Using CPU % or Energy values in Activity Monitor appears to show that running code on E cores is less efficient than on P cores. Don’t believe a word of it.
CPU
Select a test, time it, and compare the result with those from other systems. Choose whether to use a synthetic or application benchmark, and don’t forget your confirmation bias.
It delivers detailed weather forecasts for days in advance, and real-time manipulation of elaborate textured 3D models. But more mundane tasks may not get any quicker.
How can you tell how much memory is being used by the GPU when both CPU and GPU use Unified Memory? Does it matter anyway?
Current CPU % given in Activity Monitor can be misleading and has limited value for M1 models. Here’s how to improve it.
In Activity Monitor, % CPU isn’t on a scale of 0-100. In M1 Macs, it also makes no distinction between E and P cores, nor does it allow for their changing frequency.
Each additional Safari window increases WindowServer’s memory use by 1.7 MB on an Intel Mac, but 50 MB on an M1 Mac – thirty times more.
The P cores in the M1 Pro/Max CPU are managed in two groups of four, sparing load on the second group, and distributing it unevenly within each group. Its two E cores outperform the four in the M1 too.
How does unified memory make Apple Silicon Macs faster? Why can’t you add your own memory? How much do you need?
Summary and links for the latest information about what’s in the current M1 chip, from differences in caches between cores, to the Matrix Coprocessor and Fabric limitations.
The Eclectic Light Company 