You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
NVIDIA has launched Jetson AGX Xavier, a 1300 dollar chip set up for robotics. It looks like this will pave the way for far more expensive semis and increase the size of the semi market as very high value chips become common place in a wide range of applications from medical to robotics to autonomous everything. With a speed of 32 TOPS, this must represent an entirely new generation of semi technology. If semis of this cost can prove their economic worth, we could see major changes in the industry. This should give TSMC who is their partner a whole new range of products that few if any other foundries can match. One question, will chips of this power and versatility cause a fundamental change in the foundry industry? Any thoughts or comments on this would be appreciated.
Link below for reference.
Arthur - unless they can reduce power and cost quite a bit, they're in danger of looking great in the prototypes but being designed out for mass production. Not for the high end wired robots of course (who cares about a couple of thousand dollars there and downtimes at recharging stations). Maybe I should qualify my view - industrial apps, certainly, personal assistant/home/retail/hospitality/hospital/.. not so much.
Bernard, I figure we are going to see many, many more chips like this that have the power to be very productive, providing a substantial ROI, creating entirely new markets for chips, some even maybe even multiple chips of this cost ganged together to handle complex functions end to end. If this happens, what do you think the ramifications will be for the industry from EDA to final applications. I think this is the beginning of a new frontier for the industry and we will see an entirely new ecosystem built around a multitude of chips of this size, scale and cost. Any additional thoughts based on this would be appreciated. I also feel TSM, being almost always a leading edge innovator may be the leader in creating such an ecosystem.
