The upcoming GeForce RTX 4070 Super graphics card will include the entire 48MB L2 cache for AD104 processors, according to a correction from Nvidia.
The specifications sheet for the GeForce RTX 4070 Super graphics card, which Nvidia unveiled last week, indicated a 20% increase in core count over the RTX 4070. However, the L2 cache remained at 36MB.
Nvidia has now updated its product pages with the correct cache value of 48MB, bringing the Super variant on par with the RTX 4070 Ti. The company confirms that 36MB was an error on its part.
The Ada Lovelace GPU architecture memory subsystem is highly dependent on the size of the L2 cache. This issue is especially important for cards in the RTX 4060 series that have a restricted 128-bit memory bus, as it can limit the number of VRAM pool queries, resulting in poor performance at higher resolutions.
The GeForce RTX 40 series has seen increased L2 cache on all of its consumer GPUs, offering up to 96MB on AD102 (GeForce RTX 4090) and 32MB on AD107/106 (GeForce RTX 4060). The maximum cache size of the AD104, the platform upon which the RTX 4070 series is built, is 48MB.
As a result, the 48MB of L2 cache is unquestionably a wise choice for the upcoming GeForce RTX 4070 Super and a welcome increase over the non-Super model. Still, there is not much information about the kind of performance we should be expecting from the mid-range GeForce RTX 4070 Super.
With the correct L2 cache size, the graphics card should perform identically to the GeForce RTX 4070 Ti, but this remains to be seen. This may also be one of the reasons Nvidia had to discontinue its non-Super GeForce RT 4070 Ti graphics card.
Thank you! Please share your positive feedback. 🔋
How could we improve this post? Please Help us. 😔
[How-To Guides Expert]
Laiba, our tech guru at HI Digital, simplifies and crafts How-To Guides with a Google IT Support Certificate. Beyond tech, she captures the city’s aesthetics through her lens. Join her journey where tech meets creativity!
Get In Touch: Laiba@hidgt.com
