AMD designing Navi GPU with dedicated AI circuitry

Posted on Saturday, August 12 2017 @ 15:03 CEST by Thomas De Maesschalck
AMD's Radeon RX Vega 64 launches on Monday, we already have a decent idea of what to expect so the rumor mill is now turning to Navi. This future GPU architecture from AMD will be manufactured on a 7nm process. If things go well for AMD this time, the first Navi chips may launch sometime in the second half of 2018.

NVIDIA's currently shipping GV100 chip that is used by the Tesla V100 datacenter card features dedicated tensor cores for machine learning jobs and it appears AMD will do something similar. FUD Zilla learned that Navi will contain some form of dedicated artificial intelligence (AI) circuitry to improve machine learning performance.
You have to realize that it takes close to four years to finish up a chip and you have to lock down the design at some point. Vega, just like any modern GPU, runs AI algorithms just fine, but it can be compared to the Pascal generation rather than Volta.
AMD Navi slide

About the Author

Thomas De Maesschalck

Thomas has been messing with computer since early childhood and firmly believes the Internet is the best thing since sliced bread. Enjoys playing with new tech, is fascinated by science, and passionate about financial markets. When not behind a computer, he can be found with running shoes on or lifting heavy weights in the weight room.

Loading Comments