Xref: utzoo comp.lang.c++:11139 gnu.g++.help:334
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod!ncar!csn!news!grunwald
From: grunwald@foobar.colorado.edu (Dirk Grunwald)
Newsgroups: comp.lang.c++,gnu.g++.help
Subject: Re: GNU g++ not ready for anything at all. (WRONG!)
Message-ID: <1991Jan15.051733.17020@csn.org>
Date: 15 Jan 91 05:17:33 GMT
References: <1991Jan10.202317.161@ee.ualberta.ca> <16008@ogicse.ogi.edu>
	<1991Jan15.014335.13365@cs.ucla.edu>
Sender: news@csn.org
Reply-To: grunwald@foobar.colorado.edu
Distribution: na
Organization: University of Colorado at Boulder
Lines: 23
In-Reply-To: rjc@cs.ucla.edu's message of 15 Jan 91 01:43:35 GMT
Nntp-Posting-Host: foobar.colorado.edu

> While I have encountered numerous bugs in various versions of
> g++, my coding time was rarely affected very much.  In general,
> Michael's support has been incredible.  In critical cases, he
> has sent me a patch within a day (more than once in about 15
--

While this was true in the past, I must admit that I'm not certain
it's true now. Granted, Cygnus & Michael have been putting
considerable time into GCC-2, but g++ has really suffered for that
effort. It's been several months since the last release, and numerous
known bugs persist. Recent bug reports don't get much action other
than ``fixed in v2''.

Others (me included) have done some debugging and made fixes on a
variety of machines, but without knowing what bugs have and haven't
been fixed by ``G++ central'', you don't know what work you're
duplicating (& time you're wasting). There's not much incentive to
make extensive fixes.

If I could afford it, I too would use SaberC++ (which *is* available,
according to an ad I got). The cost is less than a contract for a
working g++ from Cygnus and significantly less than the cost of
hacking it myself (because my research time would go down the toilet).