[nem-en] shootout: fasta

[Micky Latowicki]

Yes, I suppose it doesn't measure much. This micro-benchmark approach
is pretty noisy. Still, I think it's good for nemerle to be visible in
this contest. Don't you?

At least the "concurrency" benchmarks show how easy (or hard) it is to
write coroutines in the language, which has some meaning, I think.
Other benchmarks measure the speed of numeric computations, array
access, etc.

One could also take a different approach and ignore speed, regarding
these as /style/ benchmarks - how clearly and succinctly does a
language express the algorithm?

Viewed this way, it might be a good idea to sacrifice some speed in
order to have clearer code. I take this approach, up to a point.

On 5/13/07, vc <vc at rsdn.ru> wrote:
> Hi,
>
> What measure this test? Speed of console output? JIT compilation time?
>
> The "shootout" is profanation. :(
>
> Vlad.
>
>
> > -----Original Message-----
> > From: devel-en-bounces at nemerle.org [mailto:devel-en-bounces at nemerle.org]
> > On Behalf Of Micky Latowicki
> > Sent: Sunday, May 13, 2007 2:10 AM
> > To: devel-en at nemerle.org
> > Subject: [nem-en] shootout: fasta
> >
> > A straightforward port, this time from C#. Not very interesting or
> > beautiful, I'm afraid. Comments and suggestions are welcome.
> >
> > Kamil: should I send the next benchmarks directly to you instead of
> > the mailing list?
> >
> > Times (N=1000000)
> > python fasta.py 1000000  : real 44.47, 44.11+0.12 (99%%)
> > mono ./csharp-build/fasta.exe 1000000  : real 1.26, 1.16+0.09 (100%%)
> > mono ./nem-build/fasta.exe 1000000  : real 1.48, 1.32+0.14 (99%%)
> >
> > Times (N=25000000)
> > mono ./csharp-build/fasta.exe 25000000  : real 31.38, 25.95+2.95 (92%%)
> > mono ./nem-build/fasta.exe 25000000  : real 35.85, 30.61+3.03 (93%%)
> >
> > Cheers.
>
>
> _______________________________________________
> https://nemerle.org/mailman/listinfo/devel-en
>