Datacolor User to User Support Group.

> No, that defies the laws of color theory; if you fill a room with
yellow 
> light, the eye adapts, and that yellow light looks white. There must
be some other 
> source before your eye can distinguish.

That can't be correct. It would mean that if I projected a yellow
image (RGB=255,255,0) my eye would adapt to it and it would look white
to me. That's obviously not the case - at least not within a few hours
(maybe if I lived in a yellow world for a few years it would ;-)

Furthermore, I know what the walls and other object in the room look
like and which color they are, so I CAN tell if they are illuminated
by a greenish or blueish light, etc.

Also, if you compare e.g. grey parts of an image with colored areas,
you can tell if the grey tends towards red, green blue, etc.
especially if you can compare different settings directly, which is
the case with the "Before" and "After" images.

> How would you explain the many satisfied projection S2P calibration
users 
> then?
Probably the colors were initially "very bad" and after calibration
just "bad". That's definitely an improvement and since everything is
relative according to your explanation, even a "bad" result can look
quite good. Maybe my PLV-70 is by default pretty good (just a tad on
the green side), so messing with the settings can just make it worse?

> projection. Is the room you are working in pitch black when the
projector is 
> off?
Of course! I have absolute control over the ambient light and I can't
see my hand in front of my face if the projector is off.

> shadow, and the Spyder is reading the whole screen, not some tiny
area in the 
> shadow.
My screen size is about 7 square meters. When the Spyder is placed 30
cm away from the screen, how can it measure the whole 7 m2? It can
only see a tiny fraction of the screen, with the shadow in the middle
covering a significant area! In just this area, there's no direct
light from the projector, just reflected light from the rest of the
room. For me it's clear that this absolutely will influence the
measured values! Hence the reports of people who are setting the
Spyder up in a way that it faces the projector, not the screen...

> ...projection rules, and not allow the eye to white point compensate
to the projected light.

But that's exactly the point! I don't need any calibration if the eye
would compensate to anything anyway. But it doesn't! That's why I want
the white point to be exactly where it should be, not where my eye
maybe moves it over time. The whole point of calibration is to make
the results physically correct and independant of subjective impressions.

Regards,

Martin