In a surprising turn of events, NVIDIA has corrected an error in its official launch presentation for the highly anticipated RTX 4070 SUPER graphics card. Rather than the initial specifications of 36 MB of L2 cache, the RTX 4070 SUPER will feature an impressive 48 MB of L2 cache. This correction comes as a relief to gamers eagerly awaiting the release of the GPU.
At the CES 2024 event, both AMD and NVIDIA unveiled new graphics cards targeted at the gaming market. AMD presented its mid-range Radeon RX 7600 XT GPU at a price of $329, while NVIDIA set its sights on offering high-performance models, including the RTX 4070 SUPER, priced at $599. This new model will replace the existing RTX 4070, which has seen a price drop to $549.
The increased L2 cache size is a significant enhancement for the RTX 4070 SUPER. L2 cache plays a vital role in the memory subsystem of GPUs, particularly for those with a limited memory bus. The new Ada Lovelace GPU architecture memory subsystem boasts up to 96 MB of L2 cache on the AD102 and 32 MB on the AD107/106. Since the RTX 4070 SUPER is based on the AD104 GPU, which already features 48 MB of total L2 cache, it is only logical for NVIDIA to use the full capacity rather than limiting it to 36 MB.
The additional cores and increased L2 cache size on the RTX 4070 SUPER are expected to deliver a noticeable improvement in gaming performance. However, it is worth noting that the default TDP of 220W may impose some limitations, as it is only 20W higher than the RTX 4070. Gamers eagerly anticipating the arrival of the RTX 4070 SUPER can mark their calendars for January 17, when it is set to hit the market with early reviews becoming available a day earlier.
With these upgrades, the RTX 4070 SUPER aims to offer an immersive gaming experience and solidify NVIDIA’s position as a leading provider of high-performance graphics cards.
The source of the article is from the blog japan-pc.jp