home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.scheme
- Path: sparky!uunet!news.claremont.edu!ucivax!ucla-cs!ucla-mic!unixg.ubc.ca!prodigy.bc.ca!news
- From: Joerg Beekmann <jbeekman@prodigy.bc.ca>
- Subject: Re: Are interpreters now as fast as compiled code used to be?
- Message-ID: <1992Dec17.191058.28471@prodigy.bc.ca>
- Sender: news@prodigy.bc.ca
- Organization: Prodigy Technologies Corp.
- X-Useragent: Nuntius v1.0
- References: <4051@mitech.com> <4065@mitech.com> <4066@mitech.com>
- <4067@mitech.com>
- Date: Thu, 17 Dec 1992 19:10:58 GMT
- Lines: 14
-
- What I would like to know is how is the equation affected by
- machines with large caches and slow main memory systems.
-
- If an interpreter could fit into the cache could the greater
- density of byte code compensate (and to what extent) for
- the need to interprete it. In other words the cache would
- be the "micro store" and the chip would become a scheme
- machine.
-
- Finally will this factor become more improtant? CPU's
- seem to be increasing in speed faster than ram (where
- the empasis is on greater size)
-
- Any thoughts??
-