oldk1331 wrote:
> > - We probaly should be testing speed after doing:
> >
> > )lisp (proclaim '(optimize (speed 3) (safety 0)))
>
> I can't believe that this is not the default! Adding it
> to my ~/.fricas.init .
On testsuite safety 0 gives something like 10-20% speedup.
It is probably largest speedup from any simple single
change. But it is still not _very_ large. OTOH safety 0
makes debugging much harder. IME during developement
the extra speed is not worth frustration when
debugging at safety 0. For releases 20% speedup/slowdown
is unlikely to be a deal breaker. But misterious crash
without any indication what went wrong may discourage
potential users. And without resonable error
message it is extremaly hard to remotely diagnose
a problem.
I used FriCAS compiled at safety 0 to perform some
largish computations, where it gave substatial
speedup. We probably should compile some selected
domains at safety 0, this should give us most
gains with limited impact on debugging.
--
Waldek Hebisch
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to fricas-devel+unsubscr...@googlegroups.com.
To post to this group, send email to fricas-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/fricas-devel.
For more options, visit https://groups.google.com/d/optout.
