The company has scheduled its special address at CES 2024 for January 8th, starting at 8 AM PDT. While there’s no explicit mention of the GeForce RTX 40 “SUPER” family of GPUs, it’s highly anticipated that these graphics cards will be a significant focus, if not the sole focus, of the event, as they are poised to make a big splash in 2024. This aligns with a recent tweet from Kopite7kimi, suggesting that NVIDIA GeForce RTX 40 “SUPER” GPUs are on track for a CES unveiling.
What fans currently know is that NVIDIA is preparing to launch at least three graphics cards as part of its GeForce RTX 40 “SUPER” Refresh lineup. These cards are expected to include the GeForce RTX 4080 SUPER, RTX 4070 Ti SUPER, and the RTX 4070 Super. It’s important to note that these naming conventions, as well as specifications and prices, are not finalized.
NVIDIA typically evaluates multiple SKUs before the official launch to prevent leaks of actual “launch” hardware, so the rumors circulating may or may not align with the final products. However, it is true that many of the rumored SKUs exist in some form and tend to surface after the official “retail” SKUs are released.
NVIDIA GeForce RTX 4080 SUPER Specification
Starting with the top-tier model, the NVIDIA GeForce RTX 4080 SUPER is rumored to feature the AD103-400 GPU with a full 10,240 cores, 320 TMUs, 112 ROPs, and 64 MB of L2 cache. The card is expected to utilize a 256-bit bus interface, suggesting a 16 GB GDDR6X memory configuration, similar to the RTX 4080 Non-SUPER. The NVIDIA GeForce RTX 4080 Non-SUPER has a TBP (Total Board Power) of 320W, so similar figures are anticipated for the 4080 SUPER.
NVIDIA GeForce RTX 4070 Ti SUPER Specifications
Moving on to the NVIDIA GeForce RTX 4070 Ti SUPER, the specifications indicate the use of AD103-275 or AD102-175 GPU SKUs with the same 8,448 cores across both variants and 48 MB of L2 cache. The GeForce RTX 4070 Ti SUPER is projected to have a TBP close to the existing 4070 Ti, around 285W.
NVIDIA GeForce RTX 4070 SUPER Specifications
Lastly, there’s the NVIDIA GeForce RTX 4070 SUPER, which is expected to come in AD104-350 or AD103-175 SKU configurations, featuring the same 7,168 cores in both variants. The GPU is set to have 48 MB of L2 cache, and it appears that this card will stick with 12 GB of GDDR6X memory. NVIDIA is likely to enhance memory speeds using better Micron DRAM, reducing the pressure on the 192-bit bus interface. The RTX 4070 SUPER is anticipated to have a 20W higher TBP compared to the existing 4070, with a TBP of 220W.
