NVIDIA RTX 40 SUPER GPUs to Debut with Enhanced Performance, January 2024 Launch

NVIDIA's GeForce RTX 40 SUPER GPU specifications have once again leaked, revealing that the RTX 4080 SUPER will use the fastest GDDR6X memory.

NVIDIA GeForce RTX 40 SUPER GPU Specs Have Leaked Yet Again

NVIDIA GeForce RTX 40 SUPER GPU specifications were initially revealed last month, covering nearly all of the data, and now Videocardz has a new addition to the specifications, pointing out that the RTX 4080 SUPER will use an insanely fast memory solution with pin speeds of 23 Gbps.

This confirmation is backed up by a retail listing of the RTX 4080 SUPER, which indicates a graphics card with a 23 Gbps effective transfer rate. The following are the detailed specifications for each of the three SUPER versions.

NVIDIA GeForce RTX 4080 SUPER Performance

The NVIDIA GeForce RTX 4080 SUPER is rumored to employ the AD103-400 GPU (PG139 SKU 355) with the full 10,240 cores, 320 TMUs, 112 ROPs, and 64 MB of L2 cache. The GPU will also be clocked slightly higher, at a 2295 MHz base and 2550 MHz boost, compared to the RTX 4080 Non-SUPER variant's 2205 MHz base and 2505 MHz boost.

The NVIDIA GeForce RTX 4080 SUPER is said to use an AD103-400 GPU (PG139 SKU 355) with 10,240 cores, 320 TMUs, 112 ROPs, and 64 MB of L2 cache. The GPU will also be clocked slightly higher, at a 2295 MHz base and 2550 MHz boost, compared to a 2205 MHz base and 2505 MHz boost on the RTX 4080 Non-SUPER model.

NVIDIA GeForce RTX 4070 Ti SUPER Specifications

The NVIDIA GeForce RTX 4070 Ti SUPER follows, with a choice of AD103-275 / AD102-175 GPU SKUs (PG141 SKU 323), 8,448 cores across both variants, and 48 MB of L2 cache. The graphics card will also include 16 GB of GDDR6 memory, representing a 4 GB VRAM improvement over the RTX 4070 Ti Non-SUPER, as well as a 256-bit bus interface. The graphics card will keep the same 285-watt TGP as the non-SUPER model.

In terms of performance, the NVIDIA GeForce RTX 4070 Ti SUPER 16 GB GPU will provide an average 15% performance jump over the Non-SUPER variant and should fight well against the Radeon RX 7900 XT at a potentially cheaper price point.

NVIDIA GeForce RTX 4070 SUPER Specifications

Finally, the NVIDIA GeForce RTX 4070 SUPER is stated to be available in AD104-350 or AD103-175 SKU (PG141 SKU 335) configurations, with the same 7,168 cores in both. An L2 cache of 48 MB is expected for the GPU. This card appears to retain the 12 GB GDDR6X memory across the same 192-bit bus interface. The RTX 4070 SUPER will have a TBP of 220W, which is 20W greater than the current 4070.

Finally, the NVIDIA GeForce RTX 4070 SUPER will be offered in either AD104-350 or AD103-175 SKU (PG141 SKU 335) configurations, with the same 7,168 cores in each. The GPU is expected to have an L2 cache of 48 MB. The 12 GB GDDR6X RAM appears to be retained on the same 192-bit bus interface on this card. The TBP of the RTX 4070 SUPER will be 220W, which is 20W higher than the existing 4070.

What We Know So Far About the NVIDIA GeForce RTX 4090 SUPER

WCCFTech has also learned that there were plans for a GeForce RTX 4090 SUPER, but they were scrapped because of current sanctions that limit the production of GPUs with above 4800 TOPs in China and other areas. There are currently no plans for an RTX 4090 SUPER or Ti GPU, but that could change if limitations relax in the following year, which appears unlikely at the moment.

The NVIDIA GeForce RTX 40 SUPER GPU family will be introduced again at CES 2024, with a launch date of January 17th for the 4070 Ti SUPER, January 24th for the 4070 Ti SUPER, and January 31st for the 4080 SUPER. In addition to these three cards, NVIDIA will release the RTX 4090 D and the RTX 3050 6 GB GPU in China in January. When these cards hit the shelves, expect a slew of personalized models.

Photo: Christian Wiediger/Unsplash