RE: The MESS at the PRESS campaign
RE: The MESS at the PRESS campaign
- Subject: RE: The MESS at the PRESS campaign
- From: Steve Upton <email@hidden>
- Date: Mon, 5 Apr 2004 18:04:32 -0800
At 12:19 AM +0200 4/6/04, Henrik Holmegaard wrote:
>
bruce fraser <email@hidden> writes :
>
>
>The relationship between the print and the proof is that the proof
>
>will show you what the document will look like on the printing condition
>
>which the Dlink uses as source. If the document contains totally incorrect
>
>separations, the proof will reflect that fact.
>
>
The device link assigns an assumed source and then converts into the proof space. How does the person preparing the proof know that the assumed source represents the objects in the page when that person did not harmonize the objects in the page in the first place ?
>
>
Don't get me wrong, there is absolutely nothing wrong with a device link conversion, if and only if all CMYK objects matching the data model / color model of the first space in the linked sequence are also in the _same_ CMYK space.
I think this is an interesting conversation and that device links bear more discussion and widespread understanding but I have a question...
What is the point of this thread?
I don't mean to be abrupt or confrontational, I just want to distill this conversation down to a single thread - if that's possible. I've heard a number of possible threads, including:
- device link profiles are potentially dangerous. The response has been that they are no more dangerous than a typical photo-mechanical proof in that they treat all incoming CMYK as "normalized" whether it is or not. This seems to be more a concern that apps and workflow tools ensure normalization prior to the proofing (or space-changing) transform that links provide, not a problem with links themselves.
- device link profiles are proprietary. My response is to say that they are not. The actual transforms that occur inside profile tables have (with the exception of colorimetric tags in non-link profiles) been left open to interpretation/execution by the profile software manufacturer. Device link tables are no different except that the "edits" tend to be performed algorithmically by the machine more than "by hand". I would not call an inkjet profile proprietary / non-ICC just because an installer edited it for a better match. The machines did their best and the human stepped in later to tweak it. If any profile looks and smells like an ICC profile from the outside, the insides are irrelevant.
- device-link profiles are bad for late-binding workflows. I would venture that they have nothing to do with them. They are either used to get people out of workflow dead-ends or heading into the purposeful dead-end of proofing.
- device link profiles are unusable in a PDF/X workflow. same as above
- device-link profiles are unnecessary as the world is moving to late-binding workflows. Most of the talks I give where I mention PDF/X-1 I find it is new to people, PDF/X-3 is REALLY new. Late-binding is a good aim point. Today, on this side of the pond, it is rare.
- device-link profiles are unnecessary as the world is moving to standardized separations. TR001 and TR004 are good starts. Thankfully TR001 was captured in a high-quality profile distributed by Adobe. I think it is fair to say that the only separation standard we see on this side of the pond is whatever Photoshop defaults to. Good as the profiles are, the CMYK produced is still inappropriate to upwards of 50% of the organizations it's handed to.
- device-link profiles reduce the over-all invert-ability of the color workflow. I think this is another way of stating the late-binding workflow problem. Truthfully, I was not aware of an overall need/desire for workflow inversion. Late-binding, sure.
- the US is woefully behind in the widespread adoption of printing standards. Yup, sure, next topic. It's true but talked to death.
<getting on soapbox>
I suppose my main concern is when users like Bill Whitfield read these posts and get thrown off course. Here's a guy who's doing his best to do his homework and find a good solution for his organization. Let's please be careful of these people by clearly stating when our public discussions are "big-picture, forward-thinking, industry-support" discussions. And make sure we don't throw the baby out with the bath water.
Bill, (and others in his position) the capabilities of profiles built and used within the ICC spec are up to the challenge of high-quality end-to-end color workflows and contract-level proofing. Do they need to be tweaked & edited sometimes? yes. Are we hoping that research and development will continue to improve them? Yes! This does NOT mean that ICC profiles or the ICC spec are irreparably flawed or bad. Not up to total reliance on machine measuring and calculation, perhaps. But not bad.
Vendors sometimes choose to go outside ICC or other specifications to improve the final result, and sometimes they do it for competitive reasons. Just like the "cowboy" printers on this side of the pond, sometimes creating solutions (or print jobs) outside of a standard makes for more exciting & compelling marketing copy. It breaks down the cohesiveness of the industry while satisfying shorter-term goals.
My experience with the ICC organization has been eye-opening. They are a group of open-minded, helpful people who have the INDUSTRY's needs at heart. They are working hard on continually improving the current spec for print purposes while expanding it to support exciting emerging technologies like digital cinema and high dynamic range imaging.
Let's applaud those who join industry organizations and support the companies (and printers) that work to support standards by buying their products. Specs like the ICC were meant to evolve as we learn how to do things better.
<getting off soapbox now>
Regards,
Steve
________________________________________________________________________
o Steve Upton CHROMiX www.chromix.com
o (hueman) 866.CHROMiX
o email@hidden 206.985.6837
o ColorGear ColorThink ColorValet ColorSmarts ProfileCentral
________________________________________________________________________
--
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.