Live data from Hacker News

M4 Mac mini's efficiency

jeffgeerling.com

571–580 of 632 posts

Re: M4 Mac mini's efficiency

#571
post #45

“If only they didn't put the power button on the bottom.” While I think Apple was off the rocker on this particular decision, I do respect their org structure that allows this type of decision to occur. Believe me, there are companies where a dozen people or more would weigh in and prevent an unpopular choice. Consensus sometimes hinders a desired result (both good and bad).

It's a way of signaling how the product should be used. Plug it in, hit the power button, put it down, and never turn it off again. For many users that's probably the only time they will ever interact with that button (or want to). I actually think it's a really good choice and shows Apple really understands design. And with the relatively low power consumption it makes sense. It's not like it's drawing a ton of powe…

I think it is really bad design. Perhaps necessary because of space restraints and in that case understandable. But that is entirely different to good design and I cannot really buy the "use case explanation".

Many leave their devices on their desk and Apple always had a problem with just letting devices turn of completely, there are regularly problems with it. And they do drain power on idle, which is a frequent complaint.

Yes, we are that insane to use a lot of Apple devices for business in some departments. MDM for phones and iPads is top for the baseline administration, but the devices are eccentric to say the least.

Re: M4 Mac mini's efficiency

#572

Earlier quoted context omitted.

Yeah, I don’t know why 8gb base models get so much hate online. 8gb is 64 billion bits of memory. If you’re writing everyday software and you need more memory than that, you’re almost certainly doing something wrong.

Seriously? What if you want to have a few browser tabs and a spreadsheet open? Or containers? My M1 routinely rests around 22gb of RAM.

How much RAM should a few browser tabs and a spreadsheet use? Spreadsheets and webpages were both invented at a time when computers had orders of magnitude less ram than they do today. And yet, Excel and Netscape navigator still worked fine. It seems to me that bigger computers have caused chrome to use more memory.

If 16gb is considered to be a "bare minimum" for RAM, well, how much ram will all those programs use next year? Or in 10 years?

That doesn't help you right now, but 22gb is ridiculous for a few browser tabs and a spreadsheet.

Re: M4 Mac mini's efficiency

#573
post #569

Earlier quoted context omitted.

But it’s not an apology, it’s the right design decision. The battery charges to a usable amount extremely quickly, and if you could plug it in all the time most would, which defeats the point.

Please explain how it's in my best interest that I must use my peripherals wirelessly. The only wireless mouse I have ever owned is in my work bag, so I have one wherever I go, it's not for regular use and I have zero problems with mouse cables, for the actual 30th year this December.

Because it’s the design of the product? Every product is designed with a specific usage in mind. This is designed to be wireless, hence all of the ways in which it enforces and enables that. The battery lasts a very long time, so even in your work bag it should be fine (although are you then plugging into many different computers to associate it?)

If you want a corded mouse (and it sounds like that’s a better fit), there are plenty of options on the market.

Re: M4 Mac mini's efficiency

#574

Earlier quoted context omitted.

I see thank you. I dream to setup something like iCloud but with open source software and hosted at home :) Not sure if there is anything like that out there.

nextcloud.com comes to mind.

Is that something that can be setup on Mac OS? Or do I need to install Linux on Mac Mini?

Re: M4 Mac mini's efficiency

#576
post #90

Earlier quoted context omitted.

Depends what 'home server' means. MacOS would need syncookies to be a viable tcp server on public IPs, IMHO, but MacOS pulled FreeBSD's TCP stack a couple months before syncookies were added, and they never rebased or otherwise added syncookies later. I haven't looked into if they pulled any scalability updates over the years, but I kind of assume they haven't, and the stack would have a lot of lock contention if you…

Is this something that you can fix by putting the server behind Cloudflare? I assume most "home server" users would do that (or a similar service provided by Apple if they go down that route).

Well, Cloudflare is kind of spendy if you want them to proxy non-http traffic.

If you put a proxy in front, and you're careful to only allow inbound connections from the proxy, you should be ok though.

Re: M4 Mac mini's efficiency

#577

Earlier quoted context omitted.

So get a stronger one. There will be always something that is better on one or another side. I have 13th gen i7, with 64 GB RAM and 2 TB ssd (and 2.5 GbE). It was 800 EUR + VAT, last year. How much would similar Mac Mini cost? Not a fair comparison either. Edit: 69W for the NUC is not TDP. It is 69W power brick that ships with the machine.

what do you get in https://browserbench.org/Speedometer3.0 ? - m4 mini is 45 it seems.

16.6 in Firefox, 19.8 in Chrome (Ubuntu 24.04, both browsers in flatpak).

Re: M4 Mac mini's efficiency

#578

Earlier quoted context omitted.

It's a way of signaling how the product should be used. Plug it in, hit the power button, put it down, and never turn it off again. For many users that's probably the only time they will ever interact with that button (or want to). I actually think it's a really good choice and shows Apple really understands design. And with the relatively low power consumption it makes sense. It's not like it's drawing a ton of powe…

I have a Mac Mini and can't remember the last time I had to manually press the button. IIRC it even reboots on its own after a power outage. I think I shut it down once for an extended vacation just to make sure appliances weren't on while I was gone and when I switched apartments. Otherwise I'd check and post my uptime from the command line. It's a launch M1 mini so I'd wager less times pressing the power button tha…

I'd never thought about it before reading this comment but I now realise I don't even know where the power button on my Mac Studio is. I used it once when I first set it up and haven't touched it since.

Re: M4 Mac mini's efficiency

#579
post #299

Earlier quoted context omitted.

It's a way of signaling how the product should be used. Plug it in, hit the power button, put it down, and never turn it off again. For many users that's probably the only time they will ever interact with that button (or want to). I actually think it's a really good choice and shows Apple really understands design. And with the relatively low power consumption it makes sense. It's not like it's drawing a ton of powe…

This is actually how I've used my M1 Macbook Pro since I got it. I never fully turn it in. It's either sleeping when plugging into my Thunderbolt 3 dock, or its sleeping on my dining room table on battery power. The efficiency is so good it never dies even if I don't use it for a day.

My work machine is an M3 Macbook Pro. I put it to sleep on a Friday, and after a three-day weekend, it's still ready to go on Tuesday with 95% of battery left.

What's irritating is that a lot of Intel laptops used to be able to get pretty close to this, back when they supported legacy sleep states. I have yet to own a newer Intel laptop that can sleep for more than 24 hours without almost completely draining the battery.

Re: M4 Mac mini's efficiency

#580

Earlier quoted context omitted.

I don’t think anything is stopping you buying a second hand / 3rd party module online. It just needs to be physically compatible with your particular generation of hardware.

I don't think this is true. If you watch the videos, dosdude1 specifically says he had to order blank NANDs for this process. Then you DFU restore the system from another mac. I have no proof, but I assume part of this DFU restore process is the new NAND chips being hardware paired in some way. Again I have no proof, but there must be reasons he claims they have to be blank NANDs

> there must be reasons he claims they have to be blank NANDs

If you are building new SSDs, it makes sense to be certain that used NAND chips aren't slipping into your supply chain.

They do have a limited number of write cycles that they are good for, after all.

Post reply on HN