With Mafia 2 coming closer to it’s release, hype atmosphere is getting hotter, and apart from other materials you can find on the web, PC Games Hardware has published decent interview with Denby Grace from 2K Czech on several Mafia’s engine features, and PhysX part specifically:

PC Games Hardware: Are there any differences between the Console and the PC Version of Mafia 2 as far as technical as well as visual aspects are concerned, for example GPU accelerated PhysX effects?

Denby Grace: You hit the nail on the head. The biggest difference is the GPU enabled physics effects. While on a system without GPU (consoles and ATI cards), the game will have a great PhysX simulation with particles and cloth, however, if you have the extra hardware we are able to push things to a whole new level which has not yet been seen in open world game before.

Our destructible environments feature thousands of physical particles which in turn can be affected by individual explosion force fields to make them move. It’s all very impressive and it’s these kinds of improvements that high end PC gamers can expect.

PC Games Hardware: We know you are using Nvidia’s PhysX API and also APEX moduls (“Clothing”). Besides PhysX support why did you decide to use Nvidia’s physics middleware instead of other physics libraries like Havok or ODE? What makes Nvidia’s SDK so suitable for your title?

Denby Grace: PhysX is a really scalable system, which is something that appealed to us. We get to integrate a base system that can be dialed up and down, dependant on what system specs the users have. We have a minimum benchmark frame rate that we want to see, then the PhysX simulation level will adjust itself based upon the type of hardware you have.

CPU vs GPU quite simply means that we can really push the boat out with that much dedicated horsepower without being concerned that it’s going to affect the performance of other areas, separating this out is certainly preferable. Going a step further is separating out the gfx from the physics with separate GPU’s (SLI configuration).

People without a GPU PhysX card will still have these effects but it won’t be quite as advanced. We were very careful in ensuring that this would not affect the game play experience, no matter what system you have. , Mafia II I think looks absolutely amazing even without the full PhysX.

PC Games Hardware: Does regular by CPU calculated physics affect visuals only or is it used for gameplay terms like enemies getting hit by shattered bits of blown-away walls and the like?

Denby Grace: No we don’t do this. We didn’t want to alienate a part of our user base with that sort of selective functionality.

Also, don’t forget to check several comparison PhysX screenshots, located at this technical review.

Besides, Gametrailers.com has uploaded an interview with Eric Liu, Nvidia content manager, covering PC version specific features, like 3DVision and GPU PhysX and APEX support.

Paper is decribing hybrid grid- and – particle based fluid solver used in latest, and technically most impressive, demo from Nvidia – Raging Rapids Ride.

Abstract:

We present a hybrid water simulation method that combines grid based and particles based approaches. Our specialized shallow water solver can handle arbitrary underlying terrain slopes, arbitrary water depth and supports wet-dry regions tracking. To treat open water scenes we introduce a method for handling non-reflecting boundary conditions. Regions of liquid that cannot be represented by the height field including breaking waves, water falls and splashing due to rigid and soft bodies interaction are automatically turned into spray, splash and foam particles.

The particles are treated as simple non-interacting point masses and they exchange mass and momentum with the height field fluid. We also present a method for procedurally adding small scale waves that are advected with the water flow. We demonstrate the effectiveness of our method in various test scene including a large flowing river along a valley with beaches, big rocks, steep cliffs and waterfalls.

We still hope that this solver will make it into next, 3.x release of PhysX SDK.

In addition, demonstrational video is available (61 mb)

“On modern CPUs, SSE can easily run 1.3-2X faster than similar x87 code” – stated Kanter.

However, TGDaily has managed to recieve commentaries from Bryan Del Rizzo, Nvidia spokesperson

[And although] our SDK does [include] some SSE code, we found [that] non-SSE code can result in higher performance than SSE in many situations. [Nevertheless], we will continue to use SSE and plan to enable it by default in future releases. That being said, not all developers want SSE enabled by default, because they still want support for older CPUs for their SW versions.

Update: official responce from Nvidia – We’re not hobbling CPU PhysX

Update #2: some more Nvidia statements at this ars technica article

Update #3: and more at Hothardware.com article “NVIDIA Sheds Light On Lack Of PhysX CPU Optimizations”

But lets get back to original article. According to David, sole reason for PhysX SDK to rely on outdated x87 instruction is to make “Nvidia GPUs looks a lot better than the CPU“. This idea was inherited other websites, like TechReport.com

The PhysX logo is intended as a selling point for games taking full advantage of Nvidia hardware, but it now may take on a stronger meaning: intentionally slow on everything else.

and Semi Accurate

In the end, there is one thing that is unquestionably clear, if you remove the de-optimizations that Nvidia inflicts only on the PC CPU version of PhysX, the GPU version would unquestionably be slower than a modern CPU.

Unfortunately, previous authors are missing few vital points: PhysX SDK is used in many games running on CPU, and physics level in those titles can be easily compared to physics content in games based on other “non crippled” physics engines, like Havok; nor there are any games, that can offer content, similar to GPU PhysX effects, but running on CPU with stable framerate.

And, most important, GPU can accelerate only few parts of PhysX code – rigid bodies, joints, raycasts, forcefields, broadphase, etc – rely purely on CPU, so what is the reason not to optimize those at the full potential, to make PhysX SDK more attractive for developers (and thus increase number of  games with GPU PhysX support) ?! Something is telling us that reason “just to make GPUs look better over CPU” is not so obvious.

And what do you think ? Tell us in comments.

Abstract:

We present a simple and fast method to add wrinkles to dynamic meshes such as simulated cloth or the skin of an animated character. To get the desired surface details, we attach a higher resolution wrinkle mesh to the coarse base mesh allowing the wrinkle vertices to deviate from their attachment positions within a limited range. The shape of the wrinkle mesh is determined by a static solver which runs in parallel to the motion of the base mesh. Our method can be used to automatically enhance a purely animated skin mesh with wrinkles which would be atedious task to do by hand.

The fact that the tessellation of the wrinkle mesh can be chosen independently of the structure of the base mesh can be used to control the look of the wrinkles. The locations of wrinkle formation can be defined by painting the maximum distance the wrinkle mesh is allowed to deviate from the base mesh. The second important application of wrinkle meshes is to add detail to simulated meshes such as cloth. Our method allows one to reduce the resolution of the simulation mesh without losing interesting surface detail. This speeds up the simulation, collision detection and handling and it reduces stretchiness. We show the efficiency and visual quality of the approach in a real-time setting.

In generally, that means more detailed cloth simulation with less resourse consumption. Previous research – Hierarchical Position Based Dynamics – was already added in PhysX SDK 2.8.3 so we believe that Wrinkle Meshes will appear in PhysX SDK or APEX Clothing module soon enough.

Also, you can download demonstrational video (63 mb)

Update: and yet another article from TweakTown – GTX 480 vs GTX 470 + GTX 260 in PhysX games.

Read the full article.

Performance numbers are great, but GTX 465 does not seems to be a good candidate for such role because of heat, noise and high power consumption. I’m personally waiting for low-end Fermi based GPUs, to reinforce my new GTX 470

Update: Another article from TweakTown. HD 5970 and HD 5870 with GTX 465, GTX 285 and GTX 260 as PhysX cards.

Hardly many people can afford such system, but at least, it is interesting read.

Update: From NGOHQ:

After deeper investigation, this issue has identified as a rare overflow bug. Looks like Nvidia PhysX System Software 9.10.0222 doesn’t cripple performance, but I’ll run more tests just to be sure

In following article we are comparing both GPU and CPU performance of PhysX effects, using most recent PhysX System Software 9.10.0222 (blamed for “ruining game experience”), previous 9.10.0129 version and several games – Mirror’s Edge 1.01 and Cryostasis Tech Demo.

System: C2Q 9400 @2.66 GHz CPU, GTX 275 GPU, 4GB RAM, Windows 7 Professional x64

MIRROR’S EDGE.

Standart “Flight Fly By” sequence was used (can be activated by “- FlybyFlight” parameter), in-game setting; 1680×1050, all high, 4xAA. Framerate measured by FRAPS 3.0.1.

As you may see, there is almost no difference in PhysX performance with different System Software versions. Moreover, our framerate pattern correlates with early benchmarks, published more than a year ago, like  Overclockers Club or Tech Report.

CRYOSTASIS TECH DEMO

In-game settings (click to view). Framerate measured by internal benchmark system.

Again, framerate with 9.10.0222 and 9.10.0129 drivers is equal. CPU PhysX performance is similar to other tests.

SUM: according to our tests, there is no difference in performance between 9.10.0222 and 9.10.0129 System Software, either for CPU or GPU calculations. We can not also verify any performance degradation compared to previous benchmarking results.

Thus, we see two explanations of original statement:

1) Performance impairment really does exist, but does occur only when non-Nvidia GPU is spotted in the system.

2) Mr. Eran “Regeneration” Badit has provided not very accurate information, either without malicious intent or with it.

Richard Huddy, AMD’s Worldwide Developer Relations Manager, Ashu Rege, Nvidia’s Senior Director of Content and Technology, and Nadeem Mohammad, Nvidia’s Director of Product Management and PhysX, are speaking out against physics engines support strategies of their rival companies. During discussion, Bullet physics SDK is opposed to PhysX SDK.

Apart from slight pro-AMD tone and some factial mistakes in engines descriptions, it’s interesting read. Conclusion we are complitely agreed with:

Whether or not Bullet takes off remains to be seen, but the next few years will certainly be an interesting challenge for both companies.

However, as we’ve already took a view at AMD and PhysX relationship history, future of that AMD-promoted GPU Bullet and it’s implementation in games (not Bullet SDK itself) isn’t looking so bright and clear for us.

It using PhysX SDK based objects with custom hybrid water simulation, utilizing both heightfield fluid solver and particle simulation.  Boat behaviour is a little choppy, escpecially when it collides with waterside surface, but water simulation looks very impressive.

Update: You can download Raging Rapids Ride demo here

Grid based shallow water flowing pass a terrain with high slope is automatically turned into particle waterfall, and than – back to heightfield water (thus, two different fluid solvers are used simultaneously).

Dynamic wakes are generated, as boat mover along the river. When solid objects impacts with water they generate waves, splashes and sprays.

Foams swirl and show off the vortices, as they are advected along the water’s velocity field.

Finally, scene includes some tearable cloth banners, sofbody based plants and floating logs (with properly calculated buoyancy).

Interesting detail – apart from fact that this demo is limited to Nvidia GPU only, it includes multi-threaded CPU implementation of all physics simulations (up to 8 threads can be used). However, even with sufficient load of all four cores for our C2Q 9400 CPU, average framerate is around 5-8 fps.

10-12 fps – on a single GTX 275. And ~30 – with dedicated GTX 260 GPU.

Sum: interesting demo, but mostly in terms of fluid simulation variety and complexity – graphics isn’t that impressive, boat and camera controls aren’t very friendly (when you are trying to control them).

But it’s good to see Heightfield Fluid solver in action finally – it was promised to be implemented into PhysX SDK for years. Keep your fingers crossed, as it may be already included in PhysX SDK 3.0.

We believe that Raging Rapids Ride demo is going to be available for public download soon enough.

Not bad, huh ? GTX 480 promises 2.5x performance gain over GTX 285 in PhysX calculations. To reinforce this statement, some benchmarking results vere revealed.

First one – for Raging Rapid Ride techno-demo, featuring complex water simulation (you can read our overview here).

Second one  – for actual GPU PhysX enhanced Dark Void title, released in January.

In addition, GTX 480’s ability to simulate and render up to 1 000 000 particles in Supersonic Sled demo was mentioned.

That is all what we have for you today

Update: PhysX performance graphs from various GTX 480/GTX 470 reviews around the web:

NVIDIA GeForce GTX 480 Video Card Review by NjnjaLane

NVIDIA GeForce GTX 480 Fermi by techPowerUp

NVIDIA’s GeForce GTX 480 and GTX 470 by AnandTech

GTX 480 Fermi reviewed, finally by Fudzilla

NVIDIA GeForce GTX 480 Graphics Card Review by HardwareHeaven

GTX480/470 GPU Review by beareyes.com.cn

GTX480/470 GPU Review by pcpop.com

NVIDIA GeForce GTX 480 Overclocking with Voltage Adjustment by Tweak Town

GTX 480 Test Part IV – PhysX and GPGPU by Extra Hardware