Date: Thu, 10 Sep 1998 10:30:54 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Tal Lavi <ranla AT post DOT tau DOT ac DOT il>
cc: djgpp AT delorie DOT com
Subject: Re: MAJOR slowdowns in translating TP7 gfx code to DJGPP2: Suplement 2
In-Reply-To: <35F743BA.366@post.tau.ac.il>
Message-ID: <Pine.SUN.3.91.980910103034.8350E-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On Wed, 9 Sep 1998, Tal Lavi wrote:

> he said that the copied test is bigger than the new test or something)

Maybe it was just too big...  Try to make it as short as you can,
without losing the important code fragments.

> a BlankPage routine, which will copy it using movedata(you said it
> will be faster, right?).

Right, movedata and dosmemput are the fastest ways of blitting large
buffers.

> but when I need to write a single pixel, or to put some shape(line,
> rectangle), on screen, I don't see a way that I could use longs instead
> of words, while i'm in 64K modes.

Sorry, I don't understand why a 64K mode prevents you from using
32-bit moves.  Why can't you combine two 16-bit words into a single
32-bit word and move it in one piece?  The code fragment you posted
suggested that you are moving the 16-bit words to consecutive
addresses, so I don't see what's preventing you from using 32-bit
words.