This is a horror story that many people in the industry could have written,
myself included (although I was usually submitting reflective art, not
digital). One answer might be to go in and work the Macs yourself (but I've
never seen a repro house that would allow that). Since I mostly worked on
the Client side of the street, I had a bit more "clout" than a 'mere'
photographer would--but I always tried to extend that clout to getting the
best reproduction of the photographer's work (for which I'd paid a princely
sum, I might add ;-) ). I got my ass kicked around a lot, too, but (if I
might boast) I gave as good as I got, most of the time. I wasn't ashamed of
most of the results (but I sure as hell heard about the others, let me tell
you)--lost a job or two in the process; that's the 'down'side.
Is there an answer? Yeah, when Profits and Repro Houses get reasonable, and
pigs fly, there probably will be. Until that happens, all you can do is the
best you can do, and hang tough. Wish I could offer better.
Best regards--LRA
>From: TonySleep@halftone.co.uk (Tony Sleep)
>Reply-To: filmscanners@halftone.co.uk
>To: filmscanners@halftone.co.uk
>Subject: Re: filmscanners: Repro issues (was Which Buggy Software?)
>Date: Tue, 17 Jul 2001 02:48 +0100 (BST)
>
>On Sun, 15 Jul 2001 19:42:49 -0500 Maris V. Lidaka, Sr.
>(mlidaka@ameritech.net) wrote:
>
> >
> > Dan's response would be that most repro houses don't use embedded color
> > profiles anyway - they do it the old-fashioned way. If he's wrong,
> > please
> > tell him ;)
>
>He's largely right, although I just had a magazine repro screw-up this week
>which seems likely to be explained as the repro house doing
>manual adjustments to a tagged image on a (gamma=1.0 by the look of it)
>input workstation which didn't speak ICC, and then sending it to an
>pagesetter which *did* - with (the now completely inappropriate) Colormatch
>RGB tag still in place that indicated, among other things, that the gamma
>was 1.8. The postmortem is continuing... fortunately, on this occasion the
>repro house concerned is keen to address the problems and open to
>discussion. I spent half of Saturday trying to figure out what had gone
>wrong and emailing the Art Ed. Fortunately both she and the editor had seen
>the scans on their own calibrated screens in PS, before they went to repro,
>else I would be getting the blame.
>
>The trouble is that even though they said they loved the pics, next time
>they might avoid the problem by giving the work to someone who turns in
>work on E6 instead. If I'd wanted to shoot it like that I'd have done so,
>but I use this stuff to get better pics in worse circumstances. It works,
>they agree - but if the repro buggers it, it's a chocolate teapot.
>
>Whatever, it's a nightmare. ICC tags are not a panacea, and can cause extra
>problems - as they seem to have done on this occasion.
>
>OTOH if you don't use them, whatever you intended the image to look like is
>out of your control entirely. You had better supply a print or tranny
>instead.
>
>Some repro houses never seem to have problems, others have been so
>disastrous I have lost clients as a result. Faced with a choice between a
>photographer and a repro house, the repro house wins, if only for
>contractual reasons.
>
>Basically Margulis is right IME. Repro houses don't need to use nor
>understand ICC, and wherever they do, it's because they have had to find
>some way of coping with 'externally supplied' scans. In UK this is rare, at
>least among repro houses working for 50,000+ circulation magazines. Yet
>this problem is not going to go away, since there are good (creative
>control) reasons for photographers to scan and supply images in dig format.
>
>Right now, it is safer to supply untagged files and trust that others in
>the chain are capable of sensible judgements about what looks right. Often
>they aren't, as printers are skilled at matching scans to images, not
>imagination.
>
>Also many repro houses want to keep every bit of scanning business, and
>have good reasons to portray photographer-supplied scans as inferior, risky
>and a route to terrible results. It doesn't help that a lot are, of course.
>But it's distressing to get clients, do a job they are happy with, and then
>lose them because the repro goes to shit. I don't know what the answer is.
>I've tried supplying Epson proofs as references, I've tried supplying
>inkjets as final artwork (I'm never totally happy with either, and this
>just isn't practical on short deadlines/email delivery, as this job was),
>I've tried tagged and untagged files. Sod's law rules, and I doubt Margulis
>has any failsafe answers either.
>
>Regards
>
>Tony Sleep
> - Online portfolio & exhibit; + film scanner info
>& comparisons
_________________________________________________________________
Get your FREE download of MSN Explorer at
