The first alleged cache and memory benchmark of the Ryzen 9 7900X has surfaced, courtesy of @9550pro on Twitter. The Zen 4 AIDA64 L3 cache benchmark shows a remarkable improvement over its predecessor as well as the 12th Gen Intel Alder Lake family. The 12-core chip recorded a memory read and write bandwidth of 1,494.8 GB/s and 1,445.7 GB/s, respectively. Meanwhile, the memory copy speed peaks at 1,476.6 GB/s with the latency coming up to 10,1ns.
— HXL (@9550pro) August 4, 2022
The Ryzen 9 7900X offers an impressive cache bandwidth gain of 50% over its predecessor while beating the Core i9-12900K by roughly the same figures. The latter records an L3 cache latency of 21.8ns, twice as more as the 10.1ns delivered by the latter. The L3 copy speed is even higher on Zen 4, beating Alder Lake by 3x.
AMD will be leaving the L3 cache size on its Zen 4 processors unchanged with single-die chips featuring 32MB and double-die SKUs 64MB of last level cache. However, the L2 cache will be doubled to 1MB per core, up from 512KB on Zen 3. That totals to a total of 80MB of L3 cache on the Ryzen 9 7950X, 12MB more than the Core i9-13900K.
