Re: Efficient bytecode design and interpretation

  Doug Bagley <usenet.2001-05-30@bagley.org> writes:
[about http://www.bagley.org/~doug/shootout/]
>Not all Gforth solutions have been optimized on my site yet. In
>particular the solutions where other languages use optimized hash
>tables, Gforth uses wordlists, and would probably do better if
>rewritten with a custom hash table implementation.


Wordlists in Gforth and other modern Forth implementations are
implemented using hash tables. I recommend using wordlists in Gforth
were you would use associative arrays in awk. However, it may be
possible to improve the efficiency of the wordlist implementation.


OTOH, the slowdown of gforth over the fastest language implementation
(gcc) is a factor of 10 for both sieve (Gforth's rank: 8 of 27) and
hash (Gforth's rank: 24 of 24), so Gforth's rank may have more to do
with how the other language implementations fare than with Gforth's
performance itself.


- anton
--
M. Anton Ertl
anton@mips.complang.tuwien.ac.at
http://www.complang.tuwien.ac.at/anton/home.html

Return to the comp.compilers page.
Search the comp.compilers archives again.