Xref: utzoo comp.compilers:1456 comp.lang.fortran:4013 comp.os.msdos.programmer:1688
Path: utzoo!attcan!uunet!ism.isc.com!haddock!ima!esegue!compilers-sender
From: srw@csun.edu (Stephen Walton)
Newsgroups: comp.compilers,comp.lang.fortran,comp.os.msdos.programmer
Subject: Re: New MS Fortran
Keywords: Fortran, review
Message-ID: <9010262250.AA19046@csunb.csun.edu>
Date: 26 Oct 90 22:50:56 GMT
References: <9010230628.AA22160@admin.ogi.edu> <1990Oct24.162529.20452@zoo.toronto.edu> <1990Oct25.010604.4796@twinsun.com>
Sender: compilers-sender@esegue.segue.boston.ma.us
Reply-To: srw@csun.edu (Stephen Walton)
Followup-To: comp.os.msdos.programmer
Organization: Compilers Central
Lines: 20
Approved: compilers@esegue.segue.boston.ma.us
In-Reply-To: <1990Oct25.211846.29118@beach.csulb.edu>

In article <1990Oct25.211846.29118@beach.csulb.edu> Jeff Sicherman writes:
>Maybe this belongs in alt.rumors, but does anybody know if there are any
>pending/possible upgrades to MS Fortran any time soon.

Yes.  Buy Lahey :-) .

Seriously, I switched from MS Fortran 5.0 to Lahey Personal Fortran 3.00
(the $99 list compiler), and the persistent floating-point stack underflow
problems I'd been having on my AT clone disappeared.  Compilation speed was
increased about a factor of 3, and run speed was not reduced at all.

The project in question, incidentally, uses the MINPACK routines, so yes,
the program is heavily floating-point oriented.  Lahey is more expensive,
but offers an education discount of 25% off list with purchase of 5 or
more compilers.
-- 
Stephen R. Walton, Dept. of Physics and Astronomy, Cal State Northridge
-- 
Send compilers articles to compilers@esegue.segue.boston.ma.us
{ima | spdcc | world}!esegue.  Meta-mail to compilers-request@esegue.