|
From: Marti M. <ma...@li...> - 2003-05-21 08:17:45
|
Hi, Always use gamma 2.2 when possible, gamma 1.0 does induce quantization errors on dark zones unless you are using 16 bits. The profiler will compensate for any gamma difference. So, you will end with almost identical results, no matter which gamma being used. Remember, profiler task is to provide *colorimetric* match, any tweaks after that can be applied by using abstract profiles on the transform chain. Regards, Marti Maria The little cms project http://www.littlecms.com ma...@li... ----- Original Message ----- From: "Richard Ross" <rs...@ea...> To: <lcm...@li...> Sent: Thursday, May 15, 2003 10:47 PM Subject: [Lcms-user] IT8 target gamma?? Hi all, This is either a stupid question or something important to understand (that I don't yet). I am calibrating an Epson consumer flatbed/transparancy scanner with a Wolf Faust 35mm film target, using LittleCMS windows profilers of course. I find that the IT8 target basically has a 'linear' gamma and I need to make a significant boost in gamma (levels/curves in Photoshop) to render a scanned image normal. More specifically I generate a scanner calibration by 1) scanning with gamma = 1 in which the raw scan looks darker and flat and 2) scanning with gamma = 2.2 or so in which the raw scan looks correct. Now assigning/converting to the appropriate generated scanner profile for these cases yields a very similar result. This is good since it means the two profiles work. However this result has an apparently low gamma (much closer to linear than 2.2). Is this typical, i.e. is the IT8 calibration linear?? Or am I doing something totally wrong. Incidentally the Epson scanners come with generic profiles seem to convert gamma=1 scans to gamma=2.2 results. Thanks for any help with this, Richard ------------------------------------------------------- Enterprise Linux Forum Conference & Expo, June 4-6, 2003, Santa Clara The only event dedicated to issues related to Linux enterprise solutions www.enterpriselinuxforum.com _______________________________________________ Lcms-user mailing list Lcm...@li... https://lists.sourceforge.net/lists/listinfo/lcms-user |