Our proof is Gracol, their proof is Gracol, and neither match

[pp411]

Our Epson proofer is calibrated and set up to a Gracol profile and our plating curves for press are aimed to match the proof.
We have a customer who sent in their proof, which differs from ours. They say it is also set to a gracol profile. Their stock is whiter, however the difference in proofs varies so much that we can't say it's the stock difference.
We scanned our ISO 12647-7 strip through the ORIS CGS certified software, and it passes. The customer does not have this strip on their proof. We were thinking of having them add this to their proof. Would this be useful? What steps would you take to determine that theirs is or isn't accurate?

 

[gordo]

We scanned our ISO 12647-7 strip through the ORIS CGS certified software, and it passes. The customer does not have this strip on their proof. We were thinking of having them add this to their proof. Would this be useful? What steps would you take to determine that theirs is or isn't accurate?

Yes! Absolutely.

 

[pp411]

Yes! Absolutely.

Thanks for the reply. So if I scan the target on their proof, and it fails as we expect, is it most likely because they haven't properly calibrated their proofer? Are there other reasons? In the few (1 or 2) other times we've encountered this with other customers they understood when we explain, so we haven't had to scan their strip to compare, but I am anticipating more questions from this customer.

 

[Stephen Marsh]

Our Epson proofer is calibrated and set up to a Gracol profile and our plating curves for press are aimed to match the proof.
We have a customer who sent in their proof, which differs from ours.

Printer’s should not have to accept third party proofs unless they are comfortable doing so, even more so “proofs” that do not contain a media wedge and or a “certification” label with the dEab values listed for the key pass/fail points. The third party proof is a guide, a desired aimpoint - it is not something that a printer has to accept as a contract proof.

They say it is also set to a gracol profile.

But do the numbers pass for the IDEAlliance media wedge?

Their stock is whiter,

Is it within tolerance, when measured with an uncut/unfiltered (UV included) spectro?

however the difference in proofs varies so much that we can’t say it’s the stock difference.

The specification allows for dEab 5 variation for many colours, with neutrals being lower.

We scanned our ISO 12647-7 strip through the ORIS CGS certified software, and it passes. The customer does not have this strip on their proof. We were thinking of having them add this to their proof. Would this be useful? What steps would you take to determine that theirs is or isn't accurate?

In a “certified” setting where specifications are the target, without a media wedge and a printed label listing the pass/fail dEab values of the wedge, a “proof” with no wedge provides no proof and is just a nice picture - even more so for a third party supplied proof.


Stephen Marsh

 

[macdevin]

Another thing that can effect the difference in proofs is to see if both proofing rips are set to honor profiles? This can have a huge effect in color. We recently had the same issue. Customer supplied proof differed greatly from ours. After looking into the issue, the result was they were supplying an PDF X/4 with a swop output intent.. Their Rip proof was not honoring the output intent and we were.

 

[mwc]

This is akin to the customer saying "It looks fine on my screen..." If they supply a proof in this way, they must have an approved test wedge on it for certification...without that you are just looking at pretty pictures without any reference to printing reproducibility.

 

[macdevin]

This is akin to the customer saying "It looks fine on my screen..." If they supply a proof in this way, they must have an approved test wedge on it for certification...without that you are just looking at pretty pictures without any reference to printing reproducibility.

The Test wedge does not tell you the whole story. In my case above their wedge passed.. Yes the wedge is needed when troubling shooting. But it's not the be all end all.

 

[mwc]

The Test wedge does not tell you the whole story. In my case above their wedge passed.. Yes the wedge is needed when troubling shooting. But it's not the be all end all.

Was this strip as large as the IDEAliance ISO 12647-7 strip? I do agree that the strip does not tell you everything...but if checked against a good reference set it should be pretty close (at least enough to flag a SWOP vs a GRACoL proof?)

 

[macdevin]

The wedge has no barring if the output intent of the PDF was used or not used.

 

[mwc]

The wedge has no barring if the output intent of the PDF was used or not used.

Yes, you are correct...I just didn't fully understand the scenario. This leaves me with a new understanding to the limitations of the wedge. I hadn't thought that a customer proof vs. our proof would vary in that way if they were both certified to the same standard. I did not think the color conversion from our rip vs theirs would play a factor, but now I see the possible pitfalls.
The Wedge simply certifies that the proofing device is outputting desired results.
If the Wedge does not go through the same conversions as the file does during the rip, then it doesn't work as an indicator of final color if the file contains objects that have been transformed.
Ideally the Wedge would be processed with the page data, but I don't know how this would work at all in production….you would have to proof each page with a wedge, as opposed to just a wedge on an imposition?

Fortunately I don't deal much with customer supplied proofs, so I hadn't thought about this much.
If you don't mind, how did you trouble-shoot your issue (customer rip not honoring output intent)?

 

[macdevin]

There are many things that can effect the difference in proofs because of rip manufactures and settings. It can be quite complex is trouble shooting and at times requires access to the customers RIP.
In this case I knew we were both using the same rip. Since the PDF that was supplied was a PDF x-4 I knew it had an output intent.. They had their profile set to swop, and were supplying a Gracol proof.
For me that's the red flag.. All I did was play out a new proof that was set not to honor the profiles..

The other issue I see often is spot colors that have transparency.. From RIP to RIP you get differences.

 

[rich apollo]

The Wedge simply certifies that the proofing device is outputting desired results.
If the Wedge does not go through the same conversions as the file does during the rip, then it doesn't work as an indicator of final color if the file contains objects that have been transformed.

Exactly.

Fortunately I don't deal much with customer supplied proofs, so I hadn't thought about this much.
If you don't mind, how did you trouble-shoot your issue (customer rip not honoring output intent)?

In a proofing sitation, you probably do not want to honor the output intent. Output intents are more of a production concern.

In this case I knew we were both using the same rip. Since the PDF that was supplied was a PDF x-4 I knew it had an output intent.. They had their profile set to swop, and were supplying a Gracol proof.

That would mean the proof is not "a Gracol proof".

The other issue I see often is spot colors that have transparency.. From RIP to RIP you get differences.

Tough to proof something for which there is no data.

 

[Stephen Marsh]

I agree with Rich, ignore output intents when performing certified proofing, the proofing condition is the output intent.


Stephen Marsh

 

[toronar]

But do the numbers pass for the IDEAlliance media wedge?
Is it within tolerance, when measured with an uncut/unfiltered (UV included) spectro?
The specification allows for dEab 5 variation for many colours, with neutrals being lower.

Emphasis mine.
You can have 2 proofs, both within tolerance. Both are dEab 4.5. One achieves this with L being +4, the other with L being -4 from the target. If you now compare the proofs between each other, you can see dEab 8. Both proofs are valid though. Now take your press work which does L on target but does b with +4 - inside tolerance, but worlds apart from both proofs.

Which is why proofs should be within 1 dEab from the target, to minimize effects like this.

 

[pp411]

This has been great information for me - thanks everyone! The customer's customer approved our proof right away. Which I guess is good in the short term, but long term I'll see if they can add the strip to the proofs if they are going to give us their proofs again. Hopefully they won't change their mind while it's on press....

 

[Correct Color]

Ideally the Wedge would be processed with the page data...

Just to add one more point...

No, you really wouldn't want to do this. The whole purpose of the wedge is simply to characterize the condition of the proofing device when it made the proof. If it's off, then if there's an issue, that's most likely it. If it's on, then that tells you the issue is elsewhere.


Mike Adams
Correct Color

 

[Stephen Marsh]

Just to add one more point...

No, you really wouldn't want to do this. The whole purpose of the wedge is simply to characterize the condition of the proofing device when it made the proof. If it's off, then if there's an issue, that's most likely it. If it's on, then that tells you the issue is elsewhere.


Mike Adams
Correct Color

Mike, are we both saying the same thing? It may vary from RIP to RIP on how things work…

The page/s and the colour bar/s should all be processed the same, so that a 40% magenta value in the file is colour managed exactly the same as a 40% magenta value in the colour bar/media wedge. The results of the colour management are then measured in the wedge to see if the numbers pass/fail against the specification being “certified”.

If the page data went through a simulation profile for the specification being certified - but the wedge did not, then the colour would be different and the wedge would be useless as the 40% magenta in the file would not match the same value in the wedge.


Stephen Marsh

 

[Correct Color]

Stephen,

Um, actually, no, we're not saying the same thing.

You're saying it correctly, and I was saying it wrong.

Of course you're right. The proofer has to be printing as characterized for the wedge to come out correctly, but at the point at which the wedge is read, those values aren't necessarily known; the only ones that are are the reference values of the wedge.

I don't work in that field all that often, and should have thought it through a little more before sounding off.


Mike

 

[curiosity]

Might want to consider this approach...

Might want to consider this approach...

We often receive "proofs" reflecting a client's expectations.
Sometimes they are so poorly produced, the difference between client supplied file and proof is irreconcilable.
That said, because clients often want what they see onscreen or what they send as a "proof", we go through the process of color editing to better resemble the client's expectations. Of course, only after a discussion of the "facts".
I've found that most clients just want good customer service. Fewer clients equate service to, in this case, color management accuracy.
IOW, "...don't tell me what I'm doing wrong, just fix it."
So...we fix it, and client gets what they want....right or wrong.