[argyllcms] Re: help with camera profile
- From: Iliah Borg <ib@xxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Sun, 5 Aug 2012 09:41:16 -0400
> I think making reliable custom ICC camera profiles is beyond the capabilities
> of most photographers. For them, it may be easier to get satisfactory colour
> with DNG profiles.
I do not see why making a DCP profile is easier. I do not see DCP profiles for
mainstream cameras behaving better. Colour and tone can't be right if the flare
and other physical effects are not compensated. Lenses need separate profiles
too. Polarizing filters also need profiles, etc. A simple ad hoc profile taken
"in the scene" is always better compared to generic profile of any level of
sophistication (hi, Edmund ;) just because it takes into account most of the
scene, lens, and camera factors.
To give you an example, a friend of mine was shooting a wedding in the church,
and he needed at least two profiles, one taken in the centre of the church, and
the other - close to altar. Those were absolutely different.
Camera sensors and camera chambers have a lot of sample variation, we saw it
many times. Canned profiles are not an answer.
I do not think applying scanner colour science to profiling digital cameras is
the right approach - this starts right from the profiling targets. Optimal
targets for matrix profiles are different from ColorCheckers. In the ideal
world each sensor needs its own target which is built-in right into the camera.
But for now we are playing table tennis with hammers and insist that it is the
only way and the right tool.
> Adobe have professionals who understand digital imaging and colour science
> better than myself and most other Argyll CMS users.
Understanding is not enough, one needs to have skills in visual perception.
--
Iliah Borg
ib@xxxxxxxxxxx
Other related posts:
