Apple M2 SoC Analysis - Worse CPU efficiency compared to the M1

[Redaktion]

The Apple M2 is here and after our reviews of the MacBook Pro 13 and the MacBook Air M2, we want to have a closer look at the new M2 chip. What changed, which improvements can you expect, and how does the increased power consumption affect the efficiency, especially compared to the previous M1 chip?

https://www.notebookcheck.net/Apple-M2-SoC-Analysis-Worse-CPU-efficiency-compared-to-the-M1.637834.0.html

[cfb]

I have an m1 mac. Just an 8gb/256gb model. After not owning apple products for over a decade because they cost too much for what you get, I call the m1 mac mini a bargain.

Quite a bit of power for the $. But I think they tried too hard to impress with their first outing.

So they cut the speed of the SSD in half with the m2, so you pretty much have to buy the 16GB or 512GB upgrades for 4x what they're worth.

Which makes the m2 mac mini is NOT a bargain at all.

Hey apple? It's best to sandbag performance before you release the product. Not after.

[M Paredes]

Very informative article, we will have to wait until M3 to see any cpu efficiency gains with the 3 nm process.

[unimpressed]

I will only buy an M2 mac if it can run Asahi LINUX well enough.

See asahilinux.org

[vitalinx7]

intel & AMD has no chance !

[Tommo53]

Using my Intel 1240P Framework laptop with adjustable settings I get much much better performance than what is shown.

At the following average package power over test duration, I obtain the following results:

Multi Core:
Intel 1240P @ 29.5W = 9667 Points = 327 Points per Watt
Intel 1240P @ 15.7W = 6193 Points = 394 Points per Watt
Apple M2 (stated in review) = 409 Points per Watt

Intel 1240P stated in Review = 258 Points per Watt

Single Core:
Intel 1240P @ 14W = 1469 Points = 104.92 Points per Watt
Intel 1240P @ 9.3W = 1244 Points = 133 Points per Watt
Apple M2 (stated in review) = 297 Points per Watt

Intel 1240P stated in Review = 64 Points per Watt

Obviously the Intel and probably AMD chips are far more efficient than shown in this article. When operating at boost clocks, the power consumption increases exponentially for a minor increase in performance. Simply turn down the CPU's and the efficiency massively increases. 

[Manuel R.]

Notebookcheck i am disappointed.

What a weak article with basic knowledge about the difference of power and energy gotten wrong.

You are seriously deducting a efficiency statement from calculating score/power (Watts) instead of energy?

You arent even telling us what kind of Watts you are dividing by. Is it mean, average, peak or some random power figure you meassured somewhere during the test?

[Nikolas M]

This is an awesome deep analysis no marketing talk article, thank you! I am looking forward to the same article for the M3.

[JohnIL]

I have used Mac's for many years even back to the PowerPC chip days. Moving to Apple silicon impressed Mac users so much because Apple always made bad choices with Intel chips based more of efficiency then performance. Especially in the MacBook Air where Apple chose Y series CPU's which were low power but also pretty bad performance. They could have used U series which would have slightly degraded battery life and probably would have required a fan. Actually Apple silicon is not that performance oriented requiring two SoC's in their higher tier Pro desktops. Clock speeds are also not that great but are somewhat overshadowed by the SoC efficiency. After using M1 Mac's for over a year, I have realized they were somewhat overhyped and were simply impressive because the previous Intel models were just so bad.