Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site brl-vgr.ARPA
Path: utzoo!watmath!clyde!floyd!harpo!seismo!brl-vgr!gwyn
From: gwyn@brl-vgr.ARPA (Doug Gwyn )
Newsgroups: net.unix-wizards
Subject: Re: VAX Floating Point Formats
Message-ID: <3227@brl-vgr.ARPA>
Date: Wed, 4-Apr-84 06:27:02 EST
Article-I.D.: brl-vgr.3227
Posted: Wed Apr  4 06:27:02 1984
Date-Received: Thu, 5-Apr-84 00:35:51 EST
References: <187@verdix.UUCP>
Organization: Ballistics Research Lab
Lines: 11

G & H floating were added to the VAX-11/780 within the first year or
so of its existence.  They were intended to be IEEE standard compatible
(I don't know if they ended up that way).  On the 780 you have to buy
some other option like user-WCS as a prerequisite; this is a botch.

I think in general if you need more dynamic range than the PDP-11
compatible floating point format, you may get into trouble by simply
using G & H instead (assuming you can do this in a simple way).
Real production code that has to worry about exponents getting too
large generally maintains its own scaling and adjusts things while
running to keep within range.