Apple M4 Mac Mini With macOS vs. Intel / AMD With Ubuntu Linux Performance

[phoronix]

Phoronix: Apple M4 Mac Mini With macOS vs. Intel / AMD With Ubuntu Linux Performance

Apple last week released their latest iMac, Mac Mini, and MacBook Pro products powered by their fourth-generation M-series Apple Silicon. The new Mac Mini in particular is interesting for under $600 starting out with the all re-designed Mac Mini with 10-core M4 and now the base model having 16GB of memory. It will take some time before there is any reasonable Linux support on the M4 hardware with Asahi Linux, but for those curious about how the M4 Mac Mini with macOS compares to AMD Ryzen and Intel Core CPUs under Linux, here are some preliminary benchmarks.

https://www.phoronix.com/review/apple-m4-intel-amd-linux

[galad]

Why x265 3.6 instead of 4.0? 4.0 has got many additional Neon optimizations.

[Steffo]

The efficiency is pretty impressive! This is the basic M4 silicon and its performance is really not bad!

[bsm117532]

You really should be reporting energy consumption in Joules, which is Watts*(time to execute).

[Michael]

Why x265 3.6 instead of 4.0? 4.0 has got many additional Neon optimizations.

Last check 4.0 build system was borked for me.

[Michael]

You really should be reporting energy consumption in Joules, which is Watts*(time to execute).

There is Joules reported as footnote on the power graphs...

[Tarnith]

Could be fun next time when doing tests like this to also do a run of ex. the 9800X3D in eco mode, or a very power limited/underclocked setting.

I suspect this would reveal a lot of the architectural efficiency that's present, but largely obscured by the higher power desktop tuning

[coder]

Keep in mind especially for the code compilation tests that it's macOS vs. Ubuntu Linux at play as well with their default toolchains for each given that most developers are sticking to the default toolchains in production.

I understand the logic of this. However, I'm primarily interested in how the CPUs compare and that really demands that you use the same version of the same toolchain on both platforms and set one of them to cross-compile, so they're both targeting the same ISA and ABI!

[bsm117532]

There is Joules reported as footnote on the power graphs...

It would be great to see this in graphs though. The quantities plotted are difficult to interpret.

[sarmad]

Looks like Apple is power limiting the M4, which is useful for a battery powered device like a laptop, but the Mac Mini is a desktop computer and it can benefit from extra performance.