A strange observation, that the last thread to complete a matrix multiplication task was always much later than others, explored to discover a different strategy used by macOS.
M1 Pro
Some threads are set to run in the background, and get allocated to the E cores. Could you run them in a VM, and effectively promote them to run on P cores instead?
Nearly 6 months ago, many thought the M2 Pro and Max would come in Macs in the autumn of this year. So why hasn’t Apple released them yet?
Three types of Disk Image, encrypted or not, tested when freshly made or used and remounted. So many variations, but only one type of Disk Image can be trusted for writing.
A Mac Studio Max has an SSD delivering up to 7.4 GB/s, and a CPU up to 50% faster than a 16-core Xeon. Why does it write an encrypted disk image at the speed of a slow hard disk?
Should you wait for Apple’s to release Macs with the M2 Max, or won’t those be available until next year? Here are some speculative dates based on the M1 release cycle.
Both P and E cores are run at different frequencies according to the load on M1 chips. This explores how macOS manages their frequencies and why.
How the E and P cores in an M1 Max chip cope with the heavy system workload after login, but still give the user the scope to run apps immediately.
AirDrop is quick, convenient and as slow as you’ll get. Ethernet all too often runs at only 1 Gb/s. Here are the fastest solutions involving M1 models.
There’s evidence to suggest that original M1 Macs write more slowly to external SSDs in some configurations. Does this extend to later models with M1 Pro or Max chips?
The Eclectic Light Company 