NVIDIA Teases GeForce RTX 40 SUPER GPUs Launch For CES 2024
NVIDIA wants you to be ready for a SUPER CES as it teases the upcoming launch of its highly anticipated GeForce RTX 40 SUPER GPUs. The company has recently updated its social media channels with a picture showcasing the iconic FE Founders Edition cooling solution featuring a white LED and an axial-tech-based cooling system. This teaser arrives just days before NVIDIA’s Special Address at CES 2024, where we can expect to get our first official look at the GeForce RTX 40 SUPER GPUs.
The flagship GPU in the lineup is the NVIDIA GeForce RTX 4080 SUPER. It is expected to utilize the AD103-400 GPU (PG139 SKU 355) with an impressive 10,240 cores, 320 TMUs, 112 ROPs, and 64 MB of L2 cache. Compared to the RTX 4080 Non-SUPER variant, the GPU will have a slightly higher clock speed of 2295 MHz base and 2550 MHz boost.
The memory configuration for the RTX 4080 SUPER remains the same as the Non-SUPER variant, with a 256-bit bus interface and 16 GB of GDDR6X memory. However, the memory solution will receive a boost to 23 Gbps, resulting in a bandwidth of up to 736 GB/s, a 3% improvement over its predecessor. The TGP (total graphics power) of the RTX 4080 SUPER is also unchanged at 320W.
In terms of performance, the RTX 4080 SUPER is anticipated to offer a modest 3-5% increase compared to the RTX 4080 Non-SUPER. However, it’s worth mentioning that custom models are expected to provide an even more substantial performance boost of around 10%. While NVIDIA has yet to disclose pricing information, the slight performance gain should position the card favorably against the Radeon RX 7900 XTX, especially if it is priced around the $1000-$1100 range.
Moving down the lineup, we have the NVIDIA GeForce RTX 4070 Ti SUPER. This GPU will be available in two variants: AD103-275 and AD102-175 (PG141 SKU 323), offering a total of 8,448 cores and 48 MB of L2 cache. Compared to the Non-SUPER variant, the RTX 4070 Ti SUPER will feature 16 GB of GDDR6 memory, a 4 GB increase. The GPU will also support a 256-bit bus interface and retain a TGP of 285W.
Performance-wise, the RTX 4070 Ti SUPER is expected to deliver an average increase of 15% over the Non-SUPER variant. This should position it strongly against the Radeon RX 7900 XT, potentially at a lower price point.
Lastly, we have the NVIDIA GeForce RTX 4070 SUPER, which will be available in AD104-350 or AD103-175 SKU (PG141 SKU 335) configurations. Both variants will feature 7,168 cores and 48 MB of L2 cache. The GPU is expected to maintain the 12 GB GDDR6X memory across a 192-bit bus interface. Notably, the RTX 4070 SUPER will have a 20W higher TBP (total board power) than the existing RTX 4070 at 220W.
In terms of performance, users can expect a 14% improvement over the RTX 4070 Non-SUPER. The RTX 4070 SUPER will also introduce the new CEM 5.1 Gen5 power connector, while existing RTX 4070 graphics cards come with either an 8-pin or 12-pin power connector.
It is important to note that plans for a GeForce RTX 4090 SUPER were abandoned due to recent sanctions limiting the release of GPUs exceeding 4800 TOPs in China and other regions. As of now, there are no plans for an RTX 4090 SUPER or Ti GPU, but this could change if restrictions ease in the coming year.
The NVIDIA GeForce RTX 40 SUPER GPU family is set to be unveiled at CES 2024, with the release dates confirmed for the 17th, 24th, and 31st of January for the 4070 Ti SUPER, 4070 Ti SUPER, and 4080 SUPER, respectively. Additionally, NVIDIA will also launch the RTX 4090 D exclusively in China and the RTX 3050 6 GB GPU in January. Expect a wide range of custom models for all these cards when they hit store shelves.
Overall, NVIDIA’s upcoming GeForce RTX 40 SUPER GPUs promise improved performance and enhanced features, aiming to provide gamers and enthusiasts with an exceptional gaming experience. CES 2024 will undoubtedly be an exciting event for technology enthusiasts, eagerly awaiting the official unveiling of these powerful graphics cards.
