Subject: Re: Jack Bingham's comments (Was ColorEyes etc)
Subject: Re: Jack Bingham's comments (Was ColorEyes etc)
- Subject: Subject: Re: Jack Bingham's comments (Was ColorEyes etc)
- From: Edmund Ronald <email@hidden>
- Date: Thu, 16 Jan 2003 12:02:03 +0100 (CET)
Mark Buckner <email@hidden> mused thusly on a rainy day:
>
As for the anticipated "Camera Raw" tools, I guess I'll have to wait and
>
see. I can't imagine them replacing good camera profiling tools, because
>
(at least the DSLR) manufacturers won't give you truly RAW camera
>
data. It all has to come through some pipeline, whether that is the
>
manufacturer's software or some plug-in for PS. Either way, something is
>
going on behind the scenenes. If I'm looking at this wrong, please clear
>
things up in more detail as to how this will work, for example with Nikon
>
or Canon DSLR files. The fact remains that every camera is different, and
>
I don't see a way to avoid the need for individual custom profiles.
No Mark, as an engineer my understanding is very different. on Nikon, NEF
really is RAW ie. you get the Bayer Matrix. You then have to deconvolve
this into some approximation of the true scene, and in fact the
open-source tools may initially be much better at that than ANY commercial
software because the deconvolution process is one of those hard problems.
My conjecture is that very little destructive processing gets done on a
professional camera operated in RAW mode - except for some possible dead
pixel removal or dark-field substraction, precisely because the
manufacturer wants later buyers of the system to be able to get better
results from the same pictures by means of upgraded software.
As to whether every camera needs to be individually characterized, I
really see no reason this can not be done at manufacture time,
simultaneously with the integrity check that is doubtless performed after
the anti-aliasing filter has been bonded to the sensor. As far as I know,
there is no reason for the sensor to "age" rapidly, although temperature
sensitivity of the CCD might be an issue introducing non-linearities there
is no reason to think that it cannot be measured once and for all for each
sensor.
Edmund
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.