From: Endlisnis <s257m AT unb DOT ca>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: MAJOR slowdowns in translating TP7 gfx code to DJGPP2
Date: Tue, 08 Sep 1998 18:01:28 -0300
Organization: NBTel Internet
Lines: 20
Message-ID: <35F59B28.35138CF4@unb.ca>
References: <35F45C78 DOT 2257 AT post DOT tau DOT ac DOT il> <35F45F0C DOT 2243BDB1 AT unb DOT ca> <35F53FD4 DOT 4D03 AT post DOT tau DOT ac DOT il>
NNTP-Posting-Host: fctnts07c34.nbnet.nb.ca
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

Tal Lavi wrote:
> > > A little highly optimized graphics aplication that run prety good on
> > > TP7 using mode 0x13, is experiencing a MAJOR slowdown using VBE2.0 in
> > > linear frame buffer mode, using two pages for ultra fast flipping.
> > > I am probably accessing the LFB memory wrong(_farsetsel, and
> > > _farnspoke*). I tried using the inline assembler, but I can't seem to
> > > get it working.
> >         Why would you use LFB for 320x200?
> No, silly.
> I have moved up to the 64K family. This thing will eventually support
> 640x480,800x600 and 1024x768 in 64K.
	Well, it is going to run twice as slow in 320x200x16bit mode (because there
is twice the memory to write to).  And it will run 9.6 times slower in
640x480x64k mode (and 25 times slower in 1024x768x64k).
-- 
     (\/) Endlisnis (\/)
          s257m AT unb DOT ca
          Endlisnis AT GeoCities DOT com
          Endlis AT nbnet DOT nb DOT ca