Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
throwaway2037
3 months ago
|
parent
|
context
|
favorite
| on:
Do not taunt happy fun branch predictor (2023)
Real question about L1 caches. For a long time, x86 (Intel & AMD) L1 caches have been pretty much pegged at 32KB. Do you know why they didn't make them larger? My guess: There is a trade-off between economics and performance.
account42
3 months ago
[–]
There is a trade-off between cache size and latency.
throwaway2037
3 months ago
|
parent
[–]
Ok, so why do the new Mx chips from Apple have an L1 cache size greater than 32KB? Did they solve some long standing design issue?
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: