G7 tolerances

[gordo]

One more question to add. Which delta E calculation method is used to evaluate the sheets. I found a document on the idealliance web site, called G7 targeted, inside the G7 & GRACoL Qualification Kit for Sheetfed Offset. On page 1, it says to use Delta E*ab, which I think is the original delta E76 calculation which doesn't take into account perceptual non-uniformities. On page 25, the column heading is delta E'ab, but there is a footnote that says something like: ISO 12647 currently uses Delta E'ab (1976) but that one is inaccurate ... so you should use one of the weighted ones instead like CMC, 1994, or 2000.

And the proof verifier spreadsheet says delta E76

So come on people, which is it?

Simple. You hire another consultant. Perhaps this one:

best, gordo

 

[David Milisock]

Even better yet what do you do when the client doesn't give a hoot about what the machine says and says they don't like it? Color is subhective so is repeatability but IMO less so.

 

[jstotz]

So come on people, which is it?

Just to clarify, the "people" I was referring to were the Idealliance people, not you all who have been most helpful.

 

[Deleted member 16349]

Simple. You hire another consultant. Perhaps this one:

best, gordo

Gordon,

Is the young lady, in the photo, missing a finger on her left hand?

Maybe she was a press operator before becoming a consultant.

 

[povertycow]

A better mouse trap still hasn't been created, just a more complicated and expensive one. Sound familiar??

 

[gordo]

Gordon,

Is the young lady, in the photo, missing a finger on her left hand?

Maybe she was a press operator before becoming a consultant.

Probably a guillotine operator in bindery. Press operators tend to get degloved.

Yuck.

gordo

 

[Deleted member 16349]

Probably a guillotine operator in bindery. Press operators tend to get degloved.

Yuck.

gordo

Not digital printing but digit cutting. Makes sense.

 

[rich apollo]

One more question to add. Which delta E calculation method is used to evaluate the sheets.

Whoa, whoa, whoa there, Folks. The original poster asked a relevant question, and now this has gone and turned into a damn-the-G7-torpedoes tirade.

Currently, ISO specifies the use of ∆E(ab) - ∆E(76) - so that's what is currently used. HOWEVER, ∆E(2000) is considered to be much more in keeping with visual perceptual differences in color, THEREFORE, there is a push to have the ISO specs updated.

One of your questions is kind of odd-duck. Asking for the before and after TRCs is like asking for a round wheel. The "after" is predefined, and publicly available. You measure what the press is doing, and then adjust it to hit the predefined TRC (NPDC) of the printing condition you want to match. This is based on density (reflectance). At the same time you impose gray balance (relative to the paper white - or not), rather than hoping that it'll happen.

As to your question about the SID variance for 5 ∆E(ab), I don't know that there is a single answer for that. Different ink formulations with different pigment loads on different substrates may yield different results. I don't know. I also don't know that it's all that relevant. Are you trying to establish some kind of window of operation? Do you feel that the 5 ∆E is too restrictive? It's not for running, but for calibration purposes. You find where you hit acceptable color and tell the guy to stay there. Acceptable variability hasn't changed. We're just looking at another way to define, and hit the target.

G7 is a methodology to achieve ISO-12647 compliance. I have had pretty good success at achieving an acceptable press to proof match with it. I have not attempted to get Master Printer certified or FOGRA certified or Chuck Norris Approved because we couldn't justify the cost. But, I'll tell you what, if our clients told us to have a freakin' goat, we'd open a stinkin' petting zoo out back. If the client wants it, it's right!

Can I satisfactorily answer the questions of the nay-sayers? No, probably not. But your skepticism doesn't make the method invalid anymore than my acceptance makes it true.

Gordo, if you really want to participate in some G7 calibrations, I'm sure you could go along with a consultant and observe/participate. I'm thinking of one guy in particular, a friend of mine who does a lot of digital presses. I'll bet he'd welcome you. Also, you could call the guys at Chromix in Seattle. I'm sure you could get plenty of information there. There are also the guys at Alder Technology in Portland. Where do you want to go? I'll find you someone.

Whatever one's feelings about IDEAlliance or the consultants that sell G7 calibration services, they do not invalidate the methodology. And, yes, some folks are going to do it better than others. A lot of folks have trouble with math. That doesn't make math wrong, does it?

 

[rich apollo]

A better mouse trap still hasn't been created, just a more complicated and expensive one. Sound familiar??

G7 calibration can be implemented for no cash out of pocket. Everything you need is freely available but the measurement device.

 

[David Milisock]

>Whoa, whoa, whoa there, Folks. The original poster asked a relevant question, and now this has gone and turned into a damn-the-G7-torpedoes tirade.

Currently, ISO specifies the use of ∆E(ab) - ∆E(76) - so that's what is currently used. HOWEVER, ∆E(2000) is considered to be much more in keeping with visual perceptual differences in color, THEREFORE, there is a push to have the ISO specs updated.

You should be a politician! I guess it depends on the definition of is?

 

[gordo]

With some snipping:

One of your questions is kind of odd-duck. Asking for the before and after TRCs is like asking for a round wheel.The "after" is predefined, and publicly available.

Possibly, but that's what I'm used to seeing - a starting point (our print condition before) and an ending point (our print condition after). So it would help me to understand.

As to your question about the SID variance for 5 ∆E(ab), I don't know that there is a single answer for that.

I don't disagree. I'd just like to have a sense of that relationship is on press as experienced by different people. If I had a press then I would probably run some density tests to see what the 5 ∆E(ab) window looked like.

Do you feel that the 5 ∆E is too restrictive? It's not for running, but for calibration purposes. You find where you hit acceptable color and tell the guy to stay there. Acceptable variability hasn't changed.

I thought that 5 ∆E(ab) was for running rather than just for calibration. My mistake. What is the acceptable variability that hasn't changed?

G7 is a methodology to achieve ISO-12647 compliance. I have had pretty good success at achieving an acceptable press to proof match with it. I have not attempted to get Master Printer certified or FOGRA certified or Chuck Norris Approved because we couldn't justify the cost.

Not Chuck Norris Approved? Well there's the problem in a nut shell. Chuck Norris doesn't build tone curves. He just stares at the plates and they curve by themselves. ;-)

Can I satisfactorily answer the questions of the nay-sayers? No, probably not. But your skepticism doesn't make the method invalid anymore than my acceptance makes it true.

The Richard Dawkins gambit. I don't think that G7 doesn't work. I'm just looking for clarity, and perhaps some evidence backing up some of the claims made for it.

Gordo, if you really want to participate in some G7 calibrations, I'm sure you could go along with a consultant and observe/participate.

Thanks. There is a consultant in Vancouver that I'm planning to visit the next time I'm in town. Thanks for the other references, I'll look them up.

best, gordo

 

[David Milisock]

>If I had a press then I would probably run some density tests to see what the 5 ∆E(ab) window looked like.

If you had a press you most likelycouldn't afford to run the tests.

 

[gordo]

>If I had a press then I would probably run some density tests to see what the 5 ∆E(ab) window looked like.

If you had a press you most likelycouldn't afford to run the tests.

I hope you're joking. Such a test would be easy to do during make ready on a live job, or over the course of a number of live jobs, or even possibly during the initial calibration run.

best, gordo

 

[David Milisock]

>I hope you're joking.
Of course I'm not joking, a test would cost the company well over $2,000 in time, material and lost production. I don't know what market you're in but 50m of an 8.5 x 11 on #100 text 4 + AQ over the same 3 panel folded is only $2,000. with a local delivery

Well just in this forum on another thread a guy admits that most of his work comes from SWOP condition but they calibrated G7, sound like colo rmanagement gone awry.

 

[gordo]

>I hope you're joking.
Of course I'm not joking, a test would cost the company well over $2,000 in time, material and lost production. I don't know what market you're in but 50m of an 8.5 x 11 on #100 text 4 + AQ over the same 3 panel folded is only $2,000. with a local delivery

I don't think you understood my post, or maybe I wasn't clear.

gordo

 

[rich apollo]

>I hope you're joking.
Of course I'm not joking, a test would cost the company well over $2,000 in time, material and lost production. I don't know what market you're in but 50m of an 8.5 x 11 on #100 text 4 + AQ over the same 3 panel folded is only $2,000. with a local delivery

Well just in this forum on another thread a guy admits that most of his work comes from SWOP condition but they calibrated G7, sound like colo rmanagement gone awry.

There is the outlay for consumables and labor. Think of it as an investment.

My first foray into color management cost the company I worked for something on the order of $40,000 in travel, consumables, hardware, software, consultants' fees, classes, et cetera. I did some calculations and found that if they saved 1 minute on each make-ready they would see a total return on investment in a year.

I'm in a heatset web plant, now. We've run all-day calibrations and testing. You wanna' talk about expense? But, I haven't heard anyone in our organization question the value of those exercises. $2000 is very little for the potential savings. I spent that out of my own pocket to attend a class on G7 implementation. What would you spend otherwise - what's your alternative? Is it simply too expensive to fingerprint a press? G7 is just fingerprinting a press with the addition of a couple of different metrics.

G7 is a methodology. You can calibrate to hit any colorspace you want. I've used it here to run heatset newsprint, SCB, and SWOP coated #5. I can teach you how to extract the relevant data for G7 calibration out of ANY ICC profile.

 

[David Milisock]

>G7 is a methodology. You can calibrate to hit any colorspace you want. I've used it here to run heatset newsprint, SCB, and SWOP coated #5. I can teach you how to extract the relevant data for G7 calibration out of ANY ICC profile

That's not the issue, the problem is that for the general printer, one that uses 30 different paper stocks a week with run lengths that are short for offset but not short enough for digital, can you make you process useful to the company? That means profitable! For the most part due to the cost of creating the profiles and the current low profit margins on work the answer has been no. What I read on this forum for the most part is technical double speak.

I'm in multiple shops on a regular basis, in my area we have had a record number of consolidations,(read print shop closings) what these plants need is a clear path for taking a multi-million dollar investment and making it repeatable in terms of being color productive, for the most part that can be most cost affectively achieved as part of the contracted equipment installation.

What profitability means in the real world of having to make a press payment every month that's equal to that of a top of the line Mercades is that they will have the Indigo and press calibrated very closely for one or two papers and the rest will fall where they fall. It also means that the profile is NOT adjusted when the paper lot changes white point, as IT WILL almost every different lot. THAT MEANS THE COLOR CHANGES! For the most part the cost of individual media based color management can only be afforded for a very select few types of papers or those used for regular runs where the run is size is sufficient or where the client is willing to pay for the calibrtion.

We need to stop BSing people, calibration is all part of the press or other hardware installation process and custom profile creation had better take into consideration two factors, profitability for the user and the fact that 99.9% of all Adobe users (which is a huge amount of the print market) are using default color management settings (BTW even CorelDRAW uses the same default CM settings as Adobe. That unfortunately means in many cases sRGB and Web coated CMYK, sometimes Adobe RGB.

 

[rich apollo]

That's not the issue, the problem is that for the general printer, one that uses 30 different paper stocks a week with run lengths that are short for offset but not short enough for digital, can you make you process useful to the company? That means profitable! For the most part due to the cost of creating the profiles and the current low profit margins on work the answer has been no. What I read on this forum for the most part is technical double speak.

Yes, David, I can make it profitable. I have 3 comments:

1) If a printer is running such a wide variety of papers, they're shooting themselves in the foot. Standardize on a few papers and they'll run much more happily.
2) G7 doesn't add or take away from the complexity of the situation you describe. If/when you change the paper, you change the color.
3) This boils back down to tolerances, or acceptable variation. If you constantly change papers, then you will have to live with greater color variation. If you standardize on some particular papers, then you'll enjoy tighter tolerances. The choice is the printers'.

And then I'll ask you what is your proposal? How would you recommend that the printer(s) deal with this issue?

I think you'll also find that I'm far too under-educated to engage in "technical double-speak". I'm a PrePress guy working in print shops. I've worked in commercial sheet-fed and heatset web.

What profitability means in the real world of having to make a press payment every month that's equal to that of a top of the line Mercades is that they will have the Indigo and press calibrated very closely for one or two papers and the rest will fall where they fall. It also means that the profile is NOT adjusted when the paper lot changes white point, as IT WILL almost every different lot. THAT MEANS THE COLOR CHANGES! For the most part the cost of individual media based color management can only be afforded for a very select few types of papers or those used for regular runs where the run is size is sufficient or where the client is willing to pay for the calibrtion.

We need to stop BSing people, calibration is all part of the press or other hardware installation process and custom profile creation had better take into consideration two factors, profitability for the user and the fact that 99.9% of all Adobe users (which is a huge amount of the print market) are using default color management settings (BTW even CorelDRAW uses the same default CM settings as Adobe. That unfortunately means in many cases sRGB and Web coated CMYK, sometimes Adobe RGB.

I don't think you're grasping the intent behind the G7 initiative. The idea is to NOT create custom press profiles, but to calibrate the system to achieve an acceptable match to standard, pre-defined colorspaces. So, for example, in my shop, I downloaded FOGRA 42L. This is a FOGRA built profile for heatset newsprint. From the profile I extracted the data I needed and then set about using G7 methodology to put our presses in line with the FOGRA 42L colorspace. I make sure that the proofs are consistent, I make sure the plates are consistent. That's what I can do.

I have done the same here for SCB and for #5 gloss coated. I cannot attune the system to every brand of paper that comes onto the pressroom floor. At times we end up running on SCA or Hi-brite; two paper-types that I have not had the chance to test. We achieve an acceptable color-match for our clients.

No, I'm not running cosmetic ads. I've been in shops that were, and yes, they run the calibration for very specific varieties of paper. The group I'm thinking of specifically, has hundreds of plating curves. They are not, however, your run-of-the-mill, commoditized print service provider. They have a clientele that is willing to pay for their efforts. In a sheetfed environment, there are tools available now that make it incredibly easy to make adjustments for specific papers. Heidelberg, Bodoni Systems, Alder Technologies, and Alwan Color are a few groups that provide tools that make these adjustments really easy to do.

 

[David Milisock]

>I don't think you're grasping the intent behind the G7 initiative.

I've grasp that intent from the beginning, to make money from the print industry.

None of these standards make a standard commercial printer any money. In fact I'm on press with a file created in Utah, Ottaw and Toranto, sent to PA for pre-press, proofed on an Onyx driven Epson 4000 with a media profile and printed on a postscript color managed Rampage driven heidelberg press. No G7 standards and the proof looks remarkable like the finished product, yes the white point is different. However the job was in house 6 hours and a total of 14 hours later on the truck.Oh by the way we made a good profit.

>I cannot attune the system to every brand of paper that comes onto the pressroom floor. At times we end up running on SCA or Hi-brite; two paper-types that I have not had the chance to test. We achieve an acceptable color-match for our clients.

Now we have heard the truth. Here's another one the designer will speck the paper, you'll print on it and deliver on time.

 

[rich apollo]

None of these standards make a standard commercial printer any money.

Once again, David, G7 is a methodology to help one achieve compliance to a color standard. It's not a standard in and of itself.

In fact I'm on press with a file created in Utah, Ottaw and Toranto, sent to PA for pre-press, proofed on an Onyx driven Epson 4000 with a media profile and printed on a postscript color managed Rampage driven heidelberg press. No G7 standards and the proof looks remarkable like the finished product, yes the white point is different. However the job was in house 6 hours and a total of 14 hours later on the truck.Oh by the way we made a good profit.

Okay, so, as a proof is a prediction of what's going to happen on press, what press conditions were the proofs predicting?

>I cannot attune the system to every brand of paper that comes onto the pressroom floor. At times we end up running on SCA or Hi-brite; two paper-types that I have not had the chance to test. We achieve an acceptable color-match for our clients.

Now we have heard the truth. Here's another one the designer will speck the paper, you'll print on it and deliver on time.

What truth is that, David?

The designer will spec the paper if you let 'em. I think it's appropriate for the designer to specify finish, weight, et cetera. I don't think it's appropriate for the designer to spec the actual paper. It's in their best interests to allow the printer to deal with the paper. We have people aware of pricing and availability. Printers are aware of various printing concerns like marking, cracking, drying, et cetera. And the printer is the one who, hopefully, has standardized his system to achieve a predictable product with certain consumables.

You don't like G7, so don't use it. I like G7 and have had success with it, so I'll continue to use it. You cannot reasonably say that it is ineffective when there is such a large number of people employing it successfully.

I would still be interested in hearing what you propose as an alternative.