Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxl!houxm!hogpc!houti!ariel!vax135!cornell!uw-beaver!tektronix!hplabs!sri-unix!jrodrig@MITRE-GATEWAY.ARPA
From: jrodrig@MITRE-GATEWAY.ARPA@sri-unix.UUCP
Newsgroups: net.micro
Subject: Re: LISP and C compiler info desired
Message-ID: <12734@sri-arpa.UUCP>
Date: Thu, 23-Aug-84 09:24:35 EDT
Article-I.D.: sri-arpa.12734
Posted: Thu Aug 23 09:24:35 1984
Date-Received: Thu, 30-Aug-84 01:11:51 EDT
Lines: 35

From:  jose rodriguez <jrodrig@MITRE-GATEWAY.ARPA>


I got Computer Innovations C86 and found it quite good (maybe a little slow)
In the other hand it comes with a very rich library with source code.
The nice thing about this is that you get several versions of the libraries:

small memory model, all dos'es, soft math
small memory model, all dos'es, 8087 math
small memory model, dos 2.0, soft math
small memory model, dos 2.0, 8087 math
large model, all dos'es, soft math
large model, all dos'es, 8087 math

(by "all dos'es" I mean versions before 2.0)

Another nice feature is that it provides a runtime environment close
to Unix -- you can have argc and argv in your main.
Also you can tinker with this environment, they provide the source code
that sets it up.

The people at Computer Innovations are quite reacheable and helpfull.
They got a BBS where gripes can be exchanged and public domain software
up/downloaded.

One thing which I don't like: the "optimizing compiler" description
of it. I have looked at the code generated and it is pretty gross. A half decent
peep-hole optimizer could shrink your code size from 10% to (I would say)
40%, 50%. (Particularly if you write lots of small routines).
 Does anyone know what optimization techniques they use?

All in all, I am quite satisfied with it.

Jose
jrodrig@mitre-gw