What do I believe: Color Eyes or Spider3 Elite

[cstanfill]

Disclaimer. I'm an enthusiastic amateur + know almost enough to be dangerous.

I've been using a DTP94b with Color Eyes Display pro for calibration for a while, and I like it. Also: my current monitor is an HP LP2475w. Meh,
not the best monitor out there but seemed to be a good choice when I bought it; I'm happy with it. It seems to qualify as a 'wide gamut' monitor
(though not pro-quality by a long shot).

So: I've read numerous opinions that the DTP94 isn't suited to wide gamut monitors, but the Spider3 does a good job. I bought the Spider3 elite package.
I profiled my monitor using the Spider3 software. It proclaimed the profile to be good. Then, for comparison sake, I fired up Color Eyes and validated
the profile. According to Color Eyes, the profile is horrible; it reports an average dE of 26.34 (!). I redid the profile using Color Eyes, and the validation
report shows an average dE of 0.70.

So: What do I believe? the Spider3 Elite software doesn't seem to provide the dE figures (is it hidden somewhere?), but it tells me that the profile
is acceptable. Color Eyes tells a completely different story; it says the profile produced by Spider3 Elite is horrific but claims to have produced a great
profile on its own. At least Color Eyes gives me numbers... but do I believe them? And is there a way to get Spider3 Elite to give me numbers
on profile quality?

For the moment I'm using the Spider3 with Color Eye's profile builder. Comments? Am I crazy/misguided/incompetent, or does the Spider3 software really produce
horribad profiles?

[trinityss]

Hi,

I have a HP LP2475w myself.
I'm using an EyeOne Display2 and Spyder3 Elite.

The panels in a LP2475w seems to be the same (more or less) as in a IP240 Quato monitor.
Use their monitor calibration software (iColor Display) that includes a software calibration matrix and the results will be more "trustfull" (the shape of the gamut will also be leaning more towards a typicall wide gamut profile).

The software also contains a calibration matrix for the DTP94, which Quato sells as well.

Let me know if I need to send my profile to you if you want to compare.

But it is still strange that the validation returns such deviated values...
- Be sure that the settings are equal
- your monitor is warmed-up (min 1hour), ideally you should make a profile with the spyder soft and immediately after do the test in ColorEyes

Maybe the ColorEyes software contains a calibration matrix (you need to make a choice between CRT and LCD, right?) and the spyder soft not...


Kr,
Steven

[jackbingham]

If the values for the other profle you loaded in coloreyes before a validation were not correct that would throw it way off. Also position of the sensor changing can have a huge impact. Monitors regrettably are no where near as uniform as we thing, or would like.  Warm up state of the monitor could be a factor. Finally I think it is fair to assume that two packages may do things differently enough to generate some of the error, and sensor handling could add to it as well.

[Scott Martin]

As Jack pointed out, the calibration settings that were used in Elite were probably different from the settings that were being validated in CEDP. And on top of that, validating in CEDP for other app calibration probably isn't trustworthy. I'm seeing some troubling variations between different Spyder 3 devices that I've never seen with the DTP94.

CEDP's iterative approach to calibration is really worthwhile. I'd calibrate to the same settings in CEDP with both devices. Comparing their validations will have value - as will some visual evaluation. Visual evaluation between Elite/S3 and CEDP/S3 is also in order.

[cstanfill]

As Jack pointed out, the calibration settings that were used in Elite were probably different from the settings that were being validated in CEDP. And on top of that, validating in CEDP for other app calibration probably isn't trustworthy. I'm seeing some troubling variations between different Spyder 3 devices that I've never seen with the DTP94.

CEDP's iterative approach to calibration is really worthwhile. I'd calibrate to the same settings in CEDP with both devices. Comparing their validations will have value - as will some visual evaluation. Visual evaluation between Elite/S3 and CEDP/S3 is also in order.

I found out the reason for the massive difference in profile quality and it was, as Jack pointed out, a difference in settings; I had created the Spyder3 Elite profile with no luminance target but erroneously set a luminance target in CEDP which, duh, resulted in a huge measured error due to a big difference in the L value. When I cleared the luminance target the differences were smaller. CDEP measures a smaller error for its own profile than for the Spyder3 Elite profile (same sensor in both cases, same setup), but the difference is within reasonable bounds. I think I'll go with CDEP because it gives me a quantitative measure of profile quantity and it is plausibly (but not definitely) producing a better profile. Is it really better or is this an artifact of the fact that I'm using CDEP to evaluate the Elite profile? No way to tell. Both profiles look good and the monitor-to-printer correspondence is probably more limited by my awful illumination setup at this point than calibration error (illumination = home office, look what the cat dragged in for light... but it would be serious work to get some proper light into this workspace).

One final tid-bit: I used the Spyder3 Studio colorimeter to profile my 3800 + see if the results were any better (I have a friend who swears this produces better results). I'm probably not very good at this, but when all was said and done my home-brew profile produced results that were visually indistinguishable from the Epson profile. Good to know that I can produce a non-awful profile if I want to use some oddball paper, but clearly not worth the effort with my level of expertise and equipment.

Anyway, as always thanks for the help.