Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!amdcad!ames!husc6!rutgers!pbox!okstate!geh
From: geh@a.cs.okstate.edu (G. E. Hedrick)
Newsgroups: comp.sys.ibm.pc
Subject: Re:  MIX C
Message-ID: <2759@okstate.UUCP>
Date: Mon, 26-Oct-87 11:29:17 EST
Article-I.D.: okstate.2759
Posted: Mon Oct 26 11:29:17 1987
Date-Received: Thu, 29-Oct-87 04:47:51 EST
Organization: Oklahoma State University
Lines: 26
Summary: Reply to request for information re. MIX C


Date:     Mon, 26 Oct 87 10:21:42 CST
From:     M J Folk <mjf@a.cs.okstate.edu>
Subject:  Re:  MIX C

I like it very much, but have had a couple of problems with it that would
cause me not to use it on a major project.  As a result of these problems,
I don't really believe I can fully trust it the way I can a tried and
tested compiler like Lattice C, which I have used a great deal in the past.

The problems I can recall encountering:  (1) short integers are one byte rather
than two bytes.  This is not illegal, but it is different I believe than virtually all other C compilers.  (2) When I tried to pass a 
short to a function, a zero was sent rather than the intended value.  When I
changed the argument to int, it worked fine.

What I like most about MIX C is the trace facility.  It is really nice for 
developmental work, and the price is right, for sure.

Another plus is the tutorial manual that comes with it.  A couple of my students have said that it was a perfect supplement to K & R.

Mike
-- 
G. E. Hedrick
Oklahoma State University
ARPA/CSNET:  hedrick@a.cs.okstate.edu
UUCP:    . . .!{cbosgd|rutgers|ihnp4|seismo|uiucdcs}!okstate!hedrick