You're totally right, but I see a lot of folks around me making absolutely bonker claims on these M1 devices. If I believed everything I'm told, I'd be expecting them to run complex After Effect rendering at 10000 frames per second.

What most people keep missing is the M1 jumped out way ahead on performance per watt on their first attempt at making their own processor.

I’ve seen the M1 Mac for less than $700. In that price range, there’s not much competition when it comes to computation, GPU performance, etc. and comes fairly close to much higher priced x86 Macs and PCs.

That’s why people are excited about the M1. You generally can’t edit 4k (and 8k) video in real-time on a machine in this price range—and what people happily paid 5x that price to get this level of performance just a few years ago.

They've been making their own chips for over a decade now. The M1 is not fundamentally different from the A series, and in fact seems to serve as a drop in replacement (in the iPad Pro line).

That being said, this is the first time they're confident enough in their own processor to rewrite their whole desktop software stack and market what are traditionally mobile-exclusive cpus as being superior to x86 in the personal computing space. And in all honesty, for good reason.

Pretty sure it wasn’t a matter of confidence, as Apple has been signaling making their own desktop processor for a while. There was an iPad launch a few years ago where Phil Schiller said that particular A-series processor had desktop-class performance like three different times.

Also, it’s the third time they’ve done a processor transition: 68K to PowerPC to Intel to ARM. There’s no doubt they’ve known for years they could do this if needed.

It wasn’t a rewrite; it was mostly a recompile, especially after Apple dropped support for 32-bit apps. iOS and macOS share many frameworks and APIs and that’s been running on ARM since 2007.

When Apple announced the switch to Intel in 2006, Jobs revealed they had Mac OS X running on Intel in the lab for years; there’s no doubt they’ve had macOS running on ARM for years as well.

Likely getting the logistics of software and hardware development to align at the scale Apple operates at is what took the time; not the lack of confidence.

This move was probably inevitable but it certainly didn’t help that Intel repeatedly missed deadlines and performance goals.

Honestly would not be surprised if it comes out at some point in the future that they launched the ARM-Mac project as soon as they'd successfully switched to Intel.

If not right after the switch to Intel, certainly when the A7 was on the drawing board, since it was the first mainstream 64-bit ARM processor.

What "but"? The claims are less bonkers with a perspective that acknowledges the significance of those physical differences.

Without the case, would the full internals (motherboard/CPU/GPU/PSU/ram/heatsinks/storage) of your desktop fit in your pants pockets? Because the M1 Mac Mini's fit in mine.

How much fan noise does your desktop produce? How much electricity does it consume? M1s are fast compared to almost everything, but they're bonkers fast compared to anything getting even remotely close to 20 hours on a battery at 3 lbs weight.

You're totally right that they are impressive in their own right, especially in their category. I'm not arguing that.

Also:

> How much fan noise does your desktop produce

Basically none since modern high end air cooling is essentially silent. Water cooler pumps are noisier though.

1- https://browser.geekbench.com/v5/cpu/compare/10508517?baseli...

2- https://browser.geekbench.com/v5/cpu/compare/9997439?baselin...

I bought an M1 MBA to replace a 2017 (might have been 2018 - I forget now) top specced Intel MBP and it cut the time to run my large Java test suite in half. Of course those are generations/years apart in technology, but the MBA doesn't have fans, barely gets warm, the battery lasts days, and it's the low end of the first generation M1.

I'll fully admit that there might be other systems out there that can do exactly that in the same fan less, portable package that I'm unaware of.

Fake news happens on HN too. Apparently. Sadly.

Hopefully it's mostly limited to discussions about Apple.

On my M1 Mini, I found it unstable and incredibly slow (for Postgres at least, using stock images from Docker Hub) and mostly quit using it. I upgraded to the latest version recently and it was still too slow.

For everything else, though, I find that the M1 Mini is significantly faster. E.g., compiling Python versions is way faster.

I see now that the official postgres images have other arch options compared to the postgis images, which are arm64 only. I wonder if that could make a difference...

And now I see that Docker Desktop actually shows a warning about potentially poor performance when using an amd64 image: "Image may have poor performance, or fail, if run via emulation."

This is good news, since I'd much rather use Docker to run postgres in dev.

Update: I switched to an official postgres image and it works great. I just had to create a derived image that installs PostGIS like so:

    FROM postgres:13-bullseye
    RUN apt-get -y install postgresql-13-postgis-3

Though as you note they're completely different classes of hardware (500W vs. 26.5W), a more Apples-to-Apples (so to speak) comparison would be native Ubuntu on both systems, though as I understand it Ubuntu for M1 is not complete and may lack GPU support and other important features.

Docker doesn't run natively on OSX, unlike Linux where it does. As a result, Docker Desktop for Mac spins up a linux VM, and this is where penalties begin to be payed.

I do wonder if Docker Desktop is smart enough to tailor the VM's architecture to the image's architecture? Moreover, is it smart enough to tailor both the VM's architecture and images architecture to the host's hardware architecture?

I'd be interested to see how these soft and hard architecture combinations actually change the performance characteristics:

    M1/ARM -> x86-VM -> ARM-image
 vs 
    M1/ARM -> ARM-VM -> ARM-image

I think it would be pretty difficult. Mac would effectively have to provide a fully compatible Linux interface in its kernel. It would probably be more likely to build a better fs-event bridge to guest systems, but considering MacOS still doesn’t have ext4 support, I’m not counting on it.