Need some advice on proofing a large magazine editorial

[elliot_n]

So best to avoid soft-proofing altogether?

[digitaldog]

So best to avoid soft-proofing altogether?

When using the correct an actual profile for the outpur process and you control the RI then no, you can and should soft proof.

[elliot_n]

I think I'm misunderstanding your test chart example. You show a situation where soft-proofing is not working (all the soft proofs look the same, but all will print differently). If the file is in RGB, not CMYK, will the soft proofs all look different?

[digitaldog]

I think I'm misunderstanding your test chart example. You show a situation where soft-proofing is not working (all the soft proofs look the same, but all will print differently). If the file is in RGB, not CMYK, will the soft proofs all look different?

Yes RGB/CMYK would look different from each other. In the CMYK example however, the three look the same but are different and would print differently due to black Gen which doesn’t “soft proof”.

[elliot_n]

One more question, and then I'll get off the OP's thread. What exactly is a CMYK profile? It's not specific to one paper is it? It's for a type of paper (coated or uncoated)? A printing company will run hundreds of different paper stocks, and it's very possible that the stock I want to print on is a stock they've never used. They can't be expected to profile all available papers, can they? After all, it's not like loading a couple of A4 sheets into an inkjet printer to make patches. Last time I asked a printer for a profile they made a big song and dance about it, but when I googled the profile they emailed me, it was just a generic profile (the top one on this page - http://www.colormanagement.org/en/isoprofile2009.html ). I was on press for that job, and at the end of the day it came down to the printer tweaking his sliders. I was happy with the final output, but it didn't feel like a properly colour-managed workflow.

[digitaldog]

One more question, and then I'll get off the OP's thread. What exactly is a CMYK profile? It's not specific to one paper is it? It's for a type of paper (coated or uncoated)? A printing company will run hundreds of different paper stocks, and it's very possible that the stock I want to print on is a stock they've never used. They can't be expected to profile all available papers, can they? After all, it's not like loading a couple of A4 sheets into an inkjet printer to make patches. Last time I asked a printer for a profile they made a big song and dance about it, but when I googled the profile they emailed me, it was just a generic profile (the top one on this page - http://www.colormanagement.org/en/isoprofile2009.html ). I was on press for that job, and at the end of the day it came down to the printer tweaking his sliders. I was happy with the final output, but it didn't feel like a properly colour-managed workflow.

It's a profile of the print process, no different than what I suspect (?) you might use when printing from a desktop printer but RGB rather than CMYK.
Yes, they can be expected to have some process to convert RGB to CMYK for any/all output types. Or something similar, a device link profile to convert CMYK to CMYK.
A 'generic profile' of ANYTHING is basically useless! If you were printing on an Epson 3880 on Luster Paper and I suggested you use a profile for an Epson 9800 and Matt paper, would you use it for soft proofing let alone converting to the output color space? I hope you wouldn’t. A press is NO different.
A non generic profile that targets a defined print process like TR001 or GRACoL2006 Coated1 if and when sent to a press that absolutely is set to produce those defined conditions isn't a 'generic' profile.
Got a 'generic profile' for a display? Equally useless. 

[elliot_n]

It's a profile of the print process, no different than what I suspect (?) you might use when printing from a desktop printer but RGB rather than CMYK.

But the profiles I use when I print with my Epson 3880, are custom profiles that I've had made for the specific paper that I'm printing on. I don't use a generic 'Epson 3880 resin-coated paper' profile or a generic 'Epson 3880 matte paper' profile.

Yes, they can be expected to have some process to convert RGB to CMYK for any/all output types. Or something similar, a device link profile to convert CMYK to CMYK.

I'm curious how they go about that. If I've simply converted to their recommended 'coated' or 'uncoated' CMYK profile, how do they further fine-tune the process so that it prints correctly on the specific paper stock I've selected. Or should I be expecting the printer to supply specific profiles for specific paper stocks?

A 'generic profile' of ANYTHING is basically useless! If you were printing on an Epson 3880 on Luster Paper and I suggested you use a profile for an Epson 9800 and Matt paper, would you use it for soft proofing let alone converting to the output color space? I hope you wouldn’t.

No, I wouldn't.

A press is NO different. A non generic profile that targets a defined print process like TR001 or GRACoL2006 Coated1 if and when sent to a press that absolutely is set to produce those defined conditions isn't a 'generic' profile.

I'm only half-following you. Printing to an inkjet printer, I can use profiles which accurately describe (and soft-proof) the subtle qualities of all sorts of different papers. But my offset printing house (that offers a much more diverse range of papers), is only offering me two profiles, either 'coated' or 'uncoated'. It seems restrictive, no?

[elliot_n]

Perhaps I shouldn't have used the word 'generic' to describe a profile such as 'ISOcoated_v2_bas.ICC ' (or the CMYK profiles that come preloaded in Photoshop). But they are not profiles of a specific press are they? Rather, they're a standard that a specific press can be calibrated to (?)

[digitaldog]

But the profiles I use when I print with my Epson 3880, are custom profiles that I've had made for the specific paper that I'm printing on. I don't use a generic 'Epson 3880 resin-coated paper' profile or a generic 'Epson 3880 matte paper' profile.

That changed nothing I've written and more, begs the question why you'd consider a generic press profile at all useful. It's not.

I'm curious how they go about that.

Pretty much the same way you do to build an RGB profile but with a lot more settings to apply (Black Gen for one).

If I've simply converted to their recommended 'coated' or 'uncoated' CMYK profile, how do they further fine-tune the process so that it prints correctly on the specific paper stock I've selected. Or should I be expecting the printer to supply specific profiles for specific paper stocks?

You seem to be describing the use of a generic profile. I told you why, especially with CMYK that's far from ideal or useful. Descriptions with actual images too!

I'm only half-following you. Printing to an inkjet printer, I can use profiles which accurately describe (and soft-proof) the subtle qualities of all sorts of different papers.

IF you demand to use the term accuracy, I'm forced to ask you to provide an accuracy metric: average, max deltaE. Further, even if you suspect it's accurate, and it may be, what's your evidence that the profile you got from a print house that doesn't promise that profile actually defines press behavior is 'accurate'?

But my offset printing house (that offers a much more diverse range of papers), is only offering me two profiles, either 'coated' or 'uncoated'. It seems restrictive, no?

Of course it does unless they only have two stocks and have stated these are custom, not generic profiles. I have no way to tell you which it is of course.

[digitaldog]

Perhaps I shouldn't have used the word 'generic' to describe a profile such as 'ISOcoated_v2_bas.ICC ' (or the CMYK profiles that come preloaded in Photoshop). But they are not profiles of a specific press are they?

You're asking a question that is impossible for me to answer. I have zero idea who you're even using! Even if I did, I'd have to get the profile, then output a big pile of color patches, measure them, and run a series of colorimetric analysis which would produce real data on accuracy. Then I'd run that test in a week to see how well their process control is. That's another can of worms. EVEN with digital presses.

[elliot_n]

I'm getting increasingly confused. On the few occasions that I've dealt with offset printers directly, I've asked for their profiles and they've just offered me two - a coated profile and an uncoated profile. Often these profiles have familiar names - they are the profiles pre-installed in Photoshop (e.g. 'Coated GRACoL2006'). Is this a bad sign? Should I be looking for a printing company that offers hundreds of profiles, made for their specific press and for the hundreds of different papers that they print on?

[digitaldog]

I'm getting increasingly confused. On the few occasions that I've dealt with offset printers directly, I've asked for their profiles and they've just offered me two - a coated profile and an uncoated profile. Often these profiles have familiar names - they are the profiles pre-installed in Photoshop (e.g. 'Coated GRACoL2006'). Is this a bad sign? Should I be looking for a printing company that offers hundreds of profiles, made for their specific press and for the hundreds of different papers that they print on?

One more question, and then I'll get off the OP's thread.

I'm confused now too.... 

[amolitor]

Is it possible the press guys are giving you generic profiles that are intended to let them usefully soft proof on their equipment, so they can see what you're shooting for, and then they work their press magic to try to hit that?

My impression is that offset presses are somewhat tempermental and that in the end it often comes down to a guy looking at proofs coming off the thing now, and tweaking the press a little too hit the target.

[aaronchan]

No offence, but you are living in a very advanced world. In here China, 90% of the printing company still only accept cmyk files, only! If you supply a rgb pdf, they simply covert it to any cmyk file by whatever default setting they are using( us swop, Japan color...etc) A lot of these factory are G7 or Fogra certified, but trust me, they are not working in these workflows at all. As long as their customer pays the money, they don’t care! Unfortunately, China is still the biggest printing country in the world......

Curves? What are they using to convert from RGB to CMYK? It may not be an ICC Profile if they really operate in the late 20th century but they still need to define a conversion. Could be the old 'classic' CMYK engine in Photoshop (doubt that, it's really hardly a production environment to do so).
I suspect the reality is many traditional off set companies have an ICC profile but simply do not want to supply it. Some incorrectly believe it contains some proprietary data their competitors might use which is really hogwash.
IF they can't define then supply a CMYK method of conversion, then it's kind of pointless and often painful to attempt any CMYK work outside their environments. Send them tagged RGB and hope for the best, get a contract proof, go to the press check (which isn't always possible and alone can cause problems).

Sent from my iPhone using Tapatalk

[samueljohnchia]

Just starting to shoot a big book project, publisher is in Japan and the printer is in Singapore.

Hi Michael, I'm based in Singapore. Which printer is printing your work? Most of the better printers here are Fogra certified and they don't use custom profiles, just the generic Fogra39 profile. It actually works quite well if they use standard process inks, run up to the right SIDs and have the correct TVI across the tone scale. Of course it wojld be great if they had custom ICC profiles but they just dont. That's the state of things here. Some printers here dont even know what CMYK profile to use, and will tell you any CMYK is fine (no its not).

[Stephen Ray]

publisher is in Japan and the printer is in Singapore.

This is key. The printer is probably very qualified by the publisher. Lots of high quality fine art publication printing use this route from Japan to Singapore.

I would like to improve my color workflow.

You might familiarize yourself with the concepts of G7 methods for the printing process. (Largely based upon measuring and controlling neutral gray balance which, because you’re a photographer, should come very easy to you.)

In the past I would get my images to where I would like them on a calibrated Eizo monitor and then send them off as flattened adobe RGB Tiffs. Results are mixed and I would like to do everything I can on my end to get them files that match what's on my screen.

Just realize and remember your monitor is an emissive device using a wider color gamut than the reflective capability of the given paper. You will need to shift your expectations. Fortunately Adobe 98 is centered upon a neutral axis and some CMYK press standards are meant to be  calibrated to aim points of neutral grays. Therefore, both you and the printer might have a common ground at least as far as neutrals, meaning no hue shifts. That leaves shadow and highlight detail along with color gamut. Try to be sure your images have data in those areas because your files will be radically transmogrified, hopefully skillfully, in the production process.

all I need is to proof

You are not actually “proofing” until your print becomes a “qualified” proof. You are merely printing a preview or a “go-by.”
Quoted from IDEAlliance (the G7 folks); “Individual proof verification requires printing the 2007 IDEAlliance ISO 12647-7 Control Strip on EVERY proof, measuring the strip with an automated spectrophotometer, and comparing the measurements to a "reference" set of CIELab values in suitable software.” Also available is a control strip to verify RGB proof prints and many Epson proofs are commonly evaluated only visually. Experienced operators know a practical proof print or go-by from something inferior or unattainable.

I hope this helps.

[elliot_n]

I'm confused now too.... 

I'm sorry to have confused you. I wonder if anyone else could help? My question, again: Previously, where I've requested CMYK profiles from printing companies they have offered me just two profiles: one for coated paper stocks, the other for uncoated paper stock. Is this problematic? Am I right to hope for a profile for the specific paper stock I am printing on? If this is the desired scenario, does anyone have any real-world experience of it actually happening?

[michaelbiondo]

Thanks,
I will see if I can find out the name of the printer.
MB

[michaelbiondo]

Hi Michael, I'm based in Singapore. Which printer is printing your work? Most of the better printers here are Fogra certified and they don't use custom profiles, just the generic Fogra39 profile. It actually works quite well if they use standard process inks, run up to the right SIDs and have the correct TVI across the tone scale. Of course it wojld be great if they had custom ICC profiles but they just dont. That's the state of things here. Some printers here dont even know what CMYK profile to use, and will tell you any CMYK is fine (no its not).

Thanks,
I will see if I can find out the name of the printer.
MB

[michaelbiondo]

This is key. The printer is probably very qualified by the publisher. Lots of high quality fine art publication printing use this route from Japan to Singapore.

You might familiarize yourself with the concepts of G7 methods for the printing process. (Largely based upon measuring and controlling neutral gray balance which, because you’re a photographer, should come very easy to you.)
Just realize and remember your monitor is an emissive device using a wider color gamut than the reflective capability of the given paper. You will need to shift your expectations. Fortunately Adobe 98 is centered upon a neutral axis and some CMYK press standards are meant to be  calibrated to aim points of neutral grays. Therefore, both you and the printer might have a common ground at least as far as neutrals, meaning no hue shifts. That leaves shadow and highlight detail along with color gamut. Try to be sure your images have data in those areas because your files will be radically transmogrified, hopefully skillfully, in the production process.
You are not actually “proofing” until your print becomes a “qualified” proof. You are merely printing a preview or a “go-by.”
Quoted from IDEAlliance (the G7 folks); “Individual proof verification requires printing the 2007 IDEAlliance ISO 12647-7 Control Strip on EVERY proof, measuring the strip with an automated spectrophotometer, and comparing the measurements to a "reference" set of CIELab values in suitable software.” Also available is a control strip to verify RGB proof prints and many Epson proofs are commonly evaluated only visually. Experienced operators know a practical proof print or go-by from something inferior or unattainable.

I hope this helps.

Thanks Stephen,

A lot of great information here and a lot to study