Live data from Hacker News

M1 Macs Review

sixcolors.com

201–210 of 391 posts

Re: M1 Macs Review

#201

The benchmark that's blown me away most so far: Marco Arment on Accidental Tech Podcast says the new Macbook Air is significantly faster to compile Overcast (a pretty big iOS app) than his 10 core $6000+ iMac Pro with 64GB RAM and a 160W CPU. https://overcast.fm/+R7DWXqqm0/1:08:11

I/O-bound, maybe? Could be misleading but still impressive

Yeah something is fishy here. Could be the new RAM which is way faster

Re: M1 Macs Review

#202
post #72
post #58

Earlier quoted context omitted.

> Linux native is not in the cards but it won't matter, it will work fine virtualized with hardly any performance penalty. If for some reason you insist on native linux, it's time to recognize that Apple hardware is not for you. How would one run a graphical Linux distro virtualized and by which software?

Parallels and VMWare Fusion would run it. Virtualbox as well. We can already do this on Intel Macs, Windows PCs, etc. Graphics are not a problem. Heck, we could since forever also run graphical macOS and Windows on virtualization apps...

Probably not VirtualBox, it's very x86-specific.

Re: M1 Macs Review

#203
post #5

After 6 years away from being "all in" on Apple, yesterday my base spec MacBook Air M1 arrived. Battery life is the key takeaway for me. Performance is fine but I'm used to a powerful desktop so it's not blowing me away but the battery life definitely is blowing me away. To run things under rosetta there's an arch command: e.g. arch -x86_64 brew install xxx - the arch preference is inherited by sub processes. I'm kin…

If a laptop's performance is on a par with a desktop's, and the fan's aren't going crazy, I'd call that a big win! My 2018 MB Pro has the fans running most of the time, and even then sometimes throttles back the CPU.

If your MBP fans are running all the time, you may want to open it up and see if it's filled with dust and lint.

Re: M1 Macs Review

#204
post #13
post #8

Earlier quoted context omitted.

The MacBook Air that GP mentioned doesn't even have a fan! That's certainly a nice benefit.

I really want the air for when I work outside in the summer. I worked from an iPad quite often with the magic keyboard, and it started throttling and sometimes overheated to the point of failure while outside in the sun. It's too cold in Chicago to test, but I would love to see if the MacBook Air can work while outside in direct hot sun - it's the only reason I'm considering the MBP.

It's too cold in Chicago to test

I used to have one of those massive 17" MacBooks (Best keyboard ever), and it used to heat up enough to turn my thighs red. Back then, Apple avoided calling them "laptops," perhaps for this reason.

Then I moved to Chicago, and found it was a wonderful machine for using outside on my balcony in cold weather because it would keep at least half of me toasty. If I started to get chilly, I'd fire up Virtual PC and snuggle in.

Re: M1 Macs Review

#205

Earlier quoted context omitted.

If we’re talking about hardware ergonomics the lack of a del key and placement of the fn key alone make me never buy a macbook again. I also hate the difference in cmd, alt and ctrl keys from other keyboards.

To be clear, my comment did not relate to either Apple or MacBooks in particular. Also, the value of each soft factor will differ from person to person. Good design highly relates to personal needs.

I guess if you redefine every feature someone cares about to be a "soft feature" then you are right. Kind of a pointless distinction though. Listing features matters because everyone cares about different features.

Re: M1 Macs Review

#206
post #154

Earlier quoted context omitted.

One thing to factor in here is that the MBA isn't cross-compiling. It's a RISC compiler writing RISC code. To be clear though, it doesn't matter in the end if the _result_ is more productivity. But it is worth noting that under the hood, the comparison is not entirely apples to apples.

Why would it matter whether you emit x86 code or ARM64 code? That code that produces either doesn't get magically faster when executed on their respective archs.

Edit: as had been pointed out here there is no x86 code generation happening so this doesn't apply.

Yes I have seen claims from people along the lines of 'x86 code has more optimization paths in mainstream compilers for historic and other reasons'. Not sure about that, certainly can't imagine such a significant difference.

Re: M1 Macs Review

#207
post #13
post #8

Earlier quoted context omitted.

The MacBook Air that GP mentioned doesn't even have a fan! That's certainly a nice benefit.

I really want the air for when I work outside in the summer. I worked from an iPad quite often with the magic keyboard, and it started throttling and sometimes overheated to the point of failure while outside in the sun. It's too cold in Chicago to test, but I would love to see if the MacBook Air can work while outside in direct hot sun - it's the only reason I'm considering the MBP.

why would you want to work under direct sunlight? you can barely see the screen in that condition

Re: M1 Macs Review

#208

I just bought a MacBook Air. I don't think it's an exaggeration to say that this redefines how we think about a laptop. Whereas previously a laptop was something that spent the bulk of its time plugged in, and only capable of short bursts of number-crunching/gaming due to thermal and battery limitations, I can literally get a good few hours of gaming untethered on this machine. And this is a base-level MacBook Air, I…

I can get a few hours of light gaming in on my Zephryus G14 on battery, and it has a vastly faster GPU than the one in the M1 such that I can also still play demanding games when tethered to a wall (it also has a VRR 120hz screen). A trick that the MacBooks can no longer pull of at all since eGPU support was killed. It's a fantastic experience, but in no way has having this laptop for the last 6 months "redefined how…

No idea why you are being downvoted, is HN really a group of Apple shills? I have the Zephyrus G14 also and love it. I'm closely watching the M1 but I really don't think it can compete with a GTX 2060. That matters a lot to me.

It also has an AMD 4900HS which while not an M1 is a fantastic processor.

Re: M1 Macs Review

#210

The benchmark that's blown me away most so far: Marco Arment on Accidental Tech Podcast says the new Macbook Air is significantly faster to compile Overcast (a pretty big iOS app) than his 10 core $6000+ iMac Pro with 64GB RAM and a 160W CPU. https://overcast.fm/+R7DWXqqm0/1:08:11

One thing to factor in here is that the MBA isn't cross-compiling. It's a RISC compiler writing RISC code. To be clear though, it doesn't matter in the end if the _result_ is more productivity. But it is worth noting that under the hood, the comparison is not entirely apples to apples.

It's not identical but no obvious reason why emitting ARM64 code from a compiler running on x86 hardware should be more onerous on ARM64 hardware.

Plus ARM64 is not really RISC in a meaningful way - it'll still be executing microcode under the hood.

Post reply on HN