[Openicc] monitor profiling

[Graeme Gill]

Hal V. Engel wrote:
> One of the interesting things is that this software was originally written to 
> support a DYI color meter that you can down load the plans for from the web 
> site.  They claim that it costs 20 to 30 Euros for parts to build as a one 
> off and is a better instrument than the Spider2.  Kind of gives you an idea 
> how inexpensive these meters are to build and how big these vendors margins 
> are since building stuff like this one off is significantly more expensive 
> than mass producing the same type of device.

I've got support for the HCFR colorimeter in the new version of Argyll
I'm working on, but unfortunately my experience with this device hasn't
been very favourable. It gave sort of reasonable results on a CRT
display, but essentially unusable results on an LCD display (black
level grossly over estimated). The suspicion on my part is that this
is due to the lack of IR filtering, and the lack of directionality in
it's sensors. The lack of any sort of factory calibration (ie. calibration
against a known standard), also limits it's usefulness for display
calibration in my opinion. Switching to the MAZet MTCSiCS sensor
could address some of these issues.

It's a neat idea, and an achievement to get such a project up and
running, and seems to work a good deal better in it's intended
purpose as a projector sensor, but as a display calibration/profiling
sensor I don't think it currently makes the grade.

The cost difference between the HCFR and something like a Spyder2 or
Eye-One display doesn't seem so great to me when you know you're
getting an instrument that is close to being a CIE observer, is setup
correctly for LCD displays, and does come with some sort of calibration
against a master instrument. Yes, you're paying the manufacturer
for the work in engineering and manufacturing the device, as well
as making them a profit, but they are actually proving something for it
that's hard to substitute.

Graeme Gill.