Are Gaming Consoles Solving PDEs for Realistic Animations?

AI Thread Summary
Gaming consoles do not typically solve partial differential equations (PDEs) for realistic animations; instead, they often use algorithms that create visually appealing effects without simulating real physics. While some advanced engines like PhysX incorporate real-life physical computations for various simulations, the primary goal in gaming is to produce visually convincing results rather than accurate physics. The hardware developed for gaming has indeed influenced numerical computing, particularly through advancements in parallel processing. However, the objectives differ significantly: gaming focuses on real-time visual fidelity, while engineering simulations prioritize accuracy over speed. This distinction highlights the unique challenges and approaches in both fields.
cytochrome
Messages
163
Reaction score
3
I was numerically solving the wave equation earlier just to produce a simple illustration of a vibrating string and my computer was working pretty hard. Then I realized how many video games nowadays have such awesome graphics with things like water, waves, motion in general, etc... Are these consoles numerically solving PDEs with obscure boundary conditions to make these animations? If that's the case, it really makes you appreciate how powerful these consoles can be (especially when user input can alter the equations and boundary conditions and run so smoothly).My question: Are gaming consoles numerically solving PDEs with obscure boundary conditions to make animations?
 
Physics news on Phys.org
cytochrome said:
My question: Are gaming consoles numerically solving PDEs with obscure boundary conditions to make animations?

Highly doubtful. As far as I am aware in most cases they don't simulate real water behavior, they rather use algorithms that make the surface "look good". Sophisticated cheating in other words.

See if this page: http://vterrain.org/Water/ doesn't help.
 
Last edited:
there was this video in which they used what they call " real physics engine "
i then went on to read more about this engine , it claims that it uses real-life physical computations to compute the movement of particles like dust , bullets and whatsoever , i don't remember the name of the video
but it was for one of UnReal games * no pun its just the name of the game :p *
also check this PhysX engine
to quote wikipedia
" It supports rigid body dynamics, soft body dynamics, ragdolls and character controllers, vehicle dynamics, volumetric fluid simulation and cloth simulation including tearing and pressurized cloth. "
 
B4ssHunter said:
there was this video in which they used what they call " real physics engine "
i then went on to read more about this engine , it claims that it uses real-life physical computations to compute the movement of particles like dust , bullets and whatsoever , i don't remember the name of the video
but it was for one of UnReal games * no pun its just the name of the game :p *
also check this PhysX engine
to quote wikipedia
" It supports rigid body dynamics, soft body dynamics, ragdolls and character controllers, vehicle dynamics, volumetric fluid simulation and cloth simulation including tearing and pressurized cloth. "


Yes that's what I was looking for. That is truly amazing if you ask me... I wonder if physicists and engineers working in computational mechanics realize that their work is also being used in video games? LOL...
 
No, they're totally oblivious to such things.

I'm curious as to what you mean by 'PDEs with obscure boundary conditions'. Most of the PDEs describing fluid flow and such have relatively simple and quite common boundary conditions.
 
SteamKing said:
No, they're totally oblivious to such things.

I'm curious as to what you mean by 'PDEs with obscure boundary conditions'. Most of the PDEs describing fluid flow and such have relatively simple and quite common boundary conditions.

Obscure was the wrong word, I meant seemingly random or a wide variety of boundary conditions. When you have user input and there's a bunch of possibilities you can get a wide array of boundary conditions. Sorry for the misuse of that word, I can't think of any boundary value problems that would arise in a videogame that aren't capable of being solved numerically...
 
SteamKing said:
No, they're totally oblivious to such things.

Not quite oblivious. A lot of "serous" numerical computing is taking a ride on the back of the hardware developed for mass-market computer gaming (e.g. the large scale parallel processing hardware in graphics cards), and the low hardware costs that come from large production runs.

But the objectives are very different even if the hardware and even some of the numerical methods are the same. A video game animation has to look right (or at least, not look too wrong) when it is computed in real time. And if it doesn't look quite right, adding some smoke, flames, and/or sound effects can easily ihide that fact!. An engineering simulation has to be right, not just look pretty, but getting answers in real time is usually irrelevant. At work, we use simulations that run for hundreds of hours, to compute what happens to something in a few milliseconds.
 
AlephZero said:
At work, we use simulations that run for hundreds of hours, to compute what happens to something in a few milliseconds.

I do the same thing with molecular dynamics simulations. It takes weeks to get a few nanoseconds in real time. I'd like to get into more classical simulations (even though molecular dynamics is primarily classical mechanics...) like fluid flow or vibrations.

Out of curiosity, what do you do?
 
Back
Top