color management & color correction,was: Color settings and upgrades
Dan Margulis <76270.1033@...>
Chris, James, Andrew,
While this thread has had valuable information it seems to have become
somewhat repetitive and unduly personal. I think the color settings thread
has played itself out. Gary was only asking how to make PS 5 behave like PS
4 and it turned into a referendum on color management. The thread on color
management and color correction may have some life left in it but I would
think that the three of you should withdraw for about 24 hours to let
things cool down. James, I don't see an objectionable personal tone in any
of Chris's posts.
I haven't participated in this one, largely because I believe the
conversation is academic, since the practical traditional side represented
by James has clearly won in the marketplace. It's worthy of note, however,
that Andrew and James basically agree on about 90% of what they are saying
and a lot of the arguments are really about semantics.
To respond briefly to the comments about my own practices: in color
correction, an overly precise knowledge of output conditions is not all
that big of an advantage. If a file is being prepared, say, for newspaper
reproduction, that's as accurate a description as is usually needed. IOW, I
would expect that having seen the actual printed product, at least 95% of
the time I wouldn't have corrected it any differently had I known more
about the specific newspaper.
Obviously, though, to cater to the <5%, it would be nice to do some kind of
finetuning. So, traditionally one edits CMYK Setup as one gains better
understanding of what the print conditions actually are. As to how many
passes are necessary to get that understanding, it wasn't clear whether
people were referring to a single image or to a group. When I calibrate to
some new condition I usually prefer a sheet of 12 or 15 typical images.
Typically I need to have that run three times before I am satisfied.
However, after two times I would certainly have a better profile than one
generated by machine.
If, however, we are discussing one image only, then I agree, after only two
passes I don't have enough to make more than a rough estimate of actual