[argyllcms] Re: Printer profiling walkthrough
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Wed, 16 Mar 2016 22:06:44 +1100
Andrew Larkin wrote:
Hi,
I thought the mpp flow was needed for
pre-conditioning the final test chart generation.
not for any workflow that results in an ICC profile.
This should be done using a first pass icc file instead?
yes - but if you are not sure what you are doing, it's best to
skip the pre-conditioning. Just create a test chart without the
targen -c option, measure the chart and create an ICC profile.
Once you have the workflow figured out, you can try re-creating
the test chart with a pre-conditioning profile.
Where there is a high degree of adaptation used in targen (1.0)
does this imply that there are no test patches outside the supplied
pre-conditioning
profile space?
Nothing to do with degree of adaptation, and it's not going to limit anything,
since matrix profiles and A2B's don't limit the gamut (they are device->CIE
values).
(Of course if you are talking CMYK profiles and your pre-conditioning profile
was
created from a chart with a very much lower total ink limit than the chart
you want to create, the pre-conditioning will not be as accurate in the device
gamut areas that have had to be extrapolated.)
In the test chart I created years ago, I included not only a substantial
number of grey ramp test patches but also a cylinder-like array of near grey
patches. I
found this useful for capturing and correcting the grey ramp's neutrality and
supporting very lightly toned monochrome prints. Is there an option for
something like
this or would I need to add these patches myself?
Nothing directly, although there are ways of achieving that type of thing.
For instance, after creating a profile, you could run an L*a*b* defined neutral
ramp through xicclu -fif to give you device values, and then manually add
them to your .ti1 file.
As our working colour space is almost
exclusively ProPhoto, this should be used as the source space for colprof?
Probably not. ProPhoto is not a good gamut definition in most circumstances,
since
it is unrealistically large. You need to come up with some strategy for
establishing
realistic image source gamuts.
For high quality, I recommend per-image source gamuts and creating per-image
device link profiles. But that workflow won't suite everyone.
Graeme Gill.
Other related posts:
