« on: September 25, 2010, 02:21:08 PM »
Nvidia’s GPU Conference just ended. Here is a list of the 10 most noteworthy products, technologies, and news items from the event.
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
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.
Added the NVIDIA Quadro 6000, 5000 and 4000 graphics cards.
Updated the specifications of the NVIDIA Quadro FX 3800 graphics card.
Updated the specifications of the NVIDIA Quadro FX 1800 and FX 1700 graphics cards.
Updated the specifications of the NVIDIA Quadro FX 580 graphics card.
Updated the specifications of the NVIDIA Quadro FX 380 LP and FX 380 graphics cards.
Removed the NVIDIA Quadro FX 5900 (Future) graphics card.
No. of cards: 126
You can download the Bullet 2.77 from the usual place at http://code.google.com/p/bullet/downloads/list (based on svn revision 2218)
The main new feature is the OpenCL and DirectCompute hardware accelerated cloth simulation.
Thanks to Lee Howes from AMD for this contribution and thanks to NVIDIA for a Fermi GTX 470 GPU for compatibility testing!
We have done a lot of work to streamline OpenCL development for various platforms. We can debug OpenCL kernels using MiniCL, a simplified OpenCL replacement that compiles kernels using the regular C/C++ compiler. The demos can be compiled using OpenCL 1.0 and 1.1 using the most recent AMD and NVIDIA OpenCL SDKs for Windows and Linux as well as Mac OSX. A lot of work went into tweaking the CMakeLists.txt for the cmake build system so it works nicely with various OpenCL SDKs cross-platform. We also used a modified version of CMake to generate Visual Studio project files that we distribute, using relative paths and environment variables etc. The OpenCL kernels are embedded in the executable by default, but they can also be loaded from disk.
There is also a DirectCompute cloth implementation for Microsoft DirectX11. You can see Lee Howes presentation at MSDN here
* OpenCL & DX11 cloth simulation as part of BulletSoftBody. See Bullet/Demos/OpenCLClothDemo and Bullet/Demos/DX11ClothDemo. This is an initial implementation, and there is no support for collision detection or other advanced features yet.
* New btParallelConstraintSolver as part of BulletMultiThreaded. Although mainly targetting PlayStation 3 SPUs it also accelerates multi-core PCs. See Bullet/Demos/MultiThreadedDemo or get the spubullet-2.77 from PS3 Devnet.
* Many more bug fixes and minor features.
Check out the precompiled Windows executables in the download section.
Some related news:
* We forked the 3ds Max plugin, replacing PhysX by Bullet. It is open source under the MIT license and it supports .bullet export. See http://dynamica.googlecode.com for source and precompiled versions.
* OgreKit is a new open source framework that makes very easy to develop using Bullet, Ogre, OpenAL and Lua. It directly reads Blender .blend files and works on Windows, Linux, Mac OSX and iPhone/iPad. See http://gamekit.googlecode.com
* Cinema 4D release 12 has improved support for Bullet, including constraints and their improved soft body implementation. Cinema 4D 12 also supports .bullet export. Download a free trial demo here: http://www.maxon.net
Feedback is welcome,
Component maker Zotac is one of NVIDIA's closest partners and it has now announced the new ZOTAC GeForce GTX 460 3DP that offers something unique, name support for up to four monitors.
Our first goal with interactive render (IR) was the full integration of our OpenGL viewport. This way, IR can be used the same way with the same navigation controls. Besides the integration with the viewport, full integration with the rest of UI is implemented, so that almost every function of Thea works in co-operation with IR. This way, we can tweak settings, apply materials, etc. with IR keep working in the background.
The second goal, was the render method itself. We wanted a new mode that would be as fast as possible especially in initial updates. Although this can be parameterized to use TR1/TR2 unbiased methods, for IR we developed two "lighter" techniques; one progressive biased and a Monte Carlo unbiased.
In today's Dispatch, we transport you to the nexus where fantasy meets technology. And at the center of it all, you'll find Intel HD Graphics, our newest integrated graphics offering and the first to include the CPU and GPU in one package. Beam me up, Scottie!
• When designing games, developing for both maximum gameplay and reduced power consumption can be a constant concern. In Star Trek* Online Goes Where No Game Has Gone Before, Jimb Esser, the lead graphics programmer for Cryptic, discusses the challenges faced in taking this beloved franchise into the world of MMO gaming, and why optimizing the game to run on processors equipped with Intel HD Graphics was an obvious choice.
• The 2010 Intel® Core™ processor family with Intel HD Graphics provides new features that improve system performance and reduce power usage, especially in mobile systems. Now, both developing and running applications when on-the-go with your laptop can give you increased performance. Read how in the Intel® HD Graphics Dynamic Frequency Technology white paper.
Indigo 2.4 is dramatically faster than Indigo 2.2, a result of numerous optimisations in the core rendering engine: vastly more efficient MLT, up to 2X faster instancing and motion blur in complex scenes, colour noise is completely eliminated, 4X faster blackbody illumination and many other substantial improvements.
Indigo shaders run 10-20X faster using Indigo 2.4 in 64 bit mode than Indigo 2.2 64 bit, a night-and-day performance difference for complex materials.
In fact, today a new Gallium3D state tracker was pushed into Mesa and it's perhaps the most interesting state tracker for this open-source graphics driver architecture yet. It's a state tracker that exposes Microsoft's DirectX 10/11 API on Linux! And it's already working and can be hooked into Wine!
...application I developed for my diploma thesis "Realistic visualization of a character in a real-time application" with Unity including:
- Skin Shading (incl. a subsurface scattering approximation)
- Hair Shading (anisotropic specular, ...)
- Cloth Simulation
- Facial animation (incl. detailed dynamic normal maps)
- and much more...
To download the application and for more information about this project, have a look at my blog at http://janbubenik.blogspot.com
Visions Of Chaos is a professional high end software application for Windows. It is simple enough for people who do not understand the mathematics behind it, but advanced enough for fractal enthusiasts to tweak and customise to their needs. It is the most complete all in one application dealing with Chaos Theory available. Every mode is written to give the best possible quality output. There are thousands of sample files included to give you an idea of what Visions Of Chaos is capable of.
Aparapi is an API for expressing data parallel workloads in Java and running those workloads on a compatible GPU if possible. Where your workload actually ends up running depends on
* whether you have the ATI Stream SDK installed
* whether you have a compatible GPU
* whether your Java parallel code avoids any constructs that would make it untranslatable to OpenCL™. These restrictions are detailed in the Aparapi Readme file.
If your code and your target platform meet all of the above, Aparapi will translate your workload to OpenCL™ and will run your workload on the GPU. Otherwise, your same code will still run as a parallel workload using a Java Thread Pool and will be able to take advantage of multiple cpu cores if you have them.