Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site cadre.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!mcnc!idis!cadre!km
From: km@cadre.UUCP
Newsgroups: net.micro.pc
Subject: Re: More C Compiler Opinions
Message-ID: <115@cadre.UUCP>
Date: Mon, 24-Dec-84 11:23:43 EST
Article-I.D.: cadre.115
Posted: Mon Dec 24 11:23:43 1984
Date-Received: Thu, 27-Dec-84 03:01:15 EST
References: <4827@fortune.UUCP>
Organization: Decision Systems Lab., Univ. of Pgh.
Lines: 24

There will probably never be a clear consensus on what the "best"
compiler is for the simple reason that programming tasks vary and
what is important to individuals varies.

I have had quite a bit of experience with both the lattice and
computer innovations products. While the lattice compiler compiles
more quickly and generates faster running code, my overwhelming
preference is for C86: with a hard disk, the difference in compile
time is not significant. What is more important for me is
portability of programs from a unix environment with a minimum of
hassle and access to the library sources (yes, there ARE bugs).
Also, programs compiled with C86 tend to always run the way you
would predict, and there are not a lot of funny i/o problems,
as with early versions of the lattice compiler. It has also been
my experience that interfacing with the operating system is
easier under C86, and there is virtually no need for coding
anything at the assembly language level. Another string point is
the C86 documentation, which is easier to use for those used to
the unix manuals.

In all fairness, my experience has been with all versions of C86,
from the earliest (running under mp/m 86) to the current, while
I have only used an earlier version of the lattice compiler.