I usually avoid hype. But the M1 is just unreal. I was back in the office for the first time in months, and since getting my M1 MBP...my overclocked i5 hackintosh with 4x RAM just disappointed. It didn’t feel slow, but it stopped feeling fast.
I just got rid of a 10 month old 16" MacBook Pro for an M1 Air. I wasn't unhappy with the performance of the 16", but the heat and noise were just misery inducing.
The thing was unusable on my lap. After 30 minutes or so just coding, it would get so hot I had to put something between it and my legs. My partner could hear the fans from the next room. And to cap it off, since Big Sur I've been unable to get more than 4 hours off the battery—usually more like 3.
Compare that to the Air: I'm losing two USB ports which is annoying, the speakers are nowhere near as good, I miss the bigger screen, and I'm down 16GB RAM (although in retrospect, I never needed 32GB for my workloads anyway). But in return, I get almost no heat, zero noise, nuts performance (my test suites run 2x faster on average, despite running on two fewer cores), and so far 10-12 hours off a charge easy. And I get real function keys instead of the Touch Bar.
Happiest I've been with a computer in a long long time.
I'm at a desk the majority of the time. But, sometimes I want to work on something casually while I'm watching TV. Or sometimes my partner is on an important call so I have to vacate our shared home office.
I don't think it's an unreasonable expectation to be able to comfortably use a laptop on your lap for short periods.
You have what many would consider the last 'good' mbp. After that one, Apple pushed to make them thinner and try to make them faster, but really went too far with the thermal envelop. It's 100% Apple's fault for the thinness, but Intel didn't do them any favors with their delays either.
I had a 2008 15" MBP, the original retina 15", and currently have a work issued 16" i9.
The 16" gets toasty, but those first two could get scalding hot. And I was concerned about that as I only had 13" in the intervening years, but the 16" is nothing like those at all. Never once had a situation where it was so hot that I had to remove it from my lap but regularly had those situations with the others.
And I'm pushing it super hard, it rarely gets better than 3-4 hours on a charge. I don't think people are remembering just how bad those earlier MBPs were.
Interesting. I have a 2015, 2017 mbp and m1 MBA sitting here. The 2017 is the worst of the bunch in terms of fan noise and heat. It is faster than the 2015, and both are 15". I don't have a 16", so maybe they fixed some of the issues on that one.
Mostly just WebStorm and/or VS Code. Neither are particularly light applications I know, but they shouldn't be taxing on a machine of that spec. For comparison, my previous machine (an entry spec 2015 13" Pro) would only spin the fans up when I was actually pushing it.
Throw in some horribly unoptimised crap like Teams and it was like sitting next to a tiny little aeroplane all day long.
Docker for mac is ridiculously bad to use. We use build containers at work, and Big Sur definitely gets mad at the VM activity & the high disk activity. My 2019 15" i9 MBP sounds like a jet engine half the time.
> (my test suites run 2x faster on average, despite running on two fewer cores)
I replaced a 2017 i7 mbp with a m1 MBA, and see the same results running my java test suites. I ran them a bunch of times thinking there had to be some error.
I use logic to make music and the noise from the fan on my 15" MBP goes beyond annoyance to making it hard to perform my task as I can't hear subtleties without cranking up the volume to compensate for the fan. As soon as the 14" drops later this year I'm on it.
Wow, I need to go listen to 16 inch, when I listened to new Air for the first time, I thought 'wow, those are surprisingly capable speakers for such a lightweight machine'
The speakers on my 16” are good enough that I don’t much miss my old school stereo with bookshelf speakers which I parted with in a move a couple years ago. They certainly aren’t as capable of that volume, but they’re loud enough to fill my small 1br house, and with excellent audio quality (probably comparable to my old setup).
Apple has consistently provided good audio iterations on its laptops, so I’d assume the same will be true on the inevitable M-series 16” replacement.
We all were complaining about the lack of innovation with iMacs, barely upgraded, MacBooks with its superficial gadget “emoji bar” as the sole innovation (and the butterfly keyboard), felt insulted with iMac Pro because it feels like “You want better? Here, pay.” It felt like a decade of maintenance, milking the cow.
All along, half of the headquarters must have been secretly working on the M1...
When they released the M1 macs without a design change, my initial take was that it was a cowardly move: they didn't want to attach a big design change and risk calling a lot of attention to what might be a flop. But I actually think it was more of a boast: showing exactly how changing only the processor changes everything in a clear A/B test scenario.
That's not the reason, it was almost certainly due to Rosetta and nothing else.
Changing the processor architecture over for the first time was going to be a massive undertaking (I'm sure there are still folks there who have nightmares about universal binaries from '06). They had no way to run a wide public beta of the hardware before releasing, so had to play it as safe as possible.
The only sane way to migrate something so fundamental is to change absolutely nothing besides the processor, so that you have a side-by-side comparison between the previous generation and the new one out in the wild to debug problems that come up. If they had added [Face ID / a touchscreen / new mechanics for the keyboard] they'd need to debug whether those changes were causing the bugs rather than the M1 changes.
That's why I've fought myself to hold of - I'd expect the next MBP to be a fundamental redesign.
I can live without a touchscreen, but please just give me a Face ID laptop Apple.
> [...], but please just give me a Face ID laptop Apple.
Why is that so much more convenient than eg the fingerprint scanner?
(Even entering my password is pretty quick for me.)
I haven't use any form of Face ID on any device so far. So I am genuinely curious. I do see the appeal of finger print auth over having to type a password or a code on the phone.
I have a Magic Keyboard for my iPad Pro that I use as a Zoom machine and for saving thoughts, doodles, and for distraction free documentation writing. Aside from that it’s also my main tablet for content ingestion.
Using it docked in the keyboard makes it feel like a laptop. FaceID for filling in forms, authenticating with SSO, it’s all basically instant and requires almost no thought from me, not even a single cycle of brain power needed.
I will say it’s not like, amazingly more convenient than the TouchID sensor on my MBA, but it’s nice to have.
It more or less reverts to the same experience of being on Chrome on any other platform where your passwords just automatically are input, without the obvious security compromises.
I actually strongly prefer the fingerprint to face-authentication (especially in the age of masks). Face ID is just so much more responsive to things like lighting conditions, and you pretty much have to look directly at your phone, so you can't casually unlock while the phone is sitting on the table while you're at a coffee date or something. I'm always touching my phone/laptop anyway when I want to use it, so I honestly see no advantage to not just using that touch itself to unlock.
FaceID uses infrared, lighting conditions do not matter.
You can disable the requiring of looking at the phone and unlocking via FaceID in settings, I regularly unlock my phone without holding it so I am "dead on looking at it".
On the controrary it does not work in complete darkness, dim light is fine.
My iPhone 12 never unlocks with Face ID in the middle of the night, and neither did my iPhone XS. I can however see a faint red light flash from the sensor when it tries, I guess this is spill-over into the visible spectrum from the IR.
That's my thinking as well. But lots of people seem to really like Face ID, and I strongly prefer to assume that there's just something I'm missing, instead of assuming those people are all idiots.
I have an iPad Pro with FaceID and the smart folio keyboard and that is a close analogy to the laptop with FaceID.
With FaceID I don’t have to “do” anything to unlock it. When I sit down in front of the iPad, it is just unlocked. I don’t have to think or act to make it happen.
With the laptop with touchID, I need to press a key to wake it and then press the right finger on the touchID key to unlock. It is a more deliberate and complex action.
I can see on phones that the differences are less, but even in the time of masks, I tend to prefer FaceID. If I’m using my phone, I’m also looking at it. The mask gets in the way but that is only a few times a day when I’m at the grocer or something like that. I know that with touchID I could pull my phone out of my pocket and unlock with my finger in one movement, but that doesn’t save much for me. You may use your phone differently that that’s fine. I think there would be value in a phone with both systems as people have different needs and preferences.
More than that - masks are an issue currently if you’re trying to unlock your phone indoors in a public place with Face ID, but I just don’t find myself in a situation where I’m wearing a mask when I need my laptop.
(I’m not going to Starbucks or sitting in an office right now, and very likely won’t be until widespread vaccination has taken place).
The convenience of Face ID is transformative to my workflow. It really does essentially take you back to when your passwords would just automatically auto-fill without any kind of checks that it’s still you, without your passwords being stolen if someone swipes your unlocked laptop and runs off.
I think it was more motivated by the fact that the first months with the new processor architecture are rough from a software point of view. While Rosetta seems to work great, there are a lot of things which didn't work initially and now slowly come around with more and more native packages appearing. A complete redesign of the machines would have attracted too many customers who would not have been ready for the bumpiness of the software in the first months.
Yeah this is what I thought at first too, but in my experience it was remarkably smooth. Even as a developer, which is probably a worst-case scenario for this type of transition, I have barely had to think about the fact that it's a different arch
I ran into this. I was making a lot of very conservative noises early on at work about potential risks to our workflows, and I've been waiting and waiting. I had thought making docker images on a different architecture to deployment would bite us, but so far it's the dog that didn't bark.
I don't know how in hell they did it, but I'm impressed so far.
More seriously, by the way, do we have comparison between the black iMac Pros and the M1? It would be ironic if iMac Pros were surpassed by the 1/4th-prices M1...
They did the same thing when they switched to Intel. Nearly every model inherited its previous design, with big design changes coming after the product line had transitioned. At the time the general consensus was that they were timid about making big visible changes that might make existing customers feel like their products had changed too much. But now it seems like the norm for Apple to stagger architecture and design changes.
If there were rumors of a design that was impossible with an Intel chip, that would have tipped their hand. Keeping mostly the same initial design meant is was easy for Intel to think Apple was just replacing like for like to save some money.
I had a touchbar Pro and I now have an M1 air. I would go for the Air.
The touchpad is slightly too big on the Pro and the palm rejection is insufficient, leading to unexpected mouse movements and occasionally unintended selection and mass-overwriting of text. This issue is magnified further with the 15" devices on which the touchpad is comically oversized.
The touchbar leads to unintended behaviors like sending half-written emails because you unconsciously brushed against it while entering a number, and only make it more difficult to do basic things like adjust brightness and volume.
The Air, on the other hand, is a well-engineered little machine that is just as usable as a 2015 Pro or Air. Pity about the limited ports. You've seen the rumors that the next Air will ship with more ports?
I have the pro. I should have got the air. I dev mobile and web and fans almost never on, so thermal throttling on air would likely be minimal. The Touch Bar is horrible, accidentally hit “back” button all the time.
We bought one of each on day one for R&D purposes. I absolutely agree. The Air is absolutely the better machine and it’s cheaper. In our benchmarks we actually saw better performance out of the Air most of the time (bursty workloads) and the only thing we could think of was that the bigger heat sink in the Air provides better cooling than in the Pro most of the time, unless you have really long compiles where the fans kick in.
You can even do a little (reversible) mod to the Air where you add some thermal pads, effectively turning the entire bottom into a giant heatsink. It'll get a little warmer to the touch than before, although not uncomfortably, but the sustained performance gains are definitely worth it.
Think of it this way: The Air is not the underpowered almost-macbook anymore. The Air is THE M1 MacBook now. The performance is all there. The magic is all there.
The Pro is like a Plus version, with some bonuses for a higher price -- no thermal throttling whatsoever, a brighter screen, the TouchBar, a bit more battery, better microphones, stuff like that. Get it only if you want those and don't mind a little heavier machine.
I got my M1 MBA recently. I also own 2015 MBA and I've to say the decision taking MBA instead of MBP is legit. I feel like I'm using iPad instead of Mac, all things are buttery smooth. I open these apps vscode-insider, Safari 6 to 9 tabs and terminal with three process in each tab. Apple Music with bluetooth.
I replaced my 2017 MBP with the M1 MBA. Only downside is the smaller screen, and fewer ports (I have a dock though). It's better in every other way.
As others mentioned, the performance of the new MBA vs new Pro is similar. I think the MBA really highlights what's great about the first iteration of the m1. So, if you're ok with the smaller screen, I would go with the MBA. If you feel yourself leaning towards the pro, I would wait for the actual pro's coming later this year.
I do mobile dev and moved from a 2018 Pro to M1 Air and super happy with it so far. All my tools work great on the M1 and most are fully supporting the M1. I always like the smaller screen so even my pro was the smaller screen.
I’m a bit worried every time I hear this. We are setting our expectations so high... doesn’t seem like it’s possible for the M2 to be as much of a leap over the M1 as the M1 was over the status quo. The M2 might be amazing, but now that we are all awake to what’s possible I doubt it’s going to blow our socks off quite as much as we hope.
But the fact that we have anything to look forward to at all is awesome. These are super exciting times in the computing space. Processors have been boring for way too long.
I think what’s so impressive about M1 though is the increase in (at least perceived) single threaded performance. We know core counts scale and that’s been the industry focus for some time. So yeah I am not expecting a huge perceived speed up from M1x/M2...but hey M1 feels nearly instantaneous I’m not really sure how big a difference a faster cpu would actually have.
So while the M1x is using the same CPU and GPU cores as the M1, the M2 will get an additional performance boost from having a newer generation of core designs.
I think we'll see more changes to other parts of the system just based on the fact that now they control the entire silicon. The next machines will probably have incredible demos with flawless "always on" face recognition that happens in microseconds and locks/unlocks your computer as you step away or sit down. Or impossibly good voice recognition that's always listening even while the machine is 'asleep' in your bag or on your desk. We'll keep seeing more and more little coprocessors, accelerators, etc. running whatever fancy machine learning model Apple engineers can dream up. And because all of this is done in the hardware it will be impossibly fast and have almost no hit to battery life. The world is their oyster now, they aren't held back by Intel's whims anymore.
Unless the M2 is a higher core version with 16-32 cores? M1 is already built on a 5nm process and you're right about the upgrades being incremental and small. But apple can bump up the core count significantly to make the m1x/m2 a big leap.
How much will the average user notice the difference between 4 and 23 high-performance cores? This will have a huge impact for highly parallel tasks like rendering, but isn't it still the case that most software still isn't designed to scale horizontally in a meaningful way?
I think the M1 is so impressive because the normal user will mostly notice the single thread and the low thread count performance, so the 4+4 configuration of the M1 hits that sweet spot. The most obvious gain would be from all the multicore loads. But for more and more heavy computing tasks, those get more common too. And I would expect any variation of the M which is tweaked towards performance rather than low power as the M1 is, to also offer a little higher clock speeds. Considering the M1 runs at 3 GHz and the most recent x86 at up to 5, there should be quite some clockspeed headroom, if Apple chooses to use it.
I feel like the M1 chips kind of broke the timeline. It's like the past ~10 years of Great Intel CPU Stagnation never happened, and we finally now get the mobile computers we thought 10 years ago that we would have in 10 years.
I think the M1X - or whatever they call the chip line with a much larger thermal envelope - will blow our socks off.
During those 10 years, mobile went huge and PC demand stagnated. You can look up corporate revenues and news coverage to verify that for yourself, if you don't believe me. It's easy to see how that market environment gave Apple a leg up. Anecdotally, I bought one or two new laptops in the 10 years prior to the pandemic, but I probably bought eight or nine different phones. And while there was a little PC innovation (mainly at the earlier part of that decade), phone hardware got sooo much better. Snapdragon SoC, waterproof flagship phones, awesome cameras, 4G, the whole nine yards. In my mind, M1 is kind of an extension of that smartphone innovation, bridged from iPad and iPhone to the Mac.
If you look at the leaps in CPU perf the iPhone/iPad has had year over year, why wouldn’t we expect this to continue with these chips which are similar in many important ways?
I’d also expect the first release was relatively conservative and restrained to ensure a low risk debut. I’m optimistic for the next few generations.
Seems like they achieve double digit performance gains in each A series chip the last few generations.
I’m excited for a 16” MacBook Pro but hoping Microsoft brings a production version of Windows 10 arm to the market. I need it for various aspects of the work I do. I can keep a spare windows box around but it’d be great to have one system.
I think a mac pro level machine will be the interesting one.
Put a bunch of these cpus in a system. I don't know if we can hope for expandability anymore, apple just might not do that anymore.
I'm going to make a prediction here: Including memory on a multi-chip module/building systems around an MCM SoC is going to become the standard for a large fraction of the personal computer market.
The performance gain from doing so in CPU<->core memory and core memory<->GPU transfers is huge, and the manufacturer can match RAM timing and performance precisely to their processor or even implement non standard ram types as they like. There are other benefits too like simplified motherboard design.
Now that Apple has taken the risk, other manufacturers will look at doing the same. Not all computers will use the SoC model, but for laptops and many desktops this will be a big win.
I don’t know how this “memory on the SOC” thing became a narrative. That isn’t responsible for any of the M1’s performance. Apple isn’t using anything fancy like HMC or HBM that can’t be done with off-package memory. It’s a regular 128 bit memory bus and standard LPDDR4x (at the top standard frequency), with slightly higher memory latency than Intel and AMD. Pretty much any HEDT x86 system has a more impressive memory system.
>I don’t know how this “memory on the SOC” thing became a narrative.
Yes it has been spreading like plague. And I had to post something similar [1] not long ago. And many more before that.
The M1 could have an Off package Quad Channel DDR4 Memory and still be as fast. The performance improvement ( from a memory perspective ) is coming from Same Memory Address Space and other similar optimisation.
Going off package will at the very least increase power dissipation. Exiting a package, going across a PCB, and entering additional packages will increase capacitance significantly as well as increase resistance and inductance. This will impact performance. If the increased capacitance does not change actual operating speed, then the buffers are supplying more current to overcome the capacitance, not to mention the potential ringing and other undesired effects from the additional parasitics. There is a penalty for going off the SOC. It is not just a narrative, it is physics.
The same memory address space choice is of course important, but its performance and power envelope is impacted by the SOC vs. separate package choice.
The combination of M1 performance and low power has happened due to a series of choices made by Apple. Forgoing user configurability and fixing memory choices at manufacture while using SOC tech made mainstream by the phone industry is one of those impactful choices. There are of course several other important choices, but it is incorrect to discard this choice as non-impactful.
If I understood the parent comment, isn't it more about shared CPU/GPU memory? This is something different which M1 has in common with game consoles and smartphones but not traditional PC's, isn't it?
no it's not the same. "Shared" means different things.
In M1, the GPU reads directly from memory written by the CPU.
In Intel/AMD, the data has to be copied from the CPU's address space to the GPU's. "Shared" only means there aren't separate main and graphics memory chips/banks. But said shared memory is segregated.
Uhm, as a games developer working on consoles.....no it isn't. You can do it like this if you wish, but generally the entire address space is accessible from either CPU or GPU. Maybe it's implemented like this on PC, but at least the architecture design on X1/XBS/PS4/PS5 allows both reads and writes from any area of memory by either cpu or gpu.
On pc by default the memory for the iGPU is a dedicated segment of RAM. There are probably tricks to read from each other RAM but not integrated like Apple has done.
The i9’s slow supported memory is a product of that being Intel’s 14nm line which uses the same architecture released in 2015. The 10nm core chips use LPDDR4x, just like the M1. The $999 Surface Laptop 3 uses it at 3733 MT/s, just a bit slower than the M1: https://www.anandtech.com/show/14933/microsoft-announces-sur.... That speed of memory is supported on the lowest end 10nm mobile i3, which appears in sub-$300 NUCs available since 2019.
The M1 is a great chip, but that has nothing to do with the location of the memory. LPDDR4x-4267 is a standard memory type. Kudos to Apple for using the highest commonly available speed bin, but it’s a standard speed bin for that type of memory.
I'd say that having massive low latency caches on die plays a larger role.
>On the cache hierarchy side of things, we’ve known for a long time that Apple’s designs are monstrous, and the A14 Firestorm cores continue this trend. Last year we had speculated that the A13 had 128KB L1 Instruction cache, similar to the 128KB L1 Data cache for which we can test for, however following Darwin kernel source dumps Apple has confirmed that it’s actually a massive 192KB instruction cache.
That’s absolutely enormous and is 3x larger than the competing Arm designs, and 6x larger than current x86 designs, which yet again might explain why Apple does extremely well in very high instruction pressure workloads, such as the popular JavaScript benchmarks.
The huge caches also appear to be extremely fast – the L1D lands in at a 3-cycle load-use latency. AMD has a 32KB 4-cycle cache, whilst Intel’s latest Sunny Cove saw a regression to 5 cycles when they grew the size to 48KB.
That cache is that big because the decode and ROB are so wide. If AMD or Intel's current designs widened the L1, it wouldn't make a difference. In fact, AMD reduced L1 cache size from 64k to 32k from Zen 1 to Zen 3.
x86 needs to find a way to scale decoders without blowing the power budget. Given that the decoders are already bigger than the integer units, I suspect that will be a hard thing to do.
I get crashes with XMP, usually right at worst possible time on a zoom call where I am presenting! And saying users should overclock just seems weird, apple works out of the box.
For whatever reason, the overall memory system on the M1 systems just seems better than intel. I really wish I could follow more details from on-die cache to how memory is actually loaded / unloaded to speeds, but every time I've looked at it a little it just seems the M1 / Apple are doing it better across the whole stack.
This may have changed since then (Mid-2020). Gamers Nexus have done an undercover sting where they found it was possible for a support agent to reject a warranty request on the basis of XMP: https://www.youtube.com/watch?v=I2gQ_bOnDx8&t=1155
Performance per Watt. Reducing the trace length makes it possible to get high frequency RAM working with acceptable power consumption. How many other ultrabooks clock their RAM at 4266MHz?
That's some weird semantics.
It is very much soldered separately, they are separate components soldered next to each other [1].
(Yeah, on a own sub-PCB, but still)
If that's not "soldered separately", then we might as well zoom out and apply that statement for the whole PC ("all components are in the same package, only the charger is separate")
Why not both? A little DRAM on SoC and the rest of the DRAM on the motherboard. Kernel in charge of "swapping". Maintains expandability while keeping most of the performance benefit.
How? You're basically describing RAM caching. Putting all the ram physically close to the processor gives a giant performance gain that's mostly lost if any of the system's RAM is "remote" on the motherboard.
Is there any concrete numbers to back the claim that "Putting all the ram physically close to the processor gives a giant performance gain"? Will be interesting to see what is the memory latency of M1 compared with a regular Intel/AMD processor.
It doesn't. As has been corrected time and time again.
The M1 has pretty high memory latency at around 100 ns [1], which is significantly higher than either AMD or Intel for typical systems. Note that physical distance between CPU and memory is rather less important for latency, as DRAM is high latency in itself, so adding a few ns at most due to wiring is not going to matter.
Just for context, the M1's latency is fairly close to other LPDDR4(x) systems (Tigerlake and Zen3 SoCs).
This is fairly typical of LPDDR4 compared to normal DDR4, part of the compromise.
Not the most scientific, but userbenchmark is useful because it has latency graphs available for millions of systems.
Yes. Thank You. If M1 discussion continues to be like this we have a possibly of stamping out M1 misinformation on HN.
But sometimes we are just lazy to provide the context or to spell out everything. These information is so readily available with a simple Google. And yet the past dozens of M1 thread this "memory" advantage thing keeps popping up.
It's a beast and it's also mega efficient. Right now on my MBA I'm running 2 displays (one external at 3008x1253 HiDPI 120hz), several apps open including VS Code and 3 browsers, playing a 4k vp9 YouTube video in Firefox, and the passively cooled M1 never crosses 40C. My other Macbook would be at 70C with the fan screaming.
My daily driver is a 2019 rMBP -- the last 15" model, with the upgraded keyboard.
I am SO TEMPTED by the M1 upgrade, especially given the trade-in valuation on my machine right now. However, it would be inconvenient to give up ports, and right NOW I still do some Windows virtualization, so I'm holding off.
May be worth noting that the battery life is much better on my M1 MBA than Intel devices, so you don't necessarily need to sacrifice one of the two ports for charging during the day.
It depends upon your use profile, but for me it leaves both available ports for peripherals not just the one that you might expect.
