![]() Although this is fiddly, the effort is worth it: with the single thread, the E cores run at a frequency of about 972 MHz, but with two threads that increases to their maximum of 2064 MHz. To discover how Activity Monitor is coming to those false results, you’ll need to run the command tool powermetrics to discover the frequencies the two E cores were running at, and their actual power consumption. So running twice the amount of code on E cores alone takes the same amount of energy, according to Activity Monitor. Running two threads has a total energy of 1940, at 970 per thread. The single thread has a total energy of 1931 units.s, at 1931 per thread. 2 threads on 2 E cores took 10.0 s, at 194% CPU and an energy value of 194.1 thread on 2 E cores took 19.5 s, at 98% CPU and an energy value of 99. ![]() Then repeat that with two threads imposing the same computational load. With an app like AsmAttic, run a single thread at minimum QoS so it runs only on the E cores. If you’ve got an M1 Pro or Max available, it’s not hard to see where this is going wrong. It’s even worse that Activity Monitor’s errors are discouraging developers from making better use of the cores in M1 chips. Given that it’s now nearly 18 months since Apple started shipping its first M1 Macs, you might think that a little surprising. This result occurs because Activity Monitor, currently version 10.14 in macOS 12.3.1, doesn’t know the difference between processors with identical cores running at fixed frequency, and Apple’s M1 chips, with two different types of core and variable frequencies for each cluster of cores. And if you believe that, you drop the idea of offering the user control over QoS, and run all that app’s code at high QoS after all. The clear conclusion is that running these eight threads on the E cores was considerably less efficient than running them on the P cores.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |