Mac Mini Power Consumption and Thermal Output Specs

It’s an interesting analysis that I don’t agree with.

The numbers from 2014 onward are rather striking

John’s making the point here that the 2018 model especially sees a significant increase in wattage. Certainly. He then draws the conclusion that it must be because Intel wasn’t able to make more efficient CPUs. I don’t think that’s it.

The CPUs in the 2018 Mac mini are a weird one-off that Intel more or less made specifically for Apple (if they weren’t exclusive, I would expect at least that the Mac mini made the biggest dent in volume): a desktop-class CPU, but in a mobile form factor. A TDP of 65W, like with many iMac models, but in the same form factor as the 15-/16-inch MacBooks Pro (which have a 45W TDP CPU).

Apple didn’t have to use this CPU. The immediate predecessor from 2014 had a 28W CPU, i.e. something they would also put in the pricier 13-inch MacBooks Pro.

Given how the 2018 model was positioned (for example, there’s a 10GigE option, and it started at $799), I suspect this was a deliberate choice of making the Mac mini higher-end than before, presumably because it’s where Ternus’s team saw more of a market fit. And since that model is still sold and the M1 Mac mini thus far doesn’t have some of those options (more external displays, 64 instead of 16 GB RAM ceiling, 10GigE, more?), I don’t think bringing up the M1 Mac mini’s power consumption is fair: we’d have to wait for the other shoe to drop, namely the higher-end mini.

Assuming, of course, that they don’t just quietly discontinue that.

Yeah, you could read it as “they had to make the 2018 model consume much more power because of Intel inefficiencies”, but I suspect, instead, that that was more of a deliberate choice to reposition it at a higher end — coupled with an unwillingness to also offer a lower-end model.

(The opposite happened, too. The 2012 Mac mini existed in a quad-core option, but the 2014 one didn’t, because due to SKU choices on Intel’s part, Apple would’ve had to go either all-quad core or do two different logic boards. Instead, they decided to go all-dual core. Maybe we’re lucky and with the M1, they’re finally willing to serve both segments. Or maybe they just haven’t decided yet.)

Unfortunately, the data provided by Apple on Mac Mini power consumption is inaccurate. The maximum numbers I have seen in the real world for 2014 and earlier minis are not anything near what Apple lists, and the idle power for the 2018 model is far lower than Apple claims - 7-9 watts at idle depending on the CPU:

https://macintouch.com/community/index.php?threads/mac-mini-and-alternatives.1587/page-8

While I've never owned one, at the other end of the table, I am extremely dubious that the G4 mini used more power than the equivalent g4 Powerbooks of the era (see https://www.anandtech.com/show/1990/6).

I owned both a titanium PowerBook G4 and an early Mini, and I completely believe the Mini used more power. The PowerBook could get a little warm. The Mini was an oven by the time it had finished booting.

The G4 Mini used a later (7450?) version of the G4 architecture. I had a 12 inch PowerBook, aluminium 2003 midway between the TiBook and Mini, and that wee fella could get as hot under the collar as Intel’s worst! So the figures are credible to me.

My M1 Air isn’t nearly as good for warming a cat in winter, let me tell you…