gray Scale done and first pass at color decoder/gamut
12-5-2008 - Second run at color and minor gray scale tweak

Set Contrast and Brightness OK.

Gray scale cal went well at 30 IRE and above. I've got an issue at 20IRE but eyeballing on the set I don't really see any issue so I've left it alone for the moment. All the rest seem to line up well with Delta E's below 3 and Gamma is almost 2.2 across the board.

Control-wise when doing color, I've left the main Color and Tint controls centered at 31 and used Perfect Color and the Service Menu for adjustments on the Decoder and xy values.

First run on Color (file Cal 12-3-2008) I set the Y values using the Perfect Color controls with no problem then started to try and get the xy points lined up on the CIE diagram. Working through the learning curve and adjusting the SM CMS controls I came up with a first pass which you can see in this file. Green, Yellow and Magenta were pretty close but Blue, Red and Cyan were off (I acknowledge the shortcoming of Blue in these models and there's not really too much that can be done on that color). I tried all sorts of stuff to get Red adjusted but with little luck - I interpreted a need to add saturation to move Red out a ways, but the Saturation control in the SM was maxed out out of the box. Adjusting Hue only made Red Salmon and Orange.

Watching some program material, there was way too much Red (sunburned folks) and Blue (I got so blue'd out I almost went blind!). Made another attempt a couple of days later (file Cal 12-5-2008) and that's where I'm at today.

I backed the Blue Perfect Color off a bunch to make it livable and wound up still being pretty close to 7% on Y.

In order to get Red under control, I backed way down on Y (majority in Perfect Color and a little using the GRL SM control) to roughly 10% - about 1/2 of the ideal target - which seemed to increase the Saturation of Red based upon the CIE chart. I was then able to use the Hue control in the SM (HRL) to rotate the Red point up a bit. My interpretation is that it is still under-saturated, but I don't know how to increase Saturation with the controls I've got. Viewing program material at these setting looks pretty good (sunburn is gone and flesh tones are pretty natural). "Pure" Reds overall still seem to be a bit "dark" if that makes sense, and increasing Y using PC doesn't really help.

Questions:

1. Anything I can try to do to get Red fixed?
2. I'm assuming Blue is what it is?
3. I noticed the color Delta E values varied from HCFR and CalMAN. Different formula?
4. Any other suggestions or comments?

Again, sorry for the long post, but I wanted to give you as much background as I could. If you need more data, please let me know.

Cheers,

Doug

 

DM WD-73736 HCFR Files.zip 3.59765625k . file
dougmcbride is offline  
Old 12-09-2008, 10:47 AM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,478
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 249
Quote:
Originally Posted by dougmcbride View Post

1. Anything I can try to do to get Red fixed?
2. I'm assuming Blue is what it is?
3. I noticed the color Delta E values varied from HCFR and CalMAN. Different formula?
4. Any other suggestions or comments?

You cannot increase saturation. That's a physical property of the display. You an only decrease it. You have substantially lowered red brightness below where it should be and the coordinate is undersaturated. Red is much too weak according to these numbers.

The blues on these displays is off. Not much you can do.

You'll have to ask the CalMAN people about that.

Yes, put the red Y back where it is supposed to be.

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

TomHuffman is offline  
Old 12-09-2008, 05:01 PM
Senior Member
 
dougmcbride's Avatar
 
Join Date: Nov 2002
Location: Auburn CA
Posts: 413
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by TomHuffman View Post

You cannot increase saturation. That's a physical property of the display. You an only decrease it. You have substantially lowered red brightness below where it should be and the coordinate is undersaturated. Red is much too weak according to these numbers.

The blues on these displays is off. Not much you can do.

You'll have to ask the CalMAN people about that.

Yes, put the red Y back where it is supposed to be.

Tom - thanks much for the reply. As far as red Y goes, could you please look at the middle file (12-3 I believe). That file has red Y at close to the correct level and I was getting major sunburn on folk's faces. Also, in trying to increase the "little" y by adjusting hue in the Service Menus (HRL), the reds then turned orange. Would increasing overall white level (contrast) and re-calibrating put me in any better position? I've seen the results of several calibrations on these sets and others have been able to get pretty close.

Sorry if these are dumb questions but I appreciate the help.

Doug
dougmcbride is offline  
Old 12-09-2008, 06:42 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,478
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 249
Quote:
Originally Posted by dougmcbride View Post

Tom - thanks much for the reply. As far as red Y goes, could you please look at the middle file (12-3 I believe). That file has red Y at close to the correct level and I was getting major sunburn on folk's faces. Also, in trying to increase the "little" y by adjusting hue in the Service Menus (HRL), the reds then turned orange. Would increasing overall white level (contrast) and re-calibrating put me in any better position? I've seen the results of several calibrations on these sets and others have been able to get pretty close.

Red saturation is controlled almost exclusively by the x coordinate, which, according to your data, is UNDERSATURATED (HRL controls red hue). I understand that you are frustrated, but you report seeing something that the data you have provided simply doesn't support. If the color decoding and white balance are correct and the red primary is undersaturated, there is no reason I can think of why you would be seeing sunburned faces. Get your y as close to 0.330 as possible. That's the best you can do.

There must be some other problem somewhere: check your connections, your meter, your software settings.

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

TomHuffman is offline  
Old 12-09-2008, 06:44 PM
AVS Special Member
 
nicholc2's Avatar
 
Join Date: Nov 2006
Location: Columbus, OH
Posts: 1,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Overall, I'm with Tom. One helpful tool (if you haven't already downloaded it) is the AccuPel Display Calibration Calculator ( http://www.accupel.com/Display%20Cal...20%28PC%29.zip ). This calculator is a very good way to help with luminance values and where the secondaries should lie depending on where the primaries lie. I'd definitely download that and use it if you haven't already.

I would just keep playing around with it and also keep in mind that depending on what you were watching, the burnt faces might have been on purpose. Like NCIS for instance has red push pretty bad. Assuming you have a bluray player, put in one of your favorite movies and watch it after you make changes.

The more you try, the more you learn, so it's never really a bad thing. Just keep track of your values like you're doing and keep up the good work.
nicholc2 is offline  
Old 12-10-2008, 12:18 PM
Senior Member
 
dougmcbride's Avatar
 
Join Date: Nov 2002
Location: Auburn CA
Posts: 413
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by TomHuffman View Post

Red saturation is controlled almost exclusively by the x coordinate, which, according to your data, is UNDERSATURATED (HRL controls red hue). I understand that you are frustrated, but you report seeing something that the data you have provided simply doesn't support. If the color decoding and white balance are correct and the red primary is undersaturated, there is no reason I can think of why you would be seeing sunburned faces. Get your y as close to 0.330 as possible. That's the best you can do.

There must be some other problem somewhere: check your connections, your meter, your software settings.

Tom - thanks again. Being new to this, I thought I may have missed some subtlety along the way but my understanding is as you describe it. On the Red Hue adjustment, I should have been more clear in that I was using that more to adjust "little y" after the 12-3-2008 exercise since it was pretty low. I'll go back and give it another try getting Y set correctly and and see how the results change.

Thanks again and Happy Holidays,

Doug
dougmcbride is offline  
Old 12-10-2008, 12:29 PM
Senior Member
 
dougmcbride's Avatar
 
Join Date: Nov 2002
Location: Auburn CA
Posts: 413
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by nicholc2 View Post

Overall, I'm with Tom. One helpful tool (if you haven't already downloaded it) is the AccuPel Display Calibration Calculator ( http://www.accupel.com/Display%20Cal...20%28PC%29.zip ). This calculator is a very good way to help with luminance values and where the secondaries should lie depending on where the primaries lie. I'd definitely download that and use it if you haven't already.

I would just keep playing around with it and also keep in mind that depending on what you were watching, the burnt faces might have been on purpose. Like NCIS for instance has red push pretty bad. Assuming you have a bluray player, put in one of your favorite movies and watch it after you make changes.

The more you try, the more you learn, so it's never really a bad thing. Just keep track of your values like you're doing and keep up the good work.

nicholc2 - thanks much for the pointer to the Accupel tool. I was not aware of that and had given some thought to how Y changes as the primaries stray from the standard, but had not idea what the magnitude of the change was. I think I read in WideScreen Review that Tom was going to follow up his original article with some coverage of that topic as well.

Also thanks for the words of encouragement. I've been trying to watch out for material that has red push and you're right, there's a bunch of it out there. Where I was seeing the flush was even on material I use as reference for being pretty natural. I'm going to keep playing and see if I can get a little closer. I'm not disappointed in the picture as it stands now, short of knowing that it could be better based upon measured data.

Cheers and Happy Holidays to you,

Doug
dougmcbride is offline  
Old 12-10-2008, 09:20 PM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi Tom. I have been playing with my new I1 Display LT and Color HCFR software as I go through the guide.

Regarding your Color Performance part of the guide...

""ΔE offers a much better approach. SMPTE has established a standard for the color accuracy of Digital Cinema, which is 4 Lab (1976) units or less. This is approximately equivalent to 2 CIE94 units or 5 CIELUV units. This seems like a reasonable tolerance. Unfortunately, we are left with the problem discussed above: which ΔE standard are we to use? SMPTE offers no guidance as to why they selected CIELAB.

Consider this oversaturated, but dim, shade of green: x0.296, y0.678, Y0.535

How far from the Rec. 709 standard does this green deviate? Using ΔE as a guide it is very hard to say. CIELUV reports that this green has a 1976 ΔE of 11.4, which, though far from perfect at just under 3 times the allowed color error, is not too bad. However, CIE94 reports that the same green exhibits a 1994 ΔE of 11.2, which is approximately six times the allowed color error. This is a huge error. Two ΔE systems report radically different results for the same color!""

When I am considering the DeltaE numbers that I record when I test my display for my Greyscale or Colors in HCFR, what am I actually looking at? If my DeltaE is 6.9 for Red for instance, is that a 1976 or 1994 number? How can I objectively evaluate that number in HCFR after my readings and know if I am good or not? Is there a "stay under this" rule?

Sorry. I'm just a little confused on this part of the process of evaluating acuracy.

This is as close as I can get on my pz85 42" without doing a service menu adjustment (will do that after I get more time). I had to break the "make red 21% of white" rule for setting the color control using red to get the DeltaE as small as I could for red.

C.

 

User controls only.zip 1.068359375k . file

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-10-2008, 10:32 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,085
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 208 Post(s)
Liked: 74
Quote:
Originally Posted by cvearl View Post

Hi Tom. I have been playing with my new I1 Display LT and Color HCFR software as I go through the guide.

Regarding your Color Performance part of the guide...

""ΔE offers a much better approach. SMPTE has established a standard for the color accuracy of Digital Cinema, which is 4 Lab (1976) units or less. This is approximately equivalent to 2 CIE94 units or 5 CIELUV units. This seems like a reasonable tolerance. Unfortunately, we are left with the problem discussed above: which ΔE standard are we to use? SMPTE offers no guidance as to why they selected CIELAB.

Consider this oversaturated, but dim, shade of green: x0.296, y0.678, Y0.535

How far from the Rec. 709 standard does this green deviate? Using ΔE as a guide it is very hard to say. CIELUV reports that this green has a 1976 ΔE of 11.4, which, though far from perfect at just under 3 times the allowed color error, is not too bad. However, CIE94 reports that the same green exhibits a 1994 ΔE of 11.2, which is approximately six times the allowed color error. This is a huge error. Two ΔE systems report radically different results for the same color!""

When I am considering the DeltaE numbers that I record when I test my display for my Greyscale or Colors in HCFR, what am I actually looking at? If my DeltaE is 6.9 for Red for instance, is that a 1976 or 1994 number? How can I objectively evaluate that number in HCFR after my readings and know if I am good or not? Is there a "stay under this" rule?

Sorry. I'm just a little confused on this part of the process of evaluating acuracy.

This is as close as I can get on my pz85 42" without doing a service menu adjustment (will do that after I get more time). I had to break the "make red 21% of white" rule for setting the color control using red to get the DeltaE as small as I could for red.

C.

I can't recall which DeltaE formula it uses. Hopefully Tom knows or you may also want to try the HCFR thread here on AVS. As for a good number to stay under, I recall hearing that number is different depending on which formula is being used - in either case I think its like 3 or 6 depending... again hopefully Tom can chime in here. These are great questions, BTW.

As a side note - under Advanced menu, Perferences - on the References tab make sure you have HDTV Rec 709 selected (assuming of course you are dialing in color for HD). IIRC it defaults to Rec 601. This will significantly impact your dE so make sure you have the right selection there.
lovingdvd is offline  
Old 12-10-2008, 10:36 PM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by lovingdvd View Post

I can't recall which DeltaE formula it uses. Hopefully Tom knows or you may also want to try the HCFR thread here on AVS. As for a good number to stay under, I recall hearing that number is different depending on which formula is being used - in either case I think its like 3 or 6 depending... again hopefully Tom can chime in here. These are great questions, BTW.

As a side note - under Advanced menu, Perferences - on the References tab make sure you have HDTV Rec 709 selected (assuming of course you are dialing in color for HD). IIRC it defaults to Rec 601. This will significantly impact your dE so make sure you have the right selection there.

Yup... Rec 709 is set in preferences. The Calibration tool is set to Eye-one disaply LT - Plasma. The meter was warmed on the set for 30 minutes and then calibrated in the dark on a black surface just before taking measurements.

Thanks for the reply. I'll check the other thread as well.

C.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-10-2008, 10:59 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,085
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 208 Post(s)
Liked: 74
Quote:
Originally Posted by cvearl View Post

Yup... Rec 709 is set in preferences. The Calibration tool is set to Eye-one disaply LT - Plasma. The meter was warmed on the set for 30 minutes and then calibrated in the dark on a black surface just before taking measurements.

Thanks for the reply. I'll check the other thread as well.

C.

If you get an answer in the other thread please post here as well.
lovingdvd is offline  
Old 12-11-2008, 04:41 AM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,478
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 249
Quote:
Originally Posted by cvearl View Post

When I am considering the DeltaE numbers that I record when I test my display for my Greyscale or Colors in HCFR, what am I actually looking at? If my DeltaE is 6.9 for Red for instance, is that a 1976 or 1994 number? How can I objectively evaluate that number in HCFR after my readings and know if I am good or not? Is there a "stay under this" rule?

HCFR uses CIELUV only. The video industry has been slow to recognize, much less adopt, any of the post-1976 standards.

For CIELUV, a reasonable set of tolerances are 2.5 for the minimum perceptable error and 5.0 as the maximum acceptable error. For real-world performance, any display whose pri/sec colors are below 10 is quite good performance, but if you have a CMS, I'd try to get that under 5.0.

BTW, in Preferences, Advanced section of HCFR there are 2 DeltaE options ("Do not use luminance" and "Use gamma reference for grayscale"). Ensure that both are unchecked. Otherwise, you will get really inaccurate numbers. HCFR includes a lot of unnecssary options that only give people the chance to screw up the results.

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

TomHuffman is offline  
Old 12-11-2008, 07:44 AM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
,
Quote:
Originally Posted by TomHuffman View Post

HCFR uses CIELUV only. The video industry has been slow to recognize, much less adopt, any of the post-1976 standards.

For CIELUV, a reasonable set of tolerances are 2.5 for the minimum perceptable error and 5.0 as the maximum acceptable error. For real-world performance, any display whose pri/sec colors are below 10 is quite good performance, but if you have a CMS, I'd try to get that under 5.0.

BTW, in Preferences, Advanced section of HCFR there are 2 DeltaE options ("Do not use luminance" and "Use gamma reference for grayscale"). Ensure that both are unchecked. Otherwise, you will get really inaccurate numbers. HCFR includes a lot of unnecssary options that only give people the chance to screw up the results.

Thank you. CIELUV. Gotcha. Thanks for those targets. I have some playing to do.

Since I have the Panasonic PZ85U (same as you just 42"), I have no CMS.

Yup both those items are unchecked in the HCFR software. Only thing I changed upson installtion was setting to Rec 709 and selecting "Eye-one Plasma". I do a 30 minute warmup and a calibration of the sensor before each session of runs. Using DVE HD Blueray for the greyscale measurements and AVS 709 for color patterns using the 100% windows. I also have Getgray and your disk from this guide.

I have noticed that the setting of Color and Tint controls of this TV, and the effect on primaries and secondaries the controls have, that it's a seasaw balancing act as far as reducing DeltaE for one color versus another or one secondary versus another...

For example in my case, with Color set to 41 my Blue's DeltaE is 4.2 and Red's is 33.2 while Color set to 34 my Blue ends up at 23.9 and my Red is now 6.9. With color control somwhere in between 41 and 34 I can balance the error between the two.

So. When I play the seasaw balancing game with the primaries and secondaries, getting them all as as equally minimized as I can, is best practice generally to lean a little towards making Red (playing with the color control) and Cyan (playing with the Tint Control) have the lower of the DeltaE's somewhat as far as viewers perception of natural color while attempting to keep the other primaries and secondaries close?

Thanks!

C.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-11-2008, 08:02 AM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,085
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 208 Post(s)
Liked: 74
Quote:
Originally Posted by cvearl View Post

...is best practice generally to lean a little towards making Red (playing with the color control) and Cyan (playing with the Tint Control) have the lower of the DeltaE's somewhat as far as viewers perception of natural color while attempting to keep the other primaries and secondaries close?

C.

Generally errors in green are considerable to be the most noticeable, followed by red, followed by blue. At least as it applies to grayscale measurements (i.e. if given a choice of an error on way or another, favor an error toward red over green etc). I assume the same would apply to primaries. Ultimately of course given the lack of controls on your unit you should go by what looks best by eye in deciding on your balancing act.
lovingdvd is offline  
Old 12-11-2008, 09:32 AM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by lovingdvd View Post

If you get an answer in the other thread please post here as well.

http://www.avsforum.com/avs-vb/showp...postcount=2460

Quote:
Originally Posted by cvearl
If my DeltaE is 6.9 for Red for instance, is that a 1976 or 1994 number? How can I objectively evaluate that number in HCFR after my readings and know if I am good or not? Is there a "stay under this" rule?

Response:
ColorHCFR uses 1976. In my opinion DE1976 is completely worthless for color due to how Y is handled. If you look at the formula it's just a distance formula, and the calculation relies entirely on the color space. The main issue is that Y clearly isn't perceptually uniform in the color space. I make this comment based on looking closely at DE calculations for SXRD with over-saturated green. The hue for the display was on-line, but the display's green was over-saturated. To even out the over-saturation, the DE1976 formula will suggest reducing Y. The issue is that the minimum DE comes at a rediculously low Y setting that looks nothing at all like the original. Based on looking at DE1976 calculations for the example display, I'd suggest it makes no sense at all for color. I figure ignore the DE calculation for color entirely and just look at xyY.


Quote:
Originally Posted by cvearl
Can the DeltaE's in ColorHCFR be converted to those used by many of you calibrators out there?

Response:
You could export the measured data and then calculate different values in a spreadsheet if you wanted.


Quote:
Originally Posted by cvearl
I had to break the "make red 21% of white" rule for setting the color control using red to get the DeltaE as small as I could for red.

Response:
In my opinion, it makes no sense to try to reduce DE1976 numbers for color. Since the xy of your primaries are off, I would probably suggest looking at what Y values the Accupel calculator suggests for your measured xy. Generally they'll be a lot closer to typical than what reducing DE1976 would suggest using.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-11-2008, 01:01 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,478
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 249
Quote:
Originally Posted by cvearl View Post

http://www.avsforum.com/avs-vb/showp...postcount=2460

ColorHCFR uses 1976. In my opinion DE1976 is completely worthless for color due to how Y is handled. If you look at the formula it's just a distance formula, and the calculation relies entirely on the color space. The main issue is that Y clearly isn't perceptually uniform in the color space. I make this comment based on looking closely at DE calculations for SXRD with over-saturated green. The hue for the display was on-line, but the display's green was over-saturated. To even out the over-saturation, the DE1976 formula will suggest reducing Y. The issue is that the minimum DE comes at a rediculously low Y setting that looks nothing at all like the original. Based on looking at DE1976 calculations for the example display, I'd suggest it makes no sense at all for color. I figure ignore the DE calculation for color entirely and just look at xyY.

CIE94 fixes this problem, though it may go a little too far in the other direction.

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

TomHuffman is offline  
Old 12-11-2008, 01:25 PM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by TomHuffman View Post

CIE94 fixes this problem, though it may go a little too far in the other direction.

So with 1976, I should just stick with your suggestion of getting the lightness of Red to be 21% that of white and to heck with the DeltaE reported in HCFR? If all Tomhuffman had at his disposal was a PZ85U and Color HCFR and an i1 Display LT?

Or am I better off with Calman for judging error delta? At least as far as me having to do less personal analysis for what setting represents the best accuracy.

I don't mind spending a couple more dollars if it's superior in that respect.

C.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-11-2008, 02:49 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,478
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 249
Quote:
Originally Posted by cvearl View Post

So with 1976, I should just stick with your suggestion of getting the lightness of Red to be 21% that of white and to heck with the DeltaE reported in HCFR? If all Tomhuffman had at his disposal was a PZ85U and Color HCFR and an i1 Display LT?

If dE is all that concerns you, just copy/paste the raw data into the spreadsheet I provide at the end of the tutorial. It has CIELUV, CIELAB, and CIE94.

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

TomHuffman is offline  
Old 12-11-2008, 03:02 PM
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,085
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 208 Post(s)
Liked: 74
Quote:
Originally Posted by TomHuffman View Post

If dE is all that concerns you, just copy/paste the raw data into the spreadsheet I provide at the end of the tutorial. It has CIELUV, CIELAB, and CIE94.

Tom - which formula do you prefer? Would it make sense for HCFR to add options to allow the user to choose which formula is used for the dE calculation? The HCFR development team is very responsible to feature requests and fixes, so if you this such a feature would be appropriate I'd like to ask them about the possibility of adding it. And if so, what's the proper request to make - "Please add an option to control whether CIELUV, CIELAB or CIE94 is used as the basis for calculating dE"?
lovingdvd is offline  
Old 12-11-2008, 03:06 PM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by TomHuffman View Post

If dE is all that concerns you, just copy/paste the raw data into the spreadsheet I provide at the end of the tutorial. It has CIELUV, CIELAB, and CIE94.

Well since I have my laptop right in front of me that should not be a problem.

In these economic times, this is sage advice. Thanks!

And akin to lovingdvd's question... what numbers are in your screenshots over in the PZ85 owners thread? CIE94? Is that Calman those shots come from?

C.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-11-2008, 03:15 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,478
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 249
I would use CIE94 for color and CIELUV for grayscale.

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

TomHuffman is offline  
Old 12-11-2008, 05:31 PM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Great!

If I am in the vacinity of these kinds of numbers from one of your threads...


ΔE (CIE94) Performance



Am I doing ok or were these just sort of hypothetical.

C.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-11-2008, 05:55 PM - Thread Starter
AVS Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 6,478
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 249
Quote:
Originally Posted by cvearl View Post

Great!

If I am in the vacinity of these kinds of numbers from one of your threads...


ΔE (CIE94) Performance



Am I doing ok or were these just sort of hypothetical.

C.

What thread is this from? The before/after numbers are nearly identical.

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

TomHuffman is offline  
Old 12-11-2008, 06:02 PM
Senior Member
 
cvearl's Avatar
 
Join Date: Sep 2008
Posts: 440
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by TomHuffman View Post

What thread is this from? The before/after numbers are nearly identical.


PZ85 Calibration Thread...

http://www.avsforum.com/avs-vb/showt...=#post14180286

Referencing the PZ85U you were showing that enabling Color Management basically only created a greater error in Cyan.

Thought these might be from your 50"

C.

Panasonic 42" Plasma TH42-PZ85U
Panasonic Blueray Player DMP-BD35
Sony 5.1 Home Theater DAV-HDX475 (wish it was Panasonic but it's ok)
cvearl is offline  
Old 12-12-2008, 01:25 PM
Member
 
MoonRabbit79's Avatar
 
Join Date: May 2008
Posts: 121
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Hi !

Does someone know the equations to calculate to secondary colors cooridnates from the absolute coordinates of the primary colors coordinates (be it Rec.709) in the CIE xyY, CIE XYZ and CIE L*C*h(a*b*) color spaces?


Thanks in advance.
MoonRabbit79 is offline  
 

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


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