NVIDIA GeForce RTX 4070 SUPER GPU AD104 chip confirmed to have full 48MB L2 cache

NVIDIA has confirmed that its upcoming GeForce RTX 4070 SUPER GPU is equipped with a full 48MB of L2 cache for its AD104 chips.

NVIDIA’s GeForce RTX 4070 SUPER packs 48MB of L2 cache on its AD104 GPU, the same as the RTX 4070 Ti.

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

For a GPU like the GeForce RTX 4070 SUPER, having a higher L2 cache is very important. It features a 192-bit memory bus interface and is designed for 1080p and 1440p gaming PCs with high refresh rates. Additional details such as higher quality textures and higher resolutions can cause bottlenecks in the memory subsystem, and even though the RTX 4070 SUPER is equipped with the latest G6X memory, there are still some scenarios where the memory can become strained and cause slowdowns. in the game. Having a large L2 cache close to the GPU ensures high-speed communication between all GPCs.

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

NVIDIA GeForce RTX 4070 SUPER GPU Specifications: Upgraded AD104 GPU With 20% More Cores

NVIDIA GeForce RTX 4070 SUPER AD104-350 or AD103-175 GPU (PG141 SKU 335) configurations with the same 7168 cores in both variants. The GPU is expected to have 36MB of L2 cache. The card will stick to 12GB of GDDR6X memory over the same 192-bit bus interface. The card for the RTX 4070 SUPER will have a 20W higher TBP than the current 4070 at 220W.

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

NVIDIA GeForce RTX 4070 SUPER GPU Performance: Faster than RTX 3090 and just 5% slower than 4070 Ti on average

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 improvements in the SUPER series. The card is also said to be faster than the frameless RTX 3090, although performance will vary on a game-by-game basis. NVIDIA also confirms that the card is only 5% slower than the $799 RTX 4070 Ti. This was replaced by the RTX 4070 Ti SUPER.

Artificial intelligence is the main talking point for the NVIDIA GeForce RTX 4070 SUPER and the RTX 40 SUPER series in general, using ML optimizations via Tensor to provide up to 2x faster performance with DLSS 3.5, 70% faster rendering and 50% faster video rendering speed . -RT and Tensor-RT LLM stack available on Windows and updated on the go.

NVIDIA GeForce RTX 4070 SUPER GPU Power Efficiency: Absolutely Smashing Perf/Watt

Another talking point for NVIDIA GeForce RT 40 SUPER GPUs like the RTX 4070 SUPER is its class-leading performance efficiency. Although the GPU is rated at 220W, the actual gaming power is around 200W on average, which is less than half of the RTX 4090’s TGP and much lower than what the Radeon RX 7900 series consumes. Even the 7800 XT is rated higher at 263W TBP. Video playback power is also 16W out of the box and 11W at idle, something AMD RDNA 3 GPUs struggle to make up for at launch.

NVIDIA GeForce RTX 4070 SUPER GPU Price and Availability

The NVIDIA GeForce RTX 4070 SUPER GPU will be available on January 17th in both Founders Edition and custom variants starting at $599. There are many custom models to choose from as shown below:

NVIDIA GeForce RTX 4070 SUPER Founders Edition Gallery:

NVIDIA GeForce RTX 40 Desktop GPU Family: