I was testing glass with PBR by using default shader in material palate Houdini “glass”. It was nice on the object itself, but the fake caustic caste on floor is a horrible. I can't accept that, is there any way to fix it?
I generated a cache map with area light for caustic and photon before the final render. And used a spotlight as direction light and a point light to create extra specular on the glass for final render.
Setting:
- Allowed all path
- mantra properties > PBR> (check on ) fake caustic support > photon map.
- Reflect limit > 6
- Glossy and diffuse limit > 4
- pixel samples > 12 x 12
I did tried to switch on shadow from Sport light and hopping it might help blur off the shadow cast, but it flatten the whole caustic. Besides, I also encountered, lights didn't really run through the glass when i place the ball inside the glass.
PBR help
6548 8 2
-
- nuvelle
- Member
- 22 posts
- Joined: 8月 2009
- Offline
-
- anamous
- Member
- 186 posts
- Joined: 1月 2006
- Offline
-
- heydabop
- Member
- 203 posts
- Joined:
- Offline
Fake caustic support should not be checked under render properties, or the properties for the material, as fake caustics are what you're seeing in that render. anamous looks like he's covered it in his scene though.
Also, if you are interested in getting caustics while using micropolygon rendering or ray tracing (not PBR), I'd be glad to explain how. However, it is more complex than using PBR.
Also, if you are interested in getting caustics while using micropolygon rendering or ray tracing (not PBR), I'd be glad to explain how. However, it is more complex than using PBR.
-
- nuvelle
- Member
- 22 posts
- Joined: 8月 2009
- Offline
I tested on same render setting by referring to anamous file, but not sure why i got lots of white dots shadow part. I found that if the area light's scale size was too small will causes white pixels in shadows shown in picture 01. Same setting, the image outcome in animus's file is clear and nice, but in my file is very low quality and white pixels in shadows.
Previously, I found it had been mention in another topic “Network Render of Caustics”, that by increasing the scale size of area light for photon cache will solve the white pixel problem. So, I re-render it again, as in picture 02, there is lot of noise and no matter how many pixel sample I increase, it just won't help much but expensive render time.
So, I began to wander and figured it out by mix matching anamous's setting with some concept of my own (not sure is it right or not), and got the effect on picture 03. But the sport lights have causing the cast sharp. Here by I attached the file, please help check, I'm not sure weather I had set wrong.
Previously, I found it had been mention in another topic “Network Render of Caustics”, that by increasing the scale size of area light for photon cache will solve the white pixel problem. So, I re-render it again, as in picture 02, there is lot of noise and no matter how many pixel sample I increase, it just won't help much but expensive render time.
So, I began to wander and figured it out by mix matching anamous's setting with some concept of my own (not sure is it right or not), and got the effect on picture 03. But the sport lights have causing the cast sharp. Here by I attached the file, please help check, I'm not sure weather I had set wrong.
Edited by - 2010年3月8日 16:20:13
-
- nuvelle
- Member
- 22 posts
- Joined: 8月 2009
- Offline
-
- nuvelle
- Member
- 22 posts
- Joined: 8月 2009
- Offline
-
- nuvelle
- Member
- 22 posts
- Joined: 8月 2009
- Offline
-
- kursad
- Member
- 88 posts
- Joined: 3月 2010
- Offline
heydabop
Fake caustic support should not be checked under render properties, or the properties for the material, as fake caustics are what you're seeing in that render. anamous looks like he's covered it in his scene though.
Also, if you are interested in getting caustics while using micropolygon rendering or ray tracing (not PBR), I'd be glad to explain how. However, it is more complex than using PBR.
Sorry for late bump on this topic. I was looking for some light into this issue of MPR and caustics. So I would be interested to know how to set up proper looking caustics in MPR.
-
- nuvelle
- Member
- 22 posts
- Joined: 8月 2009
- Offline
-
- Quick Links