On Sep 12, 2005, at 12:49 PM, [EMAIL PROTECTED] wrote:
Thanks Godfrey. I was wondering why you didn't respond immediately. You've been busy <g>. What you posted before sounded complicated because it was detailed. I tend to oversimplify. I appreciate the clarification and your evaluative tests. I've always wondered if I was losing a lot with my color space. Apparently, if there's any loss, it's minor. I have tried Colorsync color settings with Adobe 98 color space, and the results were not pleasing. I will try the North America Prepress settings one of these days, but I'm always a little leer of change when things are going well.

You're welcome. I tend to err on the side of detail and precision... it's the Mathematician/Engineer breaking out. ;-)

The only way to really know what works best is to test and experiment with the printing system you're going to be using, and with photos typical of what you're going to print. That is really the bottom line.

Regards the differences between A-RGB and sRGB colorspaces:

The test photo was an average snapshot/group photo taken in my partner's house in Tijuana ... available light, a lot of light color walls, four people's skin tones, camera triggered remotely. Not stressing the color envelope very much. I happened to be looking at it and wanted a few prints anyway...

The R-G toes on that set of comparison histograms demonstrate some clipping in those channels on the conversion to sRGB, but the effect on the photo is relatively small because the photo doesn't stress reds and greens very strongly. Only the skin tones move a little bit.

I'll probably repeat the test with some roses and leaves and a Macbeth color checker chart (I do this once in a bit to see if the inks/paper calibrations are working...) once I can find my chart again. It seems to have gone walkies since my move in April. Anyway, I suspect that the chart will show more significant deviations in the print output between the sRGB and Adobe RGB profiles. The A-RGB profile does represent a substantially larger gamut.

The crux, though, is how well the profile conversion to the printer does the job. The latest print drivers seem to be working more smoothly.CS2's print system does seem to do a better job, with less adjustment necessary, than CS did. Perhaps Adobe's continued development has improved things as well.

Godfrey



Reply via email to