<?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: Fast Radiosity using Diffuse convolved environment map</title>
	<atom:link href="http://www.softimageblog.com/archives/248/feed" rel="self" type="application/rss+xml" />
	<link>http://www.softimageblog.com/archives/248#utm_source=rss&amp;utm_medium=rss&amp;utm_campaign=fast-radiosity-using-diffuse-convoluted-environment-map</link>
	<description>People and thoughts behind Softimage in production...</description>
	<lastBuildDate>Fri, 23 Jul 2010 12:23:34 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.2</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Sarah</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17385</link>
		<dc:creator>Sarah</dc:creator>
		<pubDate>Mon, 15 Jun 2009 06:21:38 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17385</guid>
		<description>I am unable to introduce the sharpness and clearness in my work like , you have done in your textures. Can anyone tel me what&#039;s wrong i am doing? or is it something wrong with my graphic card?</description>
		<content:encoded><![CDATA[<p>I am unable to introduce the sharpness and clearness in my work like , you have done in your textures. Can anyone tel me what&#8217;s wrong i am doing? or is it something wrong with my graphic card?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: francisco</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17380</link>
		<dc:creator>francisco</dc:creator>
		<pubDate>Fri, 15 May 2009 19:25:14 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17380</guid>
		<description>Little simple question? and how do you apply all your textures and material tree to this? mix? thanks in advance</description>
		<content:encoded><![CDATA[<p>Little simple question? and how do you apply all your textures and material tree to this? mix? thanks in advance</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Michael Murphy</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17376</link>
		<dc:creator>Michael Murphy</dc:creator>
		<pubDate>Wed, 01 Apr 2009 22:36:12 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17376</guid>
		<description>&quot; Unfortunately you can’t plug anything in the global ambience parameter. So the best if to use a light that cast only ambient light. And for that you need to code it so ask your favourites shader writer to do it.&quot;

Thanks for the article. I had checked this out on your site recently. There are several techniques I&#039;ve read about recently (including yours) where an ambient only light would be very useful.
See zap&#039;s mental ray blog entry about ambient occlusion, where he notes that you can do this in 3ds Max:


Does anyone know any coders that would be willing and able to tackle this?

mm</description>
		<content:encoded><![CDATA[<p>&#8221; Unfortunately you can’t plug anything in the global ambience parameter. So the best if to use a light that cast only ambient light. And for that you need to code it so ask your favourites shader writer to do it.&#8221;</p>
<p>Thanks for the article. I had checked this out on your site recently. There are several techniques I&#8217;ve read about recently (including yours) where an ambient only light would be very useful.<br />
See zap&#8217;s mental ray blog entry about ambient occlusion, where he notes that you can do this in 3ds Max:</p>
<p>Does anyone know any coders that would be willing and able to tackle this?</p>
<p>mm</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: M</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17375</link>
		<dc:creator>M</dc:creator>
		<pubDate>Fri, 27 Mar 2009 18:45:11 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17375</guid>
		<description>Ah! That was it - I had left it at 0. I changed it to 0.001 and it all works now ;-) 

  Thanks so much!!

m.</description>
		<content:encoded><![CDATA[<p>Ah! That was it &#8211; I had left it at 0. I changed it to 0.001 and it all works now ;-) </p>
<p>  Thanks so much!!</p>
<p>m.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Harry Bardak</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17374</link>
		<dc:creator>Harry Bardak</dc:creator>
		<pubDate>Thu, 26 Mar 2009 22:45:04 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17374</guid>
		<description>And what about the max distance ? This one should be very low as well ( 0.001 )</description>
		<content:encoded><![CDATA[<p>And what about the max distance ? This one should be very low as well ( 0.001 )</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: M</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17373</link>
		<dc:creator>M</dc:creator>
		<pubDate>Thu, 26 Mar 2009 21:05:37 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17373</guid>
		<description>Hi Harry- 

   Thanks for your response. I redid the test with exactly the same model and env map as you have, the Buddha model and the beach.hdr image. I have the same settings, 1 sample and 0.01 for spread, and the .hdr has been diffuse_convoluted with the SH_plugin.  

   However, I&#039;m still getting black areas in the render. Most notably is at the feet under the robe and under the chin, and in some creases of the robe.

   Your render is all lit up properly with no black areas whatsoever?! Am wondering what is different in the setup.

 Thanks!

m.</description>
		<content:encoded><![CDATA[<p>Hi Harry- </p>
<p>   Thanks for your response. I redid the test with exactly the same model and env map as you have, the Buddha model and the beach.hdr image. I have the same settings, 1 sample and 0.01 for spread, and the .hdr has been diffuse_convoluted with the SH_plugin.  </p>
<p>   However, I&#8217;m still getting black areas in the render. Most notably is at the feet under the robe and under the chin, and in some creases of the robe.</p>
<p>   Your render is all lit up properly with no black areas whatsoever?! Am wondering what is different in the setup.</p>
<p> Thanks!</p>
<p>m.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Harry Bardak</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17372</link>
		<dc:creator>Harry Bardak</dc:creator>
		<pubDate>Thu, 26 Mar 2009 18:43:33 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17372</guid>
		<description>@Fabio 

Yep to make it work with fur you need to get a fur shader that compute normal correctly to be able to use this. When I say normally the fur should be considered like a tube and not just a ribbon. Usually the normal of the fur is bent towards a the light source. So if you a light that read a diff conv map it should work. 

@M 

I set a very small distance for the AO shader to emulate what stefano discribed and hit the environement as quick as we can but there is always a chance to hit some geo. I choose to use an AO shader because it&#039;s a lot more easy, but maybe bad for education purpose.</description>
		<content:encoded><![CDATA[<p>@Fabio </p>
<p>Yep to make it work with fur you need to get a fur shader that compute normal correctly to be able to use this. When I say normally the fur should be considered like a tube and not just a ribbon. Usually the normal of the fur is bent towards a the light source. So if you a light that read a diff conv map it should work. </p>
<p>@M </p>
<p>I set a very small distance for the AO shader to emulate what stefano discribed and hit the environement as quick as we can but there is always a chance to hit some geo. I choose to use an AO shader because it&#8217;s a lot more easy, but maybe bad for education purpose.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: M</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17371</link>
		<dc:creator>M</dc:creator>
		<pubDate>Thu, 26 Mar 2009 13:56:12 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17371</guid>
		<description>Very interesting article - Thanks for that. 

  For some reason, I always get some black areas in my renders. If the object has a lot of details, and has concave areas or pieces sticking out, some areas remain completely black. Probably because the single ray hits the object itself and not the environment. If I increase the spread from 0.01 to .5, it solves the problem but then I have to also increase the samples, which defeats the whole thing. 

 I noticed in your buddha example, you don&#039;t seem to have this problem. Am I missing something?

 Stefano, would using your method solve the problem? However I&#039;m not sure which nodes in the rendertree I should be using. 

 Any help appreciated!

m.</description>
		<content:encoded><![CDATA[<p>Very interesting article &#8211; Thanks for that. </p>
<p>  For some reason, I always get some black areas in my renders. If the object has a lot of details, and has concave areas or pieces sticking out, some areas remain completely black. Probably because the single ray hits the object itself and not the environment. If I increase the spread from 0.01 to .5, it solves the problem but then I have to also increase the samples, which defeats the whole thing. </p>
<p> I noticed in your buddha example, you don&#8217;t seem to have this problem. Am I missing something?</p>
<p> Stefano, would using your method solve the problem? However I&#8217;m not sure which nodes in the rendertree I should be using. </p>
<p> Any help appreciated!</p>
<p>m.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Stefano</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17370</link>
		<dc:creator>Stefano</dc:creator>
		<pubDate>Wed, 25 Mar 2009 09:13:34 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17370</guid>
		<description>There is an extra trick you can add, to also save the occlusion ray. Instead of the occlusion, use a constant material, with reflectivity set to Environment Only. Since the env sampling is taken along the direction specular to the incoming (eye) ray, you must then bend the normal, so to have it pointing halfway between the incoming ray and the original normal. This way, the reflected vector will match the direction of the original normal.
To do so, in the rendertree take the eye vector, negate it, add it to the normal, normalize the result and plug the normalized vector into the Bump Map slot.</description>
		<content:encoded><![CDATA[<p>There is an extra trick you can add, to also save the occlusion ray. Instead of the occlusion, use a constant material, with reflectivity set to Environment Only. Since the env sampling is taken along the direction specular to the incoming (eye) ray, you must then bend the normal, so to have it pointing halfway between the incoming ray and the original normal. This way, the reflected vector will match the direction of the original normal.<br />
To do so, in the rendertree take the eye vector, negate it, add it to the normal, normalize the result and plug the normalized vector into the Bump Map slot.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Fabio Leporelli</title>
		<link>http://www.softimageblog.com/archives/248/comment-page-1#comment-17369</link>
		<dc:creator>Fabio Leporelli</dc:creator>
		<pubDate>Wed, 25 Mar 2009 04:56:11 +0000</pubDate>
		<guid isPermaLink="false">http://www.xsi-blog.com/?p=248#comment-17369</guid>
		<description>I`ve used a very similar techinque for a project a few years ago and it was super fast and the final result was pretty good.
Unfortunately the set up I&#039;ve used didn&#039;t work with fur ... but for everything else it&#039;s just perfect.
And It&#039;s absolutely flicker free :)

Thanks a lot for the Article Harry</description>
		<content:encoded><![CDATA[<p>I`ve used a very similar techinque for a project a few years ago and it was super fast and the final result was pretty good.<br />
Unfortunately the set up I&#8217;ve used didn&#8217;t work with fur &#8230; but for everything else it&#8217;s just perfect.<br />
And It&#8217;s absolutely flicker free :)</p>
<p>Thanks a lot for the Article Harry</p>
]]></content:encoded>
	</item>
</channel>
</rss>
