Tooling up
NVIDIA and its "ecosystem partners" want you to know they've "committed to delivering the industry's broadest set of software releases to developers using GPU computing in their work."
What this means is that NVIDIA is using the spotlight of Supercomputing '09 to highlight some of the updates it and its partners have been working on when it comes to GPGPU development languages, tools and libraries.
This, of course, means CUDA, as well as some extra support for C++ and updates to accommodate NVIDIA's new Fermi architecture.
The graphics firm is also crowing about the beta release of its Nexus development environment for massively parallel computing (integrated into Microsoft Visual Studio), as well as the release of its R195 driver which sports some new extensions to its OpenCL 1.0 conformant driver and toolkit.
Using a common NVIDIA refrain, general manager of CUDA, Sanford Russell said "the only effective way to scale performance in demanding applications is to move to a parallel computing model." He went on to say CUDA made this easier thanks to the amount of support it had in the industry and the veritable "network of software consultants and training resources for massively parallel computing."
NVIDIA said it had updated its CUDA Toolkit 3.0 Beta so developers could get cracking on applications for Fermi and has included features like ECC reporting, Dual DMA Engine, Concurrent Kernel Execution and debugging support in cuda-gdb.
The firm says it has also thrown in some performance profiling for both CUDA Visual Profiler and the OpenCL Visual Profiler, as well as adding support for a new unified interoperability API for Direct3D and OpenGL, including Direct3D 11.
