(another way to skin a cat) - I dont mind sounding like a n00b

Gday,

(Fur gawdsake correct my ignorance where appropriate, how else does
one learn)

K,so in a normal colour cam, three pixels, RGB ( or BRG depending on
your coding pref) are measured and combined to create a colour tone.

When swithing to black and white in the driver, what exactly
happens? Does the logic chip take the real value (ets say 300 red,
220 blue, 250 Green) for each pixel ( which will be different due to
the colour filters,then use these numbers to represent the BW image?

My idea, though far fetched, and probably not possible with ANY hack
short of changing the ccd to an actual BW is to balance the photon
count (electron count)that each pixel would get via an alogarthm..
ie. Red would get more pixels, due to the ccd's being sensitive to
this frequency.. can we not either up the other colours values to
match this higher reading or lower (less adventagious)the red outut
so that we end up with a colour cam that behaves more like a BW?

Far fetched, thinking out loud,,

Flame me, hate me, just dont buise me.

Rat. (Garry D)
Land of Oz

- In no.spam@mail.pls, Franck SICARD <yahoo_.at._3...> wrote:
>
> Hi Greg,
>
> On Fri, Nov 14, 2003 at 03:01:18PM -0000, beekeg wrote:
> > SETPWC doen't control the B/W control AFAIK. One thing I noticed
> > when working with Qastrocam was that when the camera was set to
b/w
> > the through-put was higher than when in RGB mode when using my
old
> > Pentium I. I never noticed this under Windows so the
implementation
> > of b/w may be different between the OS's. It might be
interesting to
> > see if this is caused by the resolution of the o/p. Could be just
how
> > the drivers are written.
>
> The throughput is faster in qastrocam when you check B&W, because in
> this case a ignore all the colors value that are recieve from the
cam (i
> just look at the Y part of the YUV frames recieve. If fact the
driver
> have not a b&w mode it always work in color.
>
> For your old PI, i am working on using the SDL library for the
display
> of the frames, it will be a very big win. (i have already a working
> prototype).
>
> Franck
>
> --
> Franck SICARD <Franck.Sicard_.at._3...>
http://3demi.net/
> PGP key=DFDFA865 fingerprint=8E2E 167E 9503 461B AD76 D962 2EEE
9869 DFDF A865

---