RGB to CMYK conversion, being dependent on the printing conditions, SHOULD be
done by the
printer (meaning the company with the offset press, NOT meaning a little beige
box from
HP). But it's not a trivial task, and no one gets paid for it. Printers often
have enough
economic clout that they are able to refuse to do this free work, and it ends
up in the
hands of the designers. They often have more power than the photographers, so
the buck gets
passed again. The result is that the person who is least likely to have access
to all the
necessary information to do the job right is the person it gets foisted off on.
This is
also the person who gets blamed when the client suddenly decides to move the ad
from a
cast-coated publication to a newsprint publication and can't understand why the
photo now
looks like s - - -.
Sometimes you're better off doing corrections in a system's primary colors - if
you've got
a magenta problem, it may be more intuitive to fix with one curve in CMYK than
two curves
in RGB. Also, doing the CMYK conversion yourself gives you a chance to tweak
areas of the
image particularly compromised by the conversion process.
At the very least you need to know enough about CMYK to understand why that
bright
saturated green that "makes" a particular picture will never, ever, survive on
a 4-color
press.
Having the device do the conversion for you makes the (absolutely nuts)
assumption that the
device will get it right. Our Epson 3000 is incapable of delivering a C-0 M-0
Y-100 swatch
- even if you're printing from a CMYK, the driver converts to RGB, then back to
CMYK, in
the process introducing at least 5% cyan. Does it look OK in 95% of the images
output?
Sure! But images that want a nice saturated yellow just can't get it.
Roger Smith wrote:
> At 1:33 PM +0200 10/23/02, Anthony Atkielski wrote:
> >
> >Blind CMYK is a really bad idea, since CMYK conversions must always be done
> >with a specific output device in mind. If the real output device isn't the
> >same, image quality is always lost, and it cannot be recovered.
>
> At 1:33 PM +0200 10/23/02, Anthony Atkielski wrote:
> >It's up to the end user to convert the image for printing, since
> >every end user will have different requirements for CMYK conversion. It's
> >illogical to expect a photographer to do this, and any customer wanting it
> >must be pretty clueless about how CMYK works.
>
> At 2:28 PM +0100 10/23/02, Tony Sleep wrote:
> >Yes, I've come across this, invariably from people who do not have a clue.
> >Well, they are the boss, but you'll get the blame and lose the client when
> >they fail to listen to protestations that generic CMYK is not a good idea.
> >
> >OTOH if you insist that they'll have to tell you a bunch of parameters they
> >know nothing about for it to work properly, they will likely think you are
> >just being difficult and never use you again.
> >
> >Basically, you are stuffed whatever you do.
>
> I wrote to Tony (and others) a year or so ago about this very
> question. It seems that almost very time I am requested to submit
> colour files to a publisher they demand CMYK. Enclosed are typical
> instructions from a digital art web site, instructions which are
> required by a number of scientific journals. As a scientific
> photographer in a university biology department, I have to follow
> these guidelines for submission of articles.
>
> Regards,
> Roger Smith
>
>
>
> Color space requirements. All digital art submitted must be bitmap
> (Monochrome), grayscale, or CMYK.
>
> Graphics in the RGB color space (or Indexed color) will not separate
> correctly. They are very difficult to detect before plating or even
> going to press.
>
> It is extremely important to check every scan/file for correct color
> format before saving and submitting your work.
>
----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or
body
