The Forgetful CPU (Linux on M4)
yuka.dev
yuka.dev
I mean, the "so much better" in my experience is really the battery life.
But vs Linux, well even Linux on a non-Apple laptop tends to mean screwed up power management. The display system might be less efficient too, depending whether it's X11 or Wayland, and idk which one is the good one.
But yeah, the remaining people who needed dGPUs can now just use the integrated AS graphics, so they'll notice an even bigger difference.
And yes, Intel Iris was a good leap forward for their iGPUs at the time! But I don't see any models with Iris and a dGPU, so if you got a model with the dGPU then you would be switching between a potato iGPU and the dGPU.
> But I don't see any models with Iris and a dGPU, so if you got a model with the dGPU then you would be switching between a potato iGPU and the dGPU.
Dang, didn't notice until you pointed it out. So it's a really good thing I got the 13".
I'd love to replace it with something like a framework or a tuxedo, but they all get hot to touch. Maybe I'm in the minority, but a laptop not heating up when you turn it on is one of the most important things for it. It doesn't matter much for work, because it'll sit in a dock anyway, but I rarely use my m1 air at a table.
I think the interesting part of that sort of success is that if/when this thing will start to lose a step (local inference would be cool, 16GB of RAM is starting to get a little tight, especially for rust builds and agentic programming), I really have no serious qualms about spending $5000+ on its replacement.
For I came from almost a decade on Debian. I had become used to the ways of GNOME, I enjoyed GNOME 3 even and had no gripes about systemd.
Now I am in Windowsland. This is hell. I have no filesystem and cloud works the exact opposite way it should pretending we all see the same thing. I cannot email an attachment without closing - why is this? The concept of a shared folder has changed to everyone sharing their 'personal drives' while having no user-facing audit on what is shared with whom and via what link. Somehow a 4-digit pin is as authorative as a password to everything but a network printer that demands passwordy login. At the sane time, windows lacks handy preset printing like mac or linux - its there but buried, like everything in windows, thus not handy. The sleep key wakes the laptop and bypasses all login security. Copilot cannot be turned off unless done so by an organisation admin. At the same time, trusty fallback productivity tools like VBA fail as there is no filesystem - productivity workers are cripled using their 'everything looks like a nail if all I have is a windows hammer tool'. Teams still isn't as good for simple chat as Windows Communicator was a decade and a half ago and there isn't another option in the windows stable in sight. A 10th gen Intel chip gets overpowered by this all every hour or two. Possibly triggered by Adobe Acrobat - software that no one needs - trying to install a further part itself again.
Corporate Windows is a horror. How did it get so bad?
For delete, if you mean delete-under instead of delete-before (what the key labelled "delete" does), Fn-Delete is delete-under.
Home/End are either Cmd-Up/Down (doc begin/end) or Opt-Up/Down (line begin/end) (I haven't used a PC in a while, so I can't remember what Home/End do)
Obviously, this is not a problem for most Mac users. Personally, I've used a MacBook Pro for more than a decade, and I eventually found that so annoying that I bought an external keyboard with full size arrow keys and dedicated Del, Home, End, PgUp, PgDn keys.
In other words, things that you would have had to touch in Linux too, but because they didn't work at all, here work but until configured, do not work like you'd want them to.
It's still hilarious when defender does it's thing and the fan goes mach 1, or when the firmware updater force reboots in the middle of a meeting after you've clicked the 10 minute extend two times.
It is what it is. I'm sure we'd figure out a way to make enterprise IT horrible on any OS.
That looks like a corporation problem.
I think eventually they pushed a spyware update that improved a bit the situation but not by a lot.
Meanwhile my personal M1 Pro had truly full day battery life.
Unless one is lucky to work for small to medium size companies whose organisation hasn't yet grown to such politics and processes, with certifications, IT tickets for admin actions, and so on.
I have an M4 MacBook Pro. It gets hot. The fan kicks in, and it gets noisy. I can't say whether it is T14 noisy since I never heard one. As for the cool and fanless MacBook Air, my understanding is that Apple is quite aggressive with thermal throttling. That's perfectly acceptable, even desirable, under some workloads. Under other workloads, it is not.
They only want to sell the whole package. You might as well ask the Catholic Church to start a rap music 8-track only record club. Same probability.
The hardware is a means to an end, and nothing else to them.
If they just did motherboards with cpu/ram/pcie and let other OEMs build complete systems the story would be very different
All the capacity at all the fabs is booked up for years. They can’t make more if they want to.
So any hypothetical sales of OEM style guts would only cut into the number of Macs they can sell to existing customers. And there’s no way they would ever do that.
Higher margins?
42 people with iPhone will buy airpods, some subscriptions and apps. They also be vendor locked into Apple ecosystem and almost guaranteed to bring more iPhone sale in 3 years.
Apple must be dumb to canibalize their own market.
Now they have Apple Silicon.
I think most people here never seen the inside of sliding pizza boxes on server racks, really seeing them.
Apple doesn't have the business to business support organization to make this kind of thing work either.
Libdispatch is a burden to use for a server because of its book keeping overhead, and their history with kqueue makes it hard to convince someone to use it.
There’s no io_uring, but in practice, almost all servers run epoll anyway, don’t they?
A mega-AI workstation could.
But server? Nah.
It failed.
But I think that they learned something really valuable. As a company, they don’t have a good understanding the server market or how to sell to it. Their premium product style just didn’t have a place in the race to the bottom server market.
We had two in my grad school lab that we used as web servers. But that was the problem - it was a premium designed product that wasn’t any better of a web server than a plain (cheaper) Dell 1U box.
Maybe things would be different now that they have a really unique chip. ARM as a whole ecosystem has more buy-in from other companies than PowerPC did. Their performance per watt and GPU design might make more of a difference this time around.
But I can’t see Apple being able to release and sell a data center centric server. I think others might have a better chance.
MacOS is not a server OS either and actually maturing M* support in Linux will take years. And for inference Nvidia is years ahead of Apple.
So best Apple can hope for is some niche market.
Apple Silicon has the "best CPU" in specific dimensions of low power usage and single-core performance with up to 512GB RAM ... but not in multi-core workloads with up to 4TB of RAM.
Apple optimizing the cpu design for the best power efficiency and single-core speed is the right tradeoff when the target products are consumer devices that run on batteries such such as smartphones, tablets, and laptops.
In contrast, the Intel Xeon, AMD Threadripper, Amazon Graviton cpus will beat Apple Silicon on heavy multi-core workloads because those chips are optimized for server utilization. The tradeoff is they're not suitable for personal consumer devices.
The rumours of a potential future Apple M8 + NVIDIA seems to be targeting AI inference tasks rather than general enterprise server workloads. Maybe because of Apple's unified memory architecture has faster RAM bandwidth than Xeon/Threadripper which is important for generating tokens/sec.
The issue is that the current Apple way of designing hardware doesn't really work for servers. An individual 1U server can easily cost $50,000 or more. Soldering everything together so that if a single memory chip goes bad you have to replace the entire machine instead of that one stick is going to make customers unhappy. Likewise, they expect to be able to add storage or network ports, and expect the software to have enough openness that the new systems will work with their existing enterprise SAN or what have you.
Obviously Apple could make open server hardware, but then people might start wondering why they can't make open laptop hardware.
Maybe I should have made it Mongolian gangster rap.
Or is this just “punch buggy” for Wikipedia references.
There are talks of John Ternus potentially selling Apple Silicon for server and AI use.
https://www.macrumors.com/2026/09/16/apple-may-return-to-ser...
Cause neither Apple is bringing back OS X Server, nor Qualcom is that great on server space, other than now maybe with their Modular acquisition.
They are releasing a server cpu soon I believe. It’s called Dragonfly C1000.
I'm looking for a cheap 8 Elite phone to root it and run a special kernel module that let me run LXC containers and run arbitrary Linux ARM64 program. And run a couple of services on it, maybe I can port Kubernetes whatsoever
Also we already have the PC clones with WinIntel for the alternative experience.
Support Linux OEMs if you care.
Now it's the other way round.
Just a few short years ago apple was the most valuable company period.
How big do you want them to become?
Also, how much bigger will open harware really make them? How many people really care about linux on a mac?
That NeXT acquisition turned out great.
However I am afraid I bought a macOS Air with M1 and recently a Neo for my daughter, because, while they are excellent computers, their closed ecosystem hardware goes directly against free software projects like Linux and others as well.
Microsoft has always been much criticised but, for its own interest, it promoted from the '90 a general purpose ecosystem of hardware where free software thrived. It just got a little bad later for the GPUs where proprietary, undocumented hardware becomes the norm and Direct3D the only viable abstraction layer to use them, making it Windows only. It was this ecosystem that made free software thrive, including Linux, GCC, Python, Perl, Tcl/Tk and a lot of wonderful project that empowered people and make computer better to use.
I think we must be aware that free software needs general purpose, well documented hardware to thrive and I think we - the power users, developers, etc - we should choose responsibly what hardware we purchase to choose those that are favorable to free software. We should be also aware that in some cases this means getting a machine with lower performances or that costs a little more but there is a trade-off to be done.
Some people may say: who cares about free software, we just want the best product that works and proprietary, closed, stuff is fine. However, when we think that we are forgetting that we enjoy today a free software ecosystem that was made possible from the open hardware ecosystem of 20 - 30 years ago. We need to remember that when companies are unchallenged they inevitably begin to produce shitty products and make the user's life miserable as long as they can. This can as long as to stifle progress and significantly diminish the usefulness of a computer system.
So, no, apple will not make open hardware but we, the developers and power users, may consider to not buy their computers when they use proprietary, undocumented hardware that is hostile to free software.
The part that makes the Apple trackpad good isn’t even the software integration with the OS. It’s the actual trackpad hardware, its surface and haptics. On the software side it’s largely just good acceleration curves.
I’d argue the reason very few competitors exist is just cost and lack of understanding what feels good and why , much like your own comment does.
edit: speaking mostly about macOS here, mobile OSes are much more locked down
It was only really open to ease the transition between intel and arm.
You can’t have a development machine locked down like iOS. As long as the Mac is the development machine at apple it will need to be able to run arbitrary code.
But they'll remove support for other OS's
Made the laptop miserable to use, so I went for a MacBook instead which wouldn't have such an obvious issue be unfixed because Apple has so few SKUs to support.
Also went for it in part because I do not like Windows and macOS typically would save me when Linux support was lacking. Just had to hope for a quick port of Asahi Linux to become (unfortunately didn't happen for my M3, but thankfully it's been progressing recently).
Then there is the Antennagate on iPhone, the buggy OS releases, and plenty other examples that I can refer from what is totally unacceptable given the price point of their products.
Others aren't better, but at least they don't pretend otherwise.
Where is the challenge or the fun or feeling of accomplishment in that?
Even with years of excellent, tireless reverse-engineering, Asahi's power management for M1 is not on-par with macOS, let alone what AMD/Intel chipsets get on Linux day-1. It's just inevitable that the hacking efforts will fragment into a fast-and-loose vs slow-and-steady mentality. And both of them will be equally right, since there is no official support for any of this hardware.
I don’t understand complaining someone who spent their own time doing something isn’t using AI instead.
Is it all that different from asking why someone painted something when AI could generate an image of a painting for you?
It probably wasn’t intended as such but the sentiment that comes across to me is “why did you bother?”
There are certainly people who would always be anti-AI. But there are a lot of people like me who continually bristle at how an AI booster seems to show up in every discussion implicitly implying you’re dumb for wasting your time by not using it.
You know how people got really sick really fast on HN of people asking “why not Rust” on any discussion vaguely related to programming no matter how little it made sense? Or “A blockchain would fix that”?
Yeah. That. Again.
I don't agree. In this case it's asking AI to analyze the high-res photos of Van Gogh to learn what paint he used, what techniques, how he held his pencil and how did his strokes look like.
Sure, you could then also ask to paint one for you in similar style, but that's not what is being suggested here.