Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!cbatt!cbosgd!ucbvax!hplabs!felix!preston
From: preston@felix.UUCP (Preston L. Bannister)
Newsgroups: net.micro.atari16
Subject: Re: 68030
Message-ID: <1611@felix.UUCP>
Date: Sun, 28-Sep-86 17:42:42 EDT
Article-I.D.: felix.1611
Posted: Sun Sep 28 17:42:42 1986
Date-Received: Tue, 30-Sep-86 19:07:17 EDT
References: <860927204514.0000059A.ACYB.MA@UMass> <8609281735.AA15525@ames-nas.ARPA>
Reply-To: preston@felix.UUCP (Preston L. Bannister)
Organization: FileNet Corp., Costa Mesa, CA
Lines: 25

Probably the most significant feature of the 68030 is that the MMU is
on-chip.  This is a big win in terms of speed/cost.  It also has a data
cache in addition to the code cache.  I think there are probably other
little tweaks for more speed.

Maybe Atari should just wait until the 68030 comes out to replace the
ST. :-) On chip MMU means lower cost/simpler board, and should have
fewer wait-states with regular DRAMs.

With microprocessors that are this fast, it starts to make sense to use
Lisp for program development and to forget about "lower-level"
languages like Pascal, C, Ada... :-)

A next generation ST could have a small Lisp interpreter (like XLisp)
in ROM.  Lisp compilers are getting quite good.  The ORBIT compiler
(for Scheme) generates code that compares favorably with compilers for
more conventional languages (see "ORBIT: An Optimizing Compiler for
Scheme" in the "Proceedings of the SIGPLAN '86 Symposium on Compiler
Construction", SIGPLAN Notices v21, #7).  There are comparible compilers
for PSL and Franz Lisp (and possibly others).

========================================
Preston L. Bannister
USENET:  ucbvax!trwrb!felix!preston
BIX:     plb