<div dir="ltr"><div><div><div>Hi Brecht,<br><br></div>regarding further performance for OSL I was looking at making the structures align more closely to get rid of some of the cost associated with the shaderdata_to_shaderglobals() calls.<br>
<br>But I would need to factor out the differential3 and differential to be more just float3 and float like. <br>Also I am not sure about the alignment of
<style type="text/css">
p, li { white-space: pre-wrap; }</style>Imath::Vec3<Float> vs float3 in all cases / platforms. <br><br></div>An other thing I am already trying to implement is a shim so the svm texture-backend gets the OIIO interface that should yield the biggest performance gain it seems from my simple testing. ( testing Barcelona with missing textures on svm vs osl )<br>
</div><div><br></div><div>Not sure what your thought's on this subject are but would love to hear them.<br><br>Martijn<br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Aug 5, 2013 at 2:50 PM, Brecht Van Lommel <span dir="ltr"><<a href="mailto:brechtvanlommel@pandora.be" target="_blank">brechtvanlommel@pandora.be</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Good find, I wasn't aware there was any per thread info we could<br>
cache. I've committed this patch now with some tweaks, found about a<br>
10% performance improvement on images.blend here.<br>
<br>
Thanks,<br>
Brecht.<br>
<div><div class="h5"><br>
On Sun, Aug 4, 2013 at 5:17 PM, Martijn Berger <<a href="mailto:martijn.berger@gmail.com">martijn.berger@gmail.com</a>> wrote:<br>
> Hi fellow cyclists,<br>
><br>
> I did a little experiment to try and improve the performance of cycles/OSL<br>
> in scenes that use image textures. I tested the Barcelona scene from <a href="http://ba.org" target="_blank">ba.org</a><br>
> and tried a bit of an ugly hack to cache the OIIO perthread info.<br>
><br>
> I only have an ancient macbook pro here with me on holiday and would need to<br>
> investigate further but initial results are a ~ 15% improvement on overall<br>
> render time on this scene.<br>
><br>
> I expect the improvement to be bigger for machines with more cores but I<br>
> cannot test that from here.<br>
><br>
> I attached the patch<br>
><br>
> Martijn<br>
><br>
</div></div>> _______________________________________________<br>
> Bf-cycles mailing list<br>
> <a href="mailto:Bf-cycles@blender.org">Bf-cycles@blender.org</a><br>
> <a href="http://lists.blender.org/mailman/listinfo/bf-cycles" target="_blank">http://lists.blender.org/mailman/listinfo/bf-cycles</a><br>
><br>
_______________________________________________<br>
Bf-cycles mailing list<br>
<a href="mailto:Bf-cycles@blender.org">Bf-cycles@blender.org</a><br>
<a href="http://lists.blender.org/mailman/listinfo/bf-cycles" target="_blank">http://lists.blender.org/mailman/listinfo/bf-cycles</a><br>
</blockquote></div><br></div>