[Openicc] meta data in test chart

[Graeme Gill]

edmund ronald wrote:
> However it seems
> to me that the CLASSIC ICC profile scenario has as one great design
> weakness (among others) precisely the fact that the source gamut is
> NOT known prior to gamut mapping, as far as I know.

It may not be explicit for many profile creation tools (Argyll excepted),
but there has to be some assumption in there somewhere. Yes, it's
a weakness in the assumption that device profiles can be mixed and
matches, mitigate to some degree by V4 profiles that use the PRMG
(as long as you like a saturation-like intent, rather than a free
choice of intent).

> The reason for this design decision is -as I have come to understand
> it - is partly that for press applications, one does not wish to have
> adaptive mappings because they would destroy the coherence of work
> contributed from different source gamuts. Fine arts printers and
> photographers may have different opinions, but it would seem no one
> asked them.

I don't see that aspect as connected to your previous paragraph. If
a source gamut is a tangible part of the workflow (rather than trying
to make out that it doesn't exist, which seems to be the traditional
approach), then there is no obstacle to choosing the granularity
with which it is used - you could adapt per object, per page, per
document or even larger granularity.

Graeme Gill.