Comments on: This could work http://blog.xcski.com/2008/03/31/this-could-work Everything I used to bore people on newsgroups and mailing lists with, now in one inconvenient place. Tue, 07 Oct 2008 09:29:13 +0000 http://wordpress.org/?v=2.6.2 By: Herman Varma http://blog.xcski.com/2008/03/31/this-could-work#comment-98577 Herman Varma Fri, 25 Jul 2008 16:49:23 +0000 http://blog.xcski.com/2008/03/31/this-could-work#comment-98577 Hey long time since I heard from the guy who use to sing Barrett's Privateers over HHCode discussions and beer. Contact me through BIO in Dartmouth NS and I'll see what I can do to give you a hand. herman varma Hey long time since I heard from the guy who use to sing Barrett’s Privateers over HHCode discussions and beer. Contact me through BIO in Dartmouth NS and I’ll see what I can do to give you a hand.

herman varma

]]> By: Sarah http://blog.xcski.com/2008/03/31/this-could-work#comment-83897 Sarah Mon, 31 Mar 2008 17:01:22 +0000 http://blog.xcski.com/2008/03/31/this-could-work#comment-83897 Reminds me of a hack I made waaay back when * mandelbrot sets were newly cool * computers were slow * Pen Plotters existed! I sorted xy-vectors ( and pen color actually ) for large plotter files as part of the plot spool s/w. These could be generated really stupidly, with long pen movements between vectors. After sorting, it was kind of fun to watch it draw in Sierpinski sorted order. Sarah Reminds me of a hack I made waaay back when
* mandelbrot sets were newly cool
* computers were slow
* Pen Plotters existed!

I sorted xy-vectors ( and pen color actually ) for large plotter files as part of the plot spool s/w. These could be generated really stupidly, with long pen movements between vectors.

After sorting, it was kind of fun to watch it draw in Sierpinski sorted order.

Sarah

]]> By: Catsidhe http://blog.xcski.com/2008/03/31/this-could-work#comment-83786 Catsidhe Mon, 31 Mar 2008 01:38:40 +0000 http://blog.xcski.com/2008/03/31/this-could-work#comment-83786 The quad tree sounds like a hack I came up with lo these many years ago, when Mandelbrot sets were still cool. The PC I had to code with was slow enough that it would take a day or more to generate a set to cover the screen, so I made a couple of assumptions. I had read that the MS was connected, so my program calculated the pixels around the outside border, then quartered the area. Each quarter was checked if all pixels around the border were identical. If yes, blockfill and move on to the next quarter. If not, move on to the next quadrant, recurse. This algorithm could generate a MS in a couple of hours, by spending al its time on the fiddly bits, and not wasting time on the big flat areas. I'm still sorta proud of that code. The quad tree sounds like a hack I came up with lo these many years ago, when Mandelbrot sets were still cool.

The PC I had to code with was slow enough that it would take a day or more to generate a set to cover the screen, so I made a couple of assumptions. I had read that the MS was connected, so my program calculated the pixels around the outside border, then quartered the area. Each quarter was checked if all pixels around the border were identical. If yes, blockfill and move on to the next quarter. If not, move on to the next quadrant, recurse. This algorithm could generate a MS in a couple of hours, by spending al its time on the fiddly bits, and not wasting time on the big flat areas.

I’m still sorta proud of that code.

]]>