Hi I found in profcheck wrong readings, and do it again. Results are much better, peak error is 6.3. Thank you very much. But I have also something that I don't understand. Colors that I working with them is really different, very desaturated colors. In this case with red color space I've got very saturated red, probably again because of wrong reading. I read from ceramic tiles, and maybe this is reson because of strange colors and difficult readings. In this case printer is RGB, with CMYK printers I don't have problems at all. Thank you again Dimitrije Zivkovic On Apr 17, 2012, at 1:48 AM, Graeme Gill wrote: > Dimitrije Zivkovic wrote: >> After I read patches, and making a first profile i've got big peak error >> over 33. >> >> I try to find how to fix reading erros, and find that refine can help. >> I found also scenario for using refine, but I really don't understand. > > Hi, > before jumping into refine, I'd try and figure out why you > are getting such a large fit error. profcheck -v can point to which > test patch cause this error, and it is worth re-measuring > that patches color value. ie.e Are you sure the chart hasn't been > mis-read for instance ? > > Graeme Gill. >