NVIDIA Ada Lovelace GPUs-based GeForce RTX 40 collection graphics playing cards will probably be one of many main launches for 2022. The brand new lineup is anticipated to convey greater than 2x the efficiency improve over Ampere however may even be one of the power-hungry lineups ever launched.
NVIDIA Ada Lovelace Powered GeForce RTX 40 GPUs To Be Twice As Quick & Twice As Energy Hungry Than Ampere
The rumor as soon as once more comes from Greymon55 who has been actively sharing particulars concerning AMD and NVIDIA next-gen GPU launches. In his new tweet, the leaker claims that Ada Lovelace, NVIDIA’s subsequent structure for its GeForce RTX graphics playing cards will function double the efficiency but in addition double the ability consumption.
Double efficiency, double energy consumption, are you able to settle for it?
— Greymon55 (@greymon55) November 5, 2022
Based mostly on earlier experiences, that is trying to be the case as each AMD and NVIDIA next-gen GPUs are going to finish up providing over 2x the efficiency improve however may even improve energy enter dramatically.
The NVIDIA Ada Lovelace GPUs will energy the next-generation GeForce RTX 40 graphics playing cards that can go head-on with AMD’s RDNA 3 primarily based Radeon RX 7000 collection graphics playing cards. There’s nonetheless some hypothesis concerning using MCM by NVIDIA. The Hopper GPU, which is primarily aimed on the Datacenter & AI phase, is allegedly taping out quickly and can function an MCM structure. NVIDIA will not be utilizing an MCM design on its Ada Lovelace GPUs so they’ll maintain the normal monolithic design. The Ada Lovelace GPUs are anticipated to make the most of TSMC’s 5nm course of node and can herald a collection of key improvements, architecturally.
Beforehand rumored specs have proven us an enormous replace to the core specs. The NVIDIA AD102 “ADA GPU” seems to have 18432 CUDA Cores primarily based on the preliminary specs (which might change) offered by Kopite. That is virtually twice the cores current in Ampere which was already a large step up from Turing. A 2.2 GHz clock pace would give us 81 TFLOPs of compute efficiency (FP32). That is greater than twice the efficiency of the prevailing RTX 3090 which packs 36 TFLOPs of FP32 compute energy.
On the similar time, these enormous efficiency figures would come at the price of greater general energy consumption. NVIDIA is already anticipated to ship its RTX 3090 Ti (reportedly the primary PCIe Gen 5.0 graphics card) with a TGP of 450W. Ada Lovelace-based GeForce RTX 40 collection is anticipated to finish up with a TDP effectively over 500W and even round 600W. That is why a brand new connector sort that may assist such high-power figures is being designed for the subsequent era of graphics playing cards.
Kopite7kimi additionally hinted at some specification particulars of the NVIDIA Ada Lovelace chips some time again which you’ll learn extra about right here and take a look at the specs within the desk offered beneath:
NVIDIA CUDA GPU (RUMORED) Preliminary:
| GPU | TU102 | GA102 | AD102 |
|---|---|---|---|
| Structure | Turing | Ampere | Ada Lovelace |
| Course of | TSMC 12nm NFF | Samsung 8nm | 5nm |
| 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 |
| Theoretical TFLOPs | 16.1 | 37.6 | ~80 TFLOPs? |
| Flagship SKU | RTX 2080 Ti | RTX 3090 | RTX 4090? |
| TGP | 250W | 350W | 450-600W? |
| Launch | Sep. 2018 | Sept. 20 | 2022 (TBC) |
The NVIDIA Ada Lovelace GPU household is anticipated to convey a generational bounce just like Maxwell to Pascal. It’s anticipated to launch late in This autumn 2022 however primarily based on GPU provide and availability, it is prone to transfer again to 2023.
