From: David Jenkins <me AT jenkinsdavid DOT demon DOT co DOT uk>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: allegro quetion
Date: Sat, 16 Aug 1997 03:36:22 +0100
Organization: None
Distribution: world
Message-ID: <2Zdy$LAmIR9zEwL2@jenkinsdavid.demon.co.uk>
References: <01bca9d7$9a404fa0$5bf141c2 AT liquid-machine>
NNTP-Posting-Host: jenkinsdavid.demon.co.uk
MIME-Version: 1.0
Lines: 34
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

In article <01bca9d7$9a404fa0$5bf141c2 AT liquid-machine>, Filipe Marques
de Almeida <liquidk AT thepentagon DOT com> writes
>sorry for the dumb question but i am kind of new to djgpp and allegro
>
>is there a problem of using something like this for memory bitmaps
>
>BITMAP *bitmap;
>unsigned char *map;
>
>map = bitmap->line[0];
>
>and then using map for accessing in a linear way.
>
>my question is:
>are the memory bitmaps always one chunk (like graph mode 0x13) or is there
>the danger of the bitmaps getting fragmented

I've done something similar.
http://www.jenkinsdavid.demon.co.uk/psychle/3Dpsyc11.c

Look in the blur function.
I just create an array to copy the buffer into then after I've done all
the work using the display which is held in the array I memcpy it into
the buffer. 

BTW that's a brilliant demo, even if I do say so myself. :)
I think it's possible to optimise it even further, can anyone point me
in the right direction??


-- 
http://www.jenkinsdavid.demon.co.uk for C programmers.
David Jenkins