Date: Sun, 21 Feb 1999 13:50:07 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: "Carlos Giani_AEN2003 (M2003)" <DARKVALE AT pcc DOT tgm DOT ac DOT at>
cc: djgpp AT delorie DOT com
Subject: Re: A little question....
In-Reply-To: <D1DEE71F6F@pcc.tgm.ac.at>
Message-ID: <Pine.SUN.3.91.990221134948.6539f-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp AT delorie DOT com


On Sat, 20 Feb 1999, Carlos Giani_AEN2003 (M2003) wrote:

> I'm not sure, but could it be that accessing DOS memory (I mean the 
> first 640 kbyte) is slower than accessing all the other mem?

Not at all.  There's no difference in speed of memory access, no
matter where in the address space that memory lives.

It is true that using farptr functions each access needs 2
instructions, as opposed to 1 instruction for simple pointer
dereference.  However, unless you use farptr for processing large
buffers, the difference is usually negligible; and for large buffers,
there are better techniques.  See section 18.4 of the DJGPP FAQ list
for more details.