Re: New EyeOne ruler and soft case
Re: New EyeOne ruler and soft case
- Subject: Re: New EyeOne ruler and soft case
- From: Armand Rosenberg <email@hidden>
- Date: Thu, 3 Nov 2005 16:19:41 -0500
I came up with the same "creative interpretation of profile creation"
as Bob Frost, independently, and I firmly believe in it. If the
iterative process fails due to printer or measurement fluctuations as
Ernst Dinkla suggests, that is not a problem that can be solved by
profiling or profile editing, in principle. With averaging of
measurements, any individual print can deviate from the "average"
measurements used to create the profile, which is a problem if the
width of the measurement distribution is in the noticeable regime.
Printer inconsistency aside (since that really cannot be fully solved
by profiling), I have never seen a profile that totally satisfied my
color matching needs, no matter how expensive or sophisticated the
profile creation package and instrument (which is not to suggest that
some aren't better than others). I suspect (but cannot prove) that an
iterative process should give excellent results in very few passes. I
also suspect that some types of measurement errors could be reduced
in such a process, allowing for a less expensive (if somewhat more
time-consuming) path to a high-quality profile.
Perhaps Roger Breton can comment on his experience.
Armand
At 12:05 PM -0800 11/3/05, email@hidden wrote:
Bob Frost wrote:
> Some years ago I suggested that what we needed was a profiling program
that did the usual - print target, read target, make profile, but then
went one (or many) steps further. Print the target again, using the new
profile, read it and let the program see where it went wrong first time,
and then correct itself. You could do several iterations of print new
target with latest profile, and correct it, each time getting closer to
perfection.
I was told at the time that this was not possible, or if it was, it
would be far too expensive, but I still don't see why.
Roger Breton has written about the iterative profiling he has
done. Print Open V5. So it exists in profiling software. The
main problem could be that the fluctuations in printer output
are higher than the precision you try to create with the
iterations. And the method of measurements has to be very
consistent too. The only answer then is more averaging: more
target prints, more readings. It gets as slow as building a
pyramid and in the end you still have the inconsistency of the
printer.
--
Ernst Dinkla
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden