Date: Sat, 17 Jan 1998 17:35:04 -0800 (PST)
Message-Id: <199801180135.RAA09090@adit.ap.net>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: Michael Zanyat <ledana AT netway DOT at>, djgpp AT delorie DOT com
From: Nate Eldredge <eldredge AT ap DOT net>
Subject: Re: Fast memory
Precedence: bulk

At 08:04  1/17/1998 +0100, Michael Zanyat wrote:
>Hi again !
>
>Is there a difference in memory access speed using nearptr or other
>methods?
>Or better: What is the fastest way (memory) for virtual screens and
>VGA(13h)?
This is a topic of some controversy. Nearptr is theoretically the fastest,
but many people have found the performance gains to be negligible. Plus,
it's quite dangerous. Standard farpoke* will require a selector load on each
call, which eats many cycles. farnspoke* will work better inside tight
loops. movedata or dosmemput should be quite effective for blits.
>I found some notes on that in the FAQ but makes me not wise :-(
If you think the FAQ should be improved, please contact its maintainer at
<eliz AT is DOT elta DOT co DOT il>.

Nate Eldredge
eldredge AT ap DOT net