Re:Ampere’s Product List: 80 Cores, up to 3.3 GHz at 250 W; ...
by umanwizard on Tuesday June 30, 2020 @01:29PM
(#23692518)
I don't think that's the entire reason. Most of the common x86 instructions occurring in a normal program can be decoded to a few uops in a straightforward way, and since Sandy Bridge the decoded uops are cached anyway.
So this would only be a significant bottleneck for hot loops that are large enough that they don't fit in the uop cache.
It's definitely a real issue but it seems wrong to pin all or even most of Intel's stagnation on that.
So this would only be a significant bottleneck for hot loops that are large enough that they don't fit in the uop cache.
It's definitely a real issue but it seems wrong to pin all or even most of Intel's stagnation on that.