After 4 months of hard working, finally my bidirectional path tracer is starting to take shape.My big error is that i’m weighting my paths all wrong ,so specular surfaces appear a little bit darker and sometimes even brighter.
There is still a lot of improvement to be made,for example i can use a better heuristic to combine the paths (Balance heuristic,power heuristic…) ,but for now i’m going to stick with my current implementation and probabily move to another topic ,may be acceleration structures or BRDFs or even Fractal Flames…I don’t know
Advertisement
Please make the source available.
It´s very interesting!
Cool. I would like to see some scenes with more complex lightning to see what bipt is really capable of.
Btw. That underwater caustics would look better with smoothed normals
. Keep it up.
Thanks guys for the comments
@Alex:
The beauty of photorealistic rendering is not in coding but in the math behind it,so if you want to implement a renderer then learn how to do it and i’ll be here to assist you ,just email me and we can discuss my implementation/how to implement a BDPT.
@Michael:
Well,i liked the under water caustics but quite frankly i expected the caustics to be sharper ,so just like you said may be i should make the normals “smooth”.And i’ll definetly test more scenes in the future to see what BDPT is capable of.
I’ll keep you updated
Very very nice. !
I would be nice to see you render with “more complex” scene.
How many times for your computer to calculate those picture ?
Sorry ,i don’t remember exactly the render time but it wasn’t that high ,may be around 30 on average,but i think i can make it faster by adding multithreading and implementing a better acceleration structure like a kd-tree which is the topic of my interest right now.
Thanks for your comment btw,its nice to see you here
30 ? 30 minutes ? 30 seconds ?
30 minutes