Various leakers have begun sharing new purported details on NVIDIA’s next-generation gaming graphics cards, the GeForce RTX 40 Series. According to one leaker, Greymon55, NVIDIA’s Ampere successors will “definitely” leverage the company’s new Lovelace architecture, named after computer scientist and mathematician Ada Lovelace. Green team’s plans for its Lovelace-based GeForce RTX 40 Series are said to be finalized, with no changes expected. They will be built on one of TSMC’s 5-nanometer processes. That said, kopite7kimi suggests that there’s a chance NVIDIA could still lean on its other next-generation architecture, Hopper, instead of Lovelace for the GeForce RTX 40 Series depending on how competitive AMD’s RDNA 3-based GPUs turn out to be. Some of these models are expected to be multi-chip module (MCM) designs.
Ada Lovelace is definitely the next generation gaming card, as the entire project plan has been finalized and will not change.
— Greymon55 (@greymon55) July 24, 2021
Ada TSMC 5nm 100%, but I'm not sure if it's N5 or N5P yet.
— Greymon55 (@greymon55) July 24, 2021
-5nm(no matter TSMC or SEC)
— kopite7kimi (@kopite7kimi) May 22, 2021
-AD102 in transition or GH202 in revolution?
-RDNA3, GFX11, how does AMD reach its perf goal? Double the SIMD in CU?
-How about Intel?
If monolithic design is enough, we will meet AD102, if not we won't.
— kopite7kimi (@kopite7kimi) May 22, 2021
In terms of specifications, we have only heard about the possible internal GPU structure from Kopite already in December. It was revealed that the flagship gaming GPU (AD102) should feature as many as 144 Streaming Multiprocessors, which suggests a maximum specification of 18,432 CUDA cores.
Sources: Greymon55, kopite7kimi (via VideoCardz)