■[CUDA] CUDA 4.1 製品リリースについて（原文まま）
NVIDIA さんから、CIDA 4.1 製品リリース案内が届きました。
Dear CUDA Registered Developer,
The CUDA 4.1 production release has been posted, including a new CUDA Toolkit, SDK, Parallel Nsight and drivers.
This new release makes it easier to accelerate scientific research with GPUs with key features including:
• Re-designed Visual Profiler with automated performance analysis and expert guidance
• New LLVM-based compiler helps your applications run up to 10% faster
• 1000+ new imaging and signal processing functions in the NPP library
We’ve also added a new tri-diagonal solver, 2x faster SpMV using the ELL hybrid format, and two new random number generators to the GPU-accelerated libraries, and some great improvements to the debugging and performance analysis tools.
Join us for a Webinar covering all the great new features of this new release and its high performance GPU-accelerated libraries.
As part of our CUDA Technical Webinar series, you are also invited to join us:
“CUDA x86 – Running CUDA code on Multi-Core CPUs”, presented by
the Portland Group Inc. 10 am (Pacific Time), Tuesday, January 31, Register Now
Thanks and enjoy this great new CUDA release,
Nadeem Mohammad, NVIDIA CUDA Developer Relations