Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!cmcl2!rutgers!im4u!ut-sally!utah-cs!utah-gr!uplherc!sp7040!jose!pedro!garon
From: garon@pedro.UUCP (Garon C. Yoakum)
Newsgroups: comp.sys.ibm.pc
Subject: Re: Converting to 386 machines
Message-ID: <126@pedro.UUCP>
Date: Wed, 11-Nov-87 01:52:24 EST
Article-I.D.: pedro.126
Posted: Wed Nov 11 01:52:24 1987
Date-Received: Sun, 15-Nov-87 20:51:30 EST
References: <2032@isis.UUCP> <124@pedro.UUCP>
Organization: An As Yet Unnamed Public System
Lines: 18
Summary: 386 4G 'small' model

In article <2032@isis.UUCP>, ross@isis.UUCP (Ross McConnell) writes:
>  I have a couple of commercial software packages (civil engineering
> oriented) that currently run on PC's and AT's, and are written in MSC.
> I am fighting the 640K barrier, and am considering porting the packages
> to a 386 to take advantage of the 4G 'small' model. It seems to me that there
> are several possibilities :
> 		1) Use Xenix 386 from SCO or Microport
>  		2) Use the Phar Lap and Metaware sort of compiler/Run386 environment
> 		3) Wait for OS/2  (will it allow 4G addressing ? )

Currently, I am running MicroPort V.3/386 with the green-hills 'c' compiler.
The 4G 'small' address space works wonderfull and required zero program
modifications.

OS/2 will not allow 4G addressing and only support's the 286 addressing.
Personally, I think it's a waste of time messing with a screwed up
processor and operating system.  Take my advice, go true 386 or bag it and
go for the 68000 series processors.