I'm currently experimenting with volumetrics and falloff. In this specific example I made a simple falloff on the Z-axis and added basic noise. Whatever solution I come up with, I can't actually get what I want. I want to basically have no noise affecting the density when density is 1 (or higher). Is there a way to get this right?
volumetric noise falloff not acting they way I want
1808 4 1- rove
- Member
- 22 posts
- Joined: Jan. 2011
- Offline
- Konstantin Magnus
- Member
- 682 posts
- Joined: Sept. 2013
- Offline
Try the min function instead:
f@density = min(z, n);
https://procegen.konstantinmagnus.de/ [procegen.konstantinmagnus.de]
- tamte
- Member
- 8784 posts
- Joined: July 2007
- Offline
you are trying to multiply with noised values, even if you add your ramp to the noise, its still gonna vary using the noise values
like 0.1 + 1 = 1.1
0.5 + 1 = 1.5
...
so instead you can blend the noise and density using your ramp, which will make sure that where the ramp is 1 you get clean input density and where its 0 you'll get full noise values:
like 0.1 + 1 = 1.1
0.5 + 1 = 1.5
...
so instead you can blend the noise and density using your ramp, which will make sure that where the ramp is 1 you get clean input density and where its 0 you'll get full noise values:
f@density = lerp(basicNoise, f@density, zDensity);
Edited by tamte - Oct. 2, 2021 12:44:28
Tomas Slancik
FX Supervisor
Method Studios, NY
FX Supervisor
Method Studios, NY
- rove
- Member
- 22 posts
- Joined: Jan. 2011
- Offline
tamte
you are trying to multiply with noised values, even if you add your ramp to the noise, its still gonna vary using the noise values
like 0.1 + 1 = 1.1
0.5 + 1 = 1.5
...
so instead you can blend the noise and density using your ramp, which will make sure that where the ramp is 1 you get clean input density and where its 0 you'll get full noise values:f@density = lerp(basicNoise, f@density, zDensity);
Thanks for the clarification! This makes me feel like I should have just paid more attention during math classes way back at school. It's never too late though ;-).
- rove
- Member
- 22 posts
- Joined: Jan. 2011
- Offline
-
- Quick Links