It’s an L2, amd zen has 2x16mb L3s, with a wacky point of unification. L3 is muu... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		aey on Dec 1, 2020 \| parent \| context \| favorite \| on: Why is Apple's M1 chip so fast? It’s an L2, amd zen has 2x16mb L3s, with a wacky point of unification. L3 is muuuuch slower than L2. Apple’s L2 is huge and is the PoU for all the cores. I wouldn’t be surprised if it has direct routing to all the L1s since they are so small. 5nm doesn’t hurt.

kllrnohj on Dec 1, 2020 [–]

> amd zen has 2x16mb L3s

Zen 3 is not, it's a single 32MB L3 per chiplet.

> L3 is muuuuch slower than L2.

"L2" and "L3" are just how many layers away from the CPU it is, it's not an intrinsic "type" of cache. The speed is a function of the size.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact