NVIDIA To Launch GeForce RTX 30 SUPER ‘Ampere Refresh’ In January 2022, GeForce RTX 40 ‘Ada Lovelace’ GPUs in October 2022

Hassan Mujtaba
NVIDIA Allegedly Launching RTX 3090 SUPER, RTX 3070 Ti 16GB & RTX 2060 12GB By January 2022

A brand new rumor regarding NVIDIA's next-gen GeForce RTX 30 SUPER 'Ampere Refresh' and GeForce RTX 40 'Ada Lovelace' graphics cards has been posted by leakers, Greymon55 and Red Gaming Tech. According to the rumor, NVIDIA plans to launch two brand new GeForce graphics card lineups in 2022.

NVIDIA GeForce RTX 30 SUPER 'Ampere Refresh' & GeForce RTX 40 'Ada Lovelace' Launching In 2022, Alleges Rumor

Earlier rumors had already stated an early 2022 launch for NVIDIA's GeForce RTX 30 SUPER Ampere Refresh and a late 2022 launch for the next-gen GeForce RTX 40 'Ada Lovelace' graphics cards. Now both leakers have stated that the GeForce RTX 30 SUPER graphics card lineup is indeed headed for launch early next year while the Ada Lovelace GPUs which will be featured on the GeForce RTX 40 series will be hitting store shelves in October 2022.

Related StoryHassan Mujtaba
NVIDIA Hopper & Ampere AI GPUs Continue To Post World Records In AI Training Benchmarks

NVIDIA GeForce RTX 30 SUPER 'Ampere Refresh' Graphics Family

The NVIDIA GeForce RTX 30 SUPER Refresh will be part of 'Ampere Phase 2' which our sources state has gone into production this quarter. The GPUs will be different than the Ampere SKUs we got this and last year and will focus on design optimizations along with retaining their hardware-level block on crypto mining algorithms. Aside from that, NVIDIA is expected to utilize attach the legacy node GPUs (Samsung 8nm) to wider bus interfaces with higher VRAM capacity to drive up performance.

The lineup is expected to introduce the GA103 GPU and from the looks of things, it will be headed to both desktops and laptops. The GeForce RTX 30 SUPER family could be announced at CES 2022 and featured on several new laptops based on Intel's 12th Gen Alder Lake and AMD's Rembrandt CPU-based gaming notebooks/laptops. However, the GeForce RTX 30 SUPER lineup will be short-lived as NVIDIA is rumored to launch its next-gen graphics lineup within the same year (at least 8-9 month difference).

NVIDIA GeForce RTX 40 Next-Gen 'Ada Lovelace' Graphics Family

According to Greymon55 again, NVIDIA is expected to launch the GeForce RTX 40 series graphics card lineup based on the Ada Lovelace GPU architecture in October 2022. This would mark a year from now and at least 9 months from the GeForce RTX 30 SUPER lineup. Do remember that NVIDIA launched its GeForce RTX 30 series family almost a year after the GeForce RTX 20 SUPER graphics card refresh. The first products will be aiming at the discrete graphics card segment and hence, it is making perfect sense for the company to introduce an intermediate Ampere refresh early next year and then introduce new mobility GPUs at CES 2023.

The NVIDIA GeForce RTX 40 series 'Ada Lovelace' GPUs are rumored to hit mass production in mid of 2022 so that gives at least 3-4 months' time to build up enough stock for a proper retail launch.

The NVIDIA Ada Lovelace GPUs will power the next-generation GeForce RTX 40 graphics cards that will go head-on with AMD's RDNA 3 based Radeon RX 7000 series graphics cards. There's still some speculation regarding the use of MCM by NVIDIA. The Hopper GPU, which is primarily aimed at the Datacenter & AI segment, is allegedly taping out soon and will feature an MCM architecture. NVIDIA won't be using an MCM design on its Ada Lovelace GPUs so they will keep the traditional monolithic design.

AMD on the other hand is going on board with offering both MCM and Monolithic chips in its RDNA 3 and CDNA 2 families. The CDNA 2 GPUs will be MCM only while RDNA 3 will feature a mix of MCM and monolithic design as detailed here. It isn't confirmed which process node each company will be relying on but based on speculation, AMD's RDNA 3 & CDNA 2 family is going to be a mix of 6nm & 5nm nodes featuring new and refreshed GPUs while NVIDIA is expected to utilize TSMC's 5nm process node for its Ada Lovelace GPUs though it is not confirmed if those will be fabricated on the N5 or N5P node. Intel, on the other hand, may also rely on TSMC's 6nm process node for its own ARC Alchemist lineup of GPUs which enter mass production later this year and launch in Q1 2022.

NVIDIA / AMD / Intel GPU Roadmap (Rumor)

201920202021202220232024
AMD (RDNA 1)AMD (RDNA 2)N/AAMD (RDNA 3)N/AAMD (RDNA 4)?
NVIDIA (Turing Refresh)NVIDIA AmpereN/ANVIDIA Ampere Refresh
NVIDIA Ada Lovelace
N/ANVIDIA Hopper
N/AXe-LPN/AXe-HPG (ARC Alchemist)TBCTBC

NVIDIA CUDA GPU (RUMORED) Preliminary:

GPUTU102GA102AD102
Flagship SKURTX 2080 TiRTX 3090 TiRTX 4090?
ArchitectureTuringAmpereAda Lovelace
ProcessTSMC 12nm NFFSamsung 8nmTSMC 4N?
Die Size754mm2628mm2~600mm2
Graphics Processing Clusters (GPC)6712
Texture Processing Clusters (TPC)364272
Streaming Multiprocessors (SM)7284144
CUDA Cores46081075218432
L2 Cache6 MB6 MB96 MB
Theoretical TFLOPs 16 TFLOPs40 TFLOPs~90 TFLOPs?
Memory TypeGDDR6GDDR6XGDDR6X
Memory Capacity11 GB (2080 Ti)24 GB (3090 Ti)24 GB (4090?)
Memory Speed14 Gbps21 Gbps24 Gbps?
Memory Bandwidth616 GB/s1.008 GB/s1152 GB/s?
Memory Bus384-bit384-bit384-bit
PCIe InterfacePCIe Gen 3.0PCIe Gen 4.0PCIe Gen 4.0
TGP250W350W600W?
ReleaseSep. 2018Sept. 202H 2022 (TBC)

Which next-generation GPUs are you looking forward to the most?
Vote to see results

Deal of the Day

Comments