Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

M4 Mac mini's efficiency

jeffgeerling.com

Post reply on HN
Print view View original

451–460 of 632 posts

Re: M4 Mac mini's efficiency

#451

Post by woodson »

dcl wrote: post #368

Is there any benchmarks for these chips doing like regular 'data-sciency' CPU grunt work? Dataframe-wrangling, inverting matrices, doing large matrix, factorisations, fitting decision tree's, etc? I'm very keen on one of these, but I simply have no idea how good they are at my day to day tasks in R or Python.

It depends. If you're using Python with numpy>=2.0.0 (and macOS>=14) then you should benefit greatly from Apple's Accelerate implementation of BLAS/LAPACK routines which are behind most linear algebra operations. I'm not aware of any serious public benchmarks, though.

Re: M4 Mac mini's efficiency

#453

Post by timc3 »

rgbrgb wrote: post #10
LordKeren wrote: post #6

I can’t help but wish that Apple would provide the handful of features needed to make a Mac mini into a competent home server.

what else do you need?

What I look for is, 128GB RAM minimum, decent number of PCIe lanes because I want two fast NVMe drives, a HBA card ( though this I guess could be external ), two network ports minimum, ZFS, sane terminal, native support for containers and VMs. Native support for UPS interfacing, native support for backup of containers and VMs. And lastly a community of other users doing the same.

Dual power supplies is a nice to have.

Re: M4 Mac mini's efficiency

#454

Post by atonse »

redundantly wrote: post #193
mlsu wrote: post #98

Earlier quoted context omitted.

And I mean, it uses 4 watts idle. If you power it off overnight for 12 hours you've saved... half a glass of orange juice worth of energy.

As with all things regarding power efficiency, you have to consider the wide use of these devices, not just the individual use. If moving the power button there changes the behavior of thousands of people that would typically shut their computer down when they're not using it, that half glass of orange juice turns into thousands of gallons.

4 watts idle, how many watts on sleep?

Re: M4 Mac mini's efficiency

#455

Post by rsanheim »

barbazoo wrote: post #102

Noteworthy because related to energy efficiency across the entire product lifecycle: > Mac mini is: > Designed with more than 50% recycled content. Made with electricity sourced from 100% renewables. Shipped 50% or more with low-carbon methods. https://www.apple.com/environment/

the mac mini is the first carbon neutral mac:

* https://www.macrumors.com/2024/10/29/apple-m4-mac-mini-carbo...

* https://www.apple.com/environment/pdf/products/desktops/Mac_...

Re: M4 Mac mini's efficiency

#456

Post by edward28 »

ricardobeat wrote: post #386
layoric wrote: post #265

Earlier quoted context omitted.

In same boat, I have the 5950x with 64gb memory running PopOS and there are times I'm hitting swap a lot more than I'd like. 16 or 32gb of memory is just not feasible, and even 2TB of storage would likely cause headaches, I have a 4TB and a 2TB nvme at the moment which will come with me next upgrade. I'm leaning towards an upgrade next year to the 9950x3d if reviews pan out. Sure, it's going to be a bigger machine wi…

Memory usage is not comparable across Linux and Mac. MacOS is much better at avoiding swap, uses memory compression, shared frameworks etc. At the same time it tries to use all the memory available which makes direct system-wide comparisons not accurate. A good rule of thumb is that 8GB on Mac == 16GB on Windows/Linux.

You can also compress memory on Linux with zswap or zram.

Re: M4 Mac mini's efficiency

#457

Post by physicsguy »

switch007 wrote: post #450

Stupid question but why can't it (non Pro particularly) be powered over USB C like a MBP, if it's so power efficient? Does it have extra performance that makes full use of the 155W input? What would have it taken Apple to have given us the option to be powered by 100W USB C (eg the tech to downgrade power usage to match the power input)?

I guess the inclusion MBP battery allows drawing higher current than a USB C power supply can provide?

Re: M4 Mac mini's efficiency

#458

Post by microtonal »

nextos wrote: post #38
vetinari wrote: post #35

Earlier quoted context omitted.

NUC 14th gen with i3 is around 400 EUR with VAT, with no RAM or storage. For the other 250 EUR, surely you can get more RAM than 16 GB and more storage than 256 GB.

I use a NUC as a daily driver. The problem with NUCs is that cooling is suboptimal, the fan is small and thus noisy. It can be fixed with a third-party case, but that's at least €60-100 more for a much slower machine. Plus, you may void the guarantee by transplanting the motherboard.

It’s a shame that ASUS cancelled the NUC Extreme line. I know it’s quite a bit bigger than other NUCs. But the 13 Extreme had expandability, good cooling, and fast CPU options.

Re: M4 Mac mini's efficiency

#459

Post by jhugo »

OJFord wrote: post #285
giobox wrote: post #105

Earlier quoted context omitted.

They have variations of the program in some European countries. It's been a long time for me, but in the UK they used to just whitelist university domains (we didn't use .edu TLDs either).

We use .ac.uk though (and much of the non-USA world uses .ac.ccTLD similarly) so no need to whitelist individual university domains. I don't know about Apple, but that's a common approach. (And does irritate some where they don't use either and get missed, Canada for example.)

Some countries place no restrictions on registering domains under their .ac second level domain though.

Re: M4 Mac mini's efficiency

#460

Post by watermelon0 »

switch007 wrote: post #450

Stupid question but why can't it (non Pro particularly) be powered over USB C like a MBP, if it's so power efficient? Does it have extra performance that makes full use of the 155W input? What would have it taken Apple to have given us the option to be powered by 100W USB C (eg the tech to downgrade power usage to match the power input)?

I think most of the 155W is for USB-C power delivery, not the machine itself.

Blog post mentions that power usage during benchmark was at 42W.

Post reply on HN