On Tue, 19 Sep 2006, Peter W. Draper wrote:
> > Question is, how big should the buffer be? How long is a piece of
> > string?
>
> Well I make his message 464 characters, so that just sneaks over the
> current hard limit of 444 (500-56), so not much bigger. 1000 must look
> about right.
don't make no difference to me, but why not save the same debate in
3 years time and go for a factor of 10 increase at least? From a
position of total ignorance about this bit of AMS I'd guess that there's
unlikely to be resource issues here beyond the actual number of bytes
in the buffer itself (maybe times a small integer factor); obviously
a few kilobytes counts as pretty cheap these days.
Mark
--
Mark Taylor Astronomical Programmer Physics, Bristol University, UK
[log in to unmask] +44-117-928-8776 http://www.star.bris.ac.uk/~mbt/
|