<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Super fast triangle/rectangle intersection test</title>
	<atom:link href="http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/feed/" rel="self" type="application/rss+xml" />
	<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/</link>
	<description>Actionscript Games, Physics and Papervision3D</description>
	<lastBuildDate>Sat, 31 Jul 2010 18:57:50 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.1</generator>
	<item>
		<title>By: jun</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-1462</link>
		<dc:creator>jun</dc:creator>
		<pubDate>Mon, 09 Nov 2009 20:49:59 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-1462</guid>
		<description>lab9, in your function for calculating the line equation y=mx+c, what if xb = xa ?</description>
		<content:encoded><![CDATA[<p>lab9, in your function for calculating the line equation y=mx+c, what if xb = xa ?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jun</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-1461</link>
		<dc:creator>jun</dc:creator>
		<pubDate>Mon, 09 Nov 2009 18:05:09 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-1461</guid>
		<description>Kristof Neirynck, there is some error in your function. thanks!

#          //if all points of the triangle are on one side of the rect 
#          if((l&gt;x0 &amp;&amp; l&gt;x1 &amp;&amp; l&gt;x2)
#          &#124;&#124; (r&lt;x0 &amp;&amp; r&lt;x1 &amp;&amp; ry0 &amp;&amp; t&gt;y1 &amp;&amp; t&gt;y2)
#          &#124;&#124; (b&lt;y0 &amp;&amp; b&lt;y1 &amp;&amp; b&lt;y2)
#          ){
#              return false;
#          }</description>
		<content:encoded><![CDATA[<p>Kristof Neirynck, there is some error in your function. thanks!</p>
<p>#          //if all points of the triangle are on one side of the rect<br />
#          if((l&gt;x0 &amp;&amp; l&gt;x1 &amp;&amp; l&gt;x2)<br />
#          || (r&lt;x0 &amp;&amp; r&lt;x1 &amp;&amp; ry0 &amp;&amp; t&gt;y1 &amp;&amp; t&gt;y2)<br />
#          || (b&lt;y0 &amp;&amp; b&lt;y1 &amp;&amp; b&lt;y2)<br />
#          ){<br />
#              return false;<br />
#          }</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: eli</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-953</link>
		<dc:creator>eli</dc:creator>
		<pubDate>Fri, 26 Jun 2009 21:56:52 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-953</guid>
		<description>I did some coding today and offloaded the intersection test into a Pixel Bender filter.. naturally, it is slow as hell (takes 100 times as long) if you run it synchronously cause you have to create a new job each time the function runs..

But, if we can construct a new test that has 100000 random triangles drawn.. and the intersect test sets the triangle to green or red depending on if it is inside the Culling rectangle.. then we could make a test to take advantage of Pixel Benders ability to run asynchronously.

Though, I think that the small amount of calculations is probably just a bad fit for using a pixel bender filter.. the overhead in creating the new job may just be too much..</description>
		<content:encoded><![CDATA[<p>I did some coding today and offloaded the intersection test into a Pixel Bender filter.. naturally, it is slow as hell (takes 100 times as long) if you run it synchronously cause you have to create a new job each time the function runs..</p>
<p>But, if we can construct a new test that has 100000 random triangles drawn.. and the intersect test sets the triangle to green or red depending on if it is inside the Culling rectangle.. then we could make a test to take advantage of Pixel Benders ability to run asynchronously.</p>
<p>Though, I think that the small amount of calculations is probably just a bad fit for using a pixel bender filter.. the overhead in creating the new job may just be too much..</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jidolstar</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-952</link>
		<dc:creator>Jidolstar</dc:creator>
		<pubDate>Wed, 03 Jun 2009 07:23:07 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-952</guid>
		<description>Thank you. This article is very good.</description>
		<content:encoded><![CDATA[<p>Thank you. This article is very good.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Iain</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-951</link>
		<dc:creator>Iain</dc:creator>
		<pubDate>Sun, 31 May 2009 21:56:19 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-951</guid>
		<description>Surely we want to go the other way - only cull triangles where every point is off-screen - otherwise you see them disappearing and it looks clunky?</description>
		<content:encoded><![CDATA[<p>Surely we want to go the other way &#8211; only cull triangles where every point is off-screen &#8211; otherwise you see them disappearing and it looks clunky?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Kristof Neirynck</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-947</link>
		<dc:creator>Kristof Neirynck</dc:creator>
		<pubDate>Sun, 31 May 2009 14:30:48 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-947</guid>
		<description>Tried to implement the simplest optimizations.
Check if one of the vertexes of the triangle is inside the rectangle and check if all vertexes are on one side of the rectangle. This filters out approximately 88% of triangles.
Obviously I still need to use your lineRectangleIntersect after that.
Seems to run slightly faster.

http://wonderfl.kayac.com/code/6d3fe69ca3450a2ba452c0085aaedd6e8e339fa5</description>
		<content:encoded><![CDATA[<p>Tried to implement the simplest optimizations.<br />
Check if one of the vertexes of the triangle is inside the rectangle and check if all vertexes are on one side of the rectangle. This filters out approximately 88% of triangles.<br />
Obviously I still need to use your lineRectangleIntersect after that.<br />
Seems to run slightly faster.</p>
<p><a href="http://wonderfl.kayac.com/code/6d3fe69ca3450a2ba452c0085aaedd6e8e339fa5" rel="nofollow">http://wonderfl.kayac.com/code/6d3fe69ca3450a2ba452c0085aaedd6e8e339fa5</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: lab9</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-946</link>
		<dc:creator>lab9</dc:creator>
		<pubDate>Sun, 31 May 2009 12:47:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-946</guid>
		<description>and an another version:
http://wonderfl.kayac.com/code/1c3708bc3d5fa42c3d223cfd1091530a4aac8f14

all is done in the same function, so the code is pretty redondant, but runs really fast.</description>
		<content:encoded><![CDATA[<p>and an another version:<br />
<a href="http://wonderfl.kayac.com/code/1c3708bc3d5fa42c3d223cfd1091530a4aac8f14" rel="nofollow">http://wonderfl.kayac.com/code/1c3708bc3d5fa42c3d223cfd1091530a4aac8f14</a></p>
<p>all is done in the same function, so the code is pretty redondant, but runs really fast.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: lab9</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-945</link>
		<dc:creator>lab9</dc:creator>
		<pubDate>Sat, 30 May 2009 23:21:17 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-945</guid>
		<description>You&#039;re right, intersection is absolutely needed to make it work correctly.

I just tried an another version for intersection. It runs pretty well:

http://wonderfl.kayac.com/code/40ad55a14049d30d08a5282252f2047f27476948</description>
		<content:encoded><![CDATA[<p>You&#8217;re right, intersection is absolutely needed to make it work correctly.</p>
<p>I just tried an another version for intersection. It runs pretty well:</p>
<p><a href="http://wonderfl.kayac.com/code/40ad55a14049d30d08a5282252f2047f27476948" rel="nofollow">http://wonderfl.kayac.com/code/40ad55a14049d30d08a5282252f2047f27476948</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Seb Lee-Delisle</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-944</link>
		<dc:creator>Seb Lee-Delisle</dc:creator>
		<pubDate>Sat, 30 May 2009 22:14:22 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-944</guid>
		<description>That is some bloody good optimisation work there on the vector maths - very impressed. But yes, sadly there are several cases where no points are inside either of the two shapes, and the line intersections would still be required after all... but let&#039;s keep working!</description>
		<content:encoded><![CDATA[<p>That is some bloody good optimisation work there on the vector maths &#8211; very impressed. But yes, sadly there are several cases where no points are inside either of the two shapes, and the line intersections would still be required after all&#8230; but let&#8217;s keep working!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: lab9</title>
		<link>http://sebleedelisle.com/2009/05/super-fast-trianglerectangle-intersection-test/comment-page-1/#comment-943</link>
		<dc:creator>lab9</dc:creator>
		<pubDate>Sat, 30 May 2009 21:45:44 +0000</pubDate>
		<guid isPermaLink="false">http://www.sebleedelisle.com/?p=447#comment-943</guid>
		<description>here is the wonderfl version

http://wonderfl.kayac.com/code/9ac50d76e60fb98ba5617a71d07f7c09a17e78cc</description>
		<content:encoded><![CDATA[<p>here is the wonderfl version</p>
<p><a href="http://wonderfl.kayac.com/code/9ac50d76e60fb98ba5617a71d07f7c09a17e78cc" rel="nofollow">http://wonderfl.kayac.com/code/9ac50d76e60fb98ba5617a71d07f7c09a17e78cc</a></p>
]]></content:encoded>
	</item>
</channel>
</rss>
