archiv~1: Re: SETI Linux signal analysis

Re: SETI Linux signal analysis

David Bridgham ( dab@froghouse.org )
Fri, 27 Nov 1998 11:22:11 -0700

> Delivery-Date: Thu, 26 Nov 1998 00:53:44 -0700
> From: David Woolley <david@djwhome.demon.co.uk>
>
> Scrolling is very expensive. I seem to remember that specgram
> writes left to right and wraps round, rather than scrolling. (Text
> mode scrolling can use tricks to do with the way the hardware works,
> which are not available to anything using windows (text or
> graphic)).

Fortunately, I think most graphics cards these days include hardware
for bitblt operations. And with a reasonable integration time, the
scrolling rate should be pretty slow. Even a 1 minute integration
with a 50% overlap means only one scroll operation each 30 seconds.

> Waterfall displays are very expensive, and for a SETI station to be
> really useful, it should be operating 24x7 and rarely being watched.
> An X waterfall is even worse.

True. However for setup, examining possible hits, and debugging it's
rather useful. Already, the very crude waterfall display I've written
is showing me small signals that look to be coming from my receiver
and that I'll have to deal with somehow in the hit detector. How
about having the waterfall display run as a separate, optional,
process?

Dave