Text archives Help
- From: Carson Brownlee <
>
- To:
- Subject: [Manta] Re: Path tracer & Materials questions
- Date: Thu, 18 Mar 2010 15:35:31 -0600
-The volume material's code is hard to read. What techniques/papers is
it based on?
the volume material is a direct volume renderer and uses a vector of
discreet values from the transfer function anded with the volume values
to determine empty space for a space skipping acceleration structure.
Other than that it's fairly simple. James might be able to point you to
a specific paper. There are a few SSE versions I wrote a while back but
they only improved performance by a few percentages so they were taken
out. Aaron Knoll recently built a very fast SSE optimized CPU volume
renderer that uses some sort of intelligent sampling to get a large
performance benefit so you might want to talk to him if you are
interesting in porting it to Manta.
Carson
On Thu, 2010-03-18 at 16:45 -0400, Bo Huang wrote:
>
Seeing some interests in more advanced rendering in other thread, I will
>
bring up some questions and observations regarding the path tracer and few
>
materials.
>
>
>
-The ShadowAlgorithm samples AreaLight randomly in respect to its pdf. This
>
is inefficient for materials such as conductors, and perhaps would rarely
>
yield correct result. Perhaps we need send some hints to the
>
ShadowAlgorithm for perfectly specular materials?
>
>
-I cannot get any area light itself to render because the path tracer
>
ignores emmissive contribution. I understand we do not want to intersect
>
geometry with emissive material because the direct lighting process at the
>
current end vertex of a path is solely responsible for lighting. Hence I am
>
hesitant to change it.
>
>
>
Other little things include:
>
>
-The path tracer treats all importance terms as diffuse. That is, after
>
combining the calculations in the path tracer's traceEyeRay(), the
>
material's sampleBSDF(), and possibly AreaLight's sampling function, each
>
(brdf*cos)/pdf term in the path integral evaluates to (c*cos)/pdf, where c
>
is the diffuse reflectance.
>
>
-What type of material is considered ThinDielectric? How is the Radiative
>
Transfer equation modified? Is this related to the Kubelka-Munk model?
>
>
-For the Dielectric material attenuation function should be exp() instead
>
of Pow() right?
>
>
-The volume material's code is hard to read. What techniques/papers is it
>
based on?
>
>
>
Thanks
>
>
Bo
>
>
The information contained in this e-mail and any accompanying attachments
>
may contain information that is privileged, confidential or otherwise
>
protected from disclosure. If you are not the intended recipient of this
>
message, or if this message has been addressed to you in error, please
>
immediately alert the sender by reply e-mail and then delete this message
>
and any attachments. Any dissemination, distribution or other use of the
>
contents of this message by anyone other than the intended recipient is
>
strictly prohibited. The company accepts no liability for any damage caused
>
by any virus transmitted by this email or any attachments.
Archive powered by MHonArc 2.6.16.