Trying to do Caustics (reflected light from a Gold object) and not having much luck.
Any good tutorials on this? I couldn't find much in the docs that helped.
-Craig
Caustics
5689 9 3- craiglhoffman
- Member
- 252 posts
- Joined: July 2005
- Offline
- craiglhoffman
- Member
- 252 posts
- Joined: July 2005
- Offline
- pbowmar
- Member
- 7025 posts
- Joined: July 2005
- Offline
I did a video on caustics back for 6 or 7 (can't remember). However, as I am avoiding sleep, here's an example .hip file.
Basically, you assign a Photon Plastic shader to the Vase, set Specular prob to 1, set the rest to 0.
ASsign another to the ground, diffuse only.
Calc your photons in the Mantra ROP.
Render using the GI Light ROP.
The part I always forget is assigning the photon shader to the ground…
Cheers,
Peter B
Basically, you assign a Photon Plastic shader to the Vase, set Specular prob to 1, set the rest to 0.
ASsign another to the ground, diffuse only.
Calc your photons in the Mantra ROP.
Render using the GI Light ROP.
The part I always forget is assigning the photon shader to the ground…
Cheers,
Peter B
- craiglhoffman
- Member
- 252 posts
- Joined: July 2005
- Offline
Hi Peter!
Thanks a lot. I actually found your tutorial a while ago and it seemed like I was doing everything correctly. I went back to an old H7 version and it seems like it works properly. Could it be broken in H8? Or at least the version I am using (8.1.704)?
I can't seem to get H8 to render out a pmap larger than 1 KB.. But H7 does it fine.
Maybe it's just me.
How's it going over there? How are you liking London? Hopefully we can do lunch in Soho again soon!
-Craig
Thanks a lot. I actually found your tutorial a while ago and it seemed like I was doing everything correctly. I went back to an old H7 version and it seems like it works properly. Could it be broken in H8? Or at least the version I am using (8.1.704)?
I can't seem to get H8 to render out a pmap larger than 1 KB.. But H7 does it fine.
Maybe it's just me.
How's it going over there? How are you liking London? Hopefully we can do lunch in Soho again soon!
-Craig
- craiglhoffman
- Member
- 252 posts
- Joined: July 2005
- Offline
- jason_iversen
- Member
- 12457 posts
- Joined: July 2005
- Offline
Hey Craig,
Its not uncommon to crank the # Photons cast to a huge number like 50m. The more photons that “stick” will brighten the image up and improve the quality of the caustic.
Its not uncommon to crank the # Photons cast to a huge number like 50m. The more photons that “stick” will brighten the image up and improve the quality of the caustic.
Jason Iversen, Technology Supervisor & FX Pipeline/R+D Lead @ Weta FX
also, http://www.odforce.net [www.odforce.net]
also, http://www.odforce.net [www.odforce.net]
- pbowmar
- Member
- 7025 posts
- Joined: July 2005
- Offline
- craiglhoffman
- Member
- 252 posts
- Joined: July 2005
- Offline
It seems like I get different results in the size of the photon map if I change the Cone Angle of the light, plus Jason's response to my other posting about Photons says that the Cone Angle does control where photons go, so perhaps we do get some control.
In my case a huge majority of photons are going to get reflected off into nowhere, so a huge number is still needed to get decent caustics.
-Craig
PS: By the way, I am using Houdini 7 because 8.1 seemed to not work correctly for this. Is this just me?…
In my case a huge majority of photons are going to get reflected off into nowhere, so a huge number is still needed to get decent caustics.
-Craig
PS: By the way, I am using Houdini 7 because 8.1 seemed to not work correctly for this. Is this just me?…
- pbowmar
- Member
- 7025 posts
- Joined: July 2005
- Offline
Hi Craig,
I did this in 8.1.866, worked OK for me. Also, I believe that the cone angle does affect it but what happens is the photons cast in the “non cone angle” areas have 0 energy, which still “wastes” them IMHO. I discussed this with SESI a long time ago, so perhaps it changed
The problem is there is a relatively low upper limit (2 billion or something) on the total number you can cast so any with 0 energy don't contribute to the caustic look, meaning the sampling isn't as nice.
Cheers,
Peter B
I did this in 8.1.866, worked OK for me. Also, I believe that the cone angle does affect it but what happens is the photons cast in the “non cone angle” areas have 0 energy, which still “wastes” them IMHO. I discussed this with SESI a long time ago, so perhaps it changed
The problem is there is a relatively low upper limit (2 billion or something) on the total number you can cast so any with 0 energy don't contribute to the caustic look, meaning the sampling isn't as nice.
Cheers,
Peter B
- craiglhoffman
- Member
- 252 posts
- Joined: July 2005
- Offline
-
- Quick Links