Herb Chong wrote: > according to Bruce, the operator had no color profile to work with for > their printer so they were unable to do exact color matching.
The lab I use has a pair of D-Labs (one each of the 2 and the 3). I intend to ask them if they have colour profiles so I can do soft-proofing at home before sending them my files. I certainly hope I can get profiles from them, but who knows what goes on inside those big boxes. > i supplied > my images in sRGB profile and if they had a profile for conversion or they > sent it to me, i could have sent my files in the Agfa D-labs profile. that > should ensure better color match. The D-Lab works in the sRGB colour space. If you embed profiles in your files then it may be able to automatically convert from other spaces (Adobe RGB etc). But I'm not sure about that. > for absolute color accuracy, the operator would > have to look at my prints as reference or to have the files come in > already set to the correct color profile for the photographic printer. Also your end would have to be set up perfectly as well; ie calibrated scanner and monitor. There are always limitations, though. The simple fact that a monitor is emissive (additive) and a print is reflective (subtractive) is a headache in itself. The colour rendition of a print is dependent on the light source under which it is held. Your perception of the colour on that print varies with your brain's "compensation" for ambient conditions. > i looked at the 7x10 prints with an 8X loupe and compared the two. the dot > pattern is clearly visible in the inkjet prints but there is no dot > pattern visible in the photographic prints. The D-Lab uses a continuous tone process so there should be no dot pattern :) > it is just possible to see the scan lines from the lasers, so overall > resolution is excellent. I had two 12x18" prints done recently. Under a 10x loupe I think I can see what might be scan lines, but my eyesight isn't good enough to tell for sure. But I don't often look at prints under a loupe ;) The D-Lab prints at 400ppi so in theory it should be pretty hard to see the pixels even under a loupe, provided the original file was not resampled (100Mb file for 12x18" 400ppi at 24 bit!). > there are some differences in edge detail though. for some reason, the > D-labs prints show slightly less detail, as if a small amount of > blurring was applied. That sounds really strange. I wonder if that could be due to a lack of paper flatness, or even if its due to the laser optics. The distance from the optics to the centre of the print will be less than the distance from the optics to the edge, so it may be related to DOF. The photos on which my prints are based were "bokeh'd" at the edges so I wouldn't have noticed. > also, if there are dark edges against light backgrounds, there seems to be > a small amount of hard edges added that are not in the inkjet prints, sort > of like an small amount of unsharp mask. looking at the prints at normal > viewing distances, the D-labs output has a slight more snap to it, a > subtle added sharpness, because the edges are slightly sharper. My prints show the same thing, but they came off negatives so I would naturally expect a little unsharp masking to make up for the scanning process. I would hope that digital files were not modified by the D-Lab in this way. On the other hand, any halftone process [eg inkjet] will inevitably lose a little edge sharpness. > one of the files was blown up to about 12x18. on that print, the places > where the hard edges showed up are softened but the subtle tone changes > are lost. there seems to be some image processing taking place. Did your file have sufficient pixel count to print at 12x18? My prints, which came off negs, also show some pretty bad "bokeh'd" areas and a bit of noise but I assumed that this was due to the scanner which is only 2k x 3k and is designed primarily for speed. In my opinion 2k x 3k is pushing its luck for a 12x18" print. > i would have no trouble being satisfied with the quality of the prints i > received. owning a large format Epson printer means i can produce the > prints i want up to 12x18 with excellent quality. The only thing stopping me from buying an Epson printer is the cost. I won't be making nearly enough prints to justify its purchase: my walls just aren't big enough. If I ever sell prints, I can just pass the lab's printing charge on and its one less process that I have to look after. > color matching is the only sticking point, but that is a procedural and > training thing that the operator needs to know about if you want to blind > submit files for printing. Colour accuracy is a real pain and its not just the lab's responsibility. If you really want "right first time" prints from a file then a printer profile would be very helpful for soft-proofing at your end, provided your monitor has also been profiled. However that depends on a D-Lab profile being available, and if it is doing any automatic adjustments then the profile would not be valid (BTW you'd need separate profiles for each type of paper). I will email my lab now and see if they can supply me with profiles for their D-Lab. I won't hold my breath. But I just calibrated and profiled my monitor (with a borrowed ColorVision Spyder) and am rather keen to be able to do soft-proofing. BTW a calibrated monitor looks fantastic. The colour images on PUG look even better, but unfortunately the B&W pics look much the same ;) Cheers, - Dave http://www.digistar.com/~dmann/
