delorie.com/archives/browse.cgi | search |
Xref: | news2.mv.net comp.os.msdos.djgpp:2822 |
From: | cotool1 AT umbc DOT edu (o'toole casey) |
Newsgroups: | comp.os.msdos.djgpp |
Subject: | GRX20 speed/memory access |
Date: | 17 Apr 1996 16:32:43 -0400 |
Organization: | University of Maryland, Baltimore County |
Lines: | 21 |
Message-ID: | <4l3khb$dlu@rpc22.gl.umbc.edu> |
NNTP-Posting-Host: | rpc22.gl.umbc.edu |
NNTP-Posting-User: | cotool1 |
To: | djgpp AT delorie DOT com |
DJ-Gateway: | from newsgroup comp.os.msdos.djgpp |
Well, I was thumbing through the GRX 2.0 source and saw that it uses pokes to write pixels.. This obviously is horrendously slow given the 30 FPS I am getting just clearing a context and writing it to screen. So I figured out how to use the screen = unsigned char *(0xa0000); __enable_nearptr() screen[320*y + x + __djgpp_conventional_base] = color __disable_nearptr() Now, I would like to be able to do this but also use GRX 2.0 if need be. The problem is that when I try to write to the Ram Context it blows up on me. i am doing this: buffer = buf -> gc_baseaddr; then using buffer just like screen example above. Anyways... Anyone tried anything like this while still trying to retain the use of the GRX Library??
webmaster | delorie software privacy |
Copyright © 2019 by DJ Delorie | Updated Jul 2019 |