Re: GaMapICC
Re: GaMapICC
- Subject: Re: GaMapICC
- From: Martin Orpen <email@hidden>
- Date: Thu, 16 Oct 2008 09:03:41 +0100
On 14 Oct 2008, at 08:26, Klaus Karcher wrote:
I know that gamut mapping always has a touch of alchemy and is often
rather a matter of taste than objective science, but the loss of
details I've noticed in some situations (e.g. in saturated reds and
light cyans) seems overdone to me (I saw completely disappearing
cirrostratus clouds).
I wonder how others feel about it.
Klaus
I have also seen this problem and can show you an example.
This is a small section of an image that I am working on that is in
AdobeRGB:
<http://www.web36531.clarahost.co.uk/filedump/16665_aRGB.tif>
This is a similar section cropped from a device link conversion to ISO
Coated v2 using Argyll:
<http://www.web36531.clarahost.co.uk/filedump/16665_ISOv2_DL.tif>
And here is the same section cropped from a conversion using Argyll's
gamut mapping:
<http://www.web36531.clarahost.co.uk/filedump/16665_ISOv2GM.tif>
From a retoucher's perspective there's too little information in the
weak channels in the gamut mapped version. Without small amounts of
cyan and black you end up with no detail in the magenta.
A pixel count shows that the original RGB has 62,569 colours. The
device link conversion creates an image with 95,623 colours whereas
the gamut mapped conversion only has 51,383 colours :-(
I'm using Argyll all the time to create device link profiles and apply
them to images but I'm not convinced by the gamut mapping features at
all.
--
Martin Orpen
Idea Digital Imaging Ltd
_______________________________________________
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