Mirror & Caustics with BiDir

Please use this forum for general user support and related questions.

Moderator: coordinators

Forum rules
Please include your operating system type/version, LuxRender version and Exporter version used when submitting a support post.

Make sure you have read the Release forum thread for Release and RC (Release Candidates) builds as these threads contain information on known problems and workarounds: Test Builds Forum

Mirror & Caustics with BiDir

Postby Vutshi » Mon May 07, 2012 2:19 am

I was curious whether it is possible to see a mirror reflection of caustics with BiDir renderer or not. In theory they should be there. In practice (viewtopic.php?f=16&t=8119#p79937) it is recommended to use SPPM renderer. So I did the simplest scene I can imagine and render it in low resolution up to insane samples per pixel numbers. It was expected to see some caustics at 10^5 S/p. Here are the results.
mirror_caustics 18ksp no over.png
18 kS/p
mirror_caustics 18ksp no over.png (26.86 KiB) Viewed 873 times

mirror_caustics 200ksp no over.png
200 kS/p
mirror_caustics 200ksp no over.png (27.2 KiB) Viewed 865 times


P.S. The scene is illuminated by sun. So, if someone wants to see sun light caustics reflected in a scene one should aim for hundreds of kS/p in BiDir case.

UPDATE: SPPM produces good result almost infinitely fast compared to BiDir, but it is still a little bit blurred.
mirror_caustics2 sppm 310Mp.png
310 MPhotons
mirror_caustics2 sppm 310Mp.png (45.32 KiB) Viewed 865 times

mirror_caustics2 sppm 1.7Gp.png
1.7 GPhotons
mirror_caustics2 sppm 1.7Gp.png (44.24 KiB) Viewed 857 times

mirror_caustics2 sppm 15.35Gp.png
15.35 GPhotons
mirror_caustics2 sppm 15.35Gp.png (43.44 KiB) Viewed 810 times

SPPM settings:
startradius 1.5
photonperpass 500k
alpha 0.9
Last edited by Vutshi on Mon May 07, 2012 10:49 am, edited 4 times in total.
Vutshi
 
Posts: 53
Joined: Sat Oct 23, 2010 2:54 pm

Re: Mirror & Caustics with BiDir

Postby SATtva » Mon May 07, 2012 2:26 am

Nice test. Maybe you'll add a SPPM render as a reference?
Linux builds packager
聞くのは一時の恥、聞かぬのは一生の恥
User avatar
SATtva
Developer
 
Posts: 5500
Joined: Tue Apr 07, 2009 12:19 pm
Location: from Siberia with love

Re: Mirror & Caustics with BiDir

Postby Vutshi » Mon May 07, 2012 2:55 am

SATtva wrote:Nice test. Maybe you'll add a SPPM render as a reference?

Good idea. Now I wonder how soon SPPM will become as sharp as BiDir.
Vutshi
 
Posts: 53
Joined: Sat Oct 23, 2010 2:54 pm

Re: Mirror & Caustics with BiDir

Postby SATtva » Mon May 07, 2012 3:20 am

This depends on SPPM settings. You can lower start radius to reduce initial bias.
Linux builds packager
聞くのは一時の恥、聞かぬのは一生の恥
User avatar
SATtva
Developer
 
Posts: 5500
Joined: Tue Apr 07, 2009 12:19 pm
Location: from Siberia with love

Re: Mirror & Caustics with BiDir

Postby Vutshi » Mon May 07, 2012 3:45 am

SATtva wrote:This depends on SPPM settings. You can lower start radius to reduce initial bias.

I updated the first post with 1.7Gphotons SPPM image. It looks a bit sharper than the 310Mphotons... but the difference is so tiny. Is there any estimation on how long one needs to calculate to get image converged to the proper solution?
Vutshi
 
Posts: 53
Joined: Sat Oct 23, 2010 2:54 pm

Re: Mirror & Caustics with BiDir

Postby A-man » Mon May 07, 2012 6:38 am

In my experience, using a very small (0.25-0.5) starting radius yeilds the best result if you are going to render it for a while (12+ hours).
See Eros's post here viewtopic.php?f=8&t=5525&start=240 (He used .75, which seemed to work fairly well)
Studio XPS 9100 | 2.8 Ghz Intel 930 | 12 GB RAM | Radeon 5870 1GB
User avatar
A-man
 
Posts: 687
Joined: Sat Apr 24, 2010 4:24 pm


Return to LuxRender User Support

Who is online

Users browsing this forum: No registered users and 1 guest