[Soc-2013-dev] Weekly Report #9: Cycles Motion Blur

[Brecht Van Lommel]

On Fri, Aug 16, 2013 at 10:31 PM, Gavin Howard <gavin.d.howard at gmail.com>wrote:

> ==== What I will do next week ====
>
> At this point, I need to figure out which sampling algorithm is best for
> per-vertex sampling. That will be my target for the next week, as well as
> starting to actually **work** on the kernel.
>

I'm not sure what you mean by sampling algorithm here or why it's
per-vertex? The way I see it is that you are going to intersect a "moving
triangle" which has a number of motion samples. You are given a ray with a
time, and you would use this time as a value to interpolate between two
motion samples.

Also, before you can use data in the kernel, there's still other steps
needed. You will need to adapt BVH building code to add a "moving triangle"
type primitive, and then you need to modify the BVH data that is exported
to the kernel to include this motion data. You could already test
algorithms in the kernel by hardcoding some triangle data, depends what you
want to work on first.


> ==== Questions ====
>
> Am I correct in my assumptions in the macros above?
>

Yes.


Brecht.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.blender.org/pipermail/soc-2013-dev/attachments/20130817/550d8e39/attachment.htm