NVIDIA may skip Ada Lovelace in favor of Hopper-based RTX 40 series

Posted on Monday, Jul 26 2021 @ 19:39 CEST by Thomas De Maesschalck
NVDA logo
The last couple of days have been very fruitful in terms of rumors about both AMD's and NVIDIA's next GPU generation. VideoCardz offers an overview of some of the latest whispers about NVIDIA's GeForce RTX 40 series.

Ada Lovelace or Hopper?

First up, there seems to be consensus that NVIDIA targets Q4 2022 as the launch date of its next-generation GPUs.

Twitter leaker Greymon55 claims the NVIDIA Lovelace GPU design is ready and that NVIDIA will use TSMC's 5nm process for this chip. However, he's unsure about whether it will be N5 or N5P. Either way, this means NVIDIA is going back to TSMC for its next generation. The current Ampere-based video cards are manufactured by Samsung. Historically, NVIDIA has typically used TSMC for its high-end products.

Famous NVIDIA leaker kopite7kimi, who correctly leaked the specifications of the entire Ampere GPU lineup months in advance, is no longer certain that NVIDIA will use its Lovelace design. Should AMD's multi-chip module (MCM) based RDNA3 series perform very high, NVIDIA could opt to use its Hopper design. At the moment, it's believed that the latter is only for the datacenter market.

Tech YouTuber Moore's Law Is Dead on the other hand points out that Lovelace may also be used in NVIDIA SoCs. He claims that even if RDNA3 beats Lovelace in terms of performance, NVIDIA will outship AMD by just having a lot more volume and cheap-to-make chips.

Basically, the rumors about NVIDIA's next-generation GPUs are still very confusing.

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