[CM] Snd 20.4

bil at ccrma.Stanford.EDU bil at ccrma.Stanford.EDU
Wed May 27 05:47:46 PDT 2020


Most things are only slightly slower, but there are a few
that are as much as 4 times slower.  There are timings
at the end of s7.c (callgrind numbers for my timing tests).
I haven't looked at the standard benchmarks except pi.scm
and chudnovsky.scm (which require big integers), and they are
either #1 or #2 in speed compared to everyone else.  I need
to tie the gmp code into the rest of s7's optimizer -- just a
small matter of software!  I've spent the last 2 months
on this, and figure there's another 2 months of work to
get the last 10 percent.  (But I also need to get gtk4 to
work, and so on).

I you have a chance to try it out, please let me know how
it goes, especially any things that seem slow.



More information about the Cmdist mailing list