Color profiles to compensate for print conditions

[jstotz]

I have some questions about the use of color profiles when preparing images for different printing conditions.

I have a customer supplied image which is CMYK. It may or may not have a profile and if it does, the profile is probably SWOP or GRACoL. It probably started out as RGB, but who knows how it was converted to CMYK.

I'm going to print this image on an offset press on uncoated paper. I realize that it will loose saturation and that the darkest color will be lighter than what I could get on coated paper. I have done the G7 process on this press for this paper so I have a good profile that describes how it prints and I have a good proof that matches the press sheet.

So I run a proof of this image and the proof looks flat like it should because it is accurately predicting what the press sheet will look like. And I'm getting complaints from the customers and salespeople because the proof looks so flat.

Next I color correct the image to add contrast and saturation. Even though I can't completely fix the problem, there is usually a definite improvement. The new proof looks better, and the press run matches the new proof.

What I want to know, is if this problem can be compensated for by the correct use of profiles or am I always going to have to do manual color corrections.

Let's say I start with the original RGB image which has a correct RGB profile. I then convert it to CMYK using the uncoated press profile. Or I start with the CMYK image and convert to the uncoated press profile. Is there something built into this process that will at least partly compensate for the loss of saturation I will get on press? Or is this just adjusting for different Lab values of the colorants?

I think I understand the concept of using a profile to simulate the uncoated paper on the proofer. That's like saying "Here are the original pixel values, change them to what they would look like on the uncoated paper".

I'm much less clear on the non-simulation use of profiles to convert from RGB to CMYK or from CMYK to CMYK. Obviously, I don't want that conversion to be a simulation. I think I want it to be an anti-simulation. "Make these pixels into something that will look close to the original once the simulation (or the actual printing process) is applied to them. Do I have that right?

Thanks

 

[gordo]

Back in the steam-powered days when printers used densitometers some would set new, higher SID targets for uncoated paper in order to increase print contrast and saturation. Then use tone curves on the plates to compensate for the increased TVI. They'd profile that new print condition to drive their proofer. Some would also switch to UCR separations based on the idea that increasing the amount of ink in a separation helped to seal the paper a bit so that the ink wouldn't be absorbed so much into the substrate. They sometimes would also take a cue from the newspaper world and run a YCMK ink sequence since the Y channel in separations usually has the greatest area coverage and hence running it down first helped seal the paper.

Having your proof reflect what the press will print is the right, albeit brave, approach. But not all print buyers (or especially sales) can handle the truth. Most printers/sales prefer to show a pretty proof to the client and when the presswork looks flat explain that, obviously the client - being a professional - would obviously know what would happen on press with an uncoated paper.

best, gordo

 

[walterz]

jstotz - check your profiling software and see if it will allow you to edit the uncoated press profile you have made. Programs like MonacoProfiler will allow you to make edits including applying a curve to any/all channels. You can even apply this edit to the Inverse, Forward or both tables which means you could build a profile which will add contrast when converting to it but for proofing it will remain the same. Don't neglect to think about the naming convension as you will need to somehow remember what is unique about this particular profile you have made.

 

[jstotz]

@walterz
I can edit the profiles, but I have not tried that. I guess I was trying to find out if the act of converting an image as I described had that sort of correction intrinsicly built into the process.

 

[rich apollo]

What I want to know, is if this problem can be compensated for by the correct use of profiles or am I always going to have to do manual color corrections.

The answers are "Yes," "No," and "Maybe." :~P

First, I'm concerned. You state that "…the proof looks flat…" Maybe it's choice of words, but I would expect the proof to look "full" or "heavy" while being less saturated.

Next, I come down on the side of reseparating the file. Running something that is set up for GRACoL or SWOP conditions on an uncoated stock is not the best way to handle it, IMO. I reseparate EVERYTHING that comes into our shop. My situation is a bit more extreme because I'll receive the same file set up for GRACoL or SWOP to run on newsprint. This approach works well for me - and now that I've run it for years I would not go back. Having the file appropriately set up clears up a whole slew of headaches.

When I began generating color-accurate proofs here, I got a lot of blowback from clients. They didn't want the "dingy" proofs I was turning out. They wanted the unreasonably bright and colorful proofs that they were used to seeing. There was a period of adjustment for everyone. Once the clients saw that the proofs were representative of what they would see on press they saw value in the process.

Lastly, the whole thing depends on your clients. They are setting up the files and proofing to a condition other than how the final piece will be produced. I would encourage you to share your uncoated profile(s) with the clients. Help them to view and proof correctly up front and they will take care of the manual manipulations on their own.

I don't think I would go the route of editing the profiles unless you find that you're consistently making the same alterations. Then are those alterations being applied to all cases, or are they specific to one client.

 

[jstotz]

The answers are "Yes," "No," and "Maybe." :~P

First, I'm concerned. You state that "…the proof looks flat…" Maybe it's choice of words, but I would expect the proof to look "full" or "heavy" while being less saturated.

It doesn't look full because I've already compensated for the dot gain with a plate curve and the proof reflects that. By flat I mean less saturated and a lower maximum density.

Next, I come down on the side of reseparating the file. Running something that is set up for GRACoL or SWOP conditions on an uncoated stock is not the best way to handle it, IMO. I reseparate EVERYTHING that comes into our shop. My situation is a bit more extreme because I'll receive the same file set up for GRACoL or SWOP to run on newsprint. This approach works well for me - and now that I've run it for years I would not go back. Having the file appropriately set up clears up a whole slew of headaches.

What kind of difference does re-separating make? Does it actually increase the saturation?

When I began generating color-accurate proofs here, I got a lot of blowback from clients. They didn't want the "dingy" proofs I was turning out. They wanted the unreasonably bright and colorful proofs that they were used to seeing. There was a period of adjustment for everyone. Once the clients saw that the proofs were representative of what they would see on press they saw value in the process.

Even if I explain that the proofs match the press sheet, there is still the problem that the press sheet looks dingy. It's not that they don't like the proof, they don't like the press sheet either. That's what I'm trying to fix.

Lastly, the whole thing depends on your clients. They are setting up the files and proofing to a condition other than how the final piece will be produced. I would encourage you to share your uncoated profile(s) with the clients. Help them to view and proof correctly up front and they will take care of the manual manipulations on their own.

That will definitely depend on the customer.

I don't think I would go the route of editing the profiles unless you find that you're consistently making the same alterations. Then are those alterations being applied to all cases, or are they specific to one client.

 

[gordo]

Even if I explain that the proofs match the press sheet, there is still the problem that the press sheet looks dingy. It's not that they don't like the proof, they don't like the press sheet either. That's what I'm trying to fix.

Did you read my post?

best, gordo

 

[jstotz]

Back in the steam-powered days when printers used densitometers some would set new, higher SID targets for uncoated paper in order to increase print contrast and saturation. Then use tone curves on the plates to compensate for the increased TVI. They'd profile that new print condition to drive their proofer. Some would also switch to UCR separations based on the idea that increasing the amount of ink in a separation helped to seal the paper a bit so that the ink wouldn't be absorbed so much into the substrate. They sometimes would also take a cue from the newspaper world and run a YCMK ink sequence since the Y channel in separations usually has the greatest area coverage and hence running it down first helped seal the paper.

We have pushed the ink densities. We have not tried UCR. For UV uncoated (dry trap) we run KMCY to help the blue We haven't tried yellow first but I don't imagine that would matter much on a dry trap UV. Maybe on conventional though.

Having your proof reflect what the press will print is the right, albeit brave, approach. But not all print buyers (or especially sales) can handle the truth. Most printers/sales prefer to show a pretty proof to the client and when the presswork looks flat explain that, obviously the client - being a professional - would obviously know what would happen on press with an uncoated paper.

best, gordo

I was hoping to find a way to help with this in the color separation instead of on press but I'll keep those suggestions in mind. Thanks.

 

[rich apollo]

What kind of difference does re-separating make? Does it actually increase the saturation?

Not exactly.

My thinking is that the files are built for lower dot gains than what the uncoated sheet is yielding (even with compensation during plating). When the pressmen push the ink to increase print-contrast you see more movement in the quarter- and mid-tones than in the saturated colors.

By reseparating to the appropriate colorspace you'll have the total ink properly limited and the file will probably be somewhat lighter in the mid-tones. This will result in a bit higher print-contrast and should allow the pressmen to push the solids a little more.

Even if I explain that the proofs match the press sheet, there is still the problem that the press sheet looks dingy. It's not that they don't like the proof, they don't like the press sheet either. That's what I'm trying to fix.

I'm not sure you're gonna' win. They're expecting race-car performance from a moped.

You could try a different sheet. Mohawk always seemed to have much better ink holdout than competing uncoated sheets. You can open up your plate curves. With appropriate software you can induce the same kind of effect within the files, and you can then leave your plate curves alone.