Hacker News Reader

Comment by bgorman | Hacker News Reader

Does this open the door for PhysX acceleration on AMD cards? Why would Nvidia do this from a business point of view?

Re: business point of view

See the other link currently on the front page: https://blogs.nvidia.com/blog/2018/12/03/physx-high-fidelity.... Gives some ideas of the rationale.

Most of the time physic is not running on the GPU, it's a CPU thing.

Got into some heated debates with people about physx where from a developer standpoint it was off-loading all calculations to the CPU with the present of a physx GPU. Typically in game engines you need to pause the world and look into the physx simulation to get the object's pos/vel/mass. This causes the physx engine to stop the world to finish it's calculations so crossing the PCI bus becomes more and more expensive.

Vast majority of the time, most game engine's to solve this problem just let non interactive objects (cloud's, waves, lighting) to be calculated by the phyx sub-system, and only poking it now and then to prevent the phyx hardware from swamping to the CPU.

> Typically in game engines you need to pause the world and look into the physx simulation to get the object's pos/vel/mass. This causes the physx engine to stop the world to finish it's calculations so crossing the PCI bus becomes more and more expensive.

PhysX keeps two copies of the simulation, you read from one while PhysX is updating the other, and then they swap the pointers. You don't cross the PCI bus to get every position/velocity, that info is transferred in bulk at every step.

From a business point of view they can get developers/researchers working with PhysX early. Eventually these developers will convince their employer to shell out for other non free Nvidia products.

Maybe it's just my perception, but fewer and fewer games these days seem to be using PhysX.

Everything that is made with Unity or UE4 uses PhysX.

That said, not that many games use the GPU-accelerated parts of it; for a lot of gameplay physics CPU code path of PhysX works just fine and does not have special hardware requirements.

Yeah the GPU-accelerated part is what I was referring to. Doesn't seem like it makes much of a difference these days.

I think you're really misunderstanding how this works. The CPU handles pretty much all game logic. It needs to know the position of each car, enemy, bullet, whatever, so it can decide how to respond to things, if a car hits a player it does damage. As a result it's impossible to hand off such calculations to the GPU, so all GPU accelerated physics do is trivial visual-only effects like cloth and water physics or pieces of paper blowing in the wind. Basically anyone expecting GPU accelerated physics that has logical relevance such as physics based destructible environments is expecting too much. In other words GPU accelerated game logic physics was never the promise of PhysX. PhysX focused on making physics really easy and great from a software standpoint, and that has been an outstanding success.

The visual-only portion of GPU physics is not really that compelling which is why there isn't huge uptake. There would need to be a revolution in how games work on a fundamental level where basic game logic is calculated on the GPU to make true GPU physics happen. We might see that eventually but not anytime soon.

That doesn't seem right?

https://docs.nvidia.com/gameworks/content/gameworkslibrary/p...

Says: The GPU rigid body feature provides GPU-accelerated implementations of:

- Broad Phase

- Contact generation

- Shape and body management

- Constraint solver

These are generally only used for the visual effects, you still need these features to make the visual effects look good.

> Broad Phase, Contact generation, Shape and body management, Constraint solver

These are absolutely not only used for the visual effects, they are the fundamentals of a physics engine.

The physics of the visual effect.

GPU calculated physics works just fine when results are brought back to the Cpu.

Raycasting for game logic is cpu based as you mention because the game logic itself is on the cpu. Yet solvers and the true heavy lifting does work well on gpu.

Except, and this is the true reason we see little gpu physics, no one has spare gpu room. Thus cpu side physics wins for most games. Outside specific physics focused games giving up graphics for faster physics is not a profitable trade.

I say this as a gamedev myself who has several times made this exact decision.

The biggest problem with GPU physics is that it's a very difficult problem to tackle and you'll run into compatibility issues between the hardware vendors. It'll work on Nvidia hardware and won't work on AMD or vice versa.

Many games do have spare GPU room to spare, but since there are no good GPU-accelerated solutions for physics they don't have much of a choice.

> The biggest problem with GPU physics is that it's a very difficult problem to tackle

This is a completely meaningless statement.

> you'll run into compatibility issues between the hardware vendors

Those compatibility issues already exist in the form of DX or OpenGL drivers, and most games have to face them. Writing a sim in OpenCL would work on both Nvidia and AMD, and even on Integrated GPUs.

> Many games do have spare GPU room to spare

Many smaller games mighy but most big games do not. And those games with GPU room to spare normally have CPU to spare.

>since there are no good GPU-accelerated solutions for physics

There is - PhysX.

>This is a completely meaningless statement.

How is "difficult to implement" a meaningless statement?

>Those compatibility issues already exist in the form of DX or OpenGL drivers, and most games have to face them. Writing a sim in OpenCL would work on both Nvidia and AMD, and even on Integrated GPUs.

With completely different performance characteristics and very difficult to diagnose bugs.

>Many smaller games mighy but most big games do not. And those games with GPU room to spare normally have CPU to spare.

Maybe if you're talking about mainstream AAA single player titles, but many multiplayer titles tend to have CPU limits instead.

>There is - PhysX.

Which only works on nvidia hardware and is thus a useless solution.

What about compute shaders?

https://www.reddit.com/r/Unity3D/comments/6262d0/10_months_p...

https://www.reddit.com/r/Unity3D/comments/7ppldz/physics_sim...

I am not familiar with PhysX - must you explicitly program with PhysX against the GPU?

No. Most games use it on the CPU, as the GPU accelerated implementation is currently only available on Nvidia cards

> Everything that is made with Unity or UE4 uses PhysX

That's not true - UE4 games can be (and are) built against other physics engines.

Most engines can use other physics engines. People have also used other engines in Unity

Yeah, I should have clarified: most games using Unity or UE4 are using PhysX, since that's what these two engines are using out of the box.

Because nobody used it. It's for the same reason Intel open sourced (or at least promised to, last I checked) Thunderbolt.

PhysX is used in both the Unity and Unreal engines, representing well over 50% of games made.

This is not true at all

Which part? If you're talking about units sold rather than games made, then sure, the most popular games are often on their own engines. But that's a different measure than "games made".

"Unreal Engine 4 uses the PhysX 3.3 physics engine to drive its physical simulation calculations and perform all collision calculations. "

https://docs.unrealengine.com/en-us/Engine/Physics

"As announced at GDC’14, Unity 5.0 features an upgrade to PhysX 3.3. Let’s give it a closer look."

https://blogs.unity3d.com/2014/07/08/high-performance-physic...

Most of the stats I can find say unity is 50-60%, and UE is about 10%:

https://www.linkedin.com/pulse/unity-vs-unreal-engine-more-c...

"He said that Unity powers more than 50 percent of mobile games and about half of all PC games."

https://venturebeat.com/2017/05/25/why-unity-was-able-to-rai...

Reply on news.ycombinator.com