RE: Need some rough numbers.

["Albertson, Chris" <CAlbertson@primeadvantage.com>]

I'm interested in the other matching problem.  After
you already have an astrometric reduction for a frame.

Say you have six Mk IV frames of the same part of the
sky and now you want to "stack" six star lists to
make about 5000, six point light curves.  OK so far
as it all fits in RAM.

Now suppose you have two or three years worth of data from
four Mk IV sites and you want to make 30 million
light curves each with about 50 points but some with
1000 points.  You can't compute something like this
on demand as we are talking about a few hundred CDROMs
worth of data.  You have to start small and over the months
build on it with some kind of on-going process.  It's
this on-going process I'm interested in.  We never got
it right with the Mk III.  You'd also like to compute
cross references with many of the common star catalogs that
are in use and keep all this information all up to date.
This type of computation will run for months or years so
you want to worry about taking check points for restarting
a crashed run, backing out mistakes and all kinds of dumb
details.



> -----Original Message-----
> From: Glenn Gombert [mailto:glenngombert@hotmail.com]> 

>   Mike G's Star routine matches just fine with about 25-30 
> starts per Mark 
> IV image...I imagine that any other reduction program would 
> work equally 
> well with about that same number of stars-per-image....
>