I find the application of particle-based fluids in games particularly interesting because the player can interact with the particles in real time, and see the effects of their actions! This in turn makes simulating the particle-based fluids difficult, because it must be incredibly efficient, and still visually appealing and accurate. Here's an interesting paper that offers one solution, including C code for how they did it!
http://graphics.cs.cmu.edu/nsp/course/15-464/Fall09/papers/StamFluidforGames.pdf
orenazad
Speaking of simulations in games, this reminds me of the new smoke grenades that were presented in CS2:
https://www.youtube.com/watch?v=_y9MpNcAitQ
I'm sure those obviously aren't accurate simulations, but when I watched the video I was reminded of everything we've discussed so far! It will be interesting to further see how they interact with the environment and other dynamic objects around them. For example, will walking through them at speed shift the smoke / drag some behind you? The clips of it also dynamically filling space are also reminiscent of this.
william-fei
How would particle-based fluids be used in other fields beyond computer graphics such as in scientific or engineering simulations? I remember seeing something like this in a civ-e elective
LuxuFate
@william-fei One common thing that particle-based fluid simulations can be used for is testing dams or coastal defense against floods. This is important for hydrologic response and sediment transfer too.
jonathanlu31
For those kinds of simulation tests, how do they know the simulation matches the actual forces? It seems like most of this is modeling the general form and appearance but isn't as concerned with whether the forces are really accurate.
anzeliu
For whether the particle simulation for fluids matches the actual forces, I think it depends on where it is being used. For animation/entertainment purpose, it is more about emulating the visual behavior of fluid than to obey the physics exactly. The book, The Art of Fluid Animation, by Jos Stam, describes various artistic or visual based fluid animation techniques.
countermoe
@orenazad I was also really impressed with the new smoke grenade effect shown off for CS2, however, I'm not sure if the effect is fine-grained enough to support realistic effects such as the drag you mentioned. From what I understand, the effect is volumetric, not truly a particle-based simulation. This would mean that the effect is bound by the size of each "box" contained across each smoke grenade.
prannaypradeep999
Would the movement of fluids differ based on the viscosity of fluids? How are different particles of different types of fluid usually tracked in comparison.
sZwX74
@prannaypradeep999 I think viscosity would be something that could be taken into account when modeling the movement of fluid, and thus change how the fluid is rendered.I don't think there would be a significant change in how the particle is tracked, as the properties of the particles' behavior would change, but the actual ideas behind fluid simulation remain the same.
I find the application of particle-based fluids in games particularly interesting because the player can interact with the particles in real time, and see the effects of their actions! This in turn makes simulating the particle-based fluids difficult, because it must be incredibly efficient, and still visually appealing and accurate. Here's an interesting paper that offers one solution, including C code for how they did it! http://graphics.cs.cmu.edu/nsp/course/15-464/Fall09/papers/StamFluidforGames.pdf
Speaking of simulations in games, this reminds me of the new smoke grenades that were presented in CS2:
https://www.youtube.com/watch?v=_y9MpNcAitQ
I'm sure those obviously aren't accurate simulations, but when I watched the video I was reminded of everything we've discussed so far! It will be interesting to further see how they interact with the environment and other dynamic objects around them. For example, will walking through them at speed shift the smoke / drag some behind you? The clips of it also dynamically filling space are also reminiscent of this.
How would particle-based fluids be used in other fields beyond computer graphics such as in scientific or engineering simulations? I remember seeing something like this in a civ-e elective
@william-fei One common thing that particle-based fluid simulations can be used for is testing dams or coastal defense against floods. This is important for hydrologic response and sediment transfer too.
For those kinds of simulation tests, how do they know the simulation matches the actual forces? It seems like most of this is modeling the general form and appearance but isn't as concerned with whether the forces are really accurate.
For whether the particle simulation for fluids matches the actual forces, I think it depends on where it is being used. For animation/entertainment purpose, it is more about emulating the visual behavior of fluid than to obey the physics exactly. The book, The Art of Fluid Animation, by Jos Stam, describes various artistic or visual based fluid animation techniques.
@orenazad I was also really impressed with the new smoke grenade effect shown off for CS2, however, I'm not sure if the effect is fine-grained enough to support realistic effects such as the drag you mentioned. From what I understand, the effect is volumetric, not truly a particle-based simulation. This would mean that the effect is bound by the size of each "box" contained across each smoke grenade.
Would the movement of fluids differ based on the viscosity of fluids? How are different particles of different types of fluid usually tracked in comparison.
@prannaypradeep999 I think viscosity would be something that could be taken into account when modeling the movement of fluid, and thus change how the fluid is rendered.I don't think there would be a significant change in how the particle is tracked, as the properties of the particles' behavior would change, but the actual ideas behind fluid simulation remain the same.