[Bf-cycles] Traversal performance

[Matthew Heimlich]

Wow, that hair paper is really impressive. I know there has been some
debate as to whether or not oriented bounding boxes would provide any real
benefits to BVH traversal. Looks like that paper does some work answering
that question.


On Fri, May 30, 2014 at 1:22 PM, Marco G <digitalbath86 at hotmail.com> wrote:

> Sorry, seems links i post always get messed formatting when reading the ML
> online archives.
>
>
> http://fileadmin.cs.lth.se/graphics/research/papers/2014/drst/
>
>
> http://www.sci.utah.edu/~wald/Publications/
>
>
>
> ------------------------------
> From: digitalbath86 at hotmail.com
> To: bf-cycles at blender.org
> Date: Fri, 30 May 2014 17:07:41 +0000
> Subject: [Bf-cycles] Traversal performance
>
>
> Hi,
> couple of fresh links!
>
> Dynamic Ray Stream Traversal - 36/53% speedup
> http://fileadmin.cs.lth.se/graphics/research/papers/2014/drst/
>
> And this one, i've read from Cycles Optimization page that hairs needs to
> be more tightly bounded to avoid intersections.
> This is specifically made for hair and fur in production (intel +
> dreamworks research):
> Exploiting Local Orientation Similarity for Efficient Ray Traversal of
> Hair and Fur
> http://www.sci.utah.edu/~wald/Publications/
>
> Quickly reading through though, the latter has no support for mblur yet.
>
> Regards,
> MG
>
> _______________________________________________ Bf-cycles mailing list
> Bf-cycles at blender.org http://lists.blender.org/mailman/listinfo/bf-cycles
>
> _______________________________________________
> Bf-cycles mailing list
> Bf-cycles at blender.org
> http://lists.blender.org/mailman/listinfo/bf-cycles
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.blender.org/pipermail/bf-cycles/attachments/20140530/4212aff3/attachment.htm