Re: fast line drawing code

[dsf5454@ritvax.isc.rit.edu]

In article <2BC4DA6A.9318@news.service.uci.edu>, andrep@balboa.eng.uci.edu (Andre Prellwitz) writes:
>In article <1q1dnd$1tv@wraith.cs.uow.edu.au> u9050728@wraith.cs.uow.edu.au
>(Shane Kelvin Richards) writes:
>>
>>	I believe the FTA's line drawing routine (used in Modulae) is
>>floating around. Can someone thow it at me if they have it? or point me
>>to a ftp site where it can be found.
>
>Umm, just to nitpick here, but are you talking about the fillmode lines
>(which require fewer memory changes due to only having to change one byte
>per scan line) or the "regular" lines?  I got hold of (yanked) the FTA
>regular line-drawing algorithm from Tim Meekins' Experience 2 demo and
>ported it over to Merlin.  I have never used or tested it, though.  Maybe you
>should ask Tim for some (commented? :) code source?
>
>>+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
>>Shane Richards
>>u9050728@cs.uow.edu.au

Well, folks... I once ported over a fast algorithm for line-drawing and color
dithering... a combination of Bresenham's line-drawing algorithm and Wu
antialiasing algorithm to the GS, but never had the time to _optimize_ it in
assembly. :-( On a 486/33 I think it manipulated something like 1.25 million
pixels per second, at full speed...drawing directly to screen memory. I don't
remember how powerful it was on the GS, but it sure was _nice_ for my needs.
;-)

--Dan
.sig broken :-( -> send any email to dsf5454@ritvax.isc.rit.edu
RIT probably doesn't agree with me, either. ;-)