Nvidia confirms that Ampere is used in all new products

Nvidia sells graphics cards in a number of different product segments, where Geforce represents the gaming segment and Quadro takes up space in workstations. For heavier applications in data centers and machine learning, graphics cards in the Tesla family are used instead. With previous generations, Nvidia has applied different architectures in these product segments, where current products use the Turing architecture in the Geforce and Quadro products, while the Volta architecture takes place in the Tesla family.

However, there are exceptions to this approach, where, among other things, consumer-oriented enthusiast cards such as Nvidia Titan V instead apply Volta. During the GTC 2020 digital event, Nvidia is expected to unveil the new Ampere architecture, and according to a report from Marketwatch, the company intends to make Ampere the cornerstone of all graphics cards for the next generation.

Ampere will eventually replace Nvidia’s Turing and Volta chips with a single platform that streamlines Nvidia’s GPU lineup, Huang said in a pre-briefing with media members Wednesday. While consumers largely know Nvidia for its videogame hardware, the first launches with Ampere are aimed at AI needs in the cloud and for research.

Unquestionably, it’s the first time that we’ve unified the acceleration workload of the entire data center into one single platform

Nvidia’s CEO Jensen Huang confirms that the company is streamlining the product range by making Ampere a common technical platform. Ampere, however, first enters graphics cards aimed at data centers, servers and research that apply calculations of artificial intelligence and machine learning. These are based on the graphics circuit GA100, which is expected to be unveiled during GTC 2020.

Nvidia did not release any information Wednesday about consumer GPUs using Ampere, but when asked by a reporter in the briefing about the difference between enterprise and consumer approaches to Ampere, Huang said “there’s great overlap in the architecture, but not in the configuration”

The statement makes it even more likely that the unveiling of Ampere does not include news about the consumer series Geforce RTX 3000. When asked what distinguishes the professional variants of Ampere from the consumer variants, the answer is that they share a lot architecturally, but that the configuration differs all the more. This includes specifications such as the number of stream processors, amounts of Tensor cores and type of graphics memory.

GTC 2020 kicks off on May 14 at 15:00 and is expected to unveil the graphics architecture Ampere as part of the compact supercomputer system DGX A100.

Would you be horribly disappointed if GTC 2020 does not contain any information about the Geforce RTX 3000 series, or is a revelation of the Ampere architecture enough to satisfy your technology hunger?