Wednesday, June 22, 2005

Request for help in establishing a Monitor Calibration Quality Evaluation Protocol

As you've noticed, I write a fair bit about color management, and I'd like to get my opinions about LCD monitor calibration on a more solid footing. So, I would like to establish a test protocol that I could run on my machines to test calibrators. This is journalism, not rocket science, but I still want to be able to do things decently.

I was thinking of first adopting a methodology similar to that of Dr. Abhay Sharma for the WMU Profiling Review.

If I understand rightly, this means doing a whitepoint measurement and then computing delta E for colorchecker squares. The evaluation results can then be the avg and max delta E measured.

The instrumentation I now have is a Gretag Eyeone spectrophotometer. I wonder whether this would be sufficient to act as a reference for this purpose, seeing I'm doing journalism and not science? Also, I don't know how to do the reading: Which software can put the instrument in a state to read in the screen squares, then make measurements ? I guess I need to somehow establish the base 100L value then compute the delta E for each colorchecker square, can anyone tell me exactly how set up software to do this ? It would be nice if Gretag offered some support, eg. made their SDK available ...Maybe some existing package can already do this ?

Specialists will rightly assume I don't know what I'm doing; however, Color Management is now trickling down to consumer level so we need consumer-level testing of this type.

I am cross-posting to the blog, and the Colorsync list, so you can answer here or comment there if you wish. I would appreciate input from all the professional members of the community.

1 comment :

Lars Vinberg said...

I'm asking myself the same thing, as I just started using an Eye-One Pro with Match. Profiles seem to be good in quality but it would be nice to see a delta-E measurement.

Yesterday I built a larger display reference file, 1435 patches took an hour to measure, there was a slight increase in gamut volume and LUT curves look perhaps smoother. In terms of quality I really don't know though.