Calibrite report wrong?

[cristian]

Hello, thank you for this nice forum, with a lot of information. I have a monitor, and I calibrated it with a new calibrite Display Pro HL, on windows.

After calibration I run the validation with the color checker patches, and I get at first glance very good results, but it seems to me that they are wrong. I am attaching part of the report, and as you can see the b* is totally different between target and measured, but the DE calculated is very low... Do you have any idea of the possible reason?

Thank you

Cristian

[Damien Symonds]

Ignore that nonsense.

Does your screen match your pro lab prints, or not?

[Damien Symonds]

@cristian?

[cristian]

Hello, thanks for the answer. Actually at the moment I am not interested in the prints, but in the numbers. I am doing a scientific research for which I need a calibrated monitor, but I don't know if trusting it or not. I think I will try with another version of the software, or a third part software... I was wondering if other people met this weird behavior

[Damien Symonds]

I think you'd be better to stick with the same software, but try it on a range of screens.

[Damien Symonds]

I think perhaps you misunderstand the purpose of calibration?  Calibration exists BECAUSE screens aren't perfect.  The calibration process finds those deltas (which exist, but are different, for every single screen in the world) and corrects for them.

[cristian]

Thank you again for your answer and effort to think to a possible cause. I agree with you that screens are not perfect, I don't understand why a simple operation as calculating a DE is totally wrong (I tried it on another screen, and also updated to the new calibrite version 2.0, getting again wrong results).

I was curious to see reports from other people to see if those are correct. I am a computer scientist with a scientific approach... numbers are numbers and don't lie (usually!)

[Damien Symonds]

No, the results aren't "wrong".  They just are what they are.  That Delta column is just telling you how far each screen varies from the theoretical target that the ICC has set.

[cristian]

Hello Daniel, thank you again for your message. I made other tests, using D50 as white point for the display instead of D65 as done in the previous tests.

I am attaching the new report, and as you can see the DE calculated is again incorrect (DE between TARGET and MEASURED).

Futhermore, I found that the TARGET CIELAB values are identical to the previous report, as you can see from the first image in the firt post (and also to other two reports I made with different monitors). So it seems to me that the report is showing wrong TARGET CIELAB values (always the same - I think they should change according to the profile), but hopefully the DE calculated is between the correct DE measured and calculculated. This is of course my understanding.

If I discover something new, I will update.

Thanks,

Best,

Cristian   

[Damien Symonds]

You're searching for evidence of a problem where no problem exists.

[cristian]

Hello Damien,

why do you say there is no problem? if a validation report is wrong, I see a problem: I didn't use a free software, but I purchased something from a well-know company, and I paid it 200$, I expect it works.

Looking at first row: I made the calculation: the DE76 is 4.03, the DE2000 is 2.74, the CMC is 3.23. The report says it is 0.53. So it is wrong, therefore there is a problem.

I made other tests with other instruments and I can say the profile generated by the calibrite colorimeter is quite fine, so it seems the colorimeter does what is supposed to do, but I spent few days to figured it out. I wrote to Calibrite to ask information about this, and I received no answer. This is also a problem for me. 

Anyway, I don't want to create a case, just ask if others found this behaviour. For now I can say the profile is fine, and for me this is enough.

Thanks for follow the topic