« on: August 21, 2010, 05:15:40 PM »
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.
Mozilla hopes to release its fourth beta of Firefox 4 on Monday, adding hardware-accelerated graphics for some Windows users but leaving it turned off by default.
One way Firefox is tackling the technology is by using Windows' Direct2D interface, which can speed up the display of text and graphics on newer versions of Windows.
Bullet Continuous Collision Detection and Physics Library
This ChangeLog is incomplete, for an up-to-date list of all fixed issues see http://bullet.googlecode.com
- Added an OpenCL particle demo, running on NVidia, AMD and MiniCL
Thanks to NVidia for the original particle demo from their OpenCL SDK
- Added GPU deformable object solvers for OpenCL and DirectCompute, and a DirectX 11 cloth demo
Thanks to AMD
- Create a separate library for MiniCL,
MiniCL is a rudimentary OpenCL wrapper that allows to compile OpenCL kernels for multi-core CPU, using Win32 Threads or Posix
- Moved vectormath into Bullet/src, and added a SSE implementation
- Added a btParallelConstraintSolver, mainly for PlayStation 3 Cell SPUs (although it runs fine on CPU too)
In addition to low-level implementation changes which have improved performance across the board, Apple has also removed some implementation inefficiencies which allow us to improve visual quality, most notably in the area of GPU occlusion queries.
MemtestG80 and MemtestCL are software-based testers to test for "soft errors" in GPU memory or logic for NVIDIA CUDA-enabled GPUs (MemtestG80) or OpenCL-enabled CPUs and GPUs by any manufacturer, including both ATI and NVIDIA (MemtestCL). They use a variety of proven test patterns (some custom and some based on Memtest86) to verify the correct operation of GPU memory and logic. They are useful tools to ensure that given GPUs do not produce "silent errors" which may corrupt the results of a computation without triggering an overt error.
GPU-Z is a lightweight utility designed to give you all information about your video card and GPU.
Please note: Some versions of McAfee Antivirus report a "Generic!Artemis" infection. This is a McAfee specific false positive, GPU-Z is not infected with any virus/trojan.
* Fixed shader count on NVIDIA GTX 460, added die size & transistors
* Fixed ROP count on AMD Redwood
* Added support for ATI FirePro V7800, ATI IGP 320M, ATI FireGL X1, ATI HD 4200 variants
* Added launch on Windows startup function accessible via system menu and supporting UAC
* Autoupdater will not check for update when no Internet connection detected
* Fixed browser not starting correctly from autoupdater
* Screenshot detection that promotes GPU-Z's own screenshot upload feature
* Sensor data collection limited to 6 hours of data per sensor (~250 KB)
* Shortened Clevo PCI vendor name
Day two at Game Developers Conference Europe 2010 started for me at 9:00 am sharp, as Jerome Muffat-Meridol from Intel held his presentation "UFO invasion: DX11 and Multicore to the Rescue".
* Multi-Resolution Cloth Simulation
* Reconstructing Surfaces of Particle-Based Fluids Using Anisotropic Kernels
* Practical Animation of Compressible Flow for Shockwaves and Related Phenomena
* A Deformation Transformer for Real-Time Cloth Animation
* Example-Based Wrinkle Synthesis for Clothing Animation
* Stable Spaces for Real-Time Clothing
* Real-Time Simulation of Large Bodies of Water with Small Scale Details
* SIGGRAPH 2010 papers
* SCA 2010 papers
* Efficient Yarn-Based Cloth Simulation With Adaptive Contact Linearization
* A Novel Algorithm for Incompressible Flow Using Only A Coarse Grid Projection
* A Simple Geometric Model for Elastic Deformations
* A parallel multigrid Poisson solver for fluids simulation on large grids
* Discrete Viscous Threads
* Enhancing Fluid Animation with Adaptive, Controllable, and Intermittent Turbulence
The Data Center Solutions (DCS) team have an Oil & Gas customer that is always looking to push the envelope when it comes to getting the most out of GPGPU’s in order to deliver seismic mapping results faster. One of the best ways to do this is by increasing the GPU to server ratio. In the market today, there are a variety of servers that have 1-2 internal GPUs and there is a PCIe expansion chassis that has 4 GPUs.
What we announced today is the PowerEdge C410x PCIe expansion chassis, the first PCIe expansion chassis to connect 1-8 servers to 1-16 GPUs
KB123 - Preview Feature: ATI Stream SDK v2.2 Support For Accessing Additional Physical Memory On The GPU From OpenCL™ Applications
This article provides information about the preview feature for accessing additional physical memory on the GPU from OpenCL™ applications in the ATI Stream SDK v2.2.
The ATI Stream SDK v2.2 currently defaults to exposing 50% of the physical GPU memory to OpenCL™ applications. Certain developers may require accessing additional physical memory in order to optimize their applications when running on the GPU.
For developers who wish to experiment with increasing the amount of physical memory that is accessible to their OpenCL™ applications, the default 50% setting can be changed by setting the environment variable GPU_MAX_HEAP_SIZE to the percentage of total GPU memory that should be exposed.
For example, if you wanted to set the exposed GPU physical memory size to 75%, you need to the GPU_MAX_HEAP_SIZE environment variable to 75.
GPU_MAX_HEAP_SIZE must be set to an integer value between 0 and 100, inclusive.
It should be noted that changing the default setting for exposed GPU physical memory to the OpenCL™ application may result in unexpected behavior. This preview feature is provided solely to allow developers to experiment with accessing a larger portion of the GPU physical memory than is normally exposed.
With the introduction of OpenCL™ 1.1 support, I wanted to take a moment to address a concern some developers have expressed about OpenCL™. The developers I talk to love the fact that OpenCL™ is an open, cross-platform standard and that its evolution is not dictated by a single company. In fact, if you look at the OpenCL™ 1.1 announcement issued by Khronos, there are almost 30 companies and institutions listed, ranging from hardware and tool vendors to national laboratories.
The concern, or rather misconception, is that as an open standard OpenCL™ is slow to evolve and difficult to extend. Thankfully the Khronos OpenCL™ working group provided for a tiered extension mechanism, making OpenCL™ a nimble and extendable open standard.
Recently we released a major update to the Linux drivers for the Mali-200 and Mali-400 MP GPUs. Like many software projects we time-box our driver development, with two major releases each year. This release (r2p0) contains a bunch of exciting new features including Android support, full SMP support, performance optimizations and some important EGL extensions. We'll talk about some of these in future blog posts, but for today I wanted to tell you about the other big change we made for r2p0: we've started to release parts of the driver stack under an open source license.