Oops forgot to hit reply-to-all.. resending..<br><br><div class="gmail_quote">N-body is looking good. I am running and amd64 3000+ on ghc 6.8.1. The debian shootout is showing a huge gap between ghc 6.6 and g++ but I am not seeing that gap. One concern though is that the code doesn't look very "haskellish". So much pointer manip.
<br>
<br>
For the nbody c++ code I am getting:<br>-0.169075164<br>-0.169031665<br><br>real 0m11.168s<br>
user 0m10.891s<br>
sys 0m0.043s<br>
<br>
and for the nbody haskell code I am getting:<br>-0.169075164<br>-0.169031665<br><br>
real 0m11.595s<br>
user 0m11.422s<br>
sys 0m0.044s<div><div></div><div class="Wj3C7c"><br><br><div class="gmail_quote">On Nov 26, 2007 8:21 PM, Don Stewart <<a href="mailto:dons@galois.com" target="_blank">dons@galois.com</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
s.clover:<br><div>> In some spare time over the holidays I cooked up three shootout<br>> entries, for Fasta, the Meteor Contest, and Reverse Complement. I<br><br></div>Yay!<br><div><br>
> First up is the meteor-contest entry.<br>><br>> <a href="http://shootout.alioth.debian.org/gp4/benchmark.php" target="_blank">http://shootout.alioth.debian.org/gp4/benchmark.php</a>?<br>> test=meteor&lang=ghc&id=5
<br>><br>> This is the clear win of the bunch, with significantly improved time<br>> thanks to its translation of the better algorithm from Clean.<br><br></div>Well done! Though looks like we'll have to follow the C++ implementation
<br>to be really competitive.<br><div><br>> Next is reverse-complement.<br>><br>> <a href="http://shootout.alioth.debian.org/gp4/benchmark.php" target="_blank">http://shootout.alioth.debian.org/gp4/benchmark.php
</a>?<br>> test=revcomp&lang=ghc&id=3<br><br></div>Very good. I'm glad someone looked at that, since the old code was<br>moderately naive (first bytestring effort).<br><div><br>> Finally, there's fasta.
<br>><br>> <a href="http://shootout.alioth.debian.org/gp4/benchmark.php" target="_blank">http://shootout.alioth.debian.org/gp4/benchmark.php</a>?<br>> test=fasta&lang=ghc&id=2<br><br></div>Yeah, we should do something better here. Hmm.
<br><div><br>> p.s. It looks like they've depreciated chameneos in favor of a new<br>> version, chameneos-redux. As this was one of the places Haskell<br>> really rocked the competition, it would probably be worth updating
<br><br></div>Definitely. I note also we're beating Erlang on the new thread-ring<br>benchmark too,<br><br> <a href="http://shootout.alioth.debian.org/gp4/benchmark.php?test=threadring&lang=all" target="_blank">
http://shootout.alioth.debian.org/gp4/benchmark.php?test=threadring&lang=all</a><br><div><br>> the Haskell entry for the new benchmark. Also, the n-bodies benchmark<br>> seems like another that could be much improved.
<br><br></div>Yeah, that's a hard one.<br><font color="#888888"><br>-- Don<br></font><div><div></div><div>_______________________________________________<br>Haskell-Cafe mailing list<br><a href="mailto:Haskell-Cafe@haskell.org" target="_blank">
Haskell-Cafe@haskell.org</a><br><a href="http://www.haskell.org/mailman/listinfo/haskell-cafe" target="_blank">http://www.haskell.org/mailman/listinfo/haskell-cafe</a><br></div></div></blockquote></div><br>
</div></div></div><br>