You can achieve similar boosts by setting R6 siege and metro exodus to have exclusive access to the P cores while relagating background tasks to the e cores.
Thats why only two games are supported right now, because it takes a lot of testing to make sure the game actually benefits from this kind of optimisation, cyberpunk 2077 for example will run worse if prioritised the same way due to it being optimised well to utilise the e cores for better fps.
We haven’t yet seen benchmarks comparing APO with process lasso (at least I didn’t see any so far) but no they’re not the same, in theory APO should be more efficient given it has a specific driver and BIOS setting to do what it does.
That’s what I meant by lower level. And I assume APO is doing something more than just splitting cores, hardware unboxed mentioned using only one e core from each section or something like that for better cache usage.
I doubt it achieves the same results as APO since APO works at a lower level
They do different things tho?
Edit: isn’t APO about making the E cores work better and this just disabling them?
This can disable them yes but the main goal is to set apps running exclusively on P cores.
APO is just process lasso with tuned rulesets.
You can achieve similar boosts by setting R6 siege and metro exodus to have exclusive access to the P cores while relagating background tasks to the e cores.
Thats why only two games are supported right now, because it takes a lot of testing to make sure the game actually benefits from this kind of optimisation, cyberpunk 2077 for example will run worse if prioritised the same way due to it being optimised well to utilise the e cores for better fps.
We haven’t yet seen benchmarks comparing APO with process lasso (at least I didn’t see any so far) but no they’re not the same, in theory APO should be more efficient given it has a specific driver and BIOS setting to do what it does.
That’s what I meant by lower level. And I assume APO is doing something more than just splitting cores, hardware unboxed mentioned using only one e core from each section or something like that for better cache usage.