Thanks for visiting Imaging and Machine Vision Europe.

You're trying to access an editorial feature that is only available to logged in, registered users of Imaging and Machine Vision Europe. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Nvidia and AMD GPU support

Share this on social media:

BitFlow, the world's leading innovator of frame grabbers for machine vision and scientific imaging, is helping vision engineers realize a future where system components and software run seamlessly together by becoming the first frame grabber manufacturer to support workstation graphics processor units (GPU) from both NVIDIA and Advanced Micro Devices (AMD).

"Following our successful launch of the new BitFlow GPU Direct for Video earlier this year, we are now heralding another advance for our customers who use AMD GPUs," explained Donal Waide, Director of Sales for BitFlow. "Compatibility of our frame grabbers with these high-powered GPU designs will help BitFlow continue to dominate the corner of the market where customers are moving large volumes of data, typically with CXP cameras up to 2.5 GB/S, directly to the GPU for processing."

New software extensions co-developed by BitFlow and the two GPU makers bring the awesome processing power of NVIDIA GPU Direct for Video and AMD DirectGMA to BitFlow Cyton™, Alta™, Neon™, Karbon™ and Axion™ frame grabbers by enabling integration of the GPUs with the BitFlow Buffer Interface API (BufIn). In doing so, Bufln can acquire images and synchronize transfer directly to the memory of AMD GPUs to achieve real-time, zero latency processing.

Without the use of an extension, the CPU would be required to copy images from a BitFlow or other third-party frame grabber's buffers to GPU buffers, creating a lengthy delay of wasted cycles. This delay would impact the quality of machine vision as well as the volume of items that can be inspected per second.

Company: