a9104910@unet.univie.ac.at (Martin) wrote:
After compiling, the maxima code was faster than MMA by a factor of two,
but the lisp code was faster by a factor of eight (!). (Well, not very
thoroughly tested, though)
Perhaps we should work on the translator and the maxima implementation
rather than trying to write at such a low level. Did you look at the
generated code to try to learn why there was such a difference?
Dan Stanger