NVIDIA's GPUDirect For Video Technology Loves That Parallel Processing - HotHardware
NVIDIA's GPUDirect For Video Technology Loves That Parallel Processing

NVIDIA's GPUDirect For Video Technology Loves That Parallel Processing

Looking for some serious horsepower? NVIDIA's got you covered. The company just introduced GPUDirect for Video, a technology that enables application developers to deliver higher quality, more realistic on-air broadcoast graphics.  Taking advantage of the parallel processing power of GPUs for image processing, GPUDirect allows industry-standard video I/O devices to communicate directly with NVIDIA professional Quadro and Tesla graphics cards at ultra-low latency.

The company claims that their technology is the fastest, most deterministic way to get video to the GPU for processing, by enabling leading 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, dramatically reducing latency.

Historically, broadcasters and video production professionals had to contend with delays of as many as ten frames per second -- an amount easily visible to the eye -- when transferring video from a video I/O device to a GPU. This is because the two devices previously 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. Most importantly, these delays in aggregate significantly raised the cost and complexity of larger broadcast productions. Might sound a little bit outside of the consumer space, but then again, you may well notice the benefits of the technology in your favorite TV program in the not so distant future.

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.
+ -

"Favorite TV program?" Yeah, right. To me, that's pretty close to "favorite chronic disabling disease." Can't really say I have one.

+ -

Sounds a lot like the AMD FirePro SDI-Link technology

From Fireuser.com


At IBC 2011 AMD announced the FirePro SDI-Link for system integrators and ISVs to design fully featured SDI- and GPU-based solutions with ultra-low latency between AMD professional graphics cards and industry-standard third party SDI input/output cards (e.g. AJA, Bluefish444, Blackmagic Design, DELTACAST, DVS and Matrox).

SDI - Serial Digital Interface - is the primary digital format connection standard in the professional broadcast industry for uncompressed digital video for live feed productions (such as a live TV show), as well as for editing and monitoring video at the highest possible quality. It carries everything (video, audio, and time code) over one cable, with a bandwidth of nearly 1.5 gigabits per second (i.e. you're getting raw HD output before any compression is applied.)

Using the SDI-Link technology and the new FirePro V7900 SDI card, you get incredibly low-latency, high throughput GPU acceleration of real-time video and broadcast FX pipelines. The uncompressed digital video streams come in over an industry-standard SDI input card. The data is transferred to the GPU for real-time OpenGL/DX11/OpenCL processing (visual effects, color grading, encoding, editorial, graphics etc). Then again with virtually zero latency the processed video is passed back to the SDI card for output.

+ -

This article makes me think of Tim the Tool Man Taylor from the show Home Improvement and how he always was like "more power"...

Nice to see that they came up with something to reduce the latency in these type of situations.

Login or Register to Comment
Post a Comment
Username:   Password: