Re: basICColor_Testimages_2007
Re: basICColor_Testimages_2007
- Subject: Re: basICColor_Testimages_2007
- From: Roger Breton <email@hidden>
- Date: Thu, 22 May 2008 19:29:24 -0400
Mark,
> These appliactions are not CHANGING the gamut of the press. They are giving
> you different results
> from the same charaterisation data depending in the settings in the
> application.
I agree that, strictly speaking, gamut = Device to PCS (CMYK to Lab). My
discovery is that, when converting from Lab to Device, using different
profiles yields "gamut" contractions or expansions. In other words, the same
source colors are "mapped" more or less closer to the center by simply
selecting different profilers.
This is obvious when studying "TestingSeparations_ChannelAllocation-en.tif".
> These results may or
> may not be within the gamut of the press which can only be altered by printing
> conditions; ink, paper,
> press and process not to mention the predilictions of the press operators and
> the working environment.
> In other words they may be erroneous.
Yes, these predictions may very well be erroneous :(
> All you are seeing here are different (erroneous?) results generated by the
> application using different,
> possibly inappropriate conversion settings. What matters is that you produce
> an optimised result that
> the press is ACTUALLY able to achieve.
Hmmuh. How do you suggest we define an "optimized result" that the press is
actually able to achieve? Do you mean going on press with sample images
converted from all these profilers? To compare the results "visually"? Or
would you apply some numerical criterion?
I'm just throwing ideas in the air.
> It is quite possible to generate differen relative CMYK values (that is,
> profiles) which achieve the same
> colorimetric result.
Yes, that's one sensible explanation.
> This is why you will not find any CMYK numbers inside an
> ISO standard. You will
> only find them in profiles generated from charactersiation data that conforms
> to the tolerances in the standard.
>
> Mark Stegman
Come to think of it, this may be a part of the standard that is missing,
define "good" reproduction quality based on characterisation data. Seems
like a no-brainer to me. But one that, in practice, may prove quite
challenging and, perhaps, political too.
Nevertheless, you'll agree that these synthetic images are very useful and
we should encourage production of more such test images for mere mortals
like us?
Best regards,
Roger Breton
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden