NVIDIA confirms Ada Lovelace GPU architecture, GeForce RTX

NVIDIA has just pumped out yet another teaser for its upcoming “GeForce Beyond” special broadcast next week during GTC 2022, where we should be introduced to the next-gen Ada Lovelace GPU architecture.

The new teaser is pretty brief, with a 12-second video that slides past a GeForce RTX-powered desktop PC system and then zooms down into “Diagram for the computation by the Engine of the Numbers of Bernoulli”. This chart in particular was created by Ada Lovelace back in 1843, made for Charles Babbage’s Analytical Engine which is considered the first “non-trivial” computer program.

NVIDIA’s ANZ (Australia and New Zealand) Twitter account posted the GeForce Beyond teaser, which also included another hidden message in the form of a sticky note attached to the monitor. The sticky note has what seems to be a phone number (208) 629-7538 which some think “7538” refers to 75.38 billion transistors (recent rumors peg the flagship AD102 GPU with 75 billion+ transistors).

The sticky note attached to the monitor reads (208) 629-7538

VIEW GALLERY – 3 IMAGES

But, another way to look at it if NVIDIA were trolling us:

2 + 8 + 6 + 2 + 9 + 7 + 5 + 3 + 8 = 50

Just imagine if all of the retail packaging leaks, the purported shots of the GeForce RTX 4090 and GeForce RTX 4080 graphics cards were nothing but trolls from NVIDIA and leakers. Just imagine, if this GeForce Beyond teaser really was eluding to “50” and the GeForce RTX 5090, RTX 5080 16GB, and RTX 5080 12GB graphics cards… all on PCIe 5.0 x16 and the new PCIe 5.0-ready ATX 3.0-compatible 12VHPWR power connector.

Hmm.

We did hear rumors back in April 2022 that teased the GeForce RTX 50 series GPUs from “harukaze5719” on Twitter. I just thought I’d make the connection there, in case NVIDIA does decide to use the GeForce RTX 50 series nomenclature.

Inside, NVIDIA’s next-gen AD102 GPU is expected to have 18342 CUDA cores at its max — but that’s the full-fat AD102 — we won’t see that at launch. NVIDIA will be using a slightly cut-down AD102 GPU for its GeForce RTX 4090, with that expected to have 16384 CUDA cores, and 24GB of GDDR6X memory.

NVIDIA’s next-gen AD102 GPU will be made on TSMC’s new 4nm process node, with NVIDIA tapping the 4nm process node — which was optimized and designed exclusively for NVIDIA’s monster Hopper H100 GPU — so we should expect the same optimizations and design exclusivity for the GeForce RTX side of things with Ada Lovelace and AD102.

NVIDIA confirms Ada Lovelace GPU architecture, GeForce RTX 50 series?! 09 | TweakTown.com

The ‘Diagram for the computation by the Engine of the Numbers of Bernoulli’ chart made by Ada Lovelace back in 1843

NVIDIA should be announcing its next-gen Ada Lovelace GPU architecture at its upcoming GTC 2022 event, where the company has announced its new “GeForce Beyond” special broadcast, where we’re expecting to see the new GeForce RTX 40 series graphics cards announced (at least one: the new GeForce RTX 4090).