I had to pause some of my work getting a current-gen AMD graphics card running on the Pi 5 and testing a 192-core AmpereOne server to quickly post on the M4's efficiency.
I expected M4 to be better than M1/M2 (I haven't personally tested M3), and I hoped it would at least match the previous total-system-power efficiency king, a tiny arm SBC with an RK3588 SoC... but I didn't expect it to jump forward 32%. Efficiency gains on the Arm systems I test typically look like 2-5% year over year.
The M4 mini I just bought reaches 6.74 Gflops/W on the HPL benchmark.
UPDATE 2024-11-20: I re-ran all my tests without my 10 Gbps Ethernet plugged in, and the efficiency is even better. I also re-tested LLM performance, and made this quick video:
For the latest benchmark results, check my sbc-reviews M4 Mac mini issue.
I can get 283 Gflops at 42W, versus 264 at 66W on my M1 Max Mac Studio (for a round 4.00 Gflops/W).
The chip isn't the fastest at everything, but it's certainly the most efficient CPU I've ever tested. And that scales down to idle power, too—it hovers between 3-4W at idle—which is about the same as a Raspberry Pi.
This is total system power draw, too—not just the CPU.
And the system I bought includes 10 Gigabit Ethernet and 32 GB of RAM; most systems I've used consume 4-6W just running the 10 GbE controller!
In 1.25U of rack space, you could run three Mac minis, idling around 10W, giving almost a teraflop of CPU performance. (Not to mention there's a fast GPU/NPU, 10 GbE, and tons of high speed Thunderbolt IO in the back.)
If only they didn't put the power button on the bottom.
You can check out all my top500-benchmark results on GitHub, and follow my progress testing out the M4 mini here.
I haven't tested an M4 Pro Mac mini yet, so I'm not sure if the efficiency is any better or worse.
Comments
Do you think these could be colocated for a small-medium sized business?
you can buy vertical housings, for 8 units in 5 RU, like this:
e.g., look at mk1manufacturing
You should check out MacStadium.com they specialize in that.
A simple thing, probably best thing, is just mount 'em upside down in the rack.
The fan gets air, you get a button. The (previous) top might even sink a little heat into the shelf.
The ports don't require text labels, it's all good.
Regarding the awkward placement of the power button on the bottom… and I think I'd have to hook up my wired computer speakers to the front? yuck… is it possible to run the Mac Mini on its side? Might be more ergonomic for some setups (like mine), might improve ventilation… is this feasible or is there some technical problem with just sitting it on its side while it runs?
Everyone complain about the power button location, but what if you put the mini upside down?!
Is their a thermals deficit in doing so?
It might be best for the WiFi too.
I wonder and would love to see you load test and compare.
Will it's base variant be enough for development purposes?
> If only they didn't put the power button on the bottom.
At only 3-4 W idle, why would you ever need to turn it off? ;)
It may go to standby automatically, then when you don't have keyboard and mouse connected to it, you would want to enable it with a button.
That was my reaction to all the fuss. Mine’s unlikely ever to be switched off or needing a restart in some other way than via the OS. I suppose some people struggled to find anything else to criticise!
If I'm reading about HPL correctly, it's only using the CPU cores. If you had tasks/computational load that could use both GPU & CPU cores, that Efficiency would go up even more.
The stock unit with 16 GB ram, and gigabit ethernet idles at 3 watts. The M4 Pro Mac mini (24 GB ram, gigabit ethernet) idles at around 3.5 Watts. Not sure if it's the additional ram, the 10GbE or both that is using up an additional watt at idle.
M4 Pro Mac mini also has the ability to set the power to "High Performance" which runs the fans more aggressively to keep the CPU/GPU from Throttling.