NVIDIA today upgraded its CUDA Toolkit to version 3.0, bringing exciting times ahead for GPGPU developers as they gear up for the soon to be released Fermi-based GPUs from the manufacturer. According to NVIDIA, the new toolkit put all the tools required by developers for preparing their code for Fermi at their disposal.
Fermi has been generated lot of interest in the past few weeks and discussion boards have been awash with talks about how it can help accelerate new developments. By releasing the CUDA Toolkit 3.0, NVIDIA has provided coders with the primary means of working out performance gains from Fermi.
With the new toolkit in hand, developers can fully exploit the innovations NVIDIA has made in its Fermi architecture. Here’s a quick look at some of the key new features introduced in the CUDA Toolkit 3.0:
• Support for new GPUs based on Fermi architecture - including ECC, optimized double precision, support for linear algebra libraries such as BLAS and LAPACK, the CUDA-GDB debugger and Visual Profiler
• C++ support – delivering improved productivity with class and template inheritance
• GPGPU/Graphics interoperability – delivering Direct3D 9, 10 and 11 and OpenGL for both CUDA and OpenCL
• Improved developer tools for Linux – including the new CUDA Memory Checker that reports misalignment and out-of-bounds errors
• Tesla Compute Cluster (TCC) – improving performance and cluster management
Developers interested in getting their hands on the new toolkit can download it from here
Discuss in our forums