uncorrected color bars?

[jaed]

Hi guys, I've seen alot of people discuss the benefits of corrected vs uncorrected color bars, and I agree that using uncorrected probably gives more useable data. My question is how to run uncorrected color bars.

Currently I'm using a standard gracol curve that comes supplied with my rampage rip software and it automaticly corrects the color bar. Im wondering if there is a check box or some special file format for the color bar that would let it pass through with out getting corrected. Or do I need to set up my curves some where esle, i.e. in Adobe or some other software, so that the the color bars aren't processed at all.

thanks for any help or insight

JaeD

 

[gordo]

Hi guys, I've seen alot of people discuss the benefits of corrected vs uncorrected color bars, and I agree that using uncorrected probably gives more useable data. My question is how to run uncorrected color bars.

Currently I'm using a standard gracol curve that comes supplied with my rampage rip software and it automaticly corrects the color bar. Im wondering if there is a check box or some special file format for the color bar that would let it pass through with out getting corrected. Or do I need to set up my curves some where esle, i.e. in Adobe or some other software, so that the the color bars aren't processed at all.

thanks for any help or insight

JaeD

IMHO, the color bars should have the same curve(s) as the live image area otherwise you disconnect the bars from the presswork you sell as well as the standard that you've set yourself up to. Also, if the color bar does not have an applied curve then it will likely image with the plate's natural non-linear curve. In either case there will be a curve it's just that, unless you curve it, the color bar will have meaningless tone patches.

best, gordo

 

[Johu]

I totally second Gordo here. If your press has some sort of measuring system or even better a closed loop system your measuring system tries to control the color bars to some standard value you've set them to, but this might have a totally undesired effect on you print job colors (excluding the color bars).

 

[jaed]

let me ask you this then, say you wanted to check your print contrast values for example and your 75% patches were being cut back to say 65%, wouldn't that adversly affect the results of that test? The reason I'm asking is we don't have an accurate color managment system in place yet and I would like to check whats happening on press with known control targets. If 22% dot gain @ 50% is acceptable, what does that equate to if your 50% is imaged at lower percentage, say 35%-40%?

 

[gordo]

let me ask you this then, say you wanted to check your print contrast values for example and your 75% patches were being cut back to say 65%, wouldn't that adversly affect the results of that test? The reason I'm asking is we don't have an accurate color managment system in place yet and I would like to check whats happening on press with known control targets. If 22% dot gain @ 50% is acceptable, what does that equate to if your 50% is imaged at lower percentage, say 35%-40%?

This is not a color management issue. It is a question of simple tone reproduction.

Dot gain in this context is irrelevant. What is important is final tone reproduction. It may take you a different dot gain than me, or according to the screening that you're using, or the paper type you're printing on to get to the final target tone value. It doesn't matter. What matters is that if you have a 50% request in the file and you've targeted 65% in the final presswork for that 50% patch then 65% is what you should measure. If a cut-back curve means that the 50% request becomes 40% on the plate which becomes 65% in the press work (15% dot gain from the original request, or 25% dot gain from the dot on plate) then you've achieved the goal.

A dot gain value, however you calculate it, is of some limited interest forensically if something goes wrong in the presswork but that's about it. Print contrast in a CtP environment can be whatever you want it to be. Just apply a plate curve. It doesn't have the relevance that it had in the film-to-plate days of the past. There are better ways to determine optimal ink transfer.

Is this making any sense?

best, gordo

 

[naplajoie2000]

We run our 4/c Heidelberg color bars with the same plate curve (DGC) as is used in the job. We do run linear patches that in Esko you can assign to a different screen angle and apply a linear curve (no curve) to. These patches we use to check if plate is in spec, as we bake our plates and they do drift, and on press they can see actual press gains across the sheet along with density readings.

 

[gordo]

We run our 4/c Heidelberg color bars with the same plate curve (DGC) as is used in the job. We do run linear patches that in Esko you can assign to a different screen angle and apply a linear curve (no curve) to. These patches we use to check if plate is in spec, as we bake our plates and they do drift, and on press they can see actual press gains across the sheet along with density readings.

A point of clarification...a linear curve is not a "no curve" it is in fact a curve.
Whether the plate is linear or not does not affect the ability to check if plates are in spec.
Press gains are linked with density readings because that is how gains are calculated.

best, gordo

 

[Lukas Engqvist]

Just want to clarify what I am sure Gordo assumes. A TVI compensated colour bar, that has the same dotgain as CMYK values is the correct method. Now if you do have a device link or other colour management system, the colour bars should not go through this system (May be interesting to put them through it to make sure the inksaving is transparent, but it would not help process control).

As you already seem to understand dot gain does affect print contrast. But it is not the total print contrast that is interesting anyway. I would recommend that you can map the print contrast at any dot gain and hopefully you will see that you are at a contrast plateau where a sway in density will not produce a large sway in contrast.
If you find that varying your density within your quality tolerance gives a high big difference in print contrast you may find that you will need to look over your process so that you have more contrast stability. Variations in contrast in a print run will show.
So even if print contrast is not totally relevant mapping how it varies can be a help to get a better understanding of what does what in your process.

 

[Lukas Engqvist]

A second thought, I'll try an analogy, but this is only thinking out loud.

If your press has a too high dot gain, think of it as an overweight man... getting him a smaller suit is just going to make him look like a sausage. You can tailor a suit that makes him look OK but the big problem, that he needs to deal with is his overweight.

In the same way if your press goes wrong then you need to fix the press... in my analogy send him to the gym to get him in shape... in the press, look over all the variables and get disciplined about routines, for packing and blankets etc.

When the man is fit, the large suit you made won't fit. He'll be looking like a sack. Same with your curves, if your press gets in shape you need to tailor a "new suit".

Letts assume your press is now fit. You have your suit, or rather your colour bars, made to fit. Now when the pressman gets sloppy or something else in your flow goes wrong, your suit will no longer fit.
You now have two options. Get the press/workflow back in shape (this is the good option). Or go out fix a new curve (like getting a new suit because the girth on the old one seemed to be getting too tight).
Fixing the curve to compensate for a bad press is never going to solve the problem, or make it go away… it just means you can "get away with" a little overweight... until you get a job that requires you to be fit.

I hope the story helped more than it confused. (You may ignore it if you don't like this approach to analogous thinking). An old press may be past the stage when it can "become fit" and maybe the only thing you can do is get it a "good suit", and that may be as good as it gets... but if we are aware of the limitations of our workplace we will hopefully be wise enough not to promise what we can't deliver.

 

[jaed]

Thanks alot guys, all of the info was very helpful. It's really nice to be able to have an intelligent conversation with people about printing for a change.