RTX 4080 SUPER may be equipped with AD102
A few days ago, reports surfaced indicating NVIDIA’s preparation to unveil three GeForce RTX 40 SUPER models: the RTX 4080 SUPER, RTX 4070 Ti SUPER, and RTX 4070 SUPER. It’s highly probable that these will make their debut in 2024, and we might anticipate fresh revelations at the CES 2024 exposition early next year. While NVIDIA forwent the “SUPER” moniker in favor of “Ti” for its preceding Ampere-based GeForce RTX 30 series, a paradigm shift seems evident with the Ada Lovelace-architected GeForce RTX 40 series.
As per Benchlife, the Nvidia GeForce RTX 4080 SUPER will harness the AD102, furnished with a 20GB memory cache. Consequently, its memory bus width is expected to expand to 320-bits. This implies that the forthcoming card will align with AMD’s Radeon RX 7900 XT in terms of memory configuration. This semblance suggests a harmonized memory capacity portfolio for top-tier graphics products, ranging between 16GB, 20GB, and 24GB. Given the augmented memory allocation, the GeForce RTX 4080 SUPER might also experience a surge in CUDA core count, potentially propelling its overall prowess.
Further rumors posit that the GeForce RTX 4070 SUPER will be embedded with the AD103, elevating its memory reserves to 16GB and enhancing its memory bandwidth to 256-bits. It’s conjectured to be endowed with an augmented CUDA core array. NVIDIA is also poised to launch the new GeForce RTX 4070, outfitted with GDDR6 memory, supplanting the erstwhile GDDR6X. While other specifications ostensibly remain unaltered, a slight decrement in performance can be anticipated. However, this might be counterbalanced with a more cost-effective production, paving the way for a more attractive price point.
