RTX 4000 graphics cards will have plenty of L2 cache

According to leaked information, Nvidia’s RTX 4000 series GPUs codenamed Ada are set to make significant improvements over the RTX 3000 in terms of L2 cache.

In the last few days, a lot of information has been leaked about Nvidia’s current and unannounced products, and as the Green team tries to figure out how the revelators got access to this amount of information, more and more information is gradually coming to the web. In the latest case, remarkable details about Nvidia’s next graphic architecture codenamed Ada have been revealed.

According to VideoCards, Nvidia’s RTX 4000 series GPUs are expected to have a maximum of 96 MB of second-level cache (L2 Cache); That’s 90 MB more than the RTX 3000 series GPUs based on the Ampere architecture.

GPU block to approve unofficially presented Ada series products. This design is related to the AD102 processor. The AD102 is said to use a 600 mm square die, which is about the same size as the GA102.

RTX 4000 series GPUs are likely to have 71% more cores than the RTX 3000 series; Assuming the green team uses the full version of each chip on the destination graphics card, of course. Many important information about Nvidia’s expected processors, including clock speed and power, is not yet available; However, it is already clear that the RTX 4000 is set to make significant progress.

We will probably see a similar leap for new AMD graphics cards based on the RDNA 3 architecture.