Can't calibrate CRT and LCD to look the same - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 14 Old 08-26-2012, 05:07 PM - Thread Starter
Newbie
 
Touche's Avatar
 
Join Date: Aug 2008
Posts: 10
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
I've been trying to calibrate my displays for the last couple of days and they are driving me crazy. Calibration reports look great, but the LCD ends up with a reddish cast and the CRT a greenish one, compared to each other.

I'm using X-Rite i1 Display2 and basICColor5 software, calibrating to D65, gamma 2.2. I have tried calibrating the NEC LCD from 5500K to 7500K, but whites always look reddish vs greenish. As expected, both look ok when viewed individually.

The best I could manage was manually reducing green a bit on the CRT and doing some voodoo tinkering with LCD's red and green gain. It is somewhat acceptable now, but I've bought the calibrator specifically to make them show the same image, subjectively and objectively, only to find it very hard, if not impossible to achieve.

Any ideas how to get the same results or is it impossible with such different technologies?

Could it be a i1 Display2 defect (I bought it used, and I was expecting maybe some deviation from perfect values due to possible aging, but not different results on separate measurements)?


LCD: NEC 24wmgx3



CRT: Sony GDM-FW900




P.S. Although basICColor shows accurate colors, HCFR always shows errors in CIE triangle. Is that kind of different reporting normal?
Touche is offline  
Sponsored Links
Advertisement
 
post #2 of 14 Old 08-26-2012, 07:44 PM
AVS Special Member
 
airscapes's Avatar
 
Join Date: Dec 2008
Location: Philadelphia
Posts: 4,621
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 54 Post(s)
Liked: 111
Do a forum search on Display2 or Display Lt and you will find that these meters are not accurate and the cause of most of your trouble. You will also find that even with an accurate meter, the displays may never be exactly the same as they are different technologies and may have issues that can not be corrected, so side by side they do not look the same. For example, I have a Sony CRT that has no color management system just RGB cuts/gains in the SM and a DLP projector with no CMS .. Light blue is never the same when watching both displays at the same time even though they are both calibrated as best as possible with the adjustments available.
airscapes is online now  
post #3 of 14 Old 08-27-2012, 11:24 AM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 226
It's not so much that the meter is not accurate, it's a matter of the light spectra being very different for CRT and LCD and the meter itself does not (cannot) respond to the different light spectra from the 2 different types of displays in the same way. Chances are the CRT will look better and the LCD will be a bit "off" as described. You'd have to send the meter to some service provider, perhaps Spectracal provides this sort of service for CalMAN users... and they would characterize your meter against an LCD light source (LED or CCFL as needed). You would use the characterization data when calibrating the LCD TV and the software you are using would have to support that function (CalMAN does, not sure about others).

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #4 of 14 Old 08-27-2012, 12:43 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,612
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 23 Post(s)
Liked: 167
The other big issue is that windows doesn't do any kind of gamut control, they leave it up to ICC aware apps.

So if blue is a little different between the two monitors (view able via HCFRs gamut chart), there is absolutely no way to fix that and it will always be obvious (short of buying $1000+ video processors).

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #5 of 14 Old 08-30-2012, 08:49 AM - Thread Starter
Newbie
 
Touche's Avatar
 
Join Date: Aug 2008
Posts: 10
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
Unfortunately, I have nowhere to calibrate my Display 2 or profile it to a certain kind of display technology. I don't think it lost much precision. I don't have an objective way of verifying that, but the images do look good on the Sony CRT. My bigger problem was the discrepancy between it and the LCD. I have managed to get them closer by upping the green on the LCD and lowering it a bit on the CRT, a slightly tweaking the red and blue, and then calibrating both again choosing native white point target. It is not perfect, but close enough I guess. As you've said, the LCD and CRT are just too different.

Now, I have another question. How do I go abut calibrating LED and/or wide gamut LCDs with Display 2? I remember reading that the white point reading will be about 500K too high (or low?), but that shouldn't be much of a problem since it is easy to compensate. But what about wide gamut? Do I simply ignore the fact and calibrate the display as if it was sRGB or do some additional steps have to be made?

Thank you for the help.
Touche is offline  
post #6 of 14 Old 08-30-2012, 09:11 AM
Advanced Member
 
Smackrabbit's Avatar
 
Join Date: Sep 2001
Location: Portland, OR, USA
Posts: 893
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 30
Quote:
Originally Posted by Touche View Post

Unfortunately, I have nowhere to calibrate my Display 2 or profile it to a certain kind of display technology. I don't think it lost much precision. I don't have an objective way of verifying that, but the images do look good on the Sony CRT. My bigger problem was the discrepancy between it and the LCD. I have managed to get them closer by upping the green on the LCD and lowering it a bit on the CRT, a slightly tweaking the red and blue, and then calibrating both again choosing native white point target. It is not perfect, but close enough I guess. As you've said, the LCD and CRT are just too different.
Now, I have another question. How do I go abut calibrating LED and/or wide gamut LCDs with Display 2? I remember reading that the white point reading will be about 500K too high (or low?), but that shouldn't be much of a problem since it is easy to compensate. But what about wide gamut? Do I simply ignore the fact and calibrate the display as if it was sRGB or do some additional steps have to be made?
Thank you for the help.

It's not as simple as it being a certain amount high or low. The i1D2 has filters that are designed to let certain types of light through. However, as these have aged, and potentially at a different rate for one color than another, and you're having a different wavelength of light that will behave differently with the filters than a CRT would, there is no way to say "It's only 500K off" and have that be anything but throwing darts at a board. You could use a spectro, compare the two, and see how much the readings are off, but that would only apply to that specific screen at that specific time, and a month later might be different.

Your i1D2 should have a date available on it to determine when it was manufactured. I have one that's around 3 years old, always kept sealed in a bag as well as you can keep it, but after 2 years it has an average dE in the grayscale of 10+ compared to a calibrated i1Pro. Some colors were closer, and some were really far off, but there's no way of knowing for sure. Also, since if you use the i1D2 to calibrate the grayscale, you're going to wind up with one that is linear, but likely tinted in a certain color direction, but with how the eye handles that it will likely look correct, even if it's off by 1,000K or more. The i1D2 can do certain things, and when they are new they are relatively accurate, but by now they're almost all aged so much that you can not really trust them to any reasonable degree unless they are profiled.

Chris Heinonen
Senior Editor, Secrets of Home Theater and High Fidelity, www.hometheaterhifi.com
Displays Editor, AnandTech.com
Contributor, HDGuru.com and Wirecutter.com
ISF Level II Certified Calibrator, ReferenceHomeTheater.com
Smackrabbit is offline  
post #7 of 14 Old 08-30-2012, 09:26 AM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,172
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 79 Post(s)
Liked: 197
Quote:
Originally Posted by Doug Blackburn View Post

You'd have to send the meter to some service provider, perhaps Spectracal provides this sort of service for CalMAN users... and they would characterize your meter against an LCD light source (LED or CCFL as needed). You would use the characterization data when calibrating the LCD TV and the software you are using would have to support that function (CalMAN does, not sure about others).

I have a question. How would this characterization process differ from meter profiling against a spectro and would it produce a significantly (visible) different result?

A related question is the following: If I am profiling my C6 with my i1Pro for use on a CCFL-LCD, does it matter what tables I load onto the C6 before creating the profile (LCD, Plasma, CRT, etc.) or will the end result be the same no matter what target display type/mode the C6 is set in?
PlasmaPZ80U is offline  
post #8 of 14 Old 08-30-2012, 09:34 AM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,172
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 79 Post(s)
Liked: 197
Quote:
Originally Posted by Smackrabbit View Post

I have one that's around 3 years old, always kept sealed in a bag as well as you can keep it, but after 2 years it has an average dE in the grayscale of 10+ compared to a calibrated i1Pro. Some colors were closer, and some were really far off, but there's no way of knowing for sure. Also, since if you use the i1D2 to calibrate the grayscale, you're going to wind up with one that is linear, but likely tinted in a certain color direction, but with how the eye handles that it will likely look correct, even if it's off by 1,000K or more. The i1D2 can do certain things, and when they are new they are relatively accurate, but by now they're almost all aged so much that you can not really trust them to any reasonable degree unless they are profiled.

A few questions:

Did you put a quality dessicant in the bag?

What do you mean by 'calibrated' i1Pro?
PlasmaPZ80U is offline  
post #9 of 14 Old 08-30-2012, 11:16 AM - Thread Starter
Newbie
 
Touche's Avatar
 
Join Date: Aug 2008
Posts: 10
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 10
Quote:
Originally Posted by Smackrabbit View Post

It's not as simple as it being a certain amount high or low. The i1D2 has filters that are designed to let certain types of light through. However, as these have aged, and potentially at a different rate for one color than another, and you're having a different wavelength of light that will behave differently with the filters than a CRT would, there is no way to say "It's only 500K off" and have that be anything but throwing darts at a board. You could use a spectro, compare the two, and see how much the readings are off, but that would only apply to that specific screen at that specific time, and a month later might be different.
Your i1D2 should have a date available on it to determine when it was manufactured. I have one that's around 3 years old, always kept sealed in a bag as well as you can keep it, but after 2 years it has an average dE in the grayscale of 10+ compared to a calibrated i1Pro. Some colors were closer, and some were really far off, but there's no way of knowing for sure. Also, since if you use the i1D2 to calibrate the grayscale, you're going to wind up with one that is linear, but likely tinted in a certain color direction, but with how the eye handles that it will likely look correct, even if it's off by 1,000K or more. The i1D2 can do certain things, and when they are new they are relatively accurate, but by now they're almost all aged so much that you can not really trust them to any reasonable degree unless they are profiled.

It seems my i1D2 is from 03/2011, as that is the only date I can find on it. I don't have access to a spectro and buying one is way out of my price range, so it will have to do as long as the results look visually ok. Far from ideal, I know, but I believe it's still better than nothing. In my attempts to make my LCD and CRT similar, I've found that playing with green bias/gain changes the white point without messing with color accuracy much (at least going by calibration reports with the same meter), so I'll visually compensate with that as much as I can.

How about calibrating wide gamut monitors, if we disregard i1D2's aging? If I calibrate a wide gamut monitor using the same procedures as with standard gamut ones, will it be calibrated right or will it be waaay off (disregarding the standard oversaturated colors when viewing standard material in non ICC aware programs/OS)?


I know all this sounds bad to precision purists, and it's all just guessing and approximations without a good spectrometer, but we have to make due with what we can afford. Any help is greatly appreciated.
Touche is offline  
post #10 of 14 Old 08-30-2012, 12:05 PM
AVS Special Member
 
airscapes's Avatar
 
Join Date: Dec 2008
Location: Philadelphia
Posts: 4,621
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 54 Post(s)
Liked: 111
Quote:
Originally Posted by PlasmaPZ80U View Post

I have a question. How would this characterization process differ from meter profiling against a spectro and would it produce a significantly (visible) different result?
A related question is the following: If I am profiling my C6 with my i1Pro for use on a CCFL-LCD, does it matter what tables I load onto the C6 before creating the profile (LCD, Plasma, CRT, etc.) or will the end result be the same no matter what target display type/mode the C6 is set in?

I read Characterization = Profile
No, it does not matter what you choose for the C6 first as it is replaced by your profile created with the i1pro.
airscapes is online now  
post #11 of 14 Old 08-31-2012, 10:08 AM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 226
There is a difference between characterizing/profiling versus creating a correction matrix. In characterization/profiling, you measure one thing... a shade of gray or white and you measure the same thing with the reference meter and that ONE measurement (for red, green, and blue) is applied to the meter being characterized for ALL luminance values of red, green, and blue.

To fully correct a questionable meter against a reference meter, you would measure multiple values of red, green, and blue with both meters and create a correction matrix that "fixes" the meter being calibrated so it measures the same as the reference meter at any luminance level.

With characterization/profiling, you assume that one measurement is enough to make the meter-in-question accurate, or reasonably accurate at all luminance levels even though you measure just one (or maybe some small number of reference levels). Correction is potentially more accurate, but it's also more time consuming... potentially quite time consuming if there are a lot of measurement points involved.

Calibration is the ultimate "fix" for a questionable meter... here the meter is completely measured against standards (often by a factory technician) and in the case of colirimeters, the filters are replaced (if necessary) and the entire meter is restored to factory specifications. Many inexpensive meters don't have calibration offered for them as calibration would cost as much as making a new meter. The meter I use costs $800 to re-calibrate and frankly, each time I have it done, I wonder if I should just buy a new lower-cost meter... especially now that meters like the C5 and the newest i1 have become available. I could probably buy one of these meters every year and sell it for half-price as a used meter and be calibrating for less than I spend just to re-calibrate the high-end meter I use. On the other hand, the meter I use can't be fooled by the light source (CRT phosphors, plasma phosphors, CCFL, LED, OLED, laser, projection lamps, and anything likely to ever come along that doesn't exist now). It's difficult, perhaps impossible, to design a low-cost filter-based meter (colorimeter) that can do that. They typically need some kind of correction factor for different display types.

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #12 of 14 Old 08-31-2012, 10:47 AM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,172
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 79 Post(s)
Liked: 197
Quote:
Originally Posted by airscapes View Post

I read Characterization = Profile
No, it does not matter what you choose for the C6 first as it is replaced by your profile created with the i1pro.

Quote:
Originally Posted by Doug Blackburn View Post

There is a difference between characterizing/profiling versus creating a correction matrix. In characterization/profiling, you measure one thing... a shade of gray or white and you measure the same thing with the reference meter and that ONE measurement (for red, green, and blue) is applied to the meter being characterized for ALL luminance values of red, green, and blue.
To fully correct a questionable meter against a reference meter, you would measure multiple values of red, green, and blue with both meters and create a correction matrix that "fixes" the meter being calibrated so it measures the same as the reference meter at any luminance level.
With characterization/profiling, you assume that one measurement is enough to make the meter-in-question accurate, or reasonably accurate at all luminance levels even though you measure just one (or maybe some small number of reference levels). Correction is potentially more accurate, but it's also more time consuming... potentially quite time consuming if there are a lot of measurement points involved.
Calibration is the ultimate "fix" for a questionable meter... here the meter is completely measured against standards (often by a factory technician) and in the case of colirimeters, the filters are replaced (if necessary) and the entire meter is restored to factory specifications. Many inexpensive meters don't have calibration offered for them as calibration would cost as much as making a new meter. The meter I use costs $800 to re-calibrate and frankly, each time I have it done, I wonder if I should just buy a new lower-cost meter... especially now that meters like the C5 and the newest i1 have become available. I could probably buy one of these meters every year and sell it for half-price as a used meter and be calibrating for less than I spend just to re-calibrate the high-end meter I use. On the other hand, the meter I use can't be fooled by the light source (CRT phosphors, plasma phosphors, CCFL, LED, OLED, laser, projection lamps, and anything likely to ever come along that doesn't exist now). It's difficult, perhaps impossible, to design a low-cost filter-based meter (colorimeter) that can do that. They typically need some kind of correction factor for different display types.

Both these meters are still in spec (my i1 Pro and C6), but I want to share the following results. When I profile my C6 against the i1 Pro on a given display (like LCD or plasma) the profiled C6 reads accurately/the same vs the i1 Pro directly across the whole luminance range, not just at the 75% stim point at which the four color matrix was created. I cannot say with 100% certainty whether this behavior will remain consistent as the C6 ages, but at this time it seems a single point correction is all that is needed to make the C6 read exactly the same as the i1Pro across the entire grayscale (keeping in mind the i1Pro cannot read below 30% accurately on the average flat panel LCD or plasma).

Also, if I was to send if my C6 for re-calibration to the SpectraCal lab say 2-5 years from now, would they use a single or multi-point correction? Would they use the four color matrix method or something newer and more advanced? What about with ChromaPure/Tom Huffman and the D3 PRO (if I had that meter presently)?

What kind of colorimeter do you (@ Doug Blackburn) use for your professional calibrations and where/how often do you send it in for re-characterization/re-calibration?
PlasmaPZ80U is offline  
post #13 of 14 Old 09-01-2012, 10:51 AM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 226
I use a $14,000 Konica-Minolta CS-200 - an interesting design that employs some characteristics of filter-based meters and some characteristics of spectro meters. It generally goes back for recal every 1-2 years based on a reference measurement. If it wasn't being used in critical applications, it doesn't really drift much and it could probably go 4-5 years between calibrations. But why would you own a $14,000 meter that wasn't used for critical applications, so I'd guess my re-cal frequency is fairly typical for this meter.

Factory re-calibration of any meter involves returning it to "as new" specification in all respects. So whatever measurements and tolerances the original meter was manufactured to are met with the re-calibrated meter. If the meter is filter-based, that generally will mean replacing all the filters and "re-tuning" the meter to meet the same specs as when the meter was new.

The other thing about characterization/profiling... it doesn't improve the inherrent TOLERANCES of the meter. If the meter's spec is +/- 1 fL when measuring blue at 10 fL, it's not going to be any better after characterization or factory recalibration. (not that the meter's specs would ever be expressed that way... it's just an example)

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #14 of 14 Old 09-01-2012, 08:31 PM
Advanced Member
 
Smackrabbit's Avatar
 
Join Date: Sep 2001
Location: Portland, OR, USA
Posts: 893
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 30
Quote:
Originally Posted by PlasmaPZ80U View Post

A few questions:
Did you put a quality dessicant in the bag?
What do you mean by 'calibrated' i1Pro?

Sorry, I've been on vacation. I kept a quality desiccant (the metal pack that SpectraCal and others sell) inside a sealed bag, which was inside another bag, for the whole time with the meter, and still had the drift. I meant the i1Pro is NIST Certified, not calibrated, that was my typo. It has an average dE of 0.4 and a maximum dE of 1.0.

Chris Heinonen
Senior Editor, Secrets of Home Theater and High Fidelity, www.hometheaterhifi.com
Displays Editor, AnandTech.com
Contributor, HDGuru.com and Wirecutter.com
ISF Level II Certified Calibrator, ReferenceHomeTheater.com
Smackrabbit is offline  
Reply Display Calibration

User Tag List

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


Forum Jump: 

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

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