home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!sun-barr!olivea!mintaka.lcs.mit.edu!zurich.ai.mit.edu!jaffer
- From: jaffer@zurich.ai.mit.edu (Aubrey Jaffer)
- Newsgroups: comp.lang.scheme
- Subject: Re: Are interpreters now as fast as compiled code used to be?
- Message-ID: <JAFFER.92Dec23115358@camelot.ai.mit.edu>
- Date: 23 Dec 92 16:53:58 GMT
- References: <4051@mitech.com> <4065@mitech.com> <4066@mitech.com> <4067@mitech.com>
- <1992Dec17.191058.28471@prodigy.bc.ca>
- <JAFFER.92Dec23012204@camelot.ai.mit.edu>
- Sender: news@mintaka.lcs.mit.edu
- Organization: M.I.T. Artificial Intelligence Lab.
- Lines: 19
- In-Reply-To: jaffer@zurich.ai.mit.edu's message of Wed, 23 Dec 1992 06:22:04 GMT
-
- I retimed pi in both SCM and C.
-
- kleph.ai.mit.edu (HP 9000/350):
- (pi 1000 4) ==> 624.5
- time pi 1000 4 ==> 11.0
- ratio ==> 56.7
-
- Pixel 80 (10 Mhz 68000):
- (pi 100 5) ==> 28.25
- time pi 100 5 ==> 1.75
- ratio ==> 16.14
-
- camelot.ai.mit.edu (HP 9000/720):
- (pi 1000 4) ==> 55.85
- time pi 1000 4 ==> 3.4
- ratio ==> 16.4
-
- The moral seems to be that interpreter penalty is not too large with
- an instruction cache if that cache is large enough.
-