Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles $Revision: 1.7.0.10 $; site datacube
Path: utzoo!linus!decvax!ima!inmet!datacube!berger
From: berger@datacube
Newsgroups: net.micro.pc
Subject: Limitations to Address space
Message-ID: <44600001@datacube>
Date: Wed, 2-Apr-86 14:44:00 EST
Article-I.D.: datacube.44600001
Posted: Wed Apr  2 14:44:00 1986
Date-Received: Fri, 11-Apr-86 20:50:48 EST
Lines: 25
Nf-ID: #N:datacube:44600001:000:937
Nf-From: datacube!berger    Apr  2 14:44:00 1986


So when will people be able to escape the 1 Meg address limitation of
MS-Dos and Microsoft C on the IBM PC/AT?  Yes I know you can use bank switching
techniques.  But that continues the brain damaged traditions of the IBM PC in
general.  

I want to port code from other systems that need a large linear address
space to talk to frame buffers and other large arrays of special memory.

MS-Dos does not let you easily generate 24 bit addresses with the already
architecturally crippled 80286.  Microsoft C's concept of pointers does not
allow a pointer to memory greater than 1 Meg.

So now I have to go thru a large detour just to get things to work...

When is IBM and Microsoft going to get it toghether and at least simulate
a real computer with a large address space?

			In Frustration,
				Bob Berger 

Datacube Inc. 4 Dearborn Rd. Peabody, Ma 01960 	617-535-6644
	
{ihnp4,pyrcorp}!datacube!berger
mit-eddie!mirror!datacube!berger