If anything, Arm MacBooks have a huge advantage over my current (old) ThinkPad—battery life and performance per watt will be significantly better. I haven't read much more about the new virtualization tech that was announced at WWDC, but that's also exciting for running Linux on the Mac.

What about Arm Macs pushes you away from the platform?

not the op, but for me...

1. the big sur ui is not very attractive (i know it sounds arbitrary, but big sur just looks very non-desktop, very fischer price, and i have no alterantive since catalina wont be supported)

2. virtualization for x86_64 apps, that wont last forever, and at some point those x86 binaries wont work anymore (i already lost a ton of apps with catalina...)

3. for my work i already need vmware alot, and the idea of running even half-speed at best is a non-starter

--

that being said, as a consummer oriented device, or an iphone/ipad app developer machine i think its defintely going to do well

>1. the big sur ui is not very attractive (i know it sounds arbitrary, but big sur just looks very non-desktop, very fischer price, and i have no alterantive since catalina wont be supported)

I find these "this look fischer price/non-desktop" like hearing goth teenagers wanting something darker and more edgy -- well, the "pro" mentality equivalent, where if it doesn't look boring and beize it's not good.

It's a mighty fine look, nothing fischer prize about it, and hardly that different from the previous iteration. If anything it much less fischer prize than the original Aqua interface and its 4-5 next iterations, which was praised as a great looking ("lickable") UI.

to be fair, i dont mind the colorfulness or any of that, but the huge bezels, margins and font sizes are a huge waste of space in a desktop environment

its kind of obvious that for the need to compromise the ui of mac in order for ipad/iphone apps to look "at home" and im sure in apples calculus its worth the hit in the near term, but for this person its a bit too far...

*ive been using the big sur betas on and off for a while now

> Aqua interface and its 4-5 next iterations, which was praised as a great looking

sure, aqua got a lot better after the next iterations, so lets keep an eye out and see what happens...

>its kind of obvious that for the need to compromise the ui of mac in order for ipad/iphone apps to look "at home"

I think AS laptops are also getting touchscreens (now or on the first design refresh).

But I welcome bigger screen elements myself, it means better hit targets (Fitt's law), and less cluttered UIs -- easier on the eyes too.

As for the wasted space, it's not like they're removing menu items or icons or cutting out features to accomondate the larger buttons. Just larger looking bezels and edges.

If that's the reason why, then I guess I don't want a touch screen mac. Looking at my screen now, I would fat finger just about everything if it was a touch screen as sensitive as my iPhone (where I fat finger quite a few of things, and my fingers are really quite narrow). The mouse is a precision instrument. Your finger is not. I'd hate for the UI to suffer due to catering toward a less precise input I have no intention of ever using.

Well, a UI requiring less precision to accomondate fingers wouldn't suffer for mouse use.

If anything, it's the contrary: it would become easier to use with the higher precision mouse, because targers would be even larger.

The key insight here is that they wont make the mouse lose precision: instead, they make the targets the mouse has to hit bigger -- which, given that the mouse retains the same precision, makes it even easier to hit them!

It would suffer by loosing data density. I don't want to loose effective screen real estate where it is already limited on a laptop. I try to keep my browser windows as slim as possible for this reason.

VMWare won't run x86 Windows or Linux at all, much less at half speed: https://developer.apple.com/documentation/apple_silicon/abou...

But the server space will within a few years begin transition to ARM as well . Look at Amazons graviton ARM cloud offerings. It beats x86 on bang for buck.

Also, Apple services in Big Sur circumventing VPNs and Little Snitch is a complete nonstarter for me. I’m staying on Catalina with an Intel Mac for the foreseeable future.

You'll have to run ARM linux on the Mac. x86 Linux and Windows cannot be virtualized with Apple's Rosetta 2: https://developer.apple.com/documentation/apple_silicon/abou...

That doesn’t mean they can’t be virtualized by a third party product. We had x86 emulators on PowerPC Macs.

> What about Arm Macs pushes you away from the platform?

I need to be able to run Windows in Fusion or Parallels without emulating x86.

Speaking from experience, just do not try any fringe distro. Pick one of the top 3 popular ones - they are much better in little things that you will appreciate.

It’s good advice but I’m not sure how you count “top 3 popular distros” these days.

A lot of people into Linux seem to go through a phase where they try out a rolling release distro like Arch or Gentoo, and then after a couple years get sick of dealing with broken systems & switch to something more stable like Debian/Ubuntu/Fedora/Mint/etc. As I get older, the amount of system administration and customization of my system that I want to do is dropping rapidly. I barely even change the desktop wallpaper any more.

Community size is good approximation and that is what drives features and your ability to find solution to a problem quckly.

I would say top desktop distros (from most popular to least) are: Ubuntu, Debian, Fedora, Mint.

Anything outside of this list is for tinkers who are not easily frustrated with issues.

Yes, and I’d say more than that—tinkers who are specifically interested in tinkering with Linux desktop administration.

I’ll do a ridiculous amount of work on some hobby project, but if a system update breaks my WiFi adapter, then I’m tinkering around and facing frustrating issues with WiFi drivers, which sucks, when I really want to be tinkering around and facing frustrating issues with my hobby project, which is awesome.

It’s like: https://www.reddit.com/r/antiwork/comments/asfgcj/bad_screen...

Same reason I switched to AMD GPUs. Getting the Nvidia cards working on Linux is just a distraction from whatever I really want to do.

If you value the attention to aesthetics and UX Apple has you might want to give Pop! OS a try. It's based on Ubuntu.

Ubuntu is a pretty good substitute for macOS. My key issue with Linux is the small selection of productivity apps: markdown editors, code editors, drawing packages, password managers like 1Password. Feature rich terminal apps like iTerm2. Well designed Git UI clients like Tower etc.

The rest is pretty good. Like the core UI, file manager, virtual desktop management, configuration etc works very well. Much nicer than the Windows mess.

Cool — one of the major blockers on moving to Linux for me was fractional scaling support. I have 4K monitors but can't use them with tiny text.

I like PopOS a lot, but they didn't have fractional scaling support. It looks like they do now as of Oct 5!

I've been using 4k with linux for years. On large desktop screens I don't find fractional scaling to be necessary. The important text in apps that matter like terminals, text editors and browsers, chat clients, etc, has always supported scaling well. The only thing that lacked support were system dialogs and title bars. 2x was fine for those until fractional support came out in gnome recently.

I'm already trying out Manjaro XFCE and I'm liking it a lot.

Because Apple is switching to Arm for better performance?

Well, Apple is moving towards increasingly locked down systems. For example, it is very difficult if not impossible to install Linux on the most recent Macbooks. Previously when Apple stopped supporting a Mac, you could install Linux on it. I suspect with the ARM systems going forward, that will become even more difficult.

In other words, it is not just about better performance. It is also about Apple removing your freedoms with your own hardware in the name of 'security'. (I am not saying that there isn't increased security, but at what price?)

I would certainly love if Apple made this device friendly to Linux use. The lack of Bootcamp support makes it seem unlikely, but I can hope. The fact that they made a point of demoing Linux in a VM is (very) slightly encouraging.

The lack of Bootcamp is because Windows didn't want to support selling on Windows for ARM. https://www.theverge.com/2020/6/24/21302213/apple-silicon-ma...

It's Microsoft's fault that they don't sell Windows for ARM so including Bootcamp would have no reason (as despite us power users, the primary use case for Bootcamp (it's even in the logo) is running Windows).

If they want to lock down the system there's nothing about x86 that prevents them from doing so there. So that's a faulty argument.

At around $25 a month if your laptop lasts for 6 years

I'm not a fan of these computations. If the computer functions, you should still be able to use it, regardless of how much existing use you get out of it.

I think I got 18 months out of my last two MacBooks before they gave out. Moved to windows desktops now. Much better value propositions, especially now as we don’t have to go out any more! :)

You're very unlucky. I'm still using the same macbook pro I got in 2013.

Ouch.. I guess I was lucky.

I'm on a late 2013 which I've only actually paid for a battery replacement on, so I can anecdote believe you.

I'm not in love with macos though and there's more good options nowadays than there was in 2013.

Not at all. Some people, me included, look at hardware and it falls to bits. Really they're not particularly durable machines. Apple could do a lot better.

Thinkpads are ok though. I haven't murdered one of them. Yet. I've tried and failed several times from coffee to catapulting them and they just eat my hamfistedness.

I just bought a replacement iPad for my daughter for the one that we bought in 2013 though. Stuff can last a long time.

Nah, our household has a collection of macs none newer than 2017. My phone and iPad are from 2014 and my gf’s from 2015, and I have a watch 0. They all work fine and, except for the watch and geriatric macs, continue to get software updates.

This will be a good excuse to start a mass upgrade but I won’t be an early adopter.

Or the parent was unlucky - with some faulty production units, and the general case for most people, with hundreds of millions of units sold and the highest customer satisfaction ratings, is not that.

better performance only if your desired software runs on the new processor. if not then what do you have?

I am a light weight user in the Mac world, browsing and some games. However the few games left after the Catalina change to drop 32 bit support are still important to me. One of the publishers has preemptively dropped support for Mac and its near impossible to tell if others will continue support into AS.

I know, there is a crowd out there that loves to declare "you should not use your Mac for games" along with "Buy a system for games then". Well let me be blunt, if it cannot serve both purposes then it serves none which means leaving the platform behind.

So I look at AS with excitement and trepidation all at the same time with the realization my current Mac may well be my last.

Didn’t Wendker demo a video game running under emulation? That should work fine for your 64 bit games.

And it’s not like your old hardware will suddenly stop working. If you want to place PS2 games you need an old PS2, and this is the same.

Also, along the line of your old hardware still working: no need to upgrade any time soon.

I’m hopeful we’ll see something similar to the move Steam has made for old Pc games. With a little love, devs can usually get an old Windows 95 game running as we’ve seen in the past few years.

If the Apple silicon is powerful enough, it might not matter if there is a huge performance hit for the emulation. It’ll be an interesting next 5 years for sure.

> better performance only if your desired software runs on the new processor. if not then what do you have?

Well of course... Give it some time for the patches to come. Recompiling for multiple processor architectures is easier now than it used to be.

> I know, there is a crowd out there that loves to declare "you should not use your Mac for games" along with "Buy a system for games then". Well let me be blunt, if it cannot serve both purposes then it serves none which means leaving the platform behind.

I think you SHOULD use a Mac for games. Most people saying that are not in fact Mac users. The ARM transition isn't going to push Mac games out of existence. In fact it'll probably push Microsoft to properly treat ARM as a first class citizen so companies will be compiling their software for two different architectures anyway.

Rosetta 2 translates x86 instructions to ARM and cache result. That translation seems quite good given that they demonstrated a complex 3D game running translated code on ARM hardware much weaker than what Apple will deliver in laptops.

Honestly I think x86 games will run better on these ARM laptops. Keep in mind x86 on laptops don’t work great. Apple could outperform most of their x86 laptops with a passively cooled ARM tablet.

An ARM laptop with active cooling will destroy their x86 lineup.

Besides you also get the whole iOS games catalog now which is much larger than the macOS game selection.

The iOS catalog is moot until devs starting making real games for the platform. Apple Arcade and a few indie studios have done great, but for the most part, iOS games are micro transaction laden trash. It will take years before we see any real desktop level games developed to run exclusively on ARM macs.

I think the success of gaming on the platform will come down to game engine devs like Unreal and Unity. If they can optimize exports to target Apple silicon with minimal overhead, that’s the best bet to see gaming live on. Otherwise, I don’t see how any AAA studios can find it worth their time to build for 2 completely different architectures.

One interesting exception might be EA, as The Sims is hugely popular among casual crowds that might have a MacBook for college or something. But in the case of EA, they also will create a new Sims game and then milk it for a decade, thus paying back their R&D costs. Other studios who frequently make new games and want to push the hardware may not be as inclined to spend that with such a small install base.

> Because Apple is switching to Arm for better performance?

Because I wont be able to run Windows in Fusion without a very big performance hit.

[benchmarks needed]

They will come, when the computers are available. But Apple wouldn't invest in doing it at great expensive if it didn't produce a better more competitive product.

Look at the transition dev kit. Also, look at the intel’s iris vs apple’s gpu

Shouldn't you compare to Intel Xe for integrated graphics performance?

No, when you're trying to disrespect Intel it's customary to compare announced imaginary parts to obsolete Intel parts.

Look at what from the transition kit? The only thing available show it having single-core performance that would have been state-of-the-art ten years ago. Is there something newer and more impressive?

I couldn't transition to a Linux laptop as I don't think it has optimal battery life nor does the trackpad feel as Mac-like as I need it to be. But I could see ARM Mac laptops being used by a lot of people for connecting to real systems that are still x86. In other words, the Mac laptop is a browser machine or a display to a real work VM.

I think ARM is the nail in the coffin of the non-laptop Mac brand though.

You say that because you think ARM gives toy performance when everything actually points to Apple Silicon will be a performance beast. It may not beat the big x86 gaming rigs but I am quite confident that the highest performance laptops of any kind in the next few year will all be ARM based.

iMac and Mac Mini are also small enclosures which will benefit from ARM performance wise.

All that matters is really watt per performance and dollars relative to performance and there ARM already beats intel. To beat an x86 intel is just a matter of throwing in enough cores.

AMD will be harder as their performance relative to cost is better than ARM and intel. But there is nothing special about beating intel at this point.

Why you don’t think arm can function on iMac scale?

https://www.macrumors.com/2020/11/02/apple-smaller-mac-pro-a...

Yeah. I’m a long time Mac user. Like 17 years, I’m keen on it because UNIX and pretty.

I was looking at Arch + KDE plasma and it’s really tempting.

I returned a 16inch MBP recently as it kept crashing on shutdown. macOS on intel is a mess and so is the T2. And I’m a fanboy.

I’ve made a deal with myself. If these new Apple Si machines are blow away I’ll try again. Else I’m buying a Lenovo Legion and moving to Arch.

Not sure what to do for a phone but probably stay with iOS.

>I returned a 16inch MBP recently as it kept crashing on shutdown. macOS on intel is a mess and so is the T2. And I’m a fanboy.

macOS on Intel is a mess because a particular 16inch MBP "kept crashing on shutdown"? As if Windows and Linux running machines don't have cases where they keep running on shutdown?

macOS/T2 has bugs like all software, but this is not some general state of macOS/T2, which runs fine without "crashing on shutdown" for hundreds of millions of other people (including those with MBP16).

How about it's some particular production run / unit, and or some particular app?

(Which can happen for any vendor / OS)

Clearly I understand that.

I had 3 replacements. Each one I could reproduce a hard crash on reboot. There are many many kps.

It’s widespread.

That's fair, that's a lot of replacements.

I'm currently waiting (and cursing) Lenovo for delaying delivery of the AMD T14 I paid for in September.

I'm a bit worried for my precious free time that will be wasted on fending off the thousand-cuts onslaught of bugs, but I'm willing to try.

It's gotten to this Tim (and Jony). After ~15 uninterrupted years.

Moving from Intel to ARM is a reason to stay with Mac if you ask me. Intel has been holding computing back with their inefficient architecture, and anything that drives innovation away from x86 is a good thing in my mind.

Much as I dislike some of Apple's practices I think making ARM more mainstream is a trend I want to see continue.

Exactly, and maybe arm clouds will be more popular

No idea why this is being downvoted. Does anyone here like the X86 ISA? Sure its been made fast, but that's been quite an uphill battle against things like its complex instruction decoders and limited decoder parallelism. If the same engineering had been put into making ARM fast, ARM would be far faster than X86 is today.

I think that you're going a bit too far with 'far' faster - there isn't any evidence that this is the case and a lot of engineering effort has gone into making Arm fast - not least from Apple.

Apple aren't moving to Arm for no reason though so I think it's worth considering the reasons why this might be the case: power efficiency (probably the most important), price, ability to incorporate custom silicon, a single ISA to support, full control over the stack, access to best manufacturing process.

On the last point, I think Apple (Tim Cook especially) has probably lost faith and patience with a company that has overpromised and underdelivered for a number of years.

The fact is that x86 has 4 decades of backwards compatibility which along with the PC architecture means a reasonably open and documented platform. Every ARM system is different and they are all lacking the heritage of public documentation that the PC has. In other words, it's not about performance but freedom. I doubt Apple will ever release as much detailed information on its own chips as Intel and AMD have.

x86 is plenty fast anyway, and still wins on code density in single-threaded scalar (i.e. NOT using the vector units) branchy code of the sort commonly found in average applications.

Genuine question - what is lacking in Arm documentation that Intel provides?

Essentially all the supposedly merchant SoCs are proprietary. Instead of shipping data sheets, winbond or mediatek want to ship you a reference design with reference linux or android build that only works because of a huge binary blob.

There’s a small hardware consulting company in Palo Alto whose main differentiator is that they have a bunch of folks in China who manage to generate docs they can use. I’m guessing this involves a lot of under the table leaks and late night alcohol-fueled discussions.

Thanks, so it's really about drivers etc rather than the architecture, which makes a lot of sense. A bit surprising that there hasn't been more convergence in the Arm world.

The lack of convergence is how the companies differentiate themselves.

Will 2021 be the year of desktop ARM?

Could be. Pi and Mac announcements on the same day!

I already did. Couldn't be happier.

Don't let the door hit you on your way out...