Yeah, I was wrong about Racket: http://shootout.alioth.debian.org/u64/benchmark.php?test=all...
You're right; it looks like Racket is not in the C/LuaJIT ballpark, either. It's too bad the LuaJIT results are no longer on the web site.
Re:PyPy 1.5 Released: Catching Up
by swannodette on Sunday May 01, 2011 @03:46PM
(#2503850)
Last Alioth results I saw LuaJIT was slower than Java, and certainly not in the C ballpark at all.
http://lua-users.org/lists/lua-l/2009-10/msg01098.html shows LuaJIT beating GCC 4.3.2 on some parts of SciMark, and no more than 3× worse on any part. I'm importing the shootout CVS repository into Git, so hopefully we can make some more definitive comparisons soon.
That post is ancient ... here are newer SciMark results for LuaJIT:
Thank you very much! Looks like LuaJIT's beating GCC on almost everything now, and the JVM on most things. But that's with LuaJIT-specialized code that won't run on Lua, which is a little less exciting.
Re:PyPy 1.5 Released: Catching Up
by swannodette on Sunday May 01, 2011 @07:34PM
(#2504395)
Those overall results show that LuaJIT doesn't really compete much the JVM running under server mode.
Clojure 1.2.0 is already faster than Racket. Version 1.3.0 will probably bring it close to Go territory on those benchmarks.
For my own work I've found that getting Clojure in the Java ballpark is certainly possible.