[fitsbits] Re: top or buttom

[Steve Allen]

On Wed 2000-04-19T06:48:22 +0000, Rob Seaman hath writ:
> some other old FITS hand might even recount some boustrophedonic exploits...

well, since Rob invited it.

The first generation of Keck optical instruments adopted the Leach's
SDSU CCD controller to handle readouts from two-amplifier chips.  With
no known convention for handling such images, the software was built
to stuff the pixel streams from both amplifiers into the same FITS
image array.  Thus we preserved the spatial attributes of the image
while incurring the expense of mixing two signals with different
noise, bias and gain.  The other expense is that the pre/overscan data
sections are discontiguously displaced to the edges in a manner that
is tricky to decode.

The real-time display for this pretty much does as Rob implies; it's
writing pixels from both the right and left (though not interleaving
lines as such).

Other than that, Lick's long-standing practice, and the jargon in
general use around the folks who work with CCD signals, has been to
presume that the CCD images should be treated just as standard video
signals.  I.e., in casual discussion, sketches, drawings, early
testing, and by default, the origin of the image is at the upper left.
Early generation software often enforces this without provision for
alternative preferences, though now we largely offer a choice.

--
Steve Allen          UCO/Lick Observatory       Santa Cruz, CA 95064
sla at ucolick.org      Voice: +1 831 459 3046     http://www.ucolick.org/~sla
PGP: 1024/E46978C5   F6 78 D1 10 62 94 8F 2E    49 89 0E FE 26 B4 14 93