Camera input difference profiling

[Pascal de Bruijn]

Hi,

Quite a while ago I asked about abusing Argyll to profile the difference
between two images. For example a JPEG generated by camera firmware and
linear output from a RAW converter.

I've been rewriting my old script, and I'm running into an issue with
colprof... I uploaded a test case here:

https://encrypted.pcode.nl/files/temp/argyll/

At the end of the script colprof gives me:

No of test patches = 288
Estimating white point
colprof: Error - 1, set_icxLuLut: can't handle test points without a white
patch

Anybody a clue if I'm doing something wrong?

Regards,
Pascal de Bruijn

PS: Sorry for the double post the other day, this was related to the GMail
outage yesterday.

[Graeme Gill]

colprof: Error - 1, set_icxLuLut: can't handle test points without a white patch
Anybody a clue if I'm doing something wrong?

Hi,
the problem is as reported. Your .ti3 doesn't contain a white patch
(RGB = 100%). Since ICC profiles store data as white point relative, something
needs to set the white, and colprof looks for at least one white patch.

Graeme.