Bresenham Line Algorithm

 BBS: Inland Empire Archive
Date: 08-01-92 (17:26)             Number: 199
From: BRIAN DESSENT                Refer#: NONE
  To: RICHARD VANHOUTEN             Recvd: NO  
Subj: Bresenham Line Algorithm       Conf: (1) 80xxx

 >  BD> plot dot at (xc + x, yc + y)  ; I think there's only 4, but
 >  BD> plot dot at (xc - x, yc + y)  ; there may be 8 plotdot routines
 >  BD> plot dot at (xc + x, yc - y)
 >  BD> plot dot at (xc - x, yc - y)
 > plot dot at (xc + y, yc + x)
 > plot dot at (xc - y, yc + x)
 > plot dot at (xc + y, yc - x)
 > plot dot at (xc - y, yc - x)

Thanks, I knew it was probably 8.

 > I believe the above algorithm is also credited to Bresenham.  It
 > works with square pixels.  If your pixels aren't square, you'll need to use
 > a different algorithm.

No need to change algorithms.  If you want elipses (or to
compensate for non-square pixels), all you have to do is
modify it to loop while y > 0 (instead of while y > x), use
only the first 4 plot dot routines, and multiply the term
(2 * y) by a constant scaling factor when updating d.

And, BTW, yes I do believe both were done by Bresenham.
(both algorithyms came from the same article anyway.)

--- FMail 0.90
 * Origin: I'm not an actor, but I play one on TV. (1:3641/202)