[argyllcms] Re: Color errors on < 100% stimulus
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Mon, 13 Feb 2012 09:11:01 +1100
Kristian Jörg wrote:
> When I calibrate my Panasonic VT30 plasma (not connected to a PC) with an
> external program i.e
> ChromaPure I get dE errors in the range of 1.0 or less. The color calibration
> is done at 100%
> stimulus. But when I check the color balance at 75% stimulus I get errors of
> about 10 dE! Ouch!
> There is no way of correcting that in the TV set's controls even though it
> has a very thorough CMS
> module.
It depends so much on what sort of controls the TV has. TV and computer systems
are
really at the opposite end of the spectrum when it comes to how color is
expected
to be handled. The lowest common denominator for computer systems is pretty
simple hardware color control (the video card per channel lookup tables),
and sophisticated software driven color profiling and CMM (ICC profiles).
TV sets seem to assume that color management can only be done in hardware,
and so you are at the mercy of whatever random hardware and controls
they provide (but may not document or provide tools for). Mostly it seems
pretty dismal (an extension of the user "brightness" and "contrast" knobs.
"RGBCMY"
primary controls that don't work so well, etc.) Maybe some per channel curves
that
aren't really publicly documented, or that you'd have to write software to
access.
They do meet in the middle somewhere at the high (but rather proprietary) end -
some displays have 3D cLUT in hardware capability (things like digital cinema
displays have this capability, as well as some high end monitors), so
ICC profiling type approaches can be used to create a device link cLUT
to provide an emulation transform. Computer systems are capable
of implementing 3D cLUT devce links at speed using the GPU.
So I'm not at all surprised to hear such problems with TV calibration :-)
The TVs may simply not have the capability of doing anything other than
very simply calibration.
> The question is when profiling a PC display with Argyll (or similar profiling
> software), does it
> correct color at different levels of stimulus or only at 100%? I.e if I would
> us my TV set as a PC
> monitor would the profile correct the color balance from 0-100% stimulus?
Argyll can be used to setup an accurate emulation (within the capabilities of
the device)
of any colorspace. For displays, a combination of hardware calibration and ICC
profling & linking
is the means of doing this, but the normal means of using the full color
transforms
involves running the color through a 3D cLUT. Typically that's done in software,
although a GPU can do this in real time. While the mechanisms for applying
ICC profiles in software are fairly clear, the current means of doing similar
things with video are far from clear.
Graeme Gill.
Other related posts:
