[Fwd: Re: [AD] Interlaced blit]

["L. Ross Raszewski" <lraszewski@xxxxxxxxxx>]

Irk. have to get my replytos set up right.
-------- Original Message --------
Subject: Re: [AD] Interlaced blit
Date: Fri, 16 Jun 2000 10:55:31 -0400
From: "L. Ross Raszewski" <lraszewski@xxxxxxxxxx>
Organization: The Black Trenchcoat Society
To: Bob <ohannessian@xxxxxxxxxx>
References: <394A0E50.B30AAEBB@xxxxxxxxxx>
<394A2D5F.BE3E47BD@xxxxxxxxxxxx>

Bob wrote:
> 
> "L. Ross Raszewski" wrote:
> >
> > I've been working on a routine for interlaced blitting.  Now, lacking a
> > good familiarity with the allegro internals, i'm currently doing it in a
> > pretty naieve way.
> 
> Doesn't someone already have a extension for this ?

If someone does, I haven't been able to find it. A quick perusal of the
message archives reveald that someone had attempted, but there wsn't
enough info to suggest a course of action to me.

> Anyway, the way that I've heard it done is to play with the
> bitmap->line[], so that all odd lines are eliminated. Then line[1]
> points to line 2, line[2] point to line 4, line[3] point to line 6, etc.
> So now you have a bitmap with half the height (or so the other routines
> see it), but blit() should write it as a full bitmap.
> That way, I *think* it should work, but I'm not sure the blitter reads
> from the line[] variable at all (prolly just uses ->dat).
> 
Shouldn't that result in the final blit drawing the image squashed to
half its height, with garbage at the bottom?

I tried somethign similar, using getpixel and putpixel, and it still ran
at "right about" the same speed both ways.

Then, I tried just copying from one's line variable to the other, and my
computer crashed horribly.


Still hackin' away.


> --
> 
>  - Robert J Ohannessian


--
L. Ross Raszewski

allegro stuff:
 http://www.cs.loyola.edu/~lraszews/if/#libs