Time to update PhysX performance metrics of NVIDIA GPUs with newest title with Hardware PhysX support – Batman: Arkham City.

In following article we’ve tried to gather all the benchmarks and tests, published on the web.

[25.11.2011] Batman Arkham City im Technik-Test: Grafikkarten-, DirectX-11- und Physx-Benchmarks by PCGamesHardware.de

German article with graphics benchmarks, overview of DX 11 and PhysX features, and a superficial PhysX tests.

[25.11.2011] Batman Arkham City GPU Test by GameGPU.ru

Includes PhysX benchmarks with over 20 GPUs tested, however, only on “High” PhysX settings without Hybrid PhysX or dedicated GPU tests.

[27.11.2011] Preview: Batman Arkham City – výkon grafik pod DX11/DX9 a PhysX by DDWorld.cz

Czech preview article with very basic PhysX benchmarking part (only one GPU tested).

[28.11.2011] Want to play Batman: Arkham City with PhysX enabled? Get ready for some abysmal framerates by DSOGaming.com

Despite of provoking title, measured performance is not that bad. Also, this is the first article with benchmarks of dedicated PhysX GPUs.

[30.11.2011] Waiting for the patch: Batman: Arkham Asylum in DX9 mode by 3DNews.ru

Russian article with some PhysX/DX11 effects overview and decent PhysX testing.

[02.12.2011] Technik-Check: Batman: Arkham City im Benchmark-Test by GameStar.de

Includes some benchmarks with different PhysX settings. Results are interesting – GPU PhysX effects on CPU (AMD card) are faster that GPU PhysX on middle-end NVIDIA cards (in this case, minimum fps would be more meaning than average).

[02.12.2011] Batman Arkham City, DirectX 9 e DirectX 11 a confronto by HardwareUpgrade.it

Technology overview and minor PhysX testing.

[05.12.2011] Batman: Arkham City Gameplay Performance and IQ Review by HardOCP

Once again, disputable benchmarking methodology from HardOCP – NVIDIA cards are tested with GPU PhysX effects enabled and compared to AMD cards with PhysX disabled. Thus, it is shown what GPU PhysX takes – performance, but it is not revealed what it brings – additional effects.

[17.12.2011] Batman: Arkham City Performance Review by Playwares.com

Detailed article from a Korean website, however, PhysX part is most interesting for us.

[17.12.2011] Batman: Arkham City Graphics Breakdown & Performance Guide by GeForce.com

Performance and features overview from NVIDIA itself. Graphs are interesting, for example this one shows ineffectiveness of dedicated PhysX card usage in pair with GTX 560 Ti GPU.

[10.01.2012] Batman: Arkham City — podívejte se na PhysX a DX11 v akci by PCTuning.cz

Good technical review (in Czech) with strong benchmarking part (includes CPU tests, for example). For some reason, PhysX effects were tested only on “Medium” settings though.

– We’ll update this post as more articles emerge –

Continuing the good tradition of it’s predecessor, Batman: Arkham City provides built-in benchmark with GPU PhysX (and also DX11) effects, however this time sequence is significantly shorter and less diverse.

Update: GPU PhysX in Batman: Arkham City - review and comparison video

For those who is interested how it looks like, we have prepared following video which is showcasing the difference between minimum and maximum PhysX settings.

Meanwhile, full-scale comparison PhysX video is in the works and should be available next week.

Stay tuned.

JeGX has presented new 1.4.0 version of FluidMark, benchmarking and testing tool for GPU PhysX/Hybrid PhysX systems.

Release Highlights:

• New: added two benchmark presets (1080 and 720).

Preset:1080: with this preset, the settings are the following:: 1920×1080 fullscreen, duration of 60 sec, 60000 particles, heavy additional graphics load and multithreaded PhysX synchronized on the rendering.

Preset:720: with this preset, the settings are the following: 1280×720 fullscreen, duration of 60 sec, 30000 particles, moderate additional graphics load and multithreaded PhysX synchronized on the rendering.

• New: PhysX built-in CPU multicore option added.
• New: score submission with oZone3D.Net account.
• Update: the additional graphics load option covers now the whole screen and not only the upper-right corner.
• Update: revamp of the main startup dialog box (now a bit simpler to use).
• Update: compiled with PhysX SDK 2.8.4.6.
• Bugfix: fixed a nasty deadlock that hung FluidMark sometimes at the end the tests.

You can download PhysX FluidMark 1.4.0 from Geeks3D.com.

We remind you, that in case if you want to use FluidMark 1.40 with Hybrid PhysX Mod, you need to delete/rename PhysXDevice.dll file in application folder, due to usage of driverless PhysX SDK 2.8.4.

Alice: Madness Returns – first game with GPU PhysX support this year and title with most impressive PhysX particle effects.

To determine hardware PhysX performance patterns and GPU requirements we tried to gather all PhysX focused articles and benchmarks, available so far.

[18.06.2011] Alice: Madness Returns GPU test by GameGpu.ru

One of the first articles with proper GPU PhysX benchmarks.

According to their test, only top level NVIDIA GPUs can ensure decent framerate, while used for both graphics and PhysX calculations (however, from our experience, only most intensive PhysX scenes are affecting performance so negative).

Benchmarks of dedicated PhysX cards are also included:

As you may see, dependance is non-linear – from certain point using more powerfull GPU is not resulting in framerate increase.

[13.07.2011] Alice: Madness Returns Review by Bjorn3D.com

Standart game review with small benchmarking part

Second of all, Alice has some fantastic PhysX features, though we’ve noticed a big performance decrease when PhysX is set to High. Medium keeps most of the PhysX effects, and users can get better FPS while gaming without sacrificing too much from the effects

[18.08.2011] Alice: Madness Returns — GPU fyzika v akci by PCtuning

Checz article, includes good set of GPU and CPU benchmarks, with various PhysX settings.

– We’ll update this post as more articles emerge –

New GPU PhysX benchmark was revealed today – it is showcasing APEX integration and hardware accelerated PhysX support in upcoming online TPS game called Mars, developed by Epic China in Shanghai. Mars is the first China developed online shooting title, based on Unreal Engine 3, and is supposed to offer player fresh PvP and PvE experience, vibrant graphics, customizable avatars and weapons.

Benchmark itself is showcasing a decent level of GPU PhysX effects, especially in comparison to other games like MKZ and Jianxia 3.

This includes: various particle effects – physically simulated debris from explosions, chunks from bullet impacts, sparks, etc

Destructible environments, thanks to APEX Destructible module

high number of dynamic Cloth objects – banners, flags, torn apart by shockwaves, and flying newspapers

and Fluid simulation (for example, this fountain is using 16 000 SPH-particles).

We have performed several benchmark runs (high settings, 1680×1024, 2xMSAA), to test out the performance on our system (C2Q 9400 @2.6, 4 Gb RAM, GTX 470, Win 7 x64)

Current state of the game – open beta test. In addition, Mars fully supports NVIDIA 3D Vision technology.

You can download Mars GPU PhysX benchmark from GeForce.cn or game website (712 Mb)

Linus Tech Tips website has performed several benchmarks to find out if low-end dedicated PhysX cards are still valuable in systems with high-end graphics cards, using Mafia 2 benchmark.

8600 GTS GPU is clearly bottlenecking the whole system, resulting in lower fps than even a GTX 580, despite the fact that if single NVIDIA GPU is used in Mafia II, PhysX effects are partially running on CPU.

In conclusion, in case if you want to pair a dedicated PhysX card with GTX 580-class GPU, find a decent one – GTX 550 or “even something a little bit lower-end”, as author of this video recommends.

New version of Kombustor, MSI’s VGA burn-in test and benchmark utility, is now available.

2.0.0 release brings new graphics benchmark, that includes OpenGL 4 rendering, soft shadows, geometry instancing, tesselation and PhysX SDK based cloth and particles simulation – sort of FurMark, FluidMark and TessMark features fused together.

PhysX simulation in MSI Kombustor, running purely on C2Q CPU

Physics part is pretty robust – it is using latest PhysX SDK 2.8.4.4 and supports both GPU acceleration and multi-core CPU optimizations.

You can download MSI Kombustor 2.0.0 from here.

New version of FluidMark, benchmarking tool for GPU PhysX setups, is available for download.

Update: hotfix version 1.3.1 released

• Bugfix: online submission didn’t work properly when the nickname field was filled.
• Change: now Win32 Setup installs two shortcuts: FluidMark.exe (single GPU) and etqw.exe (SLI / CrossFire).

Release Notes:

• New: scoring system change. Now the global score is the sum of PhysX score and GraphX (3D rendering) score. The global score is more appropriate to reflect the gain brought by a dedicated PhysX card.
• New: added a new option to increase the graphics workload ([More graphics load] checkbox).
• Bugfix: fixed GPU memory allocation. Now you can run more than 700’000 SPH particles.
• Change: online submission improved.
• Change: updated GPU monitoring code with latest version of ZoomGPU (1.5.11)
• Change: removed [Heavy PhysX mode] and [Use particle count] checkboxes.

You can download FluidMark 1.3.0 from here

Jerome “JeGX” Guinot has updated his popular benchmarking tool, known as PhysX FluidMark, to version 1.2.2.

Release Notes:

Version 1.2.2 – 2010-09-20
————————–
! Change: new post processing (post-FX) effect.
! Change: scores submissions are limited to fullscreen mode + No AA + No PostFX.
! Update: compiled with PhysX SDK 2.8.4.4.
* Bugfix: the resolution 1920×1080 was not saved at the closing of FluidMark.

As you may notice, particles have changed their color from vibrant yellow to bloody red, but major improvements are related to usage of newest PhysX SDK 2.8.4.4.

1) FluidMark 1.2.2. does no require PhysX System Software to be installed, since SDK 2.8.4. is driverless.

When FluidMark 1.2.0. can only helplessly display error messages, version 1.2.2. starts without any problems.

2) Binary version of PhysX SDK 2.8.4. is compiled with SSE2 optimizations, and, thus, we can expect some performance gain while running FluidMark on CPU. But lets’s test this out:

System: C2Q 9400 CPU, GTX 470 + GTS 450 GPUs, 4Gb RAM, Win 7 x64, 260.63 GPU drivers, 9.10.0514 PSS. FluidMark Settings: 1024x768, Async mode Off, Multi-core PhysX On, 60000 particles, 60000 ms benchmarking time, 3 emitters.

Stable performance improvement – around 14 %. Of course, it may not be purely SSE merit, but related to internal SDK tweaks, but it is good to see a progress.

And what about GPU calculations ?

No actual difference between FluidMark 1.2.0 and 1.2.2.

Obviously, single GTX 470 is significantly faster than dedicated GTS 450, because FluidMark is not stressing system with graphics calculations. In actual games dedicated PhysX GPU will perform much better.

You can download FluidMark 1.2.2. from here

Our thanks to JeGX for pre-release version of FluidMark 1.2.2., provided beforehand

And yet another PhysX article has arrived today – this particular one from PClab.pl is mostly focused on GPU PhysX effects analysis and comparison.

Almost every single modern title with hardware accelerated PhysX content (except for Sacred 2 and Darkest of Days) was reviewed, and detailed description of extra PhysX content both in form of text and comparison videos were included. For example:

Mafia II

Batman: Arkham Asylum

Of course, article would be complete without huge set of benchmarks:

In addition, some space was used to explain situation with multi-threading/x87 in current PhysX SDK, and overview a concurrent physics solutions, like Bullet and Havok.

Sum:

Judging comparison video timestamps, this article was in production for several months. Amount of work is impressive – so, if you know Polish, why are you still sitting here ?