Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!rutgers!seismo!lll-lcc!styx!twg-ap!amdahl!pyramid!decwrl!decvax!tektronix!uw-beaver!fluke!ssc-vax!bcsaic!michaelm
From: michaelm@bcsaic.UUCP
Newsgroups: comp.lang.lisp
Subject: Re: LISP for IBM compatible PC's
Message-ID: <282@bcsaic.UUCP>
Date: Wed, 4-Feb-87 11:30:28 EST
Article-I.D.: bcsaic.282
Posted: Wed Feb  4 11:30:28 1987
Date-Received: Sun, 8-Feb-87 08:26:53 EST
References: <1048@sfsup.UUCP> <80300002@uiucdcsp>
Reply-To: michaelm@bcsaic.UUCP (Michael Maxwell)
Organization: Boeing Computer Services AI Center, Seattle
Lines: 27

In article <80300002@uiucdcsp> forbus@uiucdcsp.cs.uiuc.edu writes:
:
:
>If you aren't going to use Common Lisp, then I think TI's PC-SCHEME is
>clearly the implementation of choice...  Their compiler works on the PC,
>and works well (including tail recursion optimization)...
>Performance, given the machine limitations, is quite good:  TI-Scheme on an
>XT is more fun to use than GCLisp-LM on an AT...

Could you expand on this?  I skimmed the article (I don't have access to it
now). My impression was that the benchmarks showed *compiled* GCL to be
lots faster than byte-compiled Scheme; I understand there is no true
compiled Scheme for Intel-based machines.  Or did you mean TI Scheme was lots
faster than *interpreted* GCL?  (There is no interpreted/compiled distinction
in TI Scheme, right?)

BTW, I understand TI Scheme can now address up to 2 megs of expanded
(extended?) memory on an AT class machine.  That number sounds a bit odd to
me--if it can get past the 640k limit, why not up to 16 megs?  Or am I too
ignorant of the 80286 architecture?  How much of that 2 megs is left for user
programs under various conditions (w/ and w/o editor, in production versions
of code vs. development environment, etc.)?
-- 
Mike Maxwell
Boeing Advanced Technology Center
	arpa: michaelm@boeing.com
	uucp: uw-beaver!uw-june!bcsaic!michaelm