On 22 Sep 2008, at 11:46, Manlio Perillo wrote:
Don Stewart ha scritto:
Thanks to those guys who've submitted parallel programs to the
language
benchmarks game, we're climbing up the rankings, now in 3rd, and
ahead
of C :)
This is cheating, IMHO.
Some test comparisons are unfair.
The first problem is with the thread-ring benchmark.
Haskell uses the "concurrent Haskell" extension, but all other
programs (with some exceptions) uses OS threads.
This is unfair, as an example a C program can make use of the GNU
threads library, for user space threads), but there is no such
program.
Who said that C *had* to use OS threads? The point of the shootout is
to show the relative strengths of the various languages, one of the
strengths of Haskell is some excellent lightweight thread support,
this is not present in C, so C does badly on the tests that check how
well you can deal with small threads.
With parallel programs it is the same: other languages does not have
a parallel version.
Yes, and the new benchmarks are *specifically* designed to test how
fast programs are on more recent multi-core hardware, so again, the
other languages are welcome to submit parallel versions... It just
turns out that Haskell is pretty damn good at doing parallelism.
Bob
_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
