Live data from Hacker News

M1 Macs: Truth and Truthiness

daringfireball.net

191–200 of 882 posts

Re: M1 Macs: Truth and Truthiness

#191
post #33

Earlier quoted context omitted.

> In fact this makes me wonder: if Apple chips are so far ahead, why are iphones not the fastest, best smartphones by a wide margin? Apple has pretty consistently had the fastest SoCs for phones and tablets for a few years now, particularly in single core performance. Here are some benchmark results of the iPhone 12 versus some of it’s Android competitors [1]. Notice that last year’s iPhone 11 Pro Max still outperfor…

Yeah that's what puzzles me, why has chip superiority not translated into an objectively better phone? For laptops the M1 laptops are best in class in everything, battery and performance.

To a large extent, current Android flagship CPUs are "good enough" for the things people do with phones. Sure, current iPhone chips are much faster, but it doesn't matter that much for the vast majority of consumers (even for games, mobile games tend to target the lowest common denominator to a much greater extent than PC games; no-one's going to launch a mobile game that runs on the latest couple of generations of iOS and nothing else).

iPhones do likely get better longevity out of their faster hardware, mind you.

Re: M1 Macs: Truth and Truthiness

#192
post #99

It turns out that the benchmarks for M1 vs latest generation Intel & AMD CPU's are indeed overblown, and it is an incremental improvement more than a great leap forward. The source of the confusion has been the benchmarking software. To saturate one core on an Intel processor you need to run two threads, because that's the way they are designed. So the single thread benchmarks that have been used so far have been usi…

Eh? I mean, by the same token you could say that single-core benchmarks of Intel chips are invalid, because you need eight threads or whatever to saturate a POWER7's multi-way SMT.

The main purpose of single-threaded benchmarks is to approximate performance for things which are actually single-threaded.

Re: M1 Macs: Truth and Truthiness

#193
post #159
post #126

Earlier quoted context omitted.

Wait for an 8-core (or the rumored 12 core) variants. Those will embarrass your threadripper.

>Wait for an 8-core (or the rumored 12 core) variants. Those will embarrass your threadripper. It will not, my TR has 64 cores 128 Thread and you know whats the best thing? 128GB Ram...and imagine that..they aren't even soldered...and i even can exchange my GPU..or slap in 2 others. And if apple ever brings out a M1 Workstation, i can probably buy 3 other Server/Workstations with that money.

Your TR is what, 280 watts TDP? nearly 20 times more than a M1. And is 30% slower in single core.

Firestorm Cores use less than 4 watts each. A 16 firestorm core Mx will produce roughly the same multicore performance as your TR, with a less than 70 watt TDP.

And a $65 M1 dusts a 4800h that costs over 4x as much and using 1/3 the power. Top end ThreadRippers cost $2,000 to $3,000 each. You can bet an Apple Mx ThreadRipper killer isn’t going to cost more than $500 given it only needs 4x the cores of the $65 M1.

Re: M1 Macs: Truth and Truthiness

#194
post #12

I'm not sure I see most of the disagreement here. As Gruber says, Moorhead is mostly complaining about software incompatibilities. But that's part of the total package--especially with Apple which is about selling the total experience. Personally I'll probably switch sooner rather than later especially given that I don't run a lot of native apps on my laptop. But it seems pretty reasonable to critique a new architect…

Complaining about incompatibilities in software used by less than 5% of Mac users and ignoring the strong compatibility of software used by 95% of Mac users seems misleading.

Re: M1 Macs: Truth and Truthiness

#195
I don't doubt Apple's CPUs are great, but they coincide with the Mac's transition into a product I, otherwise, no longer wish to use.

I want access to my boot volume. I don't want cloud anything.

So Apple's technical prowess now makes me sad. I can either use an OS that doesn't head further every year in a direction I hate, or I can have a fast computer.

Re: M1 Macs: Truth and Truthiness

#196

I don't doubt Apple's CPUs are great, but they coincide with the Mac's transition into a product I, otherwise, no longer wish to use. I want access to my boot volume. I don't want cloud anything . So Apple's technical prowess now makes me sad. I can either use an OS that doesn't head further every year in a direction I hate, or I can have a fast computer.

Are M1 macs faster by significant factor for dev purposes though? I remember reading compiles are slow on MacOs, but don't remember what was the reason.

Re: M1 Macs: Truth and Truthiness

#197

I submitted the same exact link 8 hours earlier. This HN behavior is very frustrating. https://news.ycombinator.com/item?id=25285489

To me, this is a community of people that want to share interesting topics and information, and contribute to that sharing and discussion. The greatest prize of all this is how much accurate and useful information we all get out of this. I've become a veritable encyclopedia, a proficient and effective software engineer, a resource on many topics for friends and family, and continue to be an enthusiast of computing so…

Wholeheartedly agree and couldn’t have said it better myself.

Aside from the peer approval that you mentioned, I also enjoy the idea that I can somewhat steer the conversation and see what smart people have to say on the topics I’m most interested in.

Re: M1 Macs: Truth and Truthiness

#198
post #159

Earlier quoted context omitted.

>Wait for an 8-core (or the rumored 12 core) variants. Those will embarrass your threadripper. It will not, my TR has 64 cores 128 Thread and you know whats the best thing? 128GB Ram...and imagine that..they aren't even soldered...and i even can exchange my GPU..or slap in 2 others. And if apple ever brings out a M1 Workstation, i can probably buy 3 other Server/Workstations with that money.

Your TR is what, 280 watts TDP? nearly 20 times more than a M1. And is 30% slower in single core. Firestorm Cores use less than 4 watts each. A 16 firestorm core Mx will produce roughly the same multicore performance as your TR, with a less than 70 watt TDP. And a $65 M1 dusts a 4800h that costs over 4x as much and using 1/3 the power. Top end ThreadRippers cost $2,000 to $3,000 each. You can bet an Apple Mx ThreadRi…

How does that matter though? His machine isn't sitting on his lap or running on a battery. Electricity isn't that expensive.

Re: M1 Macs: Truth and Truthiness

#199

Earlier quoted context omitted.

Likewise, we already knew that "emulation" doesn't imply that performance will be unbearably slow. Apple’s Rosetta 2 doesn't do emulation ; it translates the x86 code into ARM code on the fly and runs it, which is why most apps run faster on M1 Macs than they do on Intel Macs--and comparably priced PCs. Now that it has happened, we will likely see more from the PC industry. Probably not. If Intel and AMD were capable…

M1 is still just a tablet SoC in a laptop form factor. We don't know what the efficiency will look like after you add all the I/O you need for a real desktop, we don't know how the memory subsystem will deal with 4 more cores that are just as fast. What's the point of having fast cores if you cannot feed them? We don't know how lower yields for bigger chips on the newest nodes will affect their costs. People act like…

M1 is still just a tablet SoC in a laptop form factor.

That wouldn't be good enough for production hardware. The developer units had a previous iPad’s A12 SoC in an enclosure for developers to test things on. For example, the A series doesn't support virtualization because that's not required on a phone or tablet. But the M1 supports this and other laptop/desktop features. Even the A14 only has 4 cores compared to the M1's 8 cores. This isn't just a tablet SoC.

We don't know what the efficiency will look like after you add all the I/O you need for a real desktop, we don't know how the memory subsystem will deal with 4 more cores that are just as fast. What's the point of having fast cores if you cannot feed them?

Perhaps you haven't been paying attention, but the Mac mini is a real desktop. And all of the M1 Macs have blazingly fast memory access. Remember, it's an SoC—the memory is on the same die as the CPUs, GPUs and they all have equal access to it. No bus to go across. That's why it's so fast.

People act like Apple changed the game or is light years ahead of everyone but they literally haven't made a proper desktop-class chip yet.

The M1 already has features no "desktop" chip has, such as 8 instruction decoders. This may not sound like a big deal, but none of the Intel or AMD chips--not the threadripper or the Zen3—has more than 4. Why this matters:

    "It is because the ability to run fast depends
    on how quickly you can fill up the ROB with
    micro-ops and with how many. The more quickly
    you fill it up and the larger it is the more
    opportunities you are given to pick instructions
    you can execute in parallel and thus improve performance.

    Machine code instructions are chopped into micro-ops
    by what we call an instruction decoder. If we have more
    decoders we can chop up more instructions in parallel
    and thus fill up the ROB faster.

    And this is where we see the huge differences.
    The biggest baddest Intel and AMD microprocessor
    cores have 4 decoders, which means they can decode
    4 instructions in parallel spitting out micro-ops.

    But Apple has a crazy 8 decoders. Not only that
    but the ROB is something like 3x larger. You can
    basically hold 3x as many instructions. No other
    mainstream chip maker has that many decoders in
    their CPUs."
So Intel or AMD will just add 8 instruction decoders to increase their throughput, right? Nope:

    "However on an x86 CPU the decoders have no clue
    where the next instruction starts. It has to
    actually analyze each instruction in order to see
    how long it is.

    The brute force way Intel and AMD deal with this
    is by simply attempting to decode instructions at
    every posssible starting points. That means we have
    to deal with lots of wrong guesses and mistakes which
    has to be discarded. This creates such a convoluted
    and complicated decoder stage, that it is really
    hard to add more decoders. But for Apple it is
    trivial in comparison to keep adding more.

    In fact adding more causes so many other problems
    that 4 decoders according to AMD itself is basically
    an upper limit for how far they can go.
    *This is what allows the M1 Firestorm cores to
    essentially process twice as many instructions as
    AMD and Intel CPUs at the same clock frequency.*"
As I've said in other threads on HN, there's no way Apple's first laptop/desktop chip should be competitive with AMD, but it is:

    "As far as I remember from performance benchmarks
    the newest AMD CPU cores, the ones called Zen3
    are slightly faster than Firestorm cores. But
    here is the kicker, that only happens because the
    Zen3 cores are clocked at 5 GHz. Firestorm cores
    are clocked at 3.2 GHz. The Zen3 is just barely
    squeezing past Firestorm despite having almost
    60% higher clock frequency.

    So why doesn’t Apple increase the clock frequency
    too? Because higher clock frequency makes the chips
    hotter. That is one of Apple’s key selling points.
    Their computers unlike Intel and AMD offerings barely
    need cooling.

    In essence one could say Firestorm cores really
    are superior to Zen3 cores. Zen3 only manages to
    stay in the game by drawing a lot more current and
    getting a lot hotter. Something Apple simply chooses
    not to do."
So yes, Apple does have catching up to do.

Come again?

The Zen3 barely beats the M1, which runs at 60% of the speed and a small fraction of the power. It can essentially process twice as many instructions at the same clock frequency. They're already ahead in many key areas, with performance per watt being the most obvious. There are issues with the x86 architecture, like instructions ranging from 1 to 15 bytes, which limits Intel and AMD from being able to process as many instructions per clock cycle as ARM processors in general and the M1 especially.

I’m quoting from the article "Why is Apple’s M1 Chip So Fast?", which has a lot more technical details: https://erik-engheim.medium.com/why-is-apples-m1-chip-so-fas...

Re: M1 Macs: Truth and Truthiness

#200
post #177
post #171

Earlier quoted context omitted.

Apple's ARM SoCs are definitely ahead, but it's not like the competition is hugely far behind. Top of the line Android phones and tablets perform well enough for most desktop tasks IMO. It's mostly the software and I/O that makes them unusable as such.

The numbers say otherwise.

The numbers show that the other chips are slower, but they don't show that they are "unusable".
Post reply on HN