Skip to main content

TSMC might bring a new 12nm manufacturing process to Nvidia's Volta GPUs

Nvidia Xavier SoC
Image used with permission by copyright holder
Nvidia’s Pascal GPU architecture has been extremely successful, with today’s GeForce GTX 10 series offering unparalleled performance at increasingly attractive prices. The company can’t rest on its laurels, however, because AMD has a new line of GPUs coming soon that could give Nvidia some real competition.

Of course, Nvidia isn’t sitting idly and already has plans for its own next generation of GPUs. Code-named Volta, the new line is expected sometime before the end of 2018 and could transition Nvidia’s production to a new 12nm process, Hexus reports.

Chip manufacturer Taiwan Semiconductor Manufacturing Company (TSMC) is expected to be tasked with the production of the Volta GPUs. Apparently, Volta is expected to be combined with Nvidia’s upcoming Xavier “supercomputer” platform to create systems-on-chip (SoCs) for self-driving cars. If the rumors are to be trusted, then Xavier will be produced on a 12nm process as well.

Get your weekly teardown of the tech behind PC gaming
Check your inbox!

There is some uncertainty about whether or not the 12nm process in question is actually as impressive as it sounds. It’s possible that TSMC’s process will remain 16nm, but that a combination of improvements to that process, subcomponent design, and marketing spin will allow the chips to merely be marketed as “12nm.”

There is no specific word on when Volta’s design and manufacturing specifications will be finalized and officially announced. Past Nvidia timelines suggest Volta should arrive sometime in 2018, but there is no precise timing nor specific information on GPU add-in cards for PC based on Volta. Xavier doesn’t have an official timeline either but Nvidia did demo the SoC at CES 2017.

Some other related information pegs TSMC as the manufacturer of Qualcomm’s upcoming Centriq 2400 server processors that are expected to be tasked to bring ARM to Microsoft’s cloud data centers. That is expected to happen by the end of 2017 and represents a real win for the ARM platform.

Mark Coppock
Mark has been a geek since MS-DOS gave way to Windows and the PalmPilot was a thing. He’s translated his love for…
A dangerous new jailbreak for AI chatbots was just discovered
the side of a Microsoft building

Microsoft has released more details about a troubling new generative AI jailbreak technique it has discovered, called "Skeleton Key." Using this prompt injection method, malicious users can effectively bypass a chatbot's safety guardrails, the security features that keeps ChatGPT from going full Taye.

Skeleton Key is an example of a prompt injection or prompt engineering attack. It's a multi-turn strategy designed to essentially convince an AI model to ignore its ingrained safety guardrails, "[causing] the system to violate its operators’ policies, make decisions unduly influenced by a user, or execute malicious instructions," Mark Russinovich, CTO of Microsoft Azure, wrote in the announcement.

Read more