I have a couple of questions on APS, as I am considering purchasing the Z3100:
1) Can APS use a target like the 1728 patch Atkinson target? If not, what is the largest target that can be used? I have found (for other printers) that best results seem to be with a moderately large number of patches. I thought 1728 was better than 918, but the 4096 patch target didn't produce a profile that was any better as far as I could tell.
2) I don't care for the Logo Colorful perceptual intent of Profilemaker, as I have found that it squashes the saturated colors too much. Prefer the Logo Chroma Plus. Would there be any problem using a profile created with the APS and reading the data from the profile into Profilemaker to regenerate the profile with my preferred rendering intent?
1) I don't think so. At least that is what I was told by HP and GM; APS is a HP product and not supported by GM. I hope they change their mind on that, or maybe I won't upgrade to the next version of Profiler Pro. As others have mentioned, the z3100 test chart is a series of large hexagon patchs; since the spectro rides above the surface of the paper, the patches need to be larger than in other charts. I don't know how GM/Xrite generated the chart, but it is not an option in Profiler Pro's test Chart Generator to generate a hex patch shape. Anybody know how it was done, or how to generate a larger hex patch chart from a reference file?
1a) This isn't exactly what you were asking, and is obvious, but one can print any standard chart that you have a reference file for (like the Atkinson charts have for i1, Spectrolino, i/o, etc) and read it on a separate spectro. That is what I do when I have the time or inclination (I use a iccolor210,no uv cut filter). But I've really been quite happy with the standard TC918 rgb chart.
2) You can load the reference file and profile made from APS into ProfileMaker 5.0.8, and generate a new profile using any setting you desire. I tried it; it works. The profiles also open in Measure Tool for comparing and averaging. I didn't try Editor, but they will open there too.
BTW, I had a rather interesting visit at the Gretag booth at PhotoExpo a couple of shows ago. In discussing with a GM engineer about the soon to be released i/o, we got on the topic of optimum patch numbers for charts. He said it was the opinion of some at GM that less was better. And he meant way less; but not less than 64! What? Yeah, he explained that the curve was smoother, and that too many patches could create 'choppy' results. Just to prove his point, he pulled out a binder with the same images printed using profiles created from different patch amounts. And then proceeded to show how much more shadow detail was visible in the print made from a profile created using 64 patches. It was true about the shadow detail; but the print was flat, dead and lifeless; no snap. My wife has a more sensitive eye than I, and she couldn't believe it. The larger patch profile print was so much better, and the shadow detail was still there, but more compressed. It was hard to believer we were all looking at the same prints at the same time. So it just goes to show that different people have different tastes. Profiling seems not to be an exact science, but also an (improving) art at this stage. For example, you like ChromaPlus, and so did I after Classic. But now I prefer the newest, Colorful. Theoretically, only the Perceptual rendering intent should have been effected, but I think there are other differences.
Does one need APS for quality prints? Not if one only uses HP papers. Since the printer can be calibrated to a known state (less than one delta E), the HP canned profiles (or other's custom profiles) should work fine. i'd bet the differences between the HP canned profile and a custom APS are going to be very small. And after all, if it is a critical photo, aren't you going to print a proof first anyway and then make fine adjustments in Photoshop? I wish I could generate a custom profile on HP media (but the free paper hasn't arrived) and compare the results to the HP APS TC918 canned profile in Measure tool. I did do a comparison between a TC918 chart, APS custom profile from the z3100, and one printed and generated by Profiler Pro (on Epson Premium Semimatte Paper), using D50-Large-Neutral Grey-Colorful settings, and got a delta2000 E of under 1 (I think it was under .6, actually), which really suprised me. And, if the calibration between printers is not consistent enough (my Epsons are not), even the ImagePrint Rip canned profiles are not going to be the ultimate, no matter how many patches or special spectros are used. But the z3100 CAN be calibrated for a paper accurately and easily; which takes into account ink density and color changes, humidity, temp, paper batch variations, head wear, and on and on. So you don't have to make large new profiles when something changes (like for my Epsons), just do a re-calibration. Wonderful.