Here's the background:
I've recently upgraded from an HPZ3100 to a Canon 8300, and I'm trying to print images that I worked up with the HP to look more or less the same on the new Canon without having to do too much (or any) further editing. I created profiles with the HP with its built-in spectrophotometer, and I bought a set of custom profiles from a profile-making service for the new Canon. Because the Canon generally has a larger gamut than the HP, printing the images to the Canon just using the new Canon profiles does not give identical results as the same files printed to the HP (I assume here that if your destination space is different, you don't get the same results with the rendering when you have a lot out of gamut colors).
Next I tried converting the image profile in PS to the old HP printer profile, then printed to the Canon with the Canon custom profile (I use absolute colorimetric when printing to the Canon under the assumption that since I've already converted to the HP printer color space I want to keep as many colors as possible unchanged when printing to the Canon. Since the Canon generally has a larger color space I shouldn't expect many, if any, colors to be changed during the rendering with abs. color.).
Here's the problem:
I can't match the colors on the Canon that I was getting with the HP. So, I started looking at some of the data in the various profiles involved, and I noticed a significant difference in the associated white point values. A lot of my printing involves very light colors, and these need to be accurately reproduced. My current theory is that if one or more of the profiles has relatively inaccurate white point, it's going to result in color casts, at least in the lighter tones, when I print to the Canon.
The white point values from the profiles for the paper with the worst color cast problem are (in Lab values):
HP profile: L: 96, a: -1, b: 1
Custom Canon profile: L: 97, a: 1, b: -7
Now, I'm assuming that even though these profiles were created for different printers, and using different profiling hardware, the paper being profiled is the same so the values for the white point should be pretty similar, but these are not the same (especially the b channel). When I print to the Canon using the proofing method of first converting to the HP profile in PS, I get a print with a yellow color cast, which seems consistent with these white point values, since to compensate for a change from b=-7 to b=1 you have to add yellow.
Here are my questions:
1) Is it an issue that the white points in these profiles are not more or less the same? Does it mean one of them is bad?
2) Could the difference in white points in these profiles explain the issue of a yellow color cast?
3) I tend to believe the custom Canon profile is the more accurate profile, since this particular paper is bright white, and it makes some sense that the b value of the white point should be negative (though I don't know if it should be that negative), so if I have a bad HP profile, my idea for a solution would be to edit the HP profile so that its white point values match that of the custom Canon profile. Is my reasoning here sound? I haven't tried this yet because I haven't found a profile editor that's not several hundred dollars, which leads me to:
4) Are there any profile editors that do not cost hundreds of dollars?