[argyllcms] Re: Comparing Input / Output ICC profiles the correct way?
- From: "Alastair M. Robinson" <profiling@xxxxxxxxxxxxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Thu, 14 Oct 2010 09:10:58 +0100
Hi :)
On 14/10/10 07:46, gklima@xxxxxxxxxxxxxx wrote:
But comparing an artificial input profile as for example Iso Coated ECI v2
which has a default RI of perceptive (what I clearly not understand, how a
single profile could have a RI. I though only the relation / translation
between two profiles could have a RI)
Yes, logically you're correct, but in practice ICC profiles have
multiple tables - one for Relative RI and one for Perceptual RI - the
Perceptual table has a certain amount of gamut mapping "baked" into it.
For V2 profiles, the source of this gamut mapping is up to the profile
creator (and can be specified using colprof's -S parameter), and for V4
profiles there's a reference "typical" print gamut which should be used.
Neither of these situations is ideal, but since generating a
gamut-mapping takes a significant amount of time, it's a sensible
compromise between slightly sub-optimal results and waiting 10 minutes
for a transform to build!
Assuming exactly the same media as used for the isocoated with also the
same spectrophotometer then it would too be absolute RI. But using a
different media with another white point then is should be relative RI?
It depends what you're trying to achieve.
For instance, last week at work we put a profiling testchart on the
press and ran both coated and uncoated paper. Having built profiles for
each, I generated proofs for some images we'd put on the same sheet, and
printed them off on high-quality inkjet paper. Using Absolute RI I get
a good side-by-side match - and by using this RI I could, given the
right software, verify that my inkjet profile's gamut is large enough to
give accurate proofs for side-by-side comparison.
On the other hand, if I wanted to show a proof to a customer, I wouldn't
want to use Absolute RI because, unless the proof's cut out, the
unprinted bright white inkjet paper around the proof would make the
proof itself look very yellow - so instead Relative RI would be
suitable. So if I wanted to verify that the inkjet's gamut is large
enough for this task, I'd compare using relative RI.
Hope that's some help.
All the best
--
Alastair M. Robinson
Other related posts:
