-----Original Message-----
From: marti.ma...@littlecms.com [mailto:marti.ma...@littlecms.com] 
Sent: 20. februar 2010 17:25
To: Esben Høgh-Rasmussen
Cc: lcms-user@lists.sourceforge.net
Subject: Re: [Lcms-user] Problems with non-perceptual intents in LCMS2

> Hi,
>
> Many thanks for reporting. No, the issue is not between keyboard and chair
> but in the code, or maybe in the spec itself. The nikon profile has only one
> table for perceptual, and then a matrix-shaper. So, what to do when the user
> requests rel.colorimetric? lcms-1 was reverting to perceptual, and I guess
> this is the required behavior. This is documented in the tutorial, page 13,
> "Rendering intents", and due to a bug, the matrix-shaper was being used
> instead.
>
> So thanks a lot. The fix takes 2 lines of code and will be in the git in few 
> hours.
>
> Thanks again
> Marti.


Hi Marti,

Thank you very much!

I updated and it looks good so far, but we will test it quite thoroughly.

Just another quick question (or maybe a feature request): is there a safe
way to apply a transform from several threads without locking?

I guess I can create several transforms and call each from its own thread,
but that may eat up a lot of memory when a large CMYK profile is used.

We anticipate 16 to 32 cores in the not-too-distant-future, and we
currently need to call cmsDoTransform() per line. I hate to take any kind
of lock inside these calls :-/

It would be nice if it is possible to use the same transform with a bit of
thread-local data and avoid the locks (or similar). Bonus points if we could
just put it on the stack and be done with it.


Kind regards,

Esben Høgh-Rasmussen Myosotis
Software Engineer
 
Phase One A/S
Roskildevej 39
DK-2000 Frederiksberg
Denmark
Phone: + 45 36 13 12 07
Fax.:    + 45 36 46 02 22
Mobile: + 45 28 51 10 79
E-mail: e...@phaseone.com
www.phaseone.com




------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Lcms-user mailing list
Lcms-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lcms-user

Reply via email to