Re: Profiler PRO
Re: Profiler PRO
- Subject: Re: Profiler PRO
- From: Rudy Vonk <email@hidden>
- Date: Fri, 31 Aug 2001 17:58:35 +0200
C. David Tobie and I will, I hope, amicably agree to differ, and I
apologize to all the other list members for the fact that my post he has
most recently replied to has not yet appeared in the list - apparently
because I ran into the same new "5K message limit" I suspect prompted
David to divide an earlier post of his into two...
email@hidden wrote:
>
Okay, this is getting more interesting now, instead of mud, we're actually
>
flinging IDEAS!
When I originally said why the ProfilerPRO method of generating CMYK
profiles is unacceptable to me (In unambiguous terms, it is true), I
thought I had given a reasonable explanation why :-)
>
There are a wide number of settings in a CMYM engine, but the end result of
>
them is color seperated to four channels instead of three, and limited to
>
certain amounts of ink in each channel, in total etc...
Semantically correct, as long as you accept that "limited to certain
amounts" includes every possibility including no limitation at all.
>
You are suggesting that seeing what the total ink capability, channel
>
capability, dot gain, bleed level, mottling level, etc... of a device may be
>
is important. I agree... and do all that with a CMYK test file before I
>
choose the settings in ProfilerPRO.
I do the same, but then only (in effect) modify the as yet unprofiled
"printer" so that it won't run, bleed or mottle any more.
>
After all, a spectro can only tell you
>
when the patches stop getting darker, not when they start to run, or bleed,
>
or mottle. This is why pre-profile ink limits are a good idea, and why limits
>
other than what the spectro decides are important. I believe that finding
>
these in advance (one way or another) is important, and eliminates a number
>
of your issues.
I think you are agreeing with me here. Only the spectro will tell me
when the patches stop getting darker - provided of course that I give it
a chance to find out :-) With the ProfilerPro approach you'll never know
because it will have stopped outputting patches before it gets there.
>
However that still leaves us with a given dark purple patch that could be
>
configured using assorted CMYK settings, many of which will not actually
>
exist under a given CMYK configuration.
Well, it would (i.e. exist under a given CMYK configuration) if you'd
let me profile for it!
>
And of a set of patches ProfilerPRO can use every one towards its profile,
>
instead of having to exclude a few hundred as out of gamut;
I do my profiles with 1,441 patches and will happily send you some of my
measurement files to see if you can find "hundreds" of readings that are
effectively out of gamut. In fact, for the kind of material Ernst is
talking about, I can easily distinguish all the "dark purples" with my
naked eyes.
>
and it is reading
>
data defined for the actual seperations it will be working with, not
>
theorizing from raw patches.
In a purely CMYK, press-oriented workflow I grant you that most of my
fundamental misgivings would be mostly irrelevant. In such an
environment, the colors you don't account for in your profile simply
will have disappeared beforehand.
>
Is other software so smart that it is
>
determining whether UCR or GRC should be used, what UCA settings is optimal,
>
What black curve and toe should be used? Nope, its still making you do all
>
this yourself. What do you use to do it? Beats me... but I use that
>
preprofile test file.
We both use a preprofile test unrelated to our profiling packages, and
*we* decide some of these options based on it.
>
When do you do this stuff? This is one of the places
>
where ProfilerPRO "thinks different", and does this stuff in advance of
>
printing the target.
No it doesn't: first you do (preprofile test) and then it inhibits
itself in favor of Photoshop.
>
Yes, it would be possible to miss some of the potential gamut of a device
>
with ProfilerPRO by choosing incorrect settings in advance, but most systems
>
require similar assumptions, at some point.
More importantly, it will sometimes (to put it mildly) be *impossible*
to capture the full gamut choosing *any* settings in advance. This, I
think, is an even shorter definition of my complaint than the previous
one :-)
--
__________________________________
Rudy Vonk
Oviedo, Spain
<email@hidden>
+34 607 354100
You can't always want what you get.
__________________________________