NVIDIA Ada Lovelace GPUs To Have A Node Advantage Over AMD RDNA 3, Rumored To Utilize TSMC 4N Process

NVIDIA’s Ada Lovelace GPUs for next-gen GeForce RTX 40 Gaming graphics playing cards can have a node benefit over AMD’s RDNA 3, as reported by Moore’s Law is Dead.

NVIDIA Ada Lovelace Gaming GPUs Rumored To Utilize TSMC’s 4N Process Node, Giving A Slight Advantage Over AMD’s RDNA 3 GPUs

From what we all know thus far, NVIDIA was expected to make the most of TSMC’s 5nm course of node for his or her Ada Lovelace GPUs that energy the next-generation gaming graphics playing cards aka GeForce RTX 40 sequence. It seems to be like the precise node has been revealed by Moore’s Law is Dead in a latest tweet. As per the most recent rumor, the NVIDIA Ada Lovelace GPUs will likely be based mostly on the TSMC 4N course of node.

AMD Ryzen 7000 CPUs & AM5 Platform Will Only Support DDR5 Memory, Comes With EXPO ‘Memory Profile’ Technology

Yep, that is the identical TSMC 4N course of node that powers the Hopper GPUs for the information heart HPC market. As for what we all know in regards to the TSMC 4N course of node, it’s a revision of their 5nm course of (to not be confused with 4nm/N4 which is a very totally different node). The TSMC 4N course of node is custom-designed completely for NVIDIA and hosts a spread of optimizations that permits for higher energy effectivity, efficiency, and a minor increase to density versus the vanilla TSMC 5nm node.

The explanation why NVIDIA could have chosen TSMC’s 4N because the candidate for its next-gen gaming GPU lineup are sort of apparent. The upcoming playing cards will be real power-hungry and NVIDIA & the corporate goes to optimize them as a lot as they will by using the 4N course of node. AMD alternatively will likely be using a mixture of TSMC 5nm and 6nm process nodes for its upcoming MCM and monolithic GPUs based mostly on the RDNA 3 graphics architecture and whereas they do not deliver the optimizations that 4N does, they are going to function an MCM method that’s anticipated to be extremely environment friendly.

So on the finish of the day, NVIDIA will get the higher node whereas AMD delivers a greater design method. At the top of the day, these will not matter a lot to end-users who solely need to play their video games on the very best {hardware} (graphics playing cards) they might get their arms on.

NVIDIA CUDA GPU (RUMORED) Preliminary:

GPU TU102 GA102 AD102
Flagship SKU RTX 2080 Ti RTX 3090 Ti RTX 4090?
Architecture Turing Ampere Ada Lovelace
Process TSMC 12nm NFF Samsung 8nm TSMC 4N?
Die Size 754mm2 628mm2 ~600mm2
Graphics Processing Clusters (GPC) 6 7 12
Texture Processing Clusters (TPC) 36 42 72
Streaming Multiprocessors (SM) 72 84 144
CUDA Cores 4608 10752 18432
L2 Cache 6 MB 6 MB 96 MB
Theoretical TFLOPs 16 TFLOPs 40 TFLOPs ~90 TFLOPs?
Memory Type GDDR6 GDDR6X GDDR6X
Memory Capacity 11 GB (2080 Ti) 24 GB (3090 Ti) 24 GB (4090?)
Memory Speed 14 Gbps 21 Gbps 24 Gbps?
Memory Bandwidth 616 GB/s 1.008 GB/s 1152 GB/s?
Memory Bus 384-bit 384-bit 384-bit
PCIe Interface PCIe Gen 3.0 PCIe Gen 4.0 PCIe Gen 4.0
TGP 250W 350W 600W?
Release Sep. 2018 Sept. 20 2H 2022 (TBC)



Source link