Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!cs.utexas.edu!uunet!proto!joe
From: joe@proto.com (Joe Huffman)
Newsgroups: comp.graphics
Subject: Re: Rayshade for DOS is here! (Achievement value only)
Message-ID: <1991Apr17.071346.1570@proto.com>
Date: 17 Apr 91 07:13:46 GMT
References: <1991Apr16.121537.7023@topaz.ucq.edu.au>
Organization: Prototronics @ Sandpoint, Idaho
Lines: 33

enzerinkp@topaz.ucq.edu.au writes:

>I'm at a bit of a loss as to where to go from here. Could anyone tell me of
>their experiences with memory problems on a UNIX platform so that I at least
>have an idea of where I'm going.

I did the port of the Zortech MSDOS compiler and graphics library to SCO
UNIX.  It allows you do things like:

  char *ptr = malloc(5000000); /* Yes -- 5 Megabytes */

and keep on running.  You may see the disk access light come on a lot as 
you access various portions of the memory but your malloc()'s don't start
failing nearly as quickly as they do under DOS.  And also they aren't 
limited to 64 Kbyte chunks like under DOS.

Similar things can be said for use of the Pharlap 386 DOS extender running
under DOS.  It's a refreshing change from the constraints of MSDOS memory.

If you were to use the Zortech graphics library (I wrote it -- yes I am 
pluggin my own product and am not in an unbiased position) you could compile
and run the code with very few (if any) changes under MSDOS, SCO UNIX, OS/2,
and the PharLap 386 DOS extender.

I don't know Rayshade but there are other problems that you may having 
that are unrelated to available memory -- like ints are 16 bits under DOS
but it may depend on haveing 32 bit ints.  Trace through program with a 
debugger or put enough printf()'s in the program to isolate where the error 
message is coming from.

Good luck...
-- 
joe@proto.com