[argyllcms] Re: Tests with DTP92 on video-projector
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Thu, 06 Apr 2006 00:33:34 +1000
Frédéric wrote:
Last night, I made a quick test to calibrate a video-projector with my
X-Rite DTP92. It seems to work, but I have some problems with light level,
and I will have to try differents distances from the screen.
Hmm. I was thinking about this sort of thing. You might get away with
pointing
it at the screen, but it may or may not work satisfactorily.
Has it is possible to make some single reading with spotread, it is easy to
adjust this distance. But how find the optimum ? I displayed a pure white
on the screen, which I think have the maximum brightness. What XYZ values
should I read on this white to obtain the best accuracy for measures during
calibration process ?
More light is better, assuming you don't overload the '92. Also, since the
RGB sensors in the instrument are distributed around the inside
of it, you probably want to keep it "flat", so as not to favour one
sensor over another.
A better optical arrangement might be to make a telephoto adapter.
A tube (painted black on the inside), with a lens at the right
point, and a thin diffuser at the focus. You want to project
a focussed image of the screen on the diffuser, and then put the
DTP92 on the diffuser. The diffuser would have to be neutral
in terms of it's effect on color. (this is just theory, I
haven't tried anything like this myself. If I could locate a source
of suitable lenses in Australia, I might give it a go.)
Graeme Gill.
Other related posts:
