An Important Note On The Alder Lake Mobile Power/Performance With Linux 5.19

[phoronix]

Phoronix: An Important Note On The Alder Lake Mobile Power/Performance With Linux 5.19

After dealing with the Intel Alder Lake P GuC firmware breakage around Linux 5.19 Git that is now to be addressed by the upstream Intel developers, it was on to performance testing the shiny new Core i7 1280P with this kernel due to be released as stable within the next two weeks... For those concerned about maximum performance, there was a glaring performance regression for this Alder Lake P on the new kernel being released as stable later this month. Well, a default change in performance/behavior at least but the power efficiency / performance-per-Watt tended to be better on this new kernel.

https://www.phoronix.com/scan.php?page=article&item=linux-519-alderlakep&num=1

[birdie]

4.2% performance lost, vs. being 16.9% more power efficient looks like a win to me. Hopefully this is adjustable at runtime.

[aufkrawall]

Michael's infamous "Scheisse!" is back, this is bad.

Well, hopefully these slowdowns (or something else) fix some ADL devices like Galaxy Books from randomly turning off...

[openminded]

That's good news. 12650H, for instance, is terribly hot in Windows and Linux 5.18, and a slight decrease in performance is better than overheating and throttling anyways.

EDIT: Ah, temperature is the same! Sad...

[Lbibass]

That's a massive efficiency increase! Like others said, hopefully it's configurable at runtime. Selecting between performance and power efficiency is always a nice thing to have.

[hamishmb]

That's an incredible efficiency increase!

[Linuxxx]

Michael

Would it be possible to do another run but with the turbo-boost frequencies disabled & also testing different CPU governors?

I have a feeling this would show even greater efficiency gains, especially with the performance governor.

BTW, the great thing about the Steam Deck is that Valve allows changing the TDP on-demand, where a lower TDP combined with amd-pstate should provide the best power-efficiency, when tested...

[birdie]

Michael

Would it be possible to do another run but with the turbo-boost frequencies disabled & also testing different CPU governors?

I have a feeling this would show even greater efficiency gains, especially with the performance governor.

BTW, the great thing about the Steam Deck is that Valve allows changing the TDP on-demand, where a lower TDP combined with amd-pstate should provide the best power-efficiency, when tested...

Pretty much all modern x86 CPUs and desktop GPUs work well beyond their highest efficiency point.

Check this TPU review: https://www.techpowerup.com/review/i...-power-limits/

Pretty jarring results, I've got to admit.

[Linuxxx]

Pretty much all modern x86 CPUs and desktop GPUs work well beyond their highest efficiency point.

Check this TPU review: https://www.techpowerup.com/review/i...-power-limits/

Pretty jarring results, I've got to admit.

True, even on my decade-old Intel i5-3210M (Ivy Bridge) disabling the meager 600 Mhz turbo-boost (2.5 to 3.1 Ghz) would make sure that the fan on my notebook would never reach its highest & loudest spin-state, even under full load with the performance governor.

With that said, disabling the turbo-boost on my desktop-class Intel i7-11700F would cut the maximum frequency from upto 4.9 Ghz down to just 2.5 Ghz, which is just too big of a hit to performance.
Thankfully, cooling these parts inside a PC tower is alot easier than inside thin 'ultrabooks'.

[kylew77]

I can't help but wonder if what OpenBSD 7.1 did was the right thing to do, when connected to the AC mains just run 100% power and when connected to just the battery then run in power save mode. It leads to some hotter laptops but better performance!