Inkjet proofing: relative or absolute?

[Colorblind]

Hi everybody, I know this topic may have been covered already (but can't find where after a quick search). I'm trying to produce Gracol/G7 and SWOP/G7 proofs on my inkjet. I'm pretty successful when I read my patches and compare them to idealliance datasets. But because my inkjet paper is a bit bluer than the white point for both, I think I should use absolute colorimetric which adds a light tint to the paper background. Is there any pros/cons versus using relative colorimetric on an inkjet paper that would be less blue (but probably much more expensive)? I'm using ApogeeX all the way so Lukas, your opinion is welcome

 

[TerryWyse]

Always use absolute, relatively speaking

Always use absolute, relatively speaking

You really should always be using absolute colorimetric rendering if your goal is to produce a proof that matches a specific specification such as GRACoL Coated1 or SWOP Coated3. Of course, it tints the background of the inkjet paper (generally) but it also affects colors as well, even the solids. If you examine/measure the primary and secondary solids with both absolute and relative colorimetric, you'll see that relative "artificially" raises the chroma (purity/"colorfulness") of these inks despite logic telling you that the difference should only be in pure whites or pastel colors.

The ONLY case to make for using relative is if the inkjet paper itself ALREADY matches exactly the target paper. In other words, if you're producing a GRACoL proof (paper L*a*b* of L*95 a*0 b*-2) then you would have to use an inkjet media that measures precisely that value.....even then, I would still produce two proofs with these two rendering intents (even though there should be no background tint) and actually verify via measurement and comparison to the GRACoL data set that it's a reasonable match....with particular attention to the solids.

Regards,
Terry Wyse

 

[meddington]

Terry is 'absolutely' right

Terry is 'absolutely' right

However, Ive always found it a bit more visually appealing to edit the proof's absolute rendering (white point) to match the actual printing stock used, which quite often differs slightly (or significantly) from the white point of particular characterization data sets. The drawback is that control strips might not measure as closely to said characterization data, but likley matches the production run a bit better visually.

 

[prepression]

A very clear and detailed explanation of color spaces including absolute, relative and perceptual can be found in this resource:

Amazon.com: Exploring Digital PrePress: The Art and Technology of Preparing Electronic Files for Printing (Design Exploration Series) (9781418012366): Reid Anderson: Books

 

[Colorblind]

Thank you guys, I think I have my answer.

 

[Lukas Engqvist]

Sorry been too bussy, yes on a proof inkjet I would normally go Absolute. In Apogee you do have an extra option that may be worth noting and it is the link exceptions. In Apogee you also have the Closed loop workflow for the proofer, to optimise the proofer rendering of a printer profile.
There is also the possibility to have profile link exceptions (so that pure colours are not managed other than dotgain compensated), You also have the option to treat CMYK as relative and Spot colours as Absolute if you like.

I would use Absolute since it is the more correct, but there have been instances where I have switched to relative because there were large areas of white, and that customer was very sensitive to the tone that was laid down in the paper. There is one more reason I may want to use relative, and that is when I have alot of images cut out to white, and I want to make sure the white is white (and not boxes of 1-2% background dither) If you use absolute on these you will have a much harder time determining if there is 1-2% boxes, because colometrically the 1% tone and the paper white simulation might be visually the same (the paper can only be made darker when matching white point)

So I would back previous posts on the theory, but in reality, well some instances you will have to decide what is it you want from the proof: the lowest delta E (most of the time the proof and the printed product will not be viewed at the same time) or do you want to catch production pitfalls such as close to zero boxes that will definately be uggly in print.

 

[Mike Strickler]

Inkjet proofing: relative or absolute?

Inkjet proofing: relative or absolute?

However, Ive always found it a bit more visually appealing to edit the proof's absolute rendering (white point) to match the actual printing stock used, which quite often differs slightly (or significantly) from the white point of particular characterization data sets. The drawback is that control strips might not measure as closely to said characterization data, but likley matches the production run a bit better visually.

If the reference profile has been edited and the control wedge is also color-managed, that is, printed through the same profile, the match to the (new) reference will be at least as good as if you had tried to match GRACoL--assuming abs. col. in both cases. Actually, having a proofing paper that is slightly bluer than the reference can be helpful, especially if it is also a bit lighter, as mostly yellow ink is added to match the more neutral reference white, and this has very little darkening effect on the background.

Mike's overall point is important: Nobody actually prints on a true GRACoL substrate, which is more of a statistical average, and not even a good average now as brighter, bluer stocks are increasingly used. Adjusting the white point of the reference to match the actual substrate will give more meaningful proofs, but it's important to discuss this with customers as they may still prefer a proof to the standards (GRACoL, SWOP3, etc.).