on 1/6/06 11:03 AM, Michael Fox Photography News Account at
> Troy, Dan,
> Thanks sincerely for your suggestions. But I'm pretty sure that the issue
> is not the RIP settings. If it were, then both profiling packages would
> have the same or similar problems. But they don't. Also, if it were the
> RIP settings, then the changes to those settings would have an effect on the
> problem. As I mentioned in my original posting, I have tried numerous ink
> limits in both the individual channels and total ink. I've even tried
> numerous papers. I even checked with the manufacturer of the RIP to make
> sure I am setting it properly (I am). While these ink limit changes
> affected the minimum L* (of course), they did not affect the three problems
> I reported. Finally, if it were a problem with my RIP settings, then others
> would not have responded (off list) with similar results from other RIPs.
> So, the three problems/observations remain:
> Monaco reverses (slightly) after reaching dMax (Gretag does not)
> Gretag consistently produces slightly lower dMax than Monaco
> Gretag consistently produces higher chroma error in the neutrals
> My posting had two purposes: to confirm that others were/are seeing this;
> and to solicit thoughts from people on this list who have knowledge of how
> the neutrals are calculated/analyzed in an effort to understand why this
> might be happening.
> As I mentioned above, others have confirmed the same problem that I've seen.
> But I'm still interested in understanding how these problems could exist.
> In other words, I can understand how one software package might extrapolate
> or interpolate gamut boundaries differently or perform gamut mapping
> differently. After all, that's the "secret sauce" for the profile software
> guys, right? But what I'm wondering is how two different packages,
> presented with the same data, would produce different neutrals for the
> RelCol intent. It seems to me that neutrals for RelCol should be
> consistent. Is it possible that, even in the RelCol intent, the software
> maker has discretion such that the profile calculation can be weighted more
> towards some other factor(s) that would give up neutrals accuracy?
Well if you really want to compare apples to apples then you shouldn't be
looking at the rel col intent. Instead you should be evaluating the abs col
intent. The relative colorimetric intent is scaled to make whites look white
and depending if you used a Gretag or X-Rite device and if had the UV filter
engaged will skew this table. I.e. Did you correct for optical brighteners
in Gretag? X-Rite doesn't have this option instead relying upon the device
to filter out UV content.
So if you really want to compare one product to the next look at the abs
col table which is basically the measurement data. Incidentally, v4 ICC spec
profiles are supposed to resolve this ambiguity in the rel table as some
manufacturers would adapt the data for that rendering intent while others
Lastly, you are going to have different neutrals built by both products
because you will not be able to create a custom curve in Gretag only select
a GCR 1-4 while Monaco allows you to create a custom black curve. Both
products will build neutrals differently because of it even if you select
the same black start and similar curve shape. US versus Swiss neutrality!
Dan B. Reid
Color Management Products & Training for Print, Internet, & Motion Graphics
http://www.rpimaging.com | Toll Free: (866) RGB-CMYK