NVIDIA introduced today NVIDIA GPUDirect for Video, a technology that enables application developers to deliver more realistic on-air graphics by permitting standard video I/O devices to communicate directly with NVIDIA professional Quadro and Tesla graphics processing units (GPUs) at ultra-low latency.
NVIDIA claims that the GPUDirect for Video technology is the fastest, most deterministic way to get video to the GPU for processing, enabling video I/O board manufacturers to take advantage of the programmability of the GPU. With GPUDirect for Video, the video I/O card is fully synchronized with the GPUs, eliminating CPU overhead and, most importantly, reducing latency.
Historically, broadcasters and video production professionals had to contend with delays of as many as ten frames -- an amount easily visible to the eye -- when transferring video from a video I/O device to a GPU, because the two devices could not communicate directly with one another to ensure synchronization. This required application developers to manage complex buffering schemes, leading to unnecessary CPU overhead and increased latency.
NVIDIA GPUDirect for Video technology will further advance the growing adoption of GPUs in film, live broadcast, and video production. Nvidi asaid that many companies such as Adobe, AJA, Autodesk, Avid and Sony are already using GPU computing to accelerate their applications.
NVIDIA GPUDirect for Video technology is available immediately, free of charge, directly from NVIDIA for video I/O manufacturers, proprietary hardware providers, and video switcher manufacturers. NVIDIA expects commercially available solutions from its partners will be on the market by late 2011. NVIDIA GPUDirect for Video is supported on Windows 7 and Linux, and is designed to work with NVIDIA Quadro 4000, Quadro 5000, and Quadro 6000 GPUs, and NVIDIA Tesla C-Series GPUs.