The NVIDIA RTX 4070 SUPER AD104 GPU features 48MB of L2 cache, not 36MB as previously claimed

The NVIDIA RTX 4070 SUPER has more L2 cache after all

In the official launch presentation, the company made a mistake.

There will be not 36MB but 48MB of L2 cache available on the RTX 4070 SUPER. NVIDIA made a mistake in the official specifications, which were shared as part of the official launch presentation. The error has not been corrected until now.

Four new cards were announced at CES 2024 for desktop gamers. AMD introduced its Radeon RX 7600 XT GPU for $329 for the mid-range segment, while NVIDIA is focusing on at least twice as expensive models, launching the RTX 4070 SUPER at $599. This model replaces the current RTX 4070, which is now discounted to $549.

The official specs listed in the review guide show 36MB of L2 cache size for the RTX 4070 SUPER model. This information was also present in the official marketing slides. However, NVIDIA has now updated the specs to ensure that this model actually uses 48MB of L2 cache. No other change was reported.

NVIDIA RTX 4070 SUPER specifications, source: NVIDIA

The L2 cache size plays a crucial role in the memory subsystem of the Ada Lovelace GPU architecture. This factor can limit VRAM pool queries, which is especially important for cards with a restricted 128-bit memory bus on the RTX 4060 series. For the RTX 40 series, NVIDIA has updated L2 cache sizes across all of its consumer GPUs, making Provides up to 96MB of L2 cache on the AD102 and 32MB on the AD107/106.

See also  SpaceX's Starship Is Ready To Fly, Just Waiting For Launch License - Ars Technica

The AD104, which the RTX 4070 series is based on, features 48MB of total L2 cache. Since the RTX 4070 SUPER uses an AD104 GPU with more SMS enabled (56 out of 60), it doesn't make sense for NVIDIA to limit that to 36MB, which is also the original RTX 4070 spec.

With that in mind, the RTX 4070 SUPER with 20% more cores and 33% more L2 cache will undoubtedly show a noticeable improvement in gaming. However, reference-spec models may be limited to the default TDP of 220W, which is only 20W more than the RTX 4070.

The RTX 4070 SUPER is scheduled to hit the market on January 17, with early reviews arriving a day earlier.

Nvidia GeForce RTX 40
VideoCardz.com short message Coda cores Display the memory bus L2 cache
M102
M103
M104
M106
M107

source: Nvidia via Wccftech



Leave a Reply

Your email address will not be published. Required fields are marked *