English version
Accueil     À propos     Téléchargement     Ressources     Contactez-nous    

Ce site est rarement mis à jour. Pour les informations les plus récentes, rendez-vous sur le nouveau site OCaml à l'adresse ocaml.org.

Browse thread
[Benchmark] NBody
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: 2005-02-13 (18:15)
From: Christophe TROESTLER <Christophe.Troestler@u...>
Subject: Re: [Caml-list] [Benchmark] NBody

Sorry for the late reply, I am catching up with email.

On Tue, 8 Feb 2005, Xavier Leroy <Xavier.Leroy@inria.fr> wrote:
> Your OCaml code is about as good as you can write.  All the unboxing
> optimizations are triggered.

Ah, thanks for telling, I was wondering about that.

> You don't say which Java implementation you used (there are several).

Sorry; Sun JDK 1.5.0.

> (Besides, being 1.3 times slower than gcc on numerical code is
> within the design envelope for OCaml.  My performance goals have
> always been "never more than twice as slow as C".)

Yes.  I am not complaining, just trying to understand (and, to say the
whole truth, I was also a bit "sad" that Java was beating my favorite
language :).

> On a "normal" (register-based) float architecture like PowerPC or
> x86_64, the OCaml-generated code is essentially identical to the
> gcc-generated one.

Oh, good, so going to 64 bits will bring us that too!

> The C translation is attached for your amusement.

Thanks for your very detailed and informative answer!

Best regards,