Hi Robert,
robert bergs wrote:
they say does not do any processing. I have been experimenting with all of them - so far 'raw' has given the best result's, although the gamma is 1.0 which is too low
IMO not generally, if 16bpp are used. Basically gamma=1 does reflect the native characterists of most scanner's sensors and A/D converters (at least approximately). However both, the LCMS measurement tool (which can handle 8bpp only), and the current qtscannerprofiler version do have an accuracy problem when they create profiles from gamma 1.0 scans.
When I wrote my question I did not have the time to explain closer. I've been asking this, because I also notice large color deviations when I attempt to scan for instance colored plastic objects like plastic cards, etc. with my scanner profile generated from an IT8 target - which otherwise works pretty well for scanning photos. Depending on the scanned object, I get cyan instead of green, or red instead of orange, or purple instead of blue, ...The piece that caused the problems was a glossy promotional flyer that was predominantly bright red.
However - though undesired - this is nomal and is a limitation of scanners in general. This effect is called scanner metamerism.
A profile is basically ony valid for scanning the same media type (with the same spectral characteristics) which was used to create the profile. Therefore a profile created from a reflective IT8 target (which is printed photochemically on photo paper) is more or less only valid for scanning photos.
If other media types are scanned with this profile, the reproduced colors may no longer be correct, and for very specific dyes/inks the error can become quite large - as a mentioned in the examples above. Even if I scan inkjet prints with my IT8 generated profile, I do get clearly visible, non-neglectible color casts in the images after applying the profile.
Depending on the scanner's light source and the spectral sensivities of the scanner's sensors, these metamerism effects may be stronger or weaker, so it largely depends on the scanner used.
IMO there also seems to be no reasonable way to compensate for these metamerism effects, except to generate a profile from a target which was printed on the same media type which needs to be scanned. Unfortunately targets on arbitrary media types are usually not available. At least I don't know a source where I could get non-photographic IT8 targets, e.g. printed with a Canon printer and Canon inks on a specific brand of paper, or printed with a HP printer and HP inks, or Epson, etc.
Anyway, as a new profile generated by Marti did improve your colors significantly, your major problem (fortunately) appears to have been a different one.
Regards, Gerhard
------------------------------------------------------- This SF.net email is sponsored by Dice.com. Did you know that Dice has over 25,000 tech jobs available today? From careers in IT to Engineering to Tech Sales, Dice has tech jobs from the best hiring companies. http://www.dice.com/index.epl?rel_code=104 _______________________________________________ Lcms-user mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/lcms-user
