More efficient solution for caustics

Generic forum to discuss Octane Render, post ideas and suggest improvements.
Forum rules
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Post Reply
MaTtY631990
Licensed Customer
Posts: 754
Joined: Thu Apr 08, 2010 8:38 pm

Will octane add any type of new kernel in the future. I think it needs more efficient methods of dealing with complex caustics in scenes. This would be great addition and welcomed by many.
User avatar
abstrax
OctaneRender Team
Posts: 5513
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

MaTtY631990 wrote:Will octane add any type of new kernel in the future. I think it needs more efficient methods of dealing with complex caustics in scenes. This would be great addition and welcomed by many.
Not in the short term. We have various ideas we would like to try out, but don't have time at the moment. For now, the best strategy on scenes with complicated lighting is to use PMC with caustic blur set to 0.01 - 0.05 and wait it out.
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
MaTtY631990
Licensed Customer
Posts: 754
Joined: Thu Apr 08, 2010 8:38 pm

Thanks for info

Would something like bidirectional pathtracing be one possibility. Classic renderers like Maxwell use this method as well as several other renderers and it would be great to have this in Octane.
User avatar
abstrax
OctaneRender Team
Posts: 5513
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

MaTtY631990 wrote:Thanks for info

Would something like bidirectional pathtracing be one possibility. Classic renderers like Maxwell use this method as well as several other renderers and it would be great to have this in Octane.
Bidirectional pathtracing is really hard to do efficiently on the GPU, which is why we haven't done it. But there are other methods we can try, when we get to it.
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
MaTtY631990
Licensed Customer
Posts: 754
Joined: Thu Apr 08, 2010 8:38 pm

abstrax wrote:
MaTtY631990 wrote:Thanks for info

Would something like bidirectional pathtracing be one possibility. Classic renderers like Maxwell use this method as well as several other renderers and it would be great to have this in Octane.
Bidirectional pathtracing is really hard to do efficiently on the GPU, which is why we haven't done it. But there are other methods we can try, when we get to it.
wow, I always wondered. I kind of thought it might have been something held back due to all the features being implemented over the past year or so. Well in any case will look forward to what a future version will bring.
Post Reply

Return to “General Discussion”