Apple M1 Max Geekbench Score | Hacker News Reader

Apple M1 Max Geekbench Score

browser.geekbench.com

488 pointsmv9880 comments

I just can't figure out what I'm missing on the "M1 is so fast" side of things. For years I worked* on an Ubuntu desktop machine I built myself. Early this year I switched to a brand new M1 mini and this this is slower and less reliable than the thing I built myself that runs Ubuntu. My Ubuntu machine had a few little issues every no and then. My Mini has weird bugs all the time. e.g. Green Screen Crashes when I have a thumbdrive plugged in. Won't wake from sleep. Loses bluetooth randomly. Not at all what I'd expect from something built by the company with unlimited funds. I would expect those issues from the Ubuntu box, but the problems were small on that thing.

*Work... Docker, Ansible, Rails apps, nothing that requires amazing super power. Everything just runs slower.

> I just can't figure out what I'm missing on the "M1 is so fast" side of things.

Two reasons:

1. M1 is a super fast laptop chip. It provides mid-range desktop performance in a laptop form factor with mostly fanless operation. No matter how you look at it, that's impressive.

2. Apple really dragged their feet on updating the old Intel Macs before the transition. People in the Mac world (excluding hackintosh) were stuck on relatively outdated x86-64 CPUs. Compared to those older CPUs, the M1 Max is a huge leap forward. Compared to modern AMD mobile parts, it's still faster but not by leaps and bounds.

But I agree that the M1 hype may be getting a little out of hand. It's fast and super power efficient, but it's mostly on par with mid-range 8-core AMD desktop CPUs from 2020. Even AMD's top mobile CPU isn't that far behind the M1 Max in Geekbench scores.

I'm very excited to get my M1 Max in a few weeks. But if these early Geekbench results are accurate, it's going to be about half as fast as my AMD desktop in code compilation (see Clang results in the detailed score breakdown). That's still mightily impressive from a low-power laptop! But I think some of the rhetoric about the M1 Max blowing away desktop CPUs is getting a little ahead of the reality.

You're missing out on the fact that Apple didn't release a 12980hk or 5980hx competitor. These are 30 watt chips that trounce the competition's 65 watt (e.g. the 12980hk and 5980hx) and beyond chips.

Hell, this Geekbench is faster than a desktop 125 watt 11900k. It's faster than a desktop 105 watt 5800x.

Apple intentionally played to the competition here. They know AMD/Intel reach some performance level X and released CPUs that perform no greater than X * 1.2. They know they are in the lead since they are paying TSMC for first dibs on 5nm, but they didn't blow their load on their first generation products.

Intel will release Alder Lake and catch up, AMD will reach Zen4 and catch up and Apple will just reach into their pocket and pull out a "oh here's a 45 watt 4nm CPU with two years of microarch upgrades" and the 2022 MBP 16 will have Geekbench scores of ~2200 and ~17000.

There's a de facto industry leader in process technology today -- TSMC. Apple is the only one willing to pay the premium. They also have a much newer microarch design (circa 2006ish) vs AMD and Intel's early 90s designs. That's a 10-20% advantage (very rough ballpark estimate). The also are on arm which is another 10-20% advantage for the frontend.

The big deal here is that this isn't going to change until Intel's process technology catches up. And, hell, I bet at that point Apple will be willing to pay enough to take first dibs there as well.

AMD will never catch up since we know they don't care to compete against Apple laptops and thus won't pay the premium for TSMC's latest art. Intel might not even care enough and let Apple have mobile/laptop market first dibs on their latest node if Apple is willing not to touch the server market. Whether or not they'd agree on the workstation MacPro vs 2 slot Xeon workstation market would be interesting.

It might be a long time before it makes sense to buy a non-Apple laptop.

The fact that this beats AMDs top laptop CPU is actually a huge deal. And that's before considering battery life and thermals.

I'll never buy an Apple computer, but I can't help but be impressed with what they've achieved here.

You need to consider the larger target group of professionals. It's really GPU capabilities that blow everything away. If you don't plan to use your MacBook Pro for video/photo editing or 3D modeling, then a M1 Pro with the same 10-core CPU and 16-core Neural Engine has all you need and costs less. Unless I'm missing something I don't think there much added benefit from the added GPU cores in your scenario, unless you want to go with the maximum configurable memory.

I disagree with this perspective. I think it's important to recognize that the M1 is a System on a Chip (SoC), not simply a CPU. Comparing the Apple M1 to "mid-range 8-core AMD desktop CPUs from 2020" is not comparing apples to apples. The M1 Max in the Geek Bench score has 10 cores whereas the AMD desktop CPUs you mention have 8-cores. That would be more of an apples to apples comparison.

Where the M1 architecture really shines is the collaboration between CPU, GPU, memory, SSD, and other components on the SoC. The components all work together within the same 5nm silicon fabric, without ever having to go out to electrical interconnects on a motherboard. Thereby saving power, heat, etc.

What you lose in repairability/upgradability, you gain in performance on every front. That tradeoff is no different than what we chose in our mobile devices. If repairability and upgradability are more important to you, then definitely don't buy a device with an Apple M1; absolutely buy a Framework laptop (https://frame.work).

> Apple really dragged their feet on updating the old Intel Macs before the transition

There was a Twitter post doing the rounds which I cannot locate now as my Twitter-search-foo is not strong enough. :-(

To summarise the gist of it: The post was made by someone on the product development team for the newly released MacBook Pro models, they referred to it as multiple years in the making.

So it may well be Apple were dragging their feet for good reason. They knew what was coming and did not want to invest further in Intel related R&D and did not want to end up with warehouses full of Intel-based devices and associated service parts.

I think people are missing the fact that it’s performance + energy efficiency where M1 blows regular x86 out of the water.

>Apple really dragged their feet on updating the old Intel Macs before the transition. People in the Mac world (excluding hackintosh) were stuck on relatively outdated x86-64 CPUs.

Maybe my expectations are different; but my 16" MacBook Pro has a Core i9-9880H, which is a 19Q2 released part - it's not exactly ancient.

> 1. M1 is a super fast laptop chip. It provides mid-range desktop performance in a laptop form factor with mostly fanless operation. No matter how you look at it, that's impressive.

Get an x86-64 laptop with a recent fastest Ryzen, install Linux on it. You're gonna see better performance for most practical things than your Mac M1, practically fanless. For half the price.

Power-per-watt remains Apple's competitive advantage and therefore battery life is 1.5-2x there.

I think the question remains, like it had before these new chips: do you want MacOS and the Apple ecosystem? If you do, they're obviously a good choice (even a great choice with these new chips). The less value you get from that ecosystem, the less you will get from these laptops. For nearly everything else, Linux will be the better choice.

From what i see the Clang difference mostly looks like its because it is a 10 core cpu (with 2 efficiency ones at that), vs a 16 core. (I'm seeing 104 klines/sec vs 180.5), the 5800x seems to hit 94-96klines/sec

Kinda wish they had a m1 max focused on extra cpu cores rather than gpu cores tbh

For reference, I’m looking to purchase a framework laptop. If I’m lucky I’m going to get 8 hours of battery life. Real number for M1 mbp reach 16+ hours, doubling the laptop of my choice.

That is crazy impressive and something that I wish I could get in the ultrabook form factor.

> Compared to those older CPUs, the M1 Max is a huge leap forward.

Is it, though? Its single-core score is roughly 2x that of the top CPU Apple put in the mid-2012 macbook pro. 2x after 10 years doesn't seem that great to me.

Maybe that's more of an indictment against intel

> were stuck on relatively outdated x86-64 CPUs.

We came to a point when X86 is faster to emulate, than to run on a more modern microarchitecture.

One point is clear: per-transistor performance of X86 is dimishing with each new core generation, and it is already losing to M1.

X86 makers will not be able to keep up for long. They will have to keep releasing bigger, and hotter chips to keep parity, until they cannot.

What’s the best Apple computer for fastest compile times nowadays?

> But I agree that the M1 hype may be getting a little out of hand.

Alder Lake will have big little core and by on 7 nm. I'm really curious if it or its successor will be good enough when compared to the current M1

Really curious to see what chips will go into the Mac Pro line next year (year after?). Will they be faster than the AMD desktop/workstation chips when they come out?

> Apple really dragged their feet

... I wonder if this was deliberate theatre for dramatic contrast?

That's Docker for Mac versus native Docker. Docker only runs on Linux, so Docker for Mac spins up a linux VM to run your containers. When you mount your Ruby or Python projects into your containers, Docker for Mac marshals tons of filesystem events over the host/guest boundary which absolutely devastates your CPU.

Docker for Mac is really just bad for your use case.

No idea what's going on with the thumb drive, bluetooth, etc.

Beyond that, it's a little silly to compare a desktop (presumably many times larger, ~500+W power supply, cooling system, etc) with a Mac mini (tiny, laptop chip, no fans, etc).

> Beyond that, it's a little silly to compare a desktop (presumably many times larger, ~500+W power supply, cooling system, etc) with a Mac mini (tiny, laptop chip, no fans, etc).

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.

I used to run Docker on an older Intel Mini and it was fine--a little slow but usable. I've also used it on other Intel Macs without major issues, including running giant Java apps and SQL Server.

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 wonder how well ARM64 is supported in the Docker community? Presumably x86 Docker images running in emulation would be even slower than ARM64 images running in a Linux VM.

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.

Every Mac mini has had a fan. The 2018 Intel version and the M1 version use a lot of the small internal space for a large fan.

> Docker for Mac is really just bad for your use case. Nah, not just for mac, docker is really just bad. period ;P

Disclaimer, I don't have an M1 Mac, but I do have a buggy ubuntu desktop and used Macs my whole technical life.

It seems that you're heavily in the minority with this. Even the weird bugs you mention are very unexpected. I've used a Mac for 15 years and never heard of an issue related to thumb drives. You may just have a lemon. See if you can just replace it (warranty, etc, not just spending more money).

It's hardly unheard of for a Mac to pick up weird issues. My wife's Macbook has a thing where the mouse cursor will just disappear when she wakes the thing up from sleep. Poking around on the internet finds other people with the same problem and no good solution (zapping PRAM doesn't help, neither did a full OS reinstall). It's just a live with it affair. The only fix is to close the lid and open it again, which isn't too bad but the issue crops up multiple times in a day and is quite annoying.

I manage a bunch of Ubuntu desktops at work and the most common issue seems to be that if you leave a machine alone for too long (a week or two), then when you log back in the DBUS or something seems to get hung up and the whole interface is mostly unusable until you log out and log back in. It can be so bad you can't even focus a window anymore or change the input focus. Next most common issue is DKMS randomly fucking up and installing a kernel without the nVidia or VirtualBox modules leaving the machine useless.

Just my N=1 anecdata, but I'm in the same boat. I got a Macbook Air from work, and I have a hard time using it compared to my Linux setup (which is saying something, since I'm using a Torvaldsforsaken Nvidia card). Here's a list of the issues I can recall off the top of my head:

- High-refresh displays cause strange green/purple artifacting

- Plugging in multiple displays just outright doesn't work

- Still a surprisingly long boot time compared to my x201 (almost a teenager now!)

- No user replaceable storage is a complete disservice when your OEM upgrades cost as much as Apple charges

- Idle temps can get a little uncomfortable when you're running several apps at once

...and the biggest one...

- A lot of software just isn't ready for ARM yet

Maybe I'm spoiled, coming from Arch Linux, but the software side of things on ARM still feels like they did in 2012 when my parents bought me a Raspberry Pi for Christmas. Sure it works, but compatibility and stability are still major sticking points for relatively common apps. Admittedly, Apple did a decent job of not breaking things any further, but without 32-bit library support it's going to be a hard pass from me. Plus, knowing that Rosetta will eventually be unsupported gives me flashbacks to watching my games library disappear after updating to Catalina.

I have both, and use both everyday.

I use an M1 MacBook Pro (16Gb Ram) for personal projects and as my standard home/travel computer. It is amazing and fast.

I use a Lenovo Carbon X1 Laptop with similar specs (i5, 16Gb Ram, m.2 ssd) for work that runs RHEL 8 (Red Hat Enterprise Linux). It's insanely fast and stable.

The overhead to run RHEL is so small it would blow your mind at the performance you get from almost nothing. Mac or Windows are crazy bloated by comparison. I know I am sparking an eternal debate by saying this, but I personally have never found Ubuntu to be as stable for a workstation (but ubuntu server is great) as RHEL is.

With that being said, I still think the M1 mac is the best computer I have ever owned. While linux is great for work, I personally enjoy the polished and more joyful experience of Mac for personal use. There are a million quality of life improvements that Mac offers that you won't get in Linux. The app ecosystem on mac is incredible.

When most people make comparisons for the M1 Mac, they are comparing windows PCs (generally Intel-based ones since Mac previously used Intel) and they compare intel-based Macs. I have never seen someone comparing it to linux performance. The speed of the M1 mac is far better than Windows and far better than old Macs. There is no question. Before my M1 mac I used a MacBook Pro with an i7, 16Gb RAM, and the upgraded dedicated graphics card. The little M1 MacBook outshines it at least 2 to 1. Best of all, the fans never turned on, and my old MacBook Pro had constant fan whine which drove me crazy.

The other incredible feat of the M1 Mac is the battery life. I run my laptop nearly exclusively on battery power now. I treat it like an iPad. You plug it in when it gets low, but I can use it for about a week between charges (I use it for 2-3 hours each day). I don't turn the screen down or modify my performance. I keep the screen fairly bright and just cruise away. I love it.

While Linux might be able to outshine on performance, it doesn't outperform with battery. My Lenovo laptop is worth about 2x my MacBook Pro. It is a premium laptop and yet running RHEL I will be lucky to get 6 hours. Compare that to ~20 hours of my MacBook.

My colleague has an M1 Mac. I have a Ubuntu desktop. My colleague always asks me to transcode videos on my machine because on her MacBook it is too slow.

yep. if things don't work contact apple support they are actually pretty decent. i had a lemon mini, randomly would go into a bootloop after os updates - had a bad mainboard so apple replaced it and it's been fine since.

Well if you actually do ‘work’ with the device, like installing software thats not in the appstore you might run into some troubles...i love my 2013 macbook air for most daily tasks, but never was there a time where i couldnt do with having a windows and linux device on hand. But yeah, thats just life. Happy to see sobering comment here that the m1 is a ‘mobile’ processor, my 12 year old pc agrees. Another question that came to mind is; what professional is gonna edit ProRes video while commuting? Is this the ultimate precarious labour creative industry machine?!

Easy, it is likely docker that is making your Mac mini slower than your old linux box?

Docker on macOS is painfully slow, because it is implemented on macOS through what amounts to a sledgehammer to the problem. Docker depends on linux kernel features so on macOS it just starts a linux virtual machine and does other high overhead compatibility tricks to get it to work. Volumes are the biggest culprit.

If you are running docker through rosetta... (don't know the state of docker on apple silicon) then that is a double whammy of compatibility layers.

Regarding bugs, yeah probably teething issues because the M1 was/is such a large departure from the norm. They should really get those things fixed pronto.

Docker for Mac Desktop supports the M1 and uses an Linux VM that is ARM.

It is not using Rosetta 2 at all.

Many folks are comparing it to previous Macs with macOS, not PCs nor other distros.

If you have an Intel Mac next to it, there's a clear noticeable difference assuming macOS is on it.

If you put W10 on the Intel Mac, it is much faster than macOS on it (from my own experience). If we could run W10 or Linux on M1, it will be much faster than Intel Mac.

Another example, Windows 10/11 on ARM in Parallels on my MBA m1 is much faster than Surface Pro X, a MS-tuned/customized device.

For me, all the gains from a W10 running faster than MacOS are lost because I have to immediately go into control panel to fix everything. Those fast computers are waiting on me to improve the user experience most of the time. I'm personally not really splitting hairs over much longer running processes if I have to walk away anyway.

I’m comparing Windows PC (i7-6700k, nvme) with MBA m1 and MBA is noticeable faster in just everything. Compilation time of JS and Java code is literally 2 times faster on MBA M1 than on desktop i7 CPU.

I have a three-year old hand-built 16-core ThreadRipper 1950X machine here with 64GB 4-channel RAM. I have a 14" MBP M1 Max on order. I just checked the Geekbench scores between these two machines:

  M1 Max: 1783 single-core | 12693 multi-core
  1950X:   901 single-core |  7409 multi-core

That's a big difference that I'm looking forward to.

Also, just checked memory bandwidth. I have 80GB/s memory bandwidth [1] on my 1950X. The MBP has 400GB/s memory bandwidth. 5x(!)

[1] https://en.wikichip.org/wiki/amd/ryzen_threadripper/1950x

*EDIT adding memory bandwidth.

> I have 80GB/s memory bandwidth [1] on my 1950X. The MBP has 400GB/s memory bandwidth. 5x(!)

Your 1950x also doesn't have a GPU. For comparison the rx6700xt, and Vega 56 have ~400GB/s. The 6 year old R9 nano has 500GB/s.

1. M1 Max has 50+ billion transistors (only a part of that count is CPU, so adjust the number downwards), 1950X (first gen 14-nm Ryzen) has 10 billion.

2. Multicore Cinebench R23 for M1 Max: 14970; Ryzen 1950X: 18780

Geekbench tests in short bursts, not taxing the CPU thermals. Cinebench is the opposite. In short, CPUs do well on some benchmarks and not so well on others.

So you are comparing a laptop that is just coming out to something 4 years old.

I share the same feeling, and I'm glad to learn I'm not alone: earlier this year I worked for a (really) sort time with a company who gave me a Macbook M1, which I was pretty excited about.

Over the course of the two weeks I worked there, I faced:

- frequent glitches when unplugging the HDMI adaptor (an Apple one, which costed an arm and a leg).

- non-compatible software (Docker, Sublimetext)

- and even a Kernel Panic while resuming from hibernation

It was like working with Linux in the 2000's, but at least with Linux I knew it was gonna be a bit rough on the edges. Given the hype found online, I wasn't at all prepared for such an experience.

> non-compatible software (Docker, Sublimetext)

What was the issue with Sublime? I use it on an M1 every day without issue.

I have had the opposite experience. Whenever a update to a kernel comes out for Ubuntu my machine apparently forgets which kernel to choose and boots up incorrectly. My M1 MacBook Air drives a Pro Display XDR without any hiccups. Its completely silent when I work with it. But, performance wise the M1 Pro and Max don't seem like they would be worth it for me to upgrade from the M1 at all. I just want a completely silent laptop and it makes a huge difference to me.

But, my workflows and workloads are slowly changing from local compilation of code to one where I will probably just SSH and or possibly figure out how to configure RDP into my Threadripper when I do actual builds and have my Macbook Air M1 with 16GB of Ram and the synergy with other Apple devices is a huge plus.

Have you talked with Apple to get a replacement?

docker is a second-class citizen on anything but linux. i'm especially amazed you bothered with the m1 mac mini if you wanted to use docker, considering how little memory it has. the memory overhead of having a VM running with only 8GB available is significant.

yes, i know the docker on mac has "native" support for the m1 cpu, that doesn't mean its not running a VM.

Yeah people forget that Docker is built to run on Linux. That is why it exists. You deploy your containers onto a Linux host that runs docker for containerization, that is the point.

The reason we have Docker for Windows and Mac is so that developers can emulate the docker environment for testing and building purposes. But ultimately it is expected that you will take your container and eventually deploy it on Linux. So Docker will always be a second class citizen on these other OS's because it is just there for emulation purposes.

Seems like your computer may have issues, you shouldn't be crashing from USB devices. Sounds like you should go get it checked out tbh.

Well could just be a software issue, for example if you installed the wrong CH430x driver to run your faux arduino clones (that didnt pay debt to ftdi) you might run into this issue.

There are an awful lot of people with USB issues on the M1. Its clearly not as mature a platform as previously.

Has there been any research to OS response times between Macos, Windows and Linux?

Ive wanted to switch to Mac many times, most recently to M1 Mac Mini, but cant get over the frustrating slowness. Every action, like opening Finder, feels like it takes so long compared to Windows even with reduce animation on. I always end up returning the mac.

Well thanks for mention it, I really feel the same. Stuff just does not open instant and perceived slowness on macOS is real. Just try simple stuff… Modern Windows just flies most of the time in comparison. Opening a file dialog in finder is a common scenario yet it is so slow and the system seems to appear limited by simple stuff like that. M1 surely improves things but the UI generally seems to prefer animation over raw speed.

I really don‘t know what the reason is but I feel that a lot of people simply ignore this. I use all 3 major systems and I cannot overlook it anymore.

Additionally, Docker is another real issue mentioned for the reasons already. Too bad that it is so common now and its slowness is a growing pain. I feel that WSL (WSLg) on Windows is huge, no comparison to run a traditional VM for the most part if you are not macOS only. Never thought I have to write this…

Meh. The M1 is a node ahead and memory is placed on package. I'm trying not to be dazzled by the tech when it comes at the price of freedom.

Ok - I get that comment for iPhone - but what is less free about the Mac vs. a PC?

I went from a 2016 top-of-the-line MacBook Pro to an M1 Mac mini, and I can't believe how much faster things are–including Docker with Rails, PostgreSQL, etc. Out of curiosity, are you running native (arm64) containers? Aside from an issue now and then with my secondary HDMI display, my machine has continued to blow me away. My development environment is almost on par with our production environment, which is a nice change.

The Mac Mini isn't super impressive, because you're comparing it to desktops where TDP is far less of a concern.

The M1 is getting a lot of attention because it's Apple's first laptop chip, and it is the fastest chip in that category by a fairly significant margin. Chips from Intel and AMD only compete with it on performance when drawing considerably more power.

MacOS just seems to have a lot of stability problems related to peripherals. I don’t think it’s specific to M1. Even on my late-era Intel MBP, Bluetooth and USB often triggers a kernel panic.

When I got my first MacBook Pro, there was actually bug in iPhone USB Tethering that caused a guaranteed kernel panic if you tried to do it. Managed to crash my laptop a few times with my iPhone 3GS.

My experience is the opposite. I love Linux, but my MacOS work laptop is just so much better.

For as long as I've used Linux it freezes randomly. This happens on all hardware configurations I've ever owned. If I don't regularly save my work, I will end up losing it.

These days I also can't leave my desktop PC suspended for too long. At some point the fans start spinning at max. No reason why.

As for my 13 inch M1 laptop, I don't even know if it has fans. I've never heard them even if I'm in a zoom meeting, sharing my screen, while compiling half the world.

Also in Linux I can't really use Zoom screen sharing. Seems to leek memory and after a while it crashes. Of course that's not Linux fault.

On my Linux machine it takes significant effort to use my AirPod's microphone.

And that's not even talking about battery life on Linux laptops.

It's always interesting how we all have such completely different experiences with our hardware and software.

Seems like you mostly have issues with Wayland. What distro are you running?

For me Zoom screen sharing works fine with Wayland, if everything runs Wayland, including the browser. Xwayland apps can't "see" Wayland apps or the desktop.

I assume the Zoom client may be Electron/Xwayland, correct?

The X to Wayland transition is mostly over soon. Probably with Ubuntu 22.04 LTS.

People also overlook the fact that Geekbench has generally favored Apple A-series CPUs. When the original M1 came out they were faster than 4800u Ryzens on Geekbench, while being slower on Cinebench. 4800u consumes more power, but is also a node behind.

Other benchmarks too have their biases. You can find plenty of discussion on this topic.

I've never owned a mac and I've never used an M1 so take this with a heavy grain of salt. Everyone I've seen talk about how fast the M1 is were coming from 4-5+ year old macs not comparing them to intel/amd chips that came out last year in a similar price range.

Well I went from a 16" 32GB 6-core i7 2019 MBP to a 16GB M1 Macbook Air and it was night and day better. So much faster, silent, cool and snappier.

Well switching hardware and OS makes it hard to tell what's responsible for the difference. I also suspect Docker and what's running in the containers might well be x86 code instead of native.

I can tell you that I have a high end mbp 16" intel-i9 with 32GB ram and it feels much slower than a new mac M1 mini. My intel-i9 runs the fan card during backups (crashplan), video conferencing, and sometimes just because even just outlook. The M1 mini on the other hand has been silent, fast, and generally a pleasure to use.

Doubling the number of fast cores, doubling (or quadrupling) the GPU and memory bandwidth should make the new MBP 14 and 16" even better.

I have a top of the line Windows 10 i7 laptop and Outlook will create more heat than any other process I am running. Now introduce Windows Modern Standby and we get a real SH*T SHOW.

I thought native M1 Linux was still a work in progress (disclaimer: I’m not a Linux geek).

Even if it’s out there, I suspect it’s not optimized.

I think standard distros are running on Rosetta2, which is an (excellent, from all reports) X86 emulator.

linux is actually ported to the M1 at this point and you can run it normally (although of course there are likely still some bugs lurking), however the graphics drivers are still a work in progress although coming along very quickly.

https://www.theregister.com/2021/10/06/asahi_linux_m1_progre...

I'm sure you're right that there are still optimizations to be done, but as you can see from Alyssa's tweet there, it is still a very fast piece of hardware for normal developer usage, as long as you don't fall into one of the scenarios that probably requires further tuning (x86 emulation probably being an obvious one, no idea if they've got Total Store Ordering working in KVM yet but that's obviously a key feature).

mega kudos to the Asahi team, the progress has been really tremendous, it's been just about 6 months since kickoff and they've made progress that some people insisted would take multiple years/was impossible at all.

Put the same thing in a fanless laptop, that seems to last forever on a charge, then you’ll understand

The Docker situation on macOS for Apple Silicon is just really, really terrible. Docker on macOS was never particularly good (compared to Linux anyway), but on Apple Silicon it is just beyond bad. A combination of how Docker spins up a VM, the state of Linux on ARM (especially on the M1) and the complexities of when you need to compile stuff that doesn’t have native ARM versions is just a mess. If Apple cared about this segment, I'd think they would contribute some resources to either improving Docker or doing native macOS containers built with Unix, but I just don’t think they do.

This is one reason I'm keeping my 2020 iMac (10-core i9, 128GB of RAM, AMD Radeon Pro 5700XT) around for a long time, even though I just ordered a 64GB 32-core GPU 14” MacBook Pro . (I have an AMD Ryzen 5900X with an Nvidia 3080 as my gaming PC too.)

If your primary interface is containers, I don’t necessarily think the M1 series machines are the best. But for lots of other tasks, it’s really just astounding.

Actually there is nothing to figure out. If limitations of macos and arm cpu is not issue for you and you want a lightweight laptop then you buy a m1 macbook. m1 macbook air first Apple product i bought and i also had latest gen ryzen laptop at that time (which was superior to intel). Simply there is nothing to compare. It is not an alternative to m1 cpu. At least 5 years behind it. Actually it looks like it will never possible for that architecture to catch up with arm for low tdp siutations. If you ever used last gen windows laptop you will know there are always bugs. And installing ubuntu on a latest gen machine. LOL. That won't make you happy even if it is possible. I also had an issue with bluetooth mouse being laggy. It seems like patched now. Bugs were really an issue with m1 cpu but i assume there shouldn't be much issues now.

>*Work... Docker, Ansible, Rails apps, nothing that requires amazing super power. Everything just runs slower.

Docker does require amazing super power on anything except Linux since only on Linux it doesn't have to rely on emulation. The overhead is 0-1% on Linux, and something way more on alternative platforms.

Besides Docker, what you're seeing is probably the fact that MacOS is way slower than Linux for almost all the things developers do in their daily lives. If the M1 port ever gets properly done, you're gonna see Linux fly on that.

M1 is fast for being a laptop CPU. This means that maybe it is not the fastest CPU in absolute terms, but if you consider performance per Watt, it is really a beast.

Although my main workstation is an Intel Ubuntu-based PC, I have also a M1 Macbook Air that I use when I travel, or when I am laying on the couch in the evening after work. That piece of hardware never gets hot, and the battery seems to last forever. Meanwhile it does whatever I throw at it; if else the limiting factor is the amount of RAM installed on the laptop.

My $1000 Windows laptop from two years ago kills my M1 MacBook Pro. Benchmarks aren't everything and there's more to speed than just raw calculations.

Is it 5 pounds and 2 inches thick? The fact is is that my M1 Macbook Air that was $1400 would cost at least $2000 from HP, Dell or Lenovo to even approach similar performance, and still have a worse screen, worse battery life, AND would still be slower. I have not seen a laptop chip from AMD or Intel in fanless, thin laptop like the Macbook Air that even gets close to the M1.

Wow, that would really be against the tide. Can you provide any more information like the specs, make and what you are doing?

Doing what?

M1 is, compared to precious editions of Macbooks, much better.

My two year old Linux desktop is a beast compared to my M1 Macmini. But I love the Macmini compared to my 2019 MBP.

Running Docker on a Mac vs on Linux is going to be a subpar experience compared to a native x86 Linux box. It would take a LOT to magic around that.

> My Mini has weird bugs all the time. e.g. Green Screen Crashes when I have a thumbdrive plugged in. Won't wake from sleep. Loses bluetooth randomly.

Either the Mini is much buggier than the MacBook Air by design, or you need to get it replaced under warranty. My MBA is the first laptop I've used for months with absolutely no hardware interaction issue.

As someone who has an Intel NUC w. Ubuntu, a Thinkpad with Windows 10, and a Macbook Pro 2015, I have a totally different experience. I spend like 5% of my total effort on OSX to make things working, the remaining 95% is spent to make things working on Windows and Ubuntu...

My AMD / Nvidia laptop running Ubuntu has had a number of issues from flash crashes to freeze ups to external monitor connectivity that have progressively gotten better as the drivers were updated. It is likely the drivers are buggy and will probably improve with time.

Not particularly defending the M1, but it might be that much of what you're using so far isn't available native and is going through Rosetta2? The other issues could be OS issues, but maybe it would be worth doing a RAM test?

Probably more to do with bug sir than the hardware. Catalina is pretty solid.

I hadn’t heard that one. I had some pain, but I can laugh now.

Currently typing this on an m1 laptop that I bought 2 weeks back. This machine blows my 16 inch Macbook out of the water. My 16 inch Macbook is pretty well spec'd, but this laptop is something else.

Let’s review: “Black. Magic. Fuckery.”—https://news.ycombinator.com/item?id=25202147

> desktop

I too have a $900 "Gaming" desktop setup from 2 years ago that should have no problem exceeding the newest M1 Max mac in terms of performance and graphics.

But it consumes 250 watts on average.

The Intel macs were extremely slow. Even a core i9 16 inch ran hot and noisy. A small xps 13 felt much faster in general use. I don't know if this was deliberate by Apple or just a side effect of the extremely anemic cooling.

So all those people who had been using macbooks felt the speed when they used a decently fast laptop. Everybody else knows that M1 laptops are fast, like non-apple laptops have been for quite some time.

macOS has always been worse at animation performance, even after the Metal switch. My M1 is buttery smooth, but you always had dropped frames even on pretty recent Mac hardware. Your other issues seem like factory defects, I've never had them.

What are the specs on your Ubuntu machine?

Even VIM is super slow on my mac book, this amazes me most hahaha

M1 is hardware. Ubuntu is software.

Your comparison makes no sense.

This is easily explained. Linux distributions don't make money if you buy a new laptop.

You more or less described how Apple hype works.

The M1 is fast like a Ferrari, it's expensive and nice in the brochure but you can't take it many places in real life let alone use more than 20% of its potential.

The answer is M1 is a great chip held back by a mediocre OS. Containers and certain other applications are just going to run faster on Linux, which has had a ton of performance tuning put into it relative to MacOS.

Here's the link to the MacBookPro18,2 OpenCL benchmark:

* M1 Max OpenCL https://browser.geekbench.com/v5/compute/3551790 [60,167 OpenCL Score]

Comparing to my current MacBook Pro (16-inch Late 2019) & my Hetzner AX101 server:

* MacBook Pro (16-inch Late 2019) vs M1 Max - CPU https://browser.geekbench.com/v5/cpu/compare/10496766?baseli... [single 163.6%, multi 188.8%]

* MacBook Pro (16-inch Late 2019) vs M1 Max - OpenCL https://browser.geekbench.com/v5/compute/compare/3551790?bas... [180.8%]

* Hetzner AX101 vs M1 Max - CPU https://browser.geekbench.com/v5/cpu/compare/10496766?baseli... [single 105.0%, multi 86.4%]

* NVIDIA GeForce RTX 2060 vs M1 Max - OpenCL https://browser.geekbench.com/v5/compute/compare/3551790?bas... [80.7%]

* NVIDIA GeForce RTX 3090 vs M1 Max - OpenCL https://browser.geekbench.com/v5/compute/compare/3551790?bas... [29.0%, boo]

I'm surprised the MacBook holds its own against the Hetzner AX101's AMD Ryzen 9 5950X 16-core CPU! The multi-core SQLite performance surprises me, I would think the M1 Max's NVMe is faster than my server's SAMSUNG MZQL23T8HCLS-00A07.

> * NVIDIA GeForce RTX 3090 vs M1 Max - OpenCL https://browser.geekbench.com/v5/compute/compare/3551790?bas... [29.0%]

30% performance for ~15% of power use in laptop form factor? that's not boo, that seems like a clear win for apple.

Performance doesn't go down linearly with power. I don't have that card to try, but maybe it would do even better at 15% of power.

Also, the 3090 costs almost like the whole macbook

It's probably possible to build a 3090 system with a 750 watt power supply though.

Since we're comparing against M1 Max, it's more fair to use a fully maxed out Macbook Pro 16-inch for comparison:

* MacBook Pro (16-inch Late 2019) vs M1 Max - CPU https://browser.geekbench.com/v5/cpu/compare/10496766?baseli... [single 145.8%, multi 176.3%]

* MacBook Pro (16-inch Late 2019) vs M1 Max - OpenCL https://browser.geekbench.com/v5/compute/compare/3551790?bas... [153.0%]

* MacBook Pro (16-inch Late 2019) vs M1 Max - Metal https://browser.geekbench.com/v5/compute/compare/3557857?bas... [167.2%]

The M1 Max still wins by a large margin, but there's a slightly smaller gap. The CPU difference is negligible but the upgraded Radeon Pro 5600M graphics card narrows the gap a bit more. It's comparable to a Radeon RX 580. The default 5500M is pretty only about 60% as powerful and surprisingly runs hotter despite being slower. (TDP 85W vs 60W it seems)

Apple compared the M1 Max to the 3080 in the Razer Blade 15 Advanced, saying it outperforms the Nvdia’s GPU [0], but Geekbench seems to disagree [1].

[0]: https://www.apple.com/newsroom/2021/10/introducing-m1-pro-an...

[1]: https://browser.geekbench.com/v5/compute/compare/3551790?bas...

Apple never claimed the M1 Max was faster than the 3080, they claimed the M1 Max was almost as fast while using 100W less power and while being dramaticly faster on battery.

I wonder what they were comparing. The Geekbench result is for an OpenCL benchmark, if they were using Metal Compute on the M1 that may explain the different results

The 2060 and 3090 on these compares are the desktop chips, right?

The 3090 is a high end desktop machine. There is no mobile version of the 3090 (https://en.m.wikipedia.org/wiki/List_of_Nvidia_graphics_proc...).

Thanks - interesting on OpenCL - presumably running on GPU? All that memory opens up some interesting possibilities.

Also I thought Apple was deprecating OpenCL in favour of Metal?

OpenCL and OpenGL have been deprecated in favor of Metal. Geekbench also has a Metal compute benchmark.

Anyone know if that “Device Memory 42.7 GB” is fixed or can be increased?

The MacBook Pro M1 Max can be ordered with either 32 GB or 64 GB of unified memory. The geekbench report shows 64 GB of memory (maxed), not sure why only 42.7 GB is usable by OpenCL--so I guess we have to assume that's the max, unless there's some software fix to get it up to 64 GB.

I'm not exactly proficient with GeekBenchery, but what I see here is that the M1 Max per core barely outperforms the M1?

https://browser.geekbench.com/v5/cpu/compare/10496766?baseli...

I think this kinda makes sense to me — the M1 Max has the same cores as the M1, just more of them and more of the performant ones, if I understand it right. The fastest work on the fastest core, when only working on a single core, is probably very similar.

Maybe a little surprised - presumably the thermal limitations on a 16 inch laptop are potentially less limiting than on a 13 inch one so that single core could be pushed to a higher frequency?

I thought I remembered that in the presentation they had souped up the individual cores too. Must be I'm misremembering.

they're the same cores as the original M1 architecturally, you just get an 8+2 on a Max instead of 4+4... and also an absolutely titanic iGPU, total transistor count is about twice that of a 3090 and almost all of it lives in the iGPU.

the only architectural difference between M1 and M1 Max in the CPU cores, besides the different combination of big/little cores, is that the M1 Max goes from quad channel to hexadecimal-channel DDR5 RAM (note a DDR5 channel is half the width of a DDR4 channel, but has longer burst and higher MT/s).

Apple's iGPU approach is real simple: unlike a console where you somewhat gimp the CPU by putting it on high-latency GDDR6 (to get enough bandwidth to feed the iGPU), they just put 16 channels of DDR5 on it, basically it's like an octochannel DDR4 server processor in terms of bandwidth. And the CPU also benefits from that as well, at least a little bit, but internally it's the same core design.

It's an absolute meme of a design, Apple just does not give a single fuck about cost here, "server class memory configuration"? sure why not, and we'll stack it on the package so it can go in a laptop.

I believe it's LPDDR5, basically a memory chip for smartphones, not DDR5.

All I wanted was an M1 that could address/use more memory. The first one could only use 16gb RAM, which was a dealbreaker for me.

Apple delivered that and so much more. I'm happy. I think there are many people like me.

Multi-display is another highly sought feature this M1 Pro/Max delivers (achievable on the M1 with a DisplayLink hub, but stability leaves something to be desired)

Tick - New cores; Tock - Scaling up the number of those same cores

I think most of the work went into the uncore portions of the SOC this time.

FYI it's the reverse. "Tock" is a new microarchitecture, and "tick" is a process shrink.

https://en.wikipedia.org/wiki/Tick–tock_model

uncore?

The A15 chips has core improvements, I suspect this is what we’ll see every year from now on 15-20% performance increase yearly for the next few years assuming no issues with TSMC…

Top intel mobile processor appears to be https://browser.geekbench.com/v5/cpu/10431820 M1 Max gives 9% boost for single-core and 34% for multicore, with similar or larger (?) TDP -- Intel is 35-45 W, M1 Max is 60W but I assume some of it (a lot?) goes to GPU. Impressive, but probably wouldn't be called revolution if came from Intel.

M1 only uses around 15-18w with all 8 cores active according to Anandtech's review (with E-cores using 1/10 the power, that's equivalent to just a little less than 4.5 big cores. I'd guess 30w for all cores would be the upper-end power limit for the cores.

Intel's "TDP" is a suggestion rather than reality. Their chips often hit 60-80 watts of peak power before hitting thermals and being forced to dial back.

At first glace yeah, but that's 45W at 2.5Ghz, but TDP isn't the power rating of the CPU. That benchmark lists the Intel CPU as up to 4.9Ghz, I would say it's actual power draw was closer to the 100W mark for CPU only.

Just to throw AMD in the mix, Ryzen 9 5980HX (35-54W).

https://browser.geekbench.com/v5/cpu/10431820

M1 Max leading 17% single core, 50% multi-core.

https://browser.geekbench.com/v5/cpu/compare/10496766?baseli...

If the TDP was really the same we’d be seeing Intel laptops with 20 hour battery lives.

If you look up other laptops with the same processor you get a very large variation. The Lenovo you point to has a 280 watt power brick, maybe the results aren’t as good when it’s not plugged in.

It might be called a revolution if Intel could get that performance boost while nearly doubling battery life and eliminating thermal throttling

I have seen far too many people making comments on MacPro "Pro" Chip.

A hypothetical 32 Core CPU and 64 Core GPU is about the max Apple could make in terms of die size reticle limit without going chiplet and TDP limit without some exotic cooling solution. Which means you cant have some imaginary 64 Core CPU and 128 Core GPU.

We can now finally have a Mac Cube, where the vast majority of the Cube will simply be a heat sink and it will still be faster than current Intel Mac Pro. I think this chip makes most sense with the A16 design on 4nm next year [1].

Interesting question would be memory, current Mac Pro support 1.5TB. A 16 Channel DDR5 would be required to feed the GPU, but that also means minimum memory on Mac Pro would be 128GB at 8GB per DIMM. One could use HBM on package, but that doesn't provide ECC Memory protection.

[1] TSMC 3nm has been postponed in case anyone not paying attention, so the whole roadmap has been shifted by a year. If this still doesn't keep those repeating Moore's law has not died yet I dont know what will.

If someone had said in 2018 "Apple is going to release a macbook pro with an ARM chip that will be faster at running x86 applications than an x86 chip while having a 20 hour batter life", then a lot of people would have responded with a lot of very clever sounding reasons why this couldn't possibly work and what the limitations would be on a hypothetical chip and how a hypothetical chip that they can imagine would behave.

I think the evidence is that Apple's chip designers don't care about anyone's preconceived and very logical sounding ideas about what can and can't be done.

So what we know is that there is going to be a Mac Pro, and that whatever it is is going to absolutely destroy the equivalent (i.e. absolute top of the range) x86 PC.

If you want to be right, take that as a fact, and work backwards from there. Any argument that starts with "I don't think it can be done" is a losing argument.

I'm so old I remember the brief time when Alpha ran x86 code faster than x86 so history is repeating. I am eagerly awaiting M1 Plaid vs. Sapphire Rapids vs. Genoa next year.

"If an elderly but distinguished scientist says that something is possible, he is almost certainly right; but if he says that it is impossible, he is very probably wrong." — Arthur C. Clarke

> Any argument that starts with "I don't think it can be done" is a losing argument.

But plenty of interesting opinions start that way -- interesting because someone with expertise thinks so, and/or because the opinion is followed by well-reasoned arguments.

OTOH, "it can't be done" is often based on current assumptions about hard limits. Those assumptions have been broken routinely and consistently for hundreds of years now.

>If you want to be right, take that as a fact...

I guess you dont know enough about Die Space and Reticle Limit on Wafer and Foundry. You are basically arguing if Apple can bypass speed of light.

It's not hard to imagine that "Jade 2C-Die" mean two compute die chiplets and "Jade 4C-Die" means four chiplets which makes 40 CPU cores and 128 GPU cores about the same size as Sapphire Rapids. It could have 256 GB RAM on a 2048-bit bus with 1,600 GB/s of bandwidth.

Chiplet isn't a silver bullet and without compromise. You will then need to figure the memory configuration and NUMA access. Basically the Jade 2C-Die on chiplet analogy doesn't make sense unless you make some significant changes to Jade-C.

Jade 4C-Die on a single die only make sense up to 40 Core ( 4x Current CPU Core or 32 HP Core and 8 HE Core ), unless there are some major cache rework there aren't any space for 128 GPU Core.

But then we also know Apple has no problem with only 2 HE Core on a MacBook Pro Laptop, why would they want to put 8 HE Core on a Desktop?

> Interesting question would be memory, current Mac Pro support 1.5TB

I wonder whether it might have on-chip RAM and slotted RAM?

Not sure what the implications to software and performance would be there, but the whole point of the new Mac Pro was an admission that the customers in that market need components they can chop and change as required. They obviously still care about that market, otherwise they would have just discontinued the Mac Pro and left it at that (instead they discontinued the iMac Pro—the intended successor to the Mac Pro—because a modular desktop is a better fit).

It would be super weird to do a total 180 on that now, especially considering they went to the effort of building things like the Afterburner cards. The move to ARM was surely in motion before work started on the latest Mac Pro design so I'm really interested to see where things go here. "Modular tower" and "tightly integrated SoC" seem completely at odds with each other.

> We can now finally have a Mac Cube

Fascinating idea. Perhaps Apple could finally match the ideal Steve Jobs dreamed of but couldn't realize in the infamous Power Mac G4 Cube.

Mac Pro is supposed to be expandable.

The most elegant solution would be to have a motherboard with cartridges. You're buying as many CPU/GPU/AI/RAM/SSD modules as you like and getting anything you want. Wanna crazy build farm? Put plenty of CPUs and some RAM. Wanna GPU farm? You've got it. Want terabytes of RAM? No problem.

Not sure if it's possible to implement. But I'd love to see it done this way.

> about the max Apple could make in terms of die size reticle limit without going chiplet

I probably have no idea what I'm talking about as I'm a software guy. But — I remember that company that made that crazy ML accelerator where an entire silicon wafer is one chip. How'd they do that? Why can't Apple and others do the same/similar?

As mentioned in your other reply. Cerebras .

Wafer Scale Engine doesn't break the reticle limit, if you search for a picture on Cerebras, you can still clearly see the square die within the wafer. From a very high level overview, they are basically doing cross interconnect with each die on wafer. Scribe line used to be physical separation for each die, now they built interconnect on top so the whole wafer becomes a huge mesh of AI chip.

It works for because you can think of each die as lots of Neutral Engine and cache only. You can easily rework around each process defects and solve the yield problem. And then you have to solve packaging, power and cooling.

>Why can't Apple and others do the same/similar?

When you have a complex SoC. Not every defect can be worked around, that is why the larger the die the lower the yield. A reason why you see in GPU they fused off certain GPU core and Memory Controller. It isn't because of market segmentation, it is necessity coming from manufacturing. To avoid this they have chiplet. Making each part a small die and interconnect them together. But as mentioned, chiplet isn't a silver bullet. Your design must have chiplet in mind for interconnect with AMD Zen. You cant just dump two pieces of M1 Max together on the same interposer and expect it to work. They could, but it wouldn't be efficient. And that sort of defeat the purpose. An analogy would be asking the current AMD Ryzen APU, which is an SoC with GPU and Memory controller, you stick two of them together and somehow expect it to work 100% faster.

Cerebras

I'd bet it's the plumbing. Getting data in and out, supplying something around 10-100 kiloamps, attaching it to anything without it ripping itself off via thermal expansion, getting the heat out - all of that sounds miserably difficult.

Like sure you can do some outrageous and expensive things to make it all work in small quantities and for huge sums of money. But you can't build a mass-market laptop that way.

I'd guess that the upper limit with present-day and near-future packaging technology for commodity hardware is ~1000 mm^2.

Doesn't HBM2E do ECC? I know Micron definitely package that at least. I'm likely misunderstanding your ECC point though.

I dont follow HBM closely so I am guessing it could be an On Die ECC correction like DDR5 instead of Full ECC in both DIMM and Memory Controller. But I could be wrong.

Wow does much better than Geekbench's prior top processor (https://browser.geekbench.com/processor-benchmarks) the Intel Core i9-11900K (https://browser.geekbench.com/processors/intel-core-i9-11900...).

10,997 for Intel i9

12,693 for M1 Max

You sorted by single core performance, then compared multi core performance. Sort by multi core performance, and you will see that the i9-11900K is nowhere near the top spot.

For example, the Ryzen 9 5950X has single/multi core scores of 1,688/16,645 - which is higher in multi core score than the M1 Max, but lower in the single core.

Interestingly, the iPhone's A15 SOC did get a newer version of Apple's big core this year.

>On an adjacent note, with a score of 7.28 in the integer suite, Apple’s A15 P-core is on equal footing with AMD’s Zen3-based Ryzen 5950X with a score of 7.29, and ahead of M1 with a score of 6.66.

https://www.anandtech.com/show/16983/the-apple-a15-soc-perfo...

On floating point, it's slightly ahead. 10.15 for the A15 vs. 9.79 for the 5950X.

Which is still not that much higher. Of the "consumer" CPUs only 5900X and 5950X score higher. And their stress power draw is about 2X of speculated M1 Max's.

That's maybe not a bad way to sort? Most of the time I'm interacting with a computer I'm waiting for some single thread to respond, so I want to maximize that, then look over a column to see if it will be adequate for bulk compute tasks as well.

Perhaps they were referencing the highest 8C chip. Certainly, a 5950X is faster, but it also has double the number of cores (counting only performance on the M1; I don't know if the 2 efficiency cores do anything on the multi-core benchmark). Not to mention the power consumption differences - one is in a laptop and the other is a desktop CPU.

Looking at a 1783/12693 on an 8-core CPU shows about a 10% scaling penalty from 1 to 8 cores - suppose a 32-core M1 came out for the Mac Pro that could scale only at 50% per core, that would still score over 28000, compared to the real-world top scorer, the 64-core 3990X scoring 25271.

The single core is second to Intel's best but the multicore is well below in the scale, comparable to Intel Xeon W-2191B or Intel Core i9-10920X, which are 18 and 12 core beasts with TDP of up to 165W.

Which means, at least for Geekbench, Apple M1 Max has a power comparable to a very powerful desktop workstation. But if you need the absolute best of the best on multicore you can get double the performance with AMD Ryzen Threadripper 3990X at 280W TDP!

Can you imagine if Apple released some beast with similar TDP? 300W Apple M1 Unleashed, the trashcan design re-imagined, with 10X power of M1 Max if can preserve similar performance per watt. That would be 5X over the best of the best.

If Apple made an iMac Pro with similar TDP to the Intel one, and keeps the performance per watt, that would mean multicore score of about 60K, which is twice of the best processor there is in the X86 World.

I suspect, these scores don't tell the full story since the Apple SoC has specialised units for processing certain kind of data and they have direct access to the data in the memory and as a result it could be unmatched by anything but at the same time it can be comically slow for some other type of processes where X86 shines.

John Siracusa had a diagram linked here that shows the die for M1 Max, and says the ultimate desktop version is basically 4 M1 Max packages. If true, that’s a 40 core CPU 128 core GPU beast, and then we can compare to the desktop 280W Ryzens.

Interestingly, the M1 Max is only a 10 core (of which only 8 are high performance). I wonder what it will look like when it’s a 20-core, or even a 64-core like the Threadripper. Imagine a 64-core M1 on an iMac or Mac Pro.

We’re in for some fun times.

I wish there were laptop-specific Geekbench rankings because right now it seems impossible to easily compare devices in the same class

The M1 Pro/Max are effectively H-class chips so you can search for 11800H, 11950H, 5800H, 5900HX, etc.

I really wonder how a single z/Architecture core would fare on this benchmark, though I imagine it's never been ported

Probably not as good as you might expect. Z machines are built for enterprise features like RAS, and performance on specific workloads.

The ultra-high-clocked IBM cpus are probably significantly faster at DB loads, and less than the best at more general benchmarks like Geekbench.

Per core performance is the most interesting metric.

Edit: for relative comparison between CPUs, per core metric is the most interesting unless you also account for heat, price and many other factors. Comparing a 56-core CPU with 10-core M1 is a meaningless comparison.

Not when building large software projects.

Just when you think things hit the top, another kid's out of the town.

What about this? https://browser.geekbench.com/v5/cpu/7421821

That’s a strangely shaped laptop, what is the battery like on it?

This is the comparison to a mid 2015 15" MacBook Pro for anyone who is curios: https://browser.geekbench.com/v5/cpu/compare/10513492?baseli...

Summary: single core performance gain is 2x whereas multi-core performance gain is 4x.

Over 6 years, 2x the performance in single-core and 4x (for 2.5x the cores, depending how you count) seems surprisingly low compared to stuff I've read (not experienced yet) with the M1.

I think a lot of the perceivable speedup in M1 Macs is less to do with these benchmarks and more to do with optimizing the OS and Apple software for the M1 and the various accelerators on the SoC (video encoding and ML).

The vast majority of that 6 years is very gradual or minimal improvements. Then M1.

Well the 2019 16 inch i9 MacBook Pro scores are 1088 and 6821 so you can see the massive uplift over just two years.

Comparison with the 2012 macbook I'm typing this on: https://browser.geekbench.com/v5/cpu/compare/10496766?baseli...

I waited for these to come out for a long time, and now "should I upgrade" isn't even a question. It's a "shut up and take my money" situation at this point, but they aren't even taking preorders over here yet. I mean, it could probably run a browser, several electron apps, AND android studio at the same time, how cool is that?!

Ha I couldn't stand the order that was, so here's the reverse:

https://browser.geekbench.com/v5/cpu/compare/10496766?baseli...

3.1% increase in single core, and 69.4% increase in multi-core.

Sorry, I think you meant to reply to this post with it: https://news.ycombinator.com/item?id=28935095

What am I missing here? The marketing presentations spoke about 200GB/sec and 400GB/sec parts. Existing CPU's generally have 10s of GB/sec. But I see these parts beating out existing parts by small margins - 100% at best.

Where is all that bandwidth going? Are none of these synthetic benchmarks stressing memory IO that much? Surely compression should benefit from 400GB/sec bandwidth?

This also raises the question how are those multiple memory channels laid out? Sequentually? Stripped? Bitwise, byte wise or word stripped?

The SoC having 400GB/s of memory bandwidth doesn't mean that any individual CPU core can saturate 400GB/s of memory bandwidth (or even if all the CPU cores combined can achieve that). The SoC's memory bus is also feeding the GPU, and GPUs tend to be _very_ memory bandwidth hungry (see other discreet GPUs pushing over 1TB/s of memory bandwidth).

The CPU performance, even where memory IO limited, is more likely limited by how well it can prefetch memory and how many in-flight reads & writes it can do. A straight memcpy benchmark might be part of this suite, but that'd also be basically the only workload where a CPU core (or multiple) could come anywhere close to hitting 400GB/s of bandwidth. Otherwise memory _latency_ will be the bigger memory IO limitation for the CPUs, and that likely hasn't drastically changed with the new M1 Pros & Max's (there may be some cache layout changes which would shift some of the numbers, but DRAM latency is likely unchanged)

For an example of this in a different product: https://www.anandtech.com/show/15578/cloud-clash-amazon-grav...

A single graviton2 CPU can "only" achieve 18-36GB/s of memory bandwidth even though the package in total can hit 200GB/s

As people stated in the announcement post, a high memory bandwidth doesn't really benefit CPUs much. That's also why AMD and Intel don't have high memory bandwidth on their cores, because it doesn't really help performance.

Where it is beneficial is the GPU; for comparison AMD and Nvidia cards often exceed 400GB/s. A desktop RTX 3090 has 900GB/s.

They have 5.2 and 10.4 TFLOPS of GPU power to feed in addition to 10 very wide cores.

Pretty impressive, all things considered. Looks like it's roughly on par with an AMD 5800X, which is a Desktop CPU with 8C/16T and a 105W TDP.

5800x seems 20% faster for single core: https://browser.geekbench.com/v5/cpu/5874365 so def not on part.

That's certainly overclocked or very high PBO offsets, at stock the 5800x gets around 1700-1800 ST.

But if you run TWO single-threaded workloads at once, the clocks dial WAY back. Meanwhile, the M1 can keep all cores at 3.2GHz pretty much indefinitely.

This is the official score: https://browser.geekbench.com/processors/amd-ryzen-7-5800x

Impressive but unbelievably expensive considering what will actually be run on it.

I kind of want one, but the 5800x for example is about 10x less expensive than a specced out MacBook Pro

Are you comparing the price of a chip to a laptop??

Workstation laptops are expensive. These new Macs are priced quite competitively. E.g. the 14“ with the full M1 Pro is $2.5k and is faster, more portable and has a much better display than a 2.7k Dell Precision 5560…

People are probably going to be reading your comparison as an objective comparison rather than an opportunistic one. For example, if you are choosing between upgrading AM4 processors on your desktop versus buying a new Macbook Pro, then of course it makes sense to compare the cost in those terms. However, the price of the M1 chip is obviously probably not that bad. Since you can’t meaningfully buy it on its own, I guess there is no fair comparison to make here anyways.

The 10-core (8 performance) M1 Max starts around $2700 in the 14" form factor.

It's hard to compare to laptops, but since we started the desktop comparison, a Ryzen 7 5800X is $450 MSRP, or about $370 current street price. Motherboards can be found for $100, but you'll more likely spend $220-250 for a good match. 32GB RAM is $150, 1TB Samsung 980 Pro is a bit under $200. Let's assume desktop RTX 3060 for the graphics (which is probably slightly more powerful than the 32-core GPU M1 Max) for MSRP $330 but street price over $700.

So we're at about $1670 for the components, before adding in case ($100), power supply ($100) and other things a laptop includes (screen, keyboard...).

IMHO, ditching Intel is a smart move for apple but a bad one for hackers. I’m sticking with older intel mbps until decent emulation scores are achieved with qemu, and probably further down the line. Hypervisors (say VBox proxmox combo) are so useful when you have to experiment/test with other platforms. Happy to see HDMI and escape key back tho, and it looks like we’ll even get a Target Display Mode equivalent! We’ll be back to my 2015 setup.:-)

My ideal M1 mbp would be this new one with 64gb ram but with a second cheap mid tier i5 with 16gb dedicated, and a rtx2070(cmon NVdia has Cuda —we need that) with 8gb, connected via internal tb3. I don’t care that it would be 700g heavier and a bit bigger or have less battery life when used to the max (idled chips should not eat a lot). It terms of bare costs thats maybe 1000$ more silicon, but it’s a world more of possibilities. What’s wrong with dedicated gpu offering— not for everyone but hackers love it and help the aura.

And —-please—- bring back a modular slot (m2?) for (additional?) storage! (Also good for hackers but not for Apple)

Target display mode on the laptop? Where'd you see that?

I read that Monterrey will Have an AirPlay based feature to do something like TDM. www.macrumors.com/2021/06/09/airplay-mac-to-mac-external-display/amp/

One big thing to consider is that this is just the first m1 Max Geekbench score, compared against the world's greatest 11900Ks. Most 11900K's are nowhere near the levels of the top preforming one.

Once Apple starts shipping M1 Max in volume, and as TSMC yields get better, you will see "golden chips" slowly start to score even higher than this one.

Golden chips?

Compared to my gen 1 M1 macbook pro: https://browser.geekbench.com/v5/cpu/compare/10514397?baseli...

I'm in the situation where I really want to pre-order a 14", but I have no idea if going with the base model would be a mistake.

Would upgrading to 32GB RAM make Xcode faster? Or would it be a waste of $400?

My opinion: the jump from 16GB to 32GB probably won't make a huge difference today, especially if your workload is already running alright in 16GB. I think it'll greatly extend the useful life of the laptop, though. For example, I wouldn't want to be using an 8GB laptop today.

RAM is something where you either have enough or you don't. When you have enough, adding more doesn't really get you anything. When you don't have enough, performance tanks.

The integrated nature of the M1 Macs gives them a much better ability to predictively swap unused content to SSD and make the most of the available RAM when things get tight while multitasking, but if you have a single task that needs 17GB and you have 16GB it's going to suffer a lot compared to if you had 32GB.

I wish Apple (and everyone else) would give up on the ultra-thin crap and make a real Pro machine that returns to upgradability at the cost of a few extra millimeters, but for now since you're stuck with what you start with I'd recommend always going at least one level above what you think you might ever need during its expected useful life.

hard to tell given that m1's address and page memory completely differently from how x86 does it. My 8GB M1 MacBook Air performs extremely well even when memory pressure is high...and it never seems to hit swap space.

Anecdotal example: I could have several Firefox tabs with active workers in the background (WhatsApp, Slack, etc.), a Zoom meeting with video conferencing on, with audio and video being routed via OBS (native), a Screen Sharing session over SSH going, and a Kubernetes cluster in Docker running, and that won't even make the MacBook hot. Nothing slows down. I could get maybe five hours out of the battery this way. Usually six.

Doing that on a maxed out Intel MacBook Pro will make it sound like a jet engine and reduce my battery life to two or three hours. It will also slow to a crawl.

I'm guessing buying a machine with 32GB of RAM is an investment into the future where workloads on m1 machines are high enough to actually give the memory a run for its money.

16GB of RAM seems so low in 2021. OTOH, hard drives are so fast on these things that maybe 16 is good enough with the SSD as overflow.

I ended up shelling out the extra $400 for 32GB, but didn't feel great about it!

Ram is ram. Don’t fall for marketing hype and think M1 means you need less ram. If you think you need it, you still need it on M1… and I say that as someone who owns an M1 Mac.

Figure out your hourly wage, how many hours a day you use your laptop, and whether $400 amortized over the life of the device is worth the "risk".

Same situation as you, looking at 14". Need to see what the Xcode benchmarks are like for the variety of M1 Pro processors on offer, to see if any upgrades from the base model are worthwhile.

If I was sticking with 16GB RAM I think I'd get a M1 Air instead. The smaller screen is the downside, but 300g lighter and substantially cheaper are good points.

Depends upon your workflow but I use xcode and android studio and 16gb isn't enough if I run a simulator or emulator. Definitely get the 32gb imo.

Unlikely to make Xcode any faster. Just look at your ram usage now and then forecast a bit to know.

I wouldn't go for the base model, since it has 6 performance cores, rather than 8.

Site is under heavy load:

1783 Single-Core Score

12693 Multi-Core Score

For comparison, M1 results:

1705 Single core, 7382 multi core

Anyone know how this compares to the M1 Pro? Curious if it's worth the additional $400. The search doesn't seem to allow exact matches.

Should be the same as the difference between the max and the pro lies in the GPU core count.

https://browser.geekbench.com/v5/cpu/10479712

Do we know of the M1 Max in the MacBook can sustain this level of performance without thermal throttling?

Mobile phones often have this issue where the benchmark looks gear until you run it for 15 minutes and the thermal throttling kicks in and kills the performance.

Thermal throttling is something I'd like to understand when comparing these against a desktop CPU.

For example, I just bought a 5900X and the Geekbench scores are very similar. My assumption is that the 5900X will be able to sustain full boost speeds indefinitely because of the additional cooling. I don't know if the MacBook can do the same.

If it can sustain mac performance without overheating inside the laptop form factor that'll definitely be a big win for Apple.

They increased the thermal cooling area inside the laptop by something like 60% if I recall, it’s a whole new design for temperature control. But they were quick to add that most of the time the fans will still not be needed because the chip rarely gets warm enough to require that.

Where is the graphics test? The M1 Max versus an NVIDIA 3080 or similar?

I'm waiting to be corrected by someone who knows GPU architecture better than me but as far as I can tell the synthetic benchmarks can trade blows with a 3070 or 80 (mobile), but the actual gaming performance isn't going to be as rosy.

Also recall that very few games needing that performance actually work on MacOS

It's probably bad, the M1 could not get 60fps on WoW so ... When I see Apple comparison I would take that with a grain of salts because the M1 is not able to run any modern game at decent fps.

In the keynote Apple said the M1 Max should be comparable to the performance of an RTX 3080 Laptop (the footnote on the graph specified the comparison was against an MSI GE76 Raider 11UH-053), which is still quite a bit below the desktop 3080.

No way it can get anywhere near 3080

Seems kind of unfair with the NVIDIA using up to 320W of power and having nearly twice the memory bandwidth. But if it runs even half as well as a 3080, that would represent amazing performance per Watt.

But does it support running x86 Docker images using hardware emulation?

This is what's holding me off buying one since I don't know whether to get 16 GB RAM (if it doesn't work) or 32 GB RAM (if it does work).

There is no hardware in the M1's for x86 emulation. Rosetta 2 does on the fly translation for JIT and caches translation for x86_864 binaries on first launch.

Docker for Mac runs a VM, inside that VM (which is Linux for ARM) if you run an x86_64 docker image it will use qemu to emulate x86_64 and run Linux Intel ELF binaries as if they were ARM.

That means that currently using Docker on macOS if there is a native ARM version available for the docker image, it will use that, but it can and will fall back to using x86_64 docker images.

That already works as-is. There is no hardware emulation though, it is all QEMU doing the work.

Single core regular M1 Macbook about the same. Multicore the Max is a lot higher.

Looks not the same to me: https://browser.geekbench.com/v5/cpu/compare/10508124?baseli...

Did I pick the wrong regular M1 Macbook?

Edit: Hmm. https://browser.geekbench.com/v5/cpu/compare/10508059?baseli...

I guess Geekbench is a little unpredictable?

And it looks like HN is pushing their capacity...

It should be about the same. It is the same micro-architecture which is why they are still called M1. There are just more of the same cores.

I’m not sure how Geekbench browser tests hold up to real world usage, but note that Apples ecosystem is WAY different compared to x86.

For starters, the chip isn’t general purpose compute, it’s more of an ASIC that is optimized to run the OS abstractions available through the official APIs (Think on screen animation such as rendering an image or detecting things in it. On x86, you implement the code and in some cases, proprietary libraries such as Intels MKL make it seemingly faster to run. On Apple Silicon, for such common use cases there are DEDICATED chip areas. Thus, an ASIC)

Is geekbench seen as a valid score? It seems that on the pc hardware sites and youtube channels I frequent that they don't seem to mention it. The only time I seem to see it is for mac stuff.

It is one of the few cross-platform benchmarks that can be run on both PCs and Macs, as well as iOS, Android, and Linux.

Not really. It's nice to run the same benchmark on all platforms but most hardware sites run a game demo that's easily repeatable and take an fps reading or run Cinebench which renders a scene.

You can find more results at https://browser.geekbench.com/v5/cpu/search?q=Apple+M1+Max

(it does, broadly, appear to be pretty comparable to Intel i9-11900K https://browser.geekbench.com/v5/cpu/search?q=Intel+Core+i9-... )

i9-11900K is 125W TDP, and the M1 is probably nowhere near that (M1 is 10-15W TDP)

I ordered an M1 Max MacBook, really looking forward to it.

It looks like the battery life is fantastic, and it's quiet and cool. Given that, what are some reasons Apple might not have upped the wattage for an even faster chip? Is it just that they are so far ahead of the competition, and these are things that can't easily be matched (fanless operation, long battery life), or is there some physical reason they wouldn't want to run the chip hotter?

Anyone know the score for Intel 12th gen Alder Lake?

1834 ST / 17370 MT [1]

But that's also for something which has a TDP of 125W [2], unsure if that's the right number for a mobile chip? Also no clue what M1 Max's TDP is either.

[1] https://browser.geekbench.com/v5/cpu/9510991

[2] https://cpu-benchmark.org/cpu/intel-core-i9-12900k/

I can't check for you due to hug but the 12900K was clocking in at about 1900 ST IIRC

Edit: Leaving this up so I can be corrected, don't think I have the right figure.

Looks like Geekbench just got the HN bug of death.

I'm envious. I like the hardware a lot. But at the same time, I can't help but think "This has to be the most powerful computer ever built to run a browser and a terminal". I can't run PyTorch with acceleration, I can't use Numba, or CUDA. It seems it will soon be possible to use Blender, which for my artsy side is a plus, and I know that a few of the other design tools I use work on MacOS with varying degrees of acceleration, so there is that.

These are not insurmountable, probably only a matter of time. Tensorflow has already been ported and optimised. https://blog.tensorflow.org/2020/11/accelerating-tensorflow-...

It’s obviously not for you, yet, then? People who need PyTorch or numba or cuda are a vanishing minority even in the tech scene.

This is the first bench I believe might be real. i saw 11542 multi core spammed across news sites a couple days back, but it didn’t align with a 1.7x boost to performance (which this result actually does). The single core was 1749 which also didn’t make sense as I’d imagine there’s maybe a tiny bit more TDP per core in the MBP 14 and 16 than the OG 13” M1 mbp. That and it was on macOS v 12.4… which is maybe nonexistent and thus fake? This score here is believable and incredible.

The single core performance is still the same as the M1, so frankly, for a laptop that's awesome, but I wish single core performance could hit the AMD Ryzen scores.

Depends what benchmark you ask. According to Passmark, the M1 is the best single core CPU no question.

All this makes me want to see Geekbench ported to other binary architectures. I'd love to see how POWER9 and POWER10 compare to Xeons and EPYCs.

How will these chips do for training neural nets? The 64 GB RAM would be awesome for larger models, so I’m willing to sacrifice some training speed.

I know this doesn't answer your question, but can I just ask (out of curiosity) why you're training ML models on a laptop?

- https://web.archive.org/web/20211020175422if_/https://browse...

- https://archive.md/zQAC3

Why does Apple not open up their hardware to other operating systems like Linux. They will already get our money from the hardware purchases, what more do they want.

I know Asahi Linux exists but without Apple's (driver) support it will never reach macOS on M1 level performance.

(If someone disagrees, please compare Nouveau with proprietary Nvidia drivers.)

Nobody wants the Linux desktop experience associated with their brand.

They have some of your money, but they want more of it.

Mac owners are more likely to buy iPads and iPhones and become embedded in the Apple ecosystem. That draws people into the app stores, which Apple gets a cut of, and into Apple Music and Apple TV...

If they're very successful they get you to sign up for Apple Pay and the Apple Card and get a chunk of your spend for things entirely unrelated to Apple.

If they just sell you the hardware and you can easily run Linux on it, you might not run macOS and move further into their ecosystem.

As cool as that would be, why would they? I mean what do they have to gain from it? Maybe a few more Linux users will buy MacBook's but not enough to impact their bottom line. Plus those users wouldn't be part of the Apple ecosystem and at least since the iPod Apple's goal has been bringing you into the ecosystem.

Nouveau is a bit of a special case because Nvidia actively blocks third party drivers from fully leveraging the capabilities of their cards. While Apple isn’t officially supporting any particular third party OS on M-series hardware, they’re also not obstructing the creation of high performance third party drivers for it.

Apple's core competency is selling hardware, with a tightly integrated ecosystem to optimize user experience with that hardware.

There is no incentive to facilitate introducing that hardware into an uncontrolled foreign ecosystem. Apple does not want users looking at their computer while muttering "this sucks" when "this" is an OS+UI famous for incomplete/buggy behavior.

(I've tried going all-in for Linux several times. The amount of "oh, you've got to tweak this..." makes it unusable.)

They did open up their hardware to other operating systems: they engineered a feature into their secure boot system which allows a Mac's owner to attest (with physical presence required) that they'd like to run a kernel not signed by Apple. That's the door Asahi Linux uses. It isn't there on iOS devices, even those also based on M1.

Everything else you're asking for would require them to spend significantly more money and/or engineering resources on supporting the port, and that just doesn't seem like a thing Apple's senior management would go for.

I’d like this but you make it sound like this would not require significant effort from them.

I suspect they would say if you want to run Linux you can do it in a VM and use the already debugged MacOS device drivers.

> Why does Apple not open up their hardware to other operating systems

It's extra effort they don't want to go to. They spent a lot of engineering and support time working with MS on boot camp, handling (what are to them) special cases and back compatibility. They really needed it at the time, but no longer need it so make no effort in making it happen. And Apple never did linux support, it's simply that linux runs on most hardware that runs windows.

Among other things, here's a major reason why it's hard: Apple supports their hardware and software for quite a long time, but is happy to break back compatibility along the way. MS considers back compatibility critical, despite the cost. I respect both positions. But getting windows running on intel Mac hardware wasn't automatic, and required developing support for bits of code designed (by MS) for all sorts of special cases. They simply needed it, so did the work, with the cooperation of MS.

Because Apple intends to sell UX, not just some hardware.

How does that compare to top of the line AMD Ryzen latest gen (site is dead for me right now)?

Single core is similar to the 5900x but multicore is more in line with 3900x.

Quite impressive for a laptop, I'm sure the power consumption will be much higher compared to the M1 but likely no where near desktop parts.

Compared to the late 2020 Macbook Air:

https://browser.geekbench.com/v5/cpu/compare/10508178?baseli...

This confirms that single-core performance is essentially the same as the basic M1.

Would like to see the Metal benchmark too, but doesn't appear yet on the Geekbench Metal page. That would be interesting to see how Geekbench scores it against many discreet GPUs.

I'm curios on the difference between the 8 cores and 10 cores for the M1 Pro. I'm definitely not going to get the Max, I'm just not the user that needs the GPU's

Someone needs to come up with realistic Mac benchmarks. Like 'seconds it takes for Finder beachball to disappear'. At this metric my M1 sees no improvement over my MB12.

Beachball disappears when Apple's developers learn to write asynchronous code. (You can't blaame the hardware for programming decisions that are fundamentally slow or stupid).

If I’m reading this right, it’s only a tiny bit faster than the existing M1 on single core. 1780 vs 1720 or so. For a lot of tasks, single core is what really matters.

Today's the day when people actually check Apple's claims with a graph that didn't originate inside Apple. Remember the reality distortion field guys?

A few days ago there were comments on HN that the Geekbench score on M1 Max was run on an operating system not optimized for M1. Has that changed with this one?

Hmm.

M1 Max = 1783 Single-Core Score 12693 Multi-Core Score

Intel Core i9-9900X 3500 MHz (10 cores) October 20th, 2021 = Single-Core Score 1078 Multi-Core Score 11305

For someone with more knowledge- are these scores dependant on the OS? Like would it have different scores if it ran Linux?

On macos, if you are running some x64 apps, they are jit'ed and emulated. On Linux, all your repo and manually compiled apps will be compiled for the native CPU so it would be fast.

Looks like it's somewhere between my AMD Ryzen 5800X desktop (paid about $800) and a 5900X.

Wondering whether it will be more profitable to work for AMD or APPLE in the next few years?

This puts it at about 3/4 the performance of a 5950X. Not bad for mobile.

I was dead set on getting a new Mac but I think I will opt for Alder Lake. It appears Apple Silicon for desktop will be pretty much constrained to mobile designs and limitations. Perhaps I will revisit if they decide to release M1 Max inside a Mac Mini but I highly doubt that will happen.

Site not loading for me :(

So can someone build a mini cloud service with these machines?

How does this compare to a Mac Pro for video editing

Should be better. The discussed how it was even better than MacPro with afterburner.

Wondering if some of the avid videogamers, in need for powerful GPUs, will consider Macbook Pros with M1 Pro or Max as a good option.

AFAIK, some of the most popular videogames run on Mac too.

Almost none of the most popular actually. No Fortnite no warzone

Some, but not nearly enough.

I wonder what Asahi Linux devs can show us.

Where are the Gflops/W?

you guys nitty grit. it's all about efficiency.

I will get one if it can run Windows and nothing comparable released by then.

Pretty underwhelming from a pure performance standpoint after all the hype from the launch.

https://browser.geekbench.com/v5/cpu/singlecore

Is that supposed to be fast guys?

Isn't that still slower than some ryzen 5 5600x? (my pc uses this but below is not my benchmark)

https://browser.geekbench.com/v5/cpu/8238216

I'm not sure how fast or good is that number.. But i've heard good things about m1 and planning to probably upgrade.

edit: Not sure what's with the downvotes, i'm genuinely curious. I'm in the apple ecosystem with many apple products, windows pc, basically a tech enthusiast without any hate towards 1 brand. geez

It's very fast for a mobile form factor, and still very fast for a desktop processor.

The benchmark you linked is grossly overclocked, as are many others people have linked throughout this thread. Which is fundamentally the problem with the Geekbench browser and people searching around for something to prove a point, when the numbers on there are completely unverified (and sometimes wholly fictitious), unproven, and often under ridiculous scenarios.

Use actual reviews from credible reviewers and extract the GB results from that. e.g. https://www.anandtech.com/show/16214/amd-zen-3-ryzen-deep-di...

And FWIW, I don't even put any credibility on these M1 results. I'll just wait until Anandtech or someone similar has a real review with competently created, reproducible benchmarks.

> Base Frequency 4.72 GHz

> Maximum Frequency 6.03 GHz

This was seriously overclocked, wouldn't be surprised if with liquid nitrogen. So probably poor comparison to laptop CPU ;)

What did I miss? Dell XPS 15 with intel CPU has way higher scores… is Geekbench not capable of working correctly on Apple Silicon?

https://browser.geekbench.com/v4/cpu/16376678

That's a Geekbench v4 score, which isn't comparable to a Geekbench v5 score.

You can see the same laptop on Geekbench v5 here, where it scores much, much lower than the M1: https://browser.geekbench.com/v5/cpu/10506812

You’re comparing Geekbench 4 and 5 scores - you can’t do that. It’s a different scale.

Here’s an example of a Geekbench 5 score: https://browser.geekbench.com/v5/cpu/10501477

Here's the actual Dell XPS 17 vs MacBook Pro: https://browser.geekbench.com/v5/cpu/compare/10502109?baseli...

Is that a legitimate GB score for that machine? Searching for scores from that machine gives me: https://browser.geekbench.com/v5/cpu/search?q=XPS+15+9510

Which shows score much lower than the linked test (maybe I'm missing something). Likewise unless it's coming from someone like AnandTech I'm skeptical of any benchmarks for the M1 Pro/Max until these machines are actually released next week.

I think you’re looking at a geekbench 4 score, vs v5 in the post. It seems that this machine performs significantly better than an XPS 15.

https://browser.geekbench.com/v5/cpu/10501477 V5 score for that

That's a v4 score, which can't be compared to a v5 score.

Geekbench 4 and Geekbench 5 are different scales.

Edit: about 5 people beat me.

Reply on news.ycombinator.com