Quote:
Originally posted by bteagarden
I wonder if there isn't something incompatible between DLP (or at least my particular X1) and the blue bar test. |
A friend and I were discussing this recently, and this is what he said:
"One disturbing thing that I learned recently, however, is that the old,
familiar method of calibrating your NTSC decoder by looking at color
bars patterns through a blue filter isn't necessarily valid on non-CRT
displays. The blue filter method assumes that the blue filter will
show exactly the level of the blue signal output from the NTSC decoder.
The modern standard NTSC color space (SMPTE-C) assumes red, green, and blue primaries that match standard CRT phosphors. The primaries on our projector are different. In particular, if you use color correction
settings to adjust the color of the primaries (to hopefully better
match the standard primaries), that actually can cause some mixing of
the projector's actual primaries when displaying standard primaries.
What this means is that green or red signals coming out of the NTSC
decoder might actually include a little bit of the projector's blue in
order to get them closer to the standard green or red, but this little
bit of blue would be seem by you through the blue filter, thus messing
up this technique of calibrating the NTSC decoder using the blue
filter. This issue is described by Guy Kuo here (although his
explanation isn't entirely consistent with mine):
http://www.avsforum.com/avs-vb/showt...96#post4161596
As he describes, if your NTSC decoder is outside your display (like in
your scaler) and is connected with an analog RGB connection, then you
can work around this by just disconnecting the red and the green
signals while doing to the NTSC calibration. Some displays support
doing this internally.
So I'm not entirely sure what this means I should do for NTSC
calibration. Should I put all of the color correction settings to zero
before calibration? Or some other value? Guy Kuo recommends putting
the primaries where they turn a corner on the CIE chart, but he doesn't
suggest where to put the secondaries. I'm feeling a bit in the dark
here.
Regarding those color correction settings, though, at least I think I
have some idea that goal is to have the respective colors read as close
as possible to certain xy coordinates on the CIE chart."