We have now posted the list of sponsored and refereed sessions that NVIDIA will be speaking at for this year's SIGGRAPH. Please bookmark http://developer.nvidia.com/object/siggraph-2009.html for up-to-date information on NVIDIA's participation in SIGGRAPH 2009.
This month's newsletter is now available for reading. This month's big news is NVIDIA's SIGGRAPH schedule and the upcoming NVIDIA GPU Technology Conference. If you're not already a subscriber to the NVIDIA Developer Newsletter, you can read it directly here http://developer.nvidia.com/object/devnews048.html
We invite you to participate in 3 new online seminars focused on GPU Computing. These seminars cover the following topics, offered at multiple times during June and early July:
-
GPU Computing - Introduction to CUDA (1.5 Hours)
-
Performance Considerations for CUDA Programming (1 Hour)
-
Advanced CUDA Optimization Techniques (1 Hour)
Please visit http://developer.nvidia.com/object/gpu_computing_online.html for more information
Last month, NVIDIA announced it has released version 2.2 of the CUDA Toolkit and SDK for GPU Computing. This latest release supports several significant new features that deliver a major leap forward in getting the most performance out of NVIDIA’s massively parallel CUDA-enabled GPUs. In addition, version 2.2 of the CUDA Toolkit includes support for Windows 7, the upcoming OS from Microsoft that embraces GPU Computing. You can find the CUDA 2.2 toolkit at http://www.nvidia.com/object/cuda_get.html
On May 12th, NVIDIA Corporation announced its OpenCL™ 1.0 drivers for Windows XP and LINUX have been submitted to the Khronos OpenCL Working Group for certification immediately after the conformance tests were approved. These pre-release drivers are now available to all NVIDIA GPU Computing registered developers.
NVIDIA was the first to provide pre-release drivers to its OpenCL Early Access Program participants in April 2009. NVIDIA remains the only vendor today with software and hardware available for developers to program with OpenCL as well as other GPU Computing environments including DirectX Compute, C with CUDA extensions and other languages.
NVIDIA’s OpenCL 1.0 drivers will support all GPU’s based on the CUDA architecture Learn More
To apply to become a GPU Computing registered developer or for more information on NVIDIA’s support for OpenCL visit www.nvidia.com/opencl
Pls note: OpenCL is a trademark of Apple Inc. used by permission by The Khronos Group.
The GPU Technology Conference, taking place September 30 to October 2 in San Jose, California, will focus on the latest breakthroughs developers, engineers, and researchers are achieving through the use of the GPU to solve the world’s most important computing challenges. Attendees will
• Learn about the seismic shifts happening in computing.
• Preview disruptive technologies to stay ahead of imminent trends.
• Get tools and techniques to impact mission critical-projects now.
• Network with experts and peers from across a broad range of fields.
Learn More> http://www.nvidia.com/gtc
We just posted a new tutorial and slide deck on Bindless Graphics for OpenGL, which can give you up to 7x performance improvements! Please read our tutorial here http://developer.nvidia.com/object/bindless_graphics.html
We're pleased to announce that Cg 2.2 ( http://is.gd/kGsR ) and NVIDIA Scene Graph SDK 5 (http://is.gd/tIyv ) are now out of beta and available for download in their final versions.
This month's developer newsletter is now available, featuring a recap of GDC, plus news of new software releases and updates. You can find it at http://developer.nvidia.com/object/devnews047.html.
This free webinar series will be offered live with two different sessions at two different times. In the event that you cannot attend during those times, we will also offer recorded copies of those sessions. The first session starts on April 15th, so don't delay!
This series will cover the basics of data parallel computing on GPU’s leveraging NVIDIA’s CUDA architecture . Tutorials will cover many topics including C for CUDA, programming to the OpenCL API , using Direct X Compute, and performance optimization techniques. These sessions will be presented by NVIDIA Developer Technology Engineering team.
Pre-registration is required if you wish to attend the live event. Please follow the links provided and registration details will be emailed back if your registration is successful.
Session 1: GPU Computing – An Introduction, 1.5 Hours + Q&A
Ideal for developers new to GPU computing, this will cover an introduction to the basics of GPU computing using the NVIDIA CUDA technology. Concepts will be illustrated with step-by-step walkthroughs of code samples, which can be readily compiled and run. Little or no prior GPU Computing experience isrequired. Topics covered include:
- Writing a small GPU computing program in C from scratch
- Data transfers
- Executing functions on the GPU
- Taking advantage of on-chip shared memory
- Coordinating CPU and GPU execution
Session Dates for "GPU Computing – An Introduction" (follow links to register)
Session 2: Performance Considerations for CUDA Programming 1.5 hour + Q&A
This is a more advance session on GPU computing and optimization considerations. Concepts will be illustrated using real code examples together with actual performance gains. Attendance of the “GPU Computing – An introduction” tutorial or some experience programming using CUDA Technology is a recommended preparation for this session. Topics covered include:
- Basic optimization techniques
- Performance measurement
- Getting the most out of CUDA’s memory system
- Execution configurations.
Session Dates for "Performance Considerations for CUDA Programming" (follow links to register)
NVIDIA Developer News Archives
