NVIDIA GeForce RTX 4070 SUPER GPU Confirmed To Feature Full 48 MB L2 Cache On AD104 Chip

NVIDIA has confirmed that the upcoming GeForce RTX 4070 SUPER GPU comes equipped with the full 48 MB L2 cache for AD104 chips.

When NVIDIA announced the GeForce RTX 4070 SUPER graphics card last week, the specifications sheet listed a 20% upgrade to the core count for the AD104 GPU over the RTX 4070 however its L2 cache was listed as the same 36 MB. Now, NVIDIA has confirmed that the 36 MB number was listed by mistake and has updated its product pages with the correct cache amount which is 48 MB, the same as the RTX 4070 Ti GPU.

Having a higher L2 cache is very important for a GPU like the GeForce RTX 4070 SUPER. It comes with a 192-bit memory bus interface and is aimed at high-refresh-rate 1080p and 1440p gaming PCs. Extra details such as higher-res textures and higher resolution itself can lead to bottlenecks within the memory subsystem and while the RTX 4070 SUPER comes equipped with the latest G6X memory, there are still certain scenarios where the memory can become strained and cause slow-downs in gaming. Having a large L2 cache near the GPU provides a high-speed interconnect across all GPCs to send data through and forth.

As such, having 48 MB of L2 cache on the NVIDIA GeForce RTX 4070 SUPER GPU is certainly a very good decision and would mark a nice improvement over the Non-SUPER variant.

The NVIDIA GeForce RTX 4070 SUPER will feature AD104-350 or AD103-175 GPU (PG141 SKU 335) configurations with the same 7,168 cores across both variants. The GPU is expected to get 36 MB of L2 cache. The card will stick with the 12 GB GDDR6X memory across the same 192-bit bus interface. For the RTX 4070 SUPER, the card will have a 20W higher TBP than the existing 4070 at 220W.

The card offers 36 Shader TFLOPs, 82 RT TFLOPs, and 568 AI TOPs while supporting all the latest NV encode technologies such as AV1 and H.264. All of these combined put the RTX 4070 within 95% of the performance of the 4070 Ti.

In terms of gaming performance, the NVIDIA GeForce RTX 4070 SUPER is said to be around 15% faster than the RTX 4070 Non-SUPER, marking the biggest uplifts in the SUPER lineup. The card is also stated to be faster than an RTX 3090 without frame-generation though the performance is going to vary on a game-to-game basis. NVIDIA also confirms that the card is just 5% slower than an RTX 4070 Ti which retailed for $799 US. That too has been replaced by the RTX 4070 Ti SUPER.

AI is a major talking point for the NVIDIA GeForce RTX 4070 SUPER and the RTX 40 SUPER series in general, delivering up to 2x faster performance with DLSS 3.5, 70% faster image generation & 50% faster video generation speeds using ML optimizations through the Tensor-RT and Tensor-RT LLM stack that has been available on Windows and is being updated on the go.

Another talking point for the NVIDIA GeForce RT 40 SUPER GPUs such as the RTX 4070 SUPER is its class-leading performance efficiency. While the GPU is rated at 220W, the actual gaming wattage is around 200W on average which is less than half the TGP of the RTX 4090 and much lower than what the Radeon RX 7900 series consumes. Even the 7800 XT is rated at a much higher 263W TBP. The video playback power is also 16W out of the box while idle power is 11W, something that the AMD RDNA 3 GPUs had trouble fixing during launch.

The NVIDIA GeForce RTX 4070 SUPER GPU will be available on the 17th of January in both Founders Edition and custom variants, starting at $599 US. There are plenty of custom models to select from such as the ones shown below:

NVIDIA GeForce RTX 4070 SUPER Founders Edition Gallery: