[ab] Rotozoom weirdness

[Anthony Fairchild]

On 12/7/06, Tomi Neste <tomi.neste at netikka.fi> wrote:
>
> Playing with the surface colorkey/alpha properties should fix it. I don't
> remember the details right now, but in general it can be a bit tricky to
> get the desired effect from the rotozoom function.

Thanks Tomi.  I've looked further into the behaviour of rotozoom and
found some postings that confirm that what I'm seeing is by design.
It seems rotozoom was designed to be very fast and the results of even
very straight-forward 90 degree rotations can sometimes be inaccurate.

The solution is to do exactly what I was doing, write my own simple
rotater.  Now I'll have to optimize my Lisp code or write my own
rotater in C and CFFI it in.  At this point I think I will just leave
my code as-is and optimize it later when everything else is done.

Thanks again,

Anthony