> And incidently; aren't sends opcodes? I mean, the difference in bc/sec
> isn't that great, but yet the number of sends show a huge increase (which is
> a good thing of course). But how is a send done if not in response to an
> opcode?
> 
> Or is bc/sec "bytecodes per second, no counting message sends"?
The tinyBenchmarks bytecodes per second number is for about the most
trivial loop the author could think of, with virtually no fullscale
message sends involved. The benchFib method on the other hand is almost
entirely message sends and thus provides something of a balancing
measure.

It would be entirely possible to make an improvement to the overall
performance of the system (as measured by larger benchmark code) that
actually lead to a lower value for the simple bytecodes/sec test. I
suspect that would be harder w.r.t the send/return test.

tim

-- 
Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
Strange OpCodes: WROM: Write to Read Only Memory