Dan Margulis Applied Color Theory

Proofing Devices

Date: Tue, 04 Feb 2003 08:59:15 -0600
 From: "Cliff White"
Subject: Proofing devices

I have a quick question regarding proofing devices. We recently made the switch to digital shooting, and have been doing our own prepress work. Now we are interested in doing our own proofing for SWOP. I apologize for coming to you without doing much research yet myself, but we only recently came to this decision, and the budget is due.

So, I was hoping that you all could make some recommendations as to good proofing devices and where to find out more info about them. (For example, I don't even know who manufactures "Matchprint." Also, if you could give me a ballpark figure as to what kind of cost we might be looking at, it would be appreciated.

The printer first and foremost needs to be a true SWOP proofing device. Once that we can take on the press run and reliably use to match on press. Next, we would like it to be networkable, so that the 2 photographers and 3 designers could all have access to it from their desktop Macs. The workload of the printer would probably only be about 100-150 prints a month, if that, varying in size from 3 inches wide to 17 inches wide and possibly bigger under isolated circumstances. Another question occurs to me at this point: Can the same printer be used to proof for different presses? Specifically, can it be used to proof for sheetfed as well as SWOP?

Any guidance is appreciated.

Thanks,

CLIFF WHITE
Staff Photographer
Missouri Department of Conservation
(573)751-4115 ext. 3854
____________________________________________________________________________

Date: Tue, 4 Feb 2003 10:42:06 -0500
 From: tbrain
Subject: RE: Proofing devices

Hi, Cliff.

I'm not exactly sure what you mean by a "true SWOP proofing device."

I think there are generally two schools of thought/design when it comes to proofers:

1) The pigments are designed to match inks on press. So, when the proofer lays down solids of Cyan, Magenta, Yellow, and Black, it mimics similar solids printed on press (and with linearization, the tints match too). I believe SWOP even specifies what the "hue" of each ink should be, within a certain tolerance (measured in LAB).

2) The color gamut of the device is large, but can be color-managed to match SWOP. I believe most inkjet proofers fall into this category. If printing a proof with no color management, the color gamut would exceed the color gamut of the press. But, with color management, the gamut is compressed so that it mimics the gamut of a printing press.

Personally, I'd rather have #2. It may not be SWOP out of the box, but it can be set up to match SWOP. More flexibility, and I think these systems are generally cheaper.

Here's a list of proofing systems that SWOP has certified:

http://www.swop.org/certification.html#cert

Thom

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Thomas Brainard
Baesman Printing Corporation
4477 Reynolds Drive
Hilliard, OH 43026
The United States of America
P: 614-771-2300
F: 614-771-2323
http://www.baesman.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
____________________________________________________________________________

Date: Tue, 4 Feb 2003 10:45:26 -0800
 From: Ray Maxwell
Subject: RE: Proofing devices

Hi Cliff,

See this URL for a list of devices that have been certified by SWOP as meeting SWOP requirements:

http://www.swop.org/certification.html#cert

There are many other proofers that include a "SWOP" profile or "TR001" that have not necessarily received SWOP certification.

Ray

Creo
Ray Maxwell | Senior Color Systems Engineer, Inkjet Printing
4225 Kincaid Street  |  Phone (604) 451-2700 ext. 2004
Burnaby, B.C.
Canada V5G 4P5
 ____________________________________________________________________________

 Date: Tue, 4 Feb 2003 12:16:49 -0700
 From: Chris Murphy
Subject: Re: Proofing devices
 
On Tuesday, February 4, 2003, at 08:42 AM, tbrain wrote:

Personally, I'd rather have #2. It may not be SWOP out of the box, but
it can be set up to match SWOP. More flexibility, and I think these
systems are generally cheaper.

And they actually do a better job of simulating actual press behavior such as print sequence and wet trap. This is something analog proofs will not do.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
____________________________________________________________________________

Date: Wed, 05 Feb 2003 09:38:41 -0500
 From: Nick Wheeler
Subject: Re: Proofing devices

on 2/4/03 9:59 AM, Cliff White wrote:

The printer first and foremost needs to be a true SWOP proofing device. Once
that we can take on the press run and reliably use to match on press. Next,
we would like it to be networkable, so that the 2 photographers and 3
designers could all have access to it from their desktop Macs.

Cliff:

I see you are a photographer and might appreciate the advice of another photographer. A photographer who has done everthing wrong, spent way too much money and is somewhat the wiser for the process.

A good press operator is like a good photographer. Just as we look at a scene and automatically know whether the lighting, contrast etc is going to work on film, a good pressman will look at a "proof" and know whether he is going to be able to hit those colors on press, in short whether he can believe it.

It sounds like some of the other traditional proofing tasks, such as looking for trapping and moire problems might not be necessary for you. So if all you need is proof of color "intent" there are any number of inexpensive desktop inkjet printers up to the task.

Just keep in mind that you need to show the press operators separations and a proof that he can "believe". It's not so much in the hardware you use as in the expertise applied to the task. Beware of high end (read expensive) solutions that may actually complicate the situation.

My recommendation is to start with the cheapest tool you can find - say an Epson C-80, learn how to confine its gamut to the SWOP color space and do a good job of proofing your files. You will have only spent a couple hundred bucks on ink, printer and paper. You will have spent a lot more on operator expertise which will pay dividends well into the future in a way that an expensive hardware purchase up front never will. You can use USB printer sharing to drive the thing over the network.

Once you have got that to work you will be well on your way to figuring out what you really need.

Beware fancy expensive solutions to the wrong problem.

Nick Wheeler
____________________________________________________________________________

Date: Wed, 05 Feb 2003 09:04:28 -0600
 From: "Cliff White"
Subject: Re: Proofing devices

When you talk about an inkjet solution, are you talking about something like an EPSON 7000 run behind some kind of RIP, like Epson's Fiery, or Kodak's Matchprint software? or something else? Or is there an even cheaper and easier solution (However, we are not so cheap that we will compromise good proofing reliability).

Thanks,

CLIFF WHITE
Staff Photographer
Missouri Department of Conservation
(573)751-4115 ext. 3854
____________________________________________________________________________

Date: Wed, 05 Feb 2003 14:01:28 -0000
 From: "Stephen Marsh "
Subject: Re: Proofing devices

Additionally, this archive has further points for consideration (ignore the heat <g>).

http: //www.ledet.com/margulis/ACT_postings/ACT-EPSON.TXT

Halftone and sans halftone proofing is another area to consider, as is the case of CTP where ideally a digital proof is generated off the same rasterized data that the plate is imaged off (I don't like the concept of a separate RIP for plates and another to generate a digi proof which supposedly reflects the plate data).

Stephen Marsh.
____________________________________________________________________________

 Date: Wed, 5 Feb 2003 10:04:10 -0700
 From: Jim Donovan
Subject: Re: Proofing devices

hello, on the ripping of files question??? we are going direct to plate with the 10-15 pubs a month and from what we are being told pdf's are on the way out and are not stable ie. stuff can change such as fonts which will result in our customers not paying for their ads. we are looking at spending a fourtune on a program called "digi page" and apperently it rips the files on the front end before we send it to the printer and it creates 1 bit tiffs.(wonder if photoshop can create 1 bit tiffs??) and their is no chance what so ever that something can "change" once at the printer. has anyone heard of this program or actually used it??? hope it doesnt "color manage" (we all know why color management failed,dan has a awsome article on it!!!! thanx again dan!!!) any advice would be very helpfull,thanx!!!!

jim donovan
____________________________________________________________________________

Date: Wed, 5 Feb 2003 11:53:12 -0500
 From: "Remaley, Dan"
Subject: RE: Proofing devices

For a more 'measurable' approach - SWOP recommends using the GATF Proof Comparator to measure the proof relative to Lab, density and dot gain (TVI). The use of this target and other user's guides for platemaking and press control are at

www.homepage.mac.com/gregbassinger/pdfsharing

Dan Remaley/GATF
412.741.6860x450
____________________________________________________________________________

Date: Wed, 05 Feb 2003 10:21:13 -0700
 From: Ron Kelly
Subject: Re: Proofing devices

Nick Wheeler wrote:

My recommendation is to start with the cheapest tool you can find [snip] and learn how to confine its gamut to the SWOP color space

FWIW here's another photographers opinion.

I agree with Nick; don't start with a jet airplane if you can use a Cessna. You're not running a printing company with hundreds of employees and you don't need a system that can serve that.

I would recommend one of the newer Epson or Canon inkjets with a slightly wider carriage (I believe the C-80 epson is only 8.5" wide?) and a bit more capacity (individually replaceable inks.) I've used Epson printers with good success but I'd probably try the Canon if starting now; Epson seems quite intent on getting you hooked up to their dedicated supplies Epson cash cow.

The RIP is the next thing. I have used Adobe PressReady for some time with good success, but it is now unsupported. I would probably try the PowerRip software; the demo looks good and it supports a wide variety of printers.

Then, do this: get a bunch of typical images that are ready to print and color corrected. Your own images are preferable to any others, but if you must you can use someone else's sample images. Assemble them in a 1 page tif file with some color swatches including neutrals, pure colors, shades, anything you want. Make it all up into a page that will serve as your color reference.

Proof this file through a commercial printer you use and trust. Critically evaluate. If neccessary, make corrections and re-proof until you are satisfied that it looks as good as it can. Burn it to CD so that it is archived and unchanging, and keep the corresponding proof.

Now you can calibrate your proofing system. Periodically, or whenever you want to check your system you re-print the calibration tif and compare it to the standard. Obviously this is most representative when sending work to the above mentioned commercial printer; others can vary but you face this problem no matter what your proofing system is.

This will get you to a certain accuracy and consistency which, depending on your expertise and diligence, can be VERY GOOD. The difference between very good and unbeatable is often less than the difference between the variation you'll find in reprinting existing plates; and once you're there the only thing that brings a job in tighter is a press check.

Just my .02.

Ron Kelly
____________________________________________________________________________

Date: Wed, 05 Feb 2003 09:52:03 -0700
 From: Andrew Rodney
Subject: Re: Proofing deviceson 2/5/03 8:04 AM, Cliff White wrote:

When you talk about an inkjet solution, are you talking about something
like an EPSON 7000 run behind some kind of RIP, like Epson's Fiery, or
Kodak's Matchprint software? or something else? Or is there an even
cheaper and easier solution (However, we are not so cheap that we will
compromise good proofing reliability).

A lot depends on what you∂re attempting to proof. ΠRough color proofsΣ (you only want to simulate on an inexpensive ink jet the color files and have no need to deal with the rest of the page) is not expensive or difficult. You∂ll need a profile for the printer/paper and you∂ll need a profile that produced the initial CMYK conversion. Photoshop will do the rest (you∂d do a CMYK to RGB conversion using an Absolute Colorimetric intent) and print. This doesn∂t take the halftone dot or the rest of the page (text/graphics) into account at all. But for many photographers who simply want to show a client how their files will appear on the contract proof, you can get amazingly close to matching the contract proof. Take baby steps and start here. The idea is NOT to forgo a contract proof but to simply predict with a high degree of accuracy what your images will look like before your client produces the contract proof.

If you want to start proofing entire pages (say from Quark or InDesign), it gets more dicey. You need a RIP for one simply to handle the images AND the rest of your page.

Andrew Rodney
____________________________________________________________________________

Date: Wed, 5 Feb 2003 15:42:14 -0700
 From: Chris Murphy
Subject: Re: Proofing devices
 
On Wednesday, February 5, 2003, at 09:52 AM, Andrew Rodney wrote:

If you want to start proofing entire pages (say from Quark or InDesign), it
gets more dicey. You need a RIP for one simply to handle the images AND the
rest of your page.

Actually InDesign has a built-in "RIP" that does a good job on non-PostScript inkjet printers. The catch is that currently InDesign will use QuickDraw drawing commands (on both OS 9 and OS X), and as such these devices are RGB, necessitating a custom RGB profile to convert the ENTIRE document at print time to this RGB space. There are possibly some other limitations to QuickDraw.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
 ____________________________________________________________________________

Date: Thu, 06 Feb 2003 00:19:50 -0600
 From: David Riecks
Subject: Re: Proofing devices

At 03:42 PM 2/5/2003 -0700, Chris Murphy wrote:

Actually InDesign has a built-in "RIP" that does a good job on
non-PostScript inkjet printers. The catch is that currently InDesign
will use QuickDraw drawing commands (on both OS 9 and OS X), and as
such these devices are RGB, necessitating a custom RGB profile to
convert the ENTIRE document at print time to this RGB space. There are
possibly some other limitations to QuickDraw.

Chris:

I'm intrigued. I know a couple of designers that would love to figure out how to do something like this. What all is involved? Printing a test file and measuring that?

Are there any places that offer a service such as this? I know that Profile City and other offer custom profiles, is this something they can do?

David  

David Riecks
http://www.riecks.com, Chicago Midwest ASMP member
 ____________________________________________________________________________

Date: Thu, 06 Feb 2003 09:25:55 -0600
 From: "Cliff White"
Subject: Re: proofing devices

Thanks for all the info on proofing devices. I have assembled enough money now to send in the budget, and I'll continue research to see what the best solution is. One more question. Our intention is to be able to use our proofs as contract proofs. We print a monthly magazine with a circulation of 450,000. So, color is pretty important. We send someone to each press check to sign off on color and registration and such, so we invest time and effort in getting it right. But, since we are a state agency, funding is limited. However, we can get money when we really need it.

My question is mainly to the printers who might be lurking on this group. Of the solutions that have been batted around: high end proofing devices, large format inkjets behind high end rips, desktop inkjets using low end rips or just CMYK profiles, etc., at what level do you become uncomfortable accepting the proof as contract? The situation that I'm hoping to avoid is bad color coming off the press and our printer blaming it on our proofs, saying that they are not matchable. At what level could we come back and say that our method has at least wide acceptance as a reliable and matcheable proof?

Thanks for all the help!

CLIFF WHITE
Staff Photographer
Missouri Department of Conservation
(573)751-4115 ext. 3854
____________________________________________________________________________

Date: Thu, 6 Feb 2003 11:07:04 -0700
 From: Chris Murphy
Subject: Re: Proofing devices

On Wednesday, February 5, 2003, at 11:19 PM, David Riecks wrote:

I'm intrigued. I know a couple of designers that would love to figure out
how to do something like this. What all is involved? Printing a test file
and measuring that?

The InDesign "RIP" does not include screening. So it's a vector to CT "RIP", effectively. Therefore you do not need to make a special profile for InDesign because the screening is still done by the driver, just like it was when producing the test target from Photoshop (for example). You simply specify a destination profile that is an RGB profile (for the non-PostScript inkjet printer).

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
____________________________________________________________________________

Date: Thu, 6 Feb 2003 11:16:43 -0700
 From: Chris Murphy
Subject: Re: proofing devices

On Thursday, February 6, 2003, at 08:25 AM, Cliff White wrote:

My question is mainly to the printers who might be lurking on this
group. Of the solutions that have been batted around: high end
proofing devices, large format inkjets behind high end rips, desktop
inkjets using low end rips or just CMYK profiles, etc., at what level do
you become uncomfortable accepting the proof as contract? The situation
that I'm hoping to avoid is bad color coming off the press and our
printer blaming it on our proofs, saying that they are not matchable.
At what level could we come back and say that our method has at least
wide acceptance as a reliable and matchable proof?

Understand the position you are putting the printer into. You are asking them to become responsible for YOUR proofing system. That's not a good situation to be in when the proof is going to become a contract. It's one thing for them to have an inkjet based proofing system, that they have control over. They are in a position to guarantee those proofs, even if its the same system you buy.

If the job involves multiple press runs, and would involve a lot of contract proofs from the printer, I would suggest you do the proofing; and then have the printer select a sampling of images/objects that they will proof: high key, low key, neutrals, pastels, saturated colors, and skin tones. If their proof of this sub sample of your job matches up in an acceptable (for both of you) fashion, there's little reason to believe there is something wrong with your proofs. I think this is a reasonable compromise unless I hear a compelling argument to the contrary.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
____________________________________________________________________________

Date: Thu, 06 Feb 2003 11:33:50 -0500
 From: Dolores Kaufman
Subject: Re: Proofing devices

As a fellow photographer I just wanted to second Andrey Rodney, Nick Wheeler and Ron Kelly's advice. If you follow Ron Kelly's procedure you'll get as close as you possibly can get to the final (Press) result without spending a bundle on fancy measuring devices, etc., which I doubt would get you any closer. I have been following a similar procedure for years and it really works. I use the latest model Epson Photo printers and a Rip (Epson Stylus Rip or Power Rip) for printing from Quark Express, though I prefer to print in RGB from Photoshop when proofing images only. Learn to use Color Sync and the proper profiles for your devices/paper (Andrew Rodney's advice on a previous post is also very good here). Above all, good color correction is the key to success and Dan's books and articles are your best resource there.

Dolores
____________________________________________________________________________

Date: Thu, 06 Feb 2003 09:29:18 -0700
 From: Andrew Rodney
Subject: Re: proofing devices

on 2/6/03 8:25 AM, Cliff White wrote:

Our intention is to be able
to use our proofs as contract proofs.

This isn∂t a technological issue but more a ΠlegalΣ issue. Will the people printing the job on press agree to contractually match the proof you create? If so, you∂re all set. If not...

Of the solutions that have been batted around: high end
proofing devices, large format inkjets behind high end rips, desktop
inkjets using low end rips or just CMYK profiles, etc., at what level do
you become uncomfortable accepting the proof as contract?

Good question. Years ago, if anyone suggested using an Iris ink jet as a Contract Proof, there would be laughter and rejection. Now that∂s done all the time. A profiled $400 Epson is just as good (perhaps better in some respects like dry down with Pigmented inks) as a very expensive Iris and the Epson will rarely spray ink all over your floor. So again, it∂s the acceptance level between the two parties more than the hardware.

Andrew Rodney
____________________________________________________________________________
 
Date: Thu, 06 Feb 2003 09:46:42 -0500
 From: Lee Clawson
Subject: Re: Proofing devices

on 2/5/03 9:38 AM, Nick Wheeler wrote:

Beware fancy expensive solutions to the wrong problem.
 
And I also agree with the idea of using a lower cost device that shows a reasonable preview of expected print without eliminating the (contract) proof from your printer. Look to saving money by minimizing changes made by your printer. I've use a Xerox Phaser 7700. It has very good tonal range and color reproduction. We run 11x17 spreads on 13x19 paper.
____________________________________________________________________________

Date: Fri, 07 Feb 2003 00:02:31 -0600
 From: David Riecks
Subject: Re: Proofing devices

At 11:07 AM 2/6/2003 -0700, Chris Murphy wrote:

The InDesign "RIP" does not include screening. So it's a vector to CT
"RIP", effectively. Therefore you do not need to make a special profile
for InDesign because the screening is still done by the driver, just
like it was when producing the test target from Photoshop (for
example). You simply specify a destination profile that is an RGB
profile (for the non-PostScript inkjet printer).

Chris:

Many of the designers I work with will place RGB images, for placement only.  I can see how they'd just send the Indesign file with placed RGB files to the inkjet.  But if they have CMYK images placed, you say that you specifiy an RGB output profile for the device. Can this situation compensated (roughly) for dot gain and press conditions... or is simply a very "rough" inkjet proof?

Would you suggest just using the stock Epson profile if sending to an Epson printer using Epson papers? Or is there a better way?

David
 
David Riecks          *   david@riecks.com
701 W. Washington St      *   Midwest/Chicago ASMP
Champaign, IL 61820       *
ph/fax 877-646-5375       *   http: //www.riecks.com/
 ____________________________________________________________________________

Date: Thu, 06 Feb 2003 22:26:36 -0800
 From: Dennis Dunbar
Subject: Re: Proofing Devices

I am wondering if including a file such as GATF's proof comparator, or Don Hutcheson's proof qualifier on the proof, just outside of your "Live" area would give printers the information they need to verify that the inkjet "prooof" is valid and can be trusted.

Dennis Dunbar

____________________________________________________________________________

Date: Fri, 07 Feb 2003 11:10:39 -0000
 From: "Stephen Marsh "
Subject: Re: Proofing devices

David Riecks wrote:

Would you suggest just using the stock Epson profile if sending to an Epson
printer using Epson papers? Or is there a better way?


David, this is my basic understanding of the simple workflow (RGB aimprints for little $$$):

You settle on a recommended ink/stock and fixed resolution and get a custom profile generated of your inkjet - cost from $100 and no need to purchase gear or have training. There are many consultants and some new validation services who can profile your printer and set you up. A validation service goes beyond just getting a custom profile and adds some insurance value to your work if you have no other backup or help, as an independent body verifies things so you do not have to have expensive measuring gear.

One such group is:

http://www.pro-file.org

This commonly presumes a RGB input workflow, but you can profile as CMYK.

The CMYK image (ICC tagged or known) is converted to the custom profile which describes the inkjet, or working RGB if that is your workflow and then later the inkjet. As the inkjet is profiled it should be a fairly easy task to simulate the CMYK original.

So apart from no halftone, the colour and tone is usually very close when all this is done right (making a larger gamut device simulate a lower gamut device). You may need different stock/ink than you normally use, which is another reason why custom profiling is required. A stock canned profile would probably not be the way to go, but I guess it depends on the device and your 'proofing' expectations.

This works great for a single image from Photoshop.

Producing page layouts may be a bit different. I guess if you profile as RGB, then the CMYK images would need to be duped and converted to RGB to fit the workflow - unless ID can convert them to RGB on the fly from a tag or presumed profile etc.

Illustration and layout software by nature are more complex to colour manage output when you have mixed objects in different flavours or modes.

Stephen Marsh.
____________________________________________________________________________

Date: Fri, 7 Feb 2003 08:42:10 -0500
 From: "John"
Subject: Re: proofing devices

Cliff White Wrote

My question is mainly to the printers who might be lurking on this
group. Of the solutions that have been batted around: high end
proofing devices, large format inkjets behind high end rips, desktop
inkjets using low end rips or just CMYK profiles, etc., at what level do
you become uncomfortable accepting the proof as contract?

Cliff

A a printer, I believe I have argued this similar point before on similar topics. There is a misconception that we (printers) cannot print anything if we did not have a proof to look at, so we know how to set the press up and how to run it. Not true. When we hang a set of plates on a press, level fountains, fit, and pull up to color using a densitiometer, the proofs are still in the job folder. We level the densities across the sheet, and level them to standards. We don't look at the proofs until dead last thing. Then, if some tweaking is warranted to match the proof closer, we may juggle some densities, or cylinder pressures, but for the most part the job color is there all ready. We actually print the same job day after day. It has solids, highlights, shadow, midtones, grey balance and trap variations. Its called a color bar. When we get it right, and its obvious when its not, the rest of the clients job that is along for the ride is usually 95% there. Now not to say there aren't the "problem children" images now and then, that during a customer OK the client determines are more important than all the others, and warrant special attention or some "press etching" to achieve, but again, the idea here is to run as close to standard densities as is possible, without have to juggle. Running inks too high or too low, or having to cut densities on half of the sheet, all have inherent consequences that can effect the runnalbilitiy of the job. ie: I've rigged and juggled to get an OK... but there is no way the job can run this way and hold any consistency.

Now, when you say you want to avoid "bad color coming off of the press and the printer blaming it on the proofs" its usually not the proofs fault at all, but the files themselves. Assuming you are dealing with a competent printer that runs to standards, more than likely when there is a problem with the color on the job, its the file and not the press. Especially if my color bar is telling me so. It doesn't make any difference if they are the printers files or the clients files. Bad color is bad color, and there is a lot of it around. Unfortunately, there was not a proof that you made, or the printer made, that showed the true end results in advance. Any client or printer that can master consistent contract proofing has the edge, and the tools to identify color problems before they get to press.

John Rawlins
____________________________________________________________________________

Date: Fri, 7 Feb 2003 10:54:20 -0700
   From: Chris Murphy
Subject: Re: Proofing devices
 
On Friday, February 7, 2003, at 04:10  AM, Stephen Marsh  wrote:

The CMYK image (ICC tagged or known) is converted to the custom
profile which describes the inkjet, or working RGB if that is your
workflow and then later the inkjet. As the inkjet is profiled it
should be a fairly easy task to simulate the CMYK original.

It's not necessary, and usually not desirable to convert a CMYK image  to an interim RGB space. Just place it, and when it comes time to print, select the inkjet printer's RGB profile, and then the necessary conversion occurs once, and only at print time.

Producing page layouts may be a bit different. I guess if you profile
as RGB, then the CMYK images would need to be duped and converted to
RGB to fit the workflow - unless ID can convert them to RGB on the
fly from a tag or presumed profile etc.

It does. If it's a tagged CMYK image, then you could get more than one conversion if the profile embedded in the CMYK image is different from the Proof Setup space. This is potentially dangerous because the proof you get will be predicated on the CMYK image(s) being *RESEPARATED*.

InDesign does not have an automatic mechanism for ignoring embedded profiles when you place them, except for unchecking "Enable Color Management." Setting the "Off" Color Management Policy only applies to InDesign *documents* not InDesign *objects*.

Illustration and layout software by nature are more complex to colour
manage output when you have mixed objects in different flavours or
modes.

You're not kidding. It took a while to figure out just exactly how InDesign behaved, and then how it interrelates with other applications.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
____________________________________________________________________________

Date: Fri, 7 Feb 2003 10:42:37 -0700
   From: Chris Murphy
Subject: Re: Proofing devices

Oh one big gotcha about InDesign (and it's an issue with QuarkXPress as well): EPS files are not supported. So whatever color space the EPS files are in, that's what they will stay in when you print. The exact path for CMYK EPS files going to a non-PostScript printer isn't one I've figured out yet. They do get converted into RGB because they have to be, but they are also rasterized by InDesign (unlike QuarkXPress which would just convert the preview if it were in CMYK, to RGB and give that to the OS for printing). Anyway, it's possible that printing EPS's will be unreliable in terms of getting something reasonable in terms of color.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
____________________________________________________________________________

Date: Fri, 7 Feb 2003 10:44:12 -0700
   From: Chris Murphy
Subject: Re: Proofing Devices

On Thursday, February 6, 2003, at 11:26 PM, Dennis Dunbar wrote:

I am wondering if including a file such as GATF's proof comparator, or
Don Hutcheson's proof qualifier on the proof, just outside of your
"Live" area would give printers the information they need to verify
that the inkjet "prooof" is valid and can be trusted.

Some printers will find that acceptable, others may want to pull some of their own contract proofs for a sampling of the job, just to be sure. After all,  if they agree a customer's proof is "contract" then
they are obligated to match it.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
____________________________________________________________________________

Date: Fri, 7 Feb 2003 10:39:00 -0700
   From: Chris Murphy
Subject: Re: Proofing devices

On Thursday, February 6, 2003, at 11:02  PM, David Riecks wrote:

Many of the designers I work with will place RGB images, for placement only.
I can see how they'd just send the Indesign file with placed RGB files to
the inkjet.

What they should do is assign the contract proofer profile as Document CMYK, then place tagged RGB images, and untagged CMYK images. The assigned Document CMYK profile is then the assumed source for all untagged CMYK images, ensuring they are only converted for proofing purposes (to the local inkjet printer). In the print dialog, in the color management section, there is a Source Space subsection and a Print Space subsection.

In Source Space, if they leave it set to "Document" then the RGB and CMYK images are converted for the inkjet printer, letting the inkjet printer do the best job it can. So for CMYK images, you get proofs, and for RGB images, you get bright, brilliant inkjet printer color.

But if in Source Space you set it to Proof (Document CMYK then appears),  it will convert RGB content to the Document CMYK space. Now everything is Document CMYK.

In Print Space, select the profile for the inkjet printer (which will be an RGB profile), and select either Relative Colorimetric or Absolute Colorimetric rendering. More details on this behavior starting on page 351 of Real World Color Management.

But if they have CMYK images placed, you say that you specifiy an RGB
output profile for the device. Can this situation compensated (roughly) for
dot gain and press conditions... or is simply  a very "rough" inkjet proof?

It does compensate for dot gain and press conditions. If the profiles are accurate for the actual behavior of the press (or contract proofing system) and for the inkjet printer, and the inkjet printer is capable of producing press colors (it encompasses the gamut of the press/contract proof), then you will get something that is near contract quality. The better the profiles, the better the results. The more average they are, the more average the results.

Would you suggest just using the stock Epson profile if sending to an Epson
printer using Epson papers? Or is there a better way?

Ideally it would be a custom profile. But the canned profiles for the latest Epson printers are not too bad. I've had reasonable results for Epson printers that have shipped in the last 18-24 months.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
____________________________________________________________________________

Date: Sat, 08 Feb 2003 07:48:53 -0000
   From: "Stephen Marsh "
Subject: Re: Proofing devices

It's not necessary, and usually not desirable to convert a CMYK image
to an interim RGB space. Just place it, and when it comes time to
print, select the inkjet printer's RGB profile, and then the necessary
conversion occurs once, and only at print time.

Thanks Chris - this does sound better, I was thinking of one of the three ways to make colour managed prints to an inkjet - with the colour managed printing of later versions of Photoshop things are nicer as you suggest.

It has been a while since I used ID, that was back in the v1 to v1.5 crossover and I did not get the time to really push this aspect - I was just trying to make it work in the traditional manual human colour managed CMYK PostScript workflow, without entering really new water such as colour managing proofs or seps at the application level.

You're not kidding. It took a while to figure out just exactly how
InDesign behaved, and then how it interrelates with other applications.

I seem to have the ability to make severe understatements. <g>

I have not made use of ICC colour managed workflows in illustration or layout for press for producing seps, although I have briefly looked into them for inhouse proofing in these situations. From the sounds of things InDesign needs a lot of homework if you want to employ ICC CM to any extent, but so does any bit of software.

Stephen Marsh.
____________________________________________________________________________

Date: Sat, 08 Feb 2003 12:55:43 +0000
   From: Steve Smith
Subject: Re: Proofing devices

If you want EPS or PDF files to be colour managed by InDesign a cheat is to give them a transparency of 99.9%. This will round up to 100% on output but it forces the file through InDesigns colour management system.

Steve
____________________________________________________________________________

Date: Sat, 08 Feb 2003 12:18:38 -0500
   From: Dolores Kaufman
Subject: Re: Proofing devices

on 2/8/03 7:55 AM, Steve Smith wrote:

If you want EPS or PDF files to be colour managed by InDesign a cheat is to
give them a transparency of 99.9%. This will round up to 100% on output but
it forces the file through InDesigns colour management system.

Do you happen to know if this works with Quark Express as well? I like using EPS files because of the ability to choose a JPG preview.

Dolores
____________________________________________________________________________

Date: Sat, 8 Feb 2003 11:39:01 -0700
   From: Chris Murphy
Subject: Re: Proofing devices

On Saturday, February 8, 2003, at 10:18  AM, Dolores Kaufman wrote:

Do you happen to know if this works with Quark Express as well? I like using
EPS files because of the ability to choose a JPG preview.

QuarkXPress doesn't do transparency. So unfortunately not.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor
____________________________________________________________________________

Date: Sat, 8 Feb 2003 11:35:10 -0700
   From: Chris Murphy
Subject: Re: Proofing devices

On Saturday, February 8, 2003, at 12:48  AM, Stephen Marsh wrote:

It has been a while since I used ID, that was back in the v1 to v1.5
crossover and I did not get the time to really push this aspect

Good choice. I think it would have been somewhere between "interesting but not useful" and a total waste of time. We didn't pay any attention to ID 1.x in RWCM. ID2's color management is very much like Photoshop (and Illustrator).

I have not made use of ICC colour managed workflows in illustration
or layout for press for producing seps, although I have briefly
looked into them for inhouse proofing in these situations. From the
sounds of things InDesign needs a lot of homework if you want to
employ ICC CM to any extent, but so does any bit of software.

The complexity is due in  part to the UI which I don't think draws a clear distinction between what settings affect the document and what settings affect objects. Sometimes the settings can affect both. Once you understand InDesign's point of view (it's weird to put it that way) however, it makes sense. But I still wish there was something that could be done to make it more intuitive with a manual or book.

The fact of the matter is, page layout is a complicated thing. You can place multiple file types, some of which do and don't support embedded profiles; some of which ID 2 can and can't convert. So you simply have to know what this file types are: TIFF, JPEG, PSD for sure. PICT might work. PDF and native AI files should work, but don't convert properly when printing (they do convert properly when exporting a PDF however). EPS behavior is just odd: ID ignores embedded ICC profiles, but honors embedded CSA's (PostScript source profiles, most commonly from Photoshop) for on-screen preview only, but ultimately does not convert the EPS file at print or export time.

So if you imagine the amount of effort you put into color management for a single Photoshop document (for some people it's minor, for others it's a little more than minor), and then understand an ID2 document is a collection of many documents (called objects) from Photoshop and Illustrator (just to name two), as well as native objects such as text, lines, shapes, backgrounds, etc.

Chris Murphy
Color Remedies (TM)
www.colorremedies.com/realworldcolor

Adobe Photoshop training classes are taught in the US by Sterling Ledet & Associates, Inc.