[Maxima] ./configure options and speed of execution.
macrakis at alum.mit.edu
Wed Jun 13 09:39:09 CDT 2012
I second Robert's comment. Speed is only one consideration among many.
Correctness, error recovery, quality of debugging tools, availability of
useful functionality (notably FFI), etc. etc. are all more important.
For that matter, speed can be measured in many different ways -- speed of
large symbolic calculations, speed of (compiled) numerical calculations,
interactive responsiveness, speed of graphing functions interactively, etc.
For the vast majority of our users (even power users), I am pretty sure
that *any* of the supported versions is fast enough. Showing that version
A is 3x faster on a benchmark than version B doesn't mean that the user
experience is any better on version A.
On Wed, Jun 13, 2012 at 10:22 AM, Robert Dodier <robert.dodier at gmail.com>wrote:
> On 2012-06-13, Rupert Swarbrick <rswarbrick at gmail.com> wrote:
> > Maxima devs: Is there a good reason to use Clisp as the default on Linux
> > boxes? Maybe we could switch to GCL or SBCL (I'd prefer the latter)
> Well, the default choice should be one which is widely available,
> stable, lacking idiosyncrasies, etc. If you would like to survey the
> Lisp world and identify an alternative to Clisp, by all means, please go
> ahead. I am willing to second the motion if you are willing to do the
> Robert Dodier
> Maxima mailing list
> Maxima at math.utexas.edu
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Maxima