Comments on: Ginger section http://toxiclibs.org/2010/01/ginger-section/ Building blocks for computational design Fri, 27 Aug 2010 09:41:48 +0100 hourly 1 http://wordpress.org/?v=3.0.1 By: Bob http://toxiclibs.org/2010/01/ginger-section/comment-page-1/#comment-470 Bob Wed, 21 Jul 2010 19:14:26 +0000 http://toxiclibs.postspectacular.com/?p=61#comment-470 That's great thanks so much! That’s great thanks so much!

]]> By: toxi http://toxiclibs.org/2010/01/ginger-section/comment-page-1/#comment-465 toxi Wed, 21 Jul 2010 12:00:03 +0000 http://toxiclibs.postspectacular.com/?p=61#comment-465 No need for that (in case of the AABB style clipping as above). The volumetric space itself is a cubic grid and what you're seeing is simply the full space and the fact that he didn't close the sides of the volume. You can also see the effect in the BoxFluid demo which comes with the libraries, only there I chose to close the sides by default (can be toggled on/off though). Online demo of that is here: http://www.openprocessing.org/visuals/?visualID=9673 As for clipping using a freely defined plane or other form: that's something still outstanding (unfortunately), but since the resulting mesh is a standard triangle mesh you could at least do a simple culling of entire polygons by using the classifyPoint() method of the Plane class. If one (or all) vertices of a triangle are on the "wrong" side of the plane, simply don't render it (or build a new mesh which only contains the triangles which are valid). That approach will create fuzzy borders for non-axis-aligned planes, but is better than nothing. If you want to spend more effort you can then also use intersectsRay() from triangle vertices outside to find the correct point on the clipping plane... No need for that (in case of the AABB style clipping as above). The volumetric space itself is a cubic grid and what you’re seeing is simply the full space and the fact that he didn’t close the sides of the volume. You can also see the effect in the BoxFluid demo which comes with the libraries, only there I chose to close the sides by default (can be toggled on/off though). Online demo of that is here: http://www.openprocessing.org/visuals/?visualID=9673

As for clipping using a freely defined plane or other form: that’s something still outstanding (unfortunately), but since the resulting mesh is a standard triangle mesh you could at least do a simple culling of entire polygons by using the classifyPoint() method of the Plane class. If one (or all) vertices of a triangle are on the “wrong” side of the plane, simply don’t render it (or build a new mesh which only contains the triangles which are valid). That approach will create fuzzy borders for non-axis-aligned planes, but is better than nothing. If you want to spend more effort you can then also use intersectsRay() from triangle vertices outside to find the correct point on the clipping plane…

]]> By: Bob http://toxiclibs.org/2010/01/ginger-section/comment-page-1/#comment-464 Bob Wed, 21 Jul 2010 08:01:24 +0000 http://toxiclibs.postspectacular.com/?p=61#comment-464 I'm curious, what method(s) would be a starting point for clipping the volume? Here it looks like a simple AABB is doing the clipping. intersectsRay? Very cool though. I’m curious, what method(s) would be a starting point for clipping the volume? Here it looks like a simple AABB is doing the clipping. intersectsRay?

Very cool though.

]]> By: echoechonoisenoise http://toxiclibs.org/2010/01/ginger-section/comment-page-1/#comment-5 echoechonoisenoise Fri, 15 Jan 2010 21:31:16 +0000 http://toxiclibs.postspectacular.com/?p=61#comment-5 thank you toxi. your libs are a kind of breakthrough in the community no? a little bit more here: http://echoechonoisenoise.wordpress.com/ thank you toxi. your libs are a kind of breakthrough in the community no?
a little bit more here: http://echoechonoisenoise.wordpress.com/

]]>