Archive for 2010
September Beta version of Unreal Development Kit (UDK), a free version of the Unreal Engine 3, is available for download.
This month’s build of UDK features significant additions, including:
· UDK users now have access to the gameplay profiler tool.
· Matinee’s movement tracks can now be split into individual translation and rotation components.
· Users can bulk edit texture properties within the Content Browser.
Apart from other usefull changes, September UDK is upgrading PhysX SDK integration with newest features of 2.8.4 SDK, like driverless destribution, SSE2 optimizations and enhanced cloth solver.
Also, we have noticed full pack of NVIDIA APEX .dlls – when APEX will be released to public, you’ll be able to use it with UDK right away.
· ‘PhysXDestructible’ has been removed.
· APEX destructibles are replacing them.
You can download September UDK Beta here or here
As for APEX Toolset release (without it APEX integration into UDK is useless for regular developer), several sources are indicating that this is going to happen somewhere in October
As good tradition, several interesting physical demos were presented during GTC Day 1 Keynote by Tony Tamasi, Senior Vice President Content and Technology in NVIDIA.
First one was showing some high-fidelity smoke simulation, with particles interacting fully with characters, producing nice fluid and turbulent behaviour.
Jerome “JeGX” Guinot has updated his popular benchmarking tool, known as PhysX FluidMark, to version 1.2.2.
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 18.104.22.168.
* 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 22.214.171.124.
According to Golem.de upcoming 3DMark 2011 benchmark won’t rely on PhysX integration, but use some in-house physics engine, based on DX 11 Compute Shaders.
Several Futuremark products like 3D mark 06 and Shattered Horizon game are using PhysX SDK, but only 3DMark Vantage features GPU/PPU accelerated physics, which can affect final CPU score.
From our opinion, this is natural change – solution, that brings substantial benefits to one of the GPU manufacturers, can hardly fit into unbiased benchmark, which 3DMark claims to be.
NVIDIA Physx Review Part 2 by TecnoGaming
This video, showcasing GPU PhysX usage in several games like Warmonger, Cryostasis and Mafia II, is supplementing “NVIDIA PhysX” article from TecnoGaming.com
PhysX Teapot Dryer by Phrogz42
Another PhysX plug-in 2.06 for 3ds Max demonstration, this time – rigid body simulation with 1331 teapots in a clothing dryer, each teapot is a simple convex hull with 16 vertices.
Interesting technical article, called “PhysX: An easy target?“, was posted by user Bohemiq Scali at Window Live blogs.
First part of the entry is dedicated to brief overview of PhysX vs AMD’s physics solutions topic, similar to our “AMD and PhysX: History of the Problem” article, and can be read briefly.
But second part is focused on recent PhysX and x87 theme, and original “PhysX87: Software Deficiency” article by David Kanter. Original statement of mr. Kanter sounds like “SSE can easily run 1.3-2X faster than similar x87 code“, and that’s where Scali gives him a full pack of criticism:
Kanter then makes claims about the gains that can be had from converting the code to SSE. He claims that SSE could quadruple performance in theory, and in reality a boost of more than 2x would be possible. Kanter claims that a modern optimizing compiler can easily vectorize the code for SSE automatically, and such gains could be had from just a recompile.
So nVidia is just leaving all this performance on the table. What’s more, if PhysX would indeed be 2-4 times faster on CPU, it would actually be a threat to GPU-accelerated physics. Kanter claims that PhysX is hobbled on the CPU, and that nVidia is deliberately doing this to make GPU physics look good.
while, actually, “magic” SSE powers were a little exaggerated, since recent tests (#1; #2) with no-doubt SSE optimized Bullet physics engine have shown that
In synthetic tests, there is about 8% to be gained from recompiling. This is nowhere near the 2-4x figure that Kanter was using. In fact, 8% faster PhysX processing would mean even less than 8% higher framerates in games, since PhysX is not the only CPU-intensive task in a game.
Perhaps the net gain in framerate would be closer to 3-4%, depending on the game. In other words, recompiling PhysX with SSE would not make CPUs threaten GPU physics. Not even close. The difference would be lost in the margin of error, most likely.
but in spite of this
Kanter’s article, wrong as it may be, is linked on many news sites and forums all over the web, and many discussions ensue. Most people buy into Kanter’s article, and some sites make even more bold claims than Kanter himself, referring to Kanter’s article as ‘absolute proof’ of nVidia’s evil actions. This is exactly what AMD needs.
You may found Scali’s article biased (AMD conspiracy theory and stuff), but it is worth a read as it has common sense. Give it a glimpse, and share your thoughts.
Also, don’t forget that PhysX SDK 2.8.4 already includes SSE2 compiler option, and should be included into next release of FluidMark, so we’re hoping to perform some tests soon.
Time to expand auditory a little – as we are adding Russian branch of PhysX News section. From now on you can easily switch between languages using this small panel:
I, fellow Zogrim, gonna be responcible for Russian version as well.
Original NVIDIA PhysX Fluid Demo is nice indeed, but it is not using fluid simulation to it’s full potential – there is only two scenes, level geometry is way too basic and water flow is too simple.
That’s why our fellow reader, known as AquaGeneral, desided to pimp Fluid Demo a little, buy replacing original models and, thus, providing more interesting usage of fluid sim.
[08.01.2011] Update: new model – “Two Story Building” – available. Mod reuploaded.
Update #2: If you want to create custom level to yourself, please refer to PhysX Fluid Demo Modding Guide.
Three new variants of initial “Pumping Station or something” scene were added:
1 – “Building”
2 – “Pool”
Siemens PLM Software, a business unit of the Siemens Industry Automation Division, has announced a new integrated machine design solution aimed at creating value for companies that develop and market machine tools and production machines – Mechatronics Concept Designer.
For physics simulation MCD solution is relaying on PhysX SDK:
Mechatronics Concept Designer also includes a state-of-the-art modeling and simulation capability based on NVIDIA PhysX technology. This physics engine, developed with the PhysX SDK from NVIDIA, is similar to the software technology used in many of today’s modern video games. This groundbreaking approach to simulation makes it easy to quickly create and interactively validate alternative design concepts.
In addition, the user is able to interact with the digital machine model while the simulation is running, providing the ability to test the effects of different inputs in real time. The ability to model real-world physical behavior in the virtual world, based on simplified mathematical models, enables early concept verification that helps detect and correct errors when they are least expensive to resolve.
Actually, this is not the first time PhysX SDK is used for serious simulation frameworks, for example, you can recall Microsoft Robotics Developer Studio, also featuring PhysX integration.
Several nice videos were uploaded on YouTube by user named Phrogz42 (and further finding are indicating that Gavin Kistner, Product Designer for PhysX Max and Maya plug-ins, is hiding behing this nick) – they are showcasing usage of PhysX 3ds Max plug-in for simulation of several complex obejcts – mostly chains of various types.
But much more interesting details were revealed by autor today, as responce to our question – PhysX plug-in v2.06 Alpha (current version, available for public, is 2.01) was used to create those demonstrations.
Most simulations were running on sub-realtime performance, due to highly increased iterations count, but stability is impressive, not to mention that almost no joints were used – each chain link is compound obejct, consisting of several rigid body primitives/convex meshes.
Take a look at description of “PhysX Chain Braid” video – “each chain link is a single rigid body with eight automatically-derived convex hulls comprising 160 verts (per link)” – it seems convex decomposition algoritm has finally made it to PhysX plug-in.
We’ll keep an eye on Phrogz42’s channel and will let you know, if something interesting will appear.