Hacker News Reader

Comment by VortexDream | Hacker News Reader

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.

View on HN

Don't get me wrong: It's impressive and I have huge respect for it. I also bought one.

However, it would be surprising if Apple's new 5nm chip didn't beat AMD's older 7nm chip at this point. Apple specifically bought out all of TSMC's 5nm capacity for themselves while AMD was stuck on 7nm (for now).

It will be interesting to see how AMD's new 6000 series mobile chips perform. According to rumors they might be launched in the next few months.

This definitely is a factor. Another thing that people frequently overlook is how competitive Zen 2 is with M1: the 4800u stands toe-to-toe with the M1 in a lot of benchmarks, and consistently beats it in multicore performance.

Make no mistake, the M1 is a truly solid processor. It has seriously stiff competition though, and I get the feeling x86 won't be dead for another half decade or so. By then, Apple will be competing with RISC-V desktop processors with 10x the performance-per-watt, and once again they'll inevitably shift their success metrics to some other arbitrary number ("The 2031 Macbook Pro Max XS has the highest dollars-per-keycap ratio out of any of the competing Windows machines we could find!")

AMD is already planning not only 20% YoY performance improvements for x86 but now has 30x efficiency plan for 2025. I think x86 is in it for much longer than a decade.

Intel otoh, depends on if they can gut all the MBAs infesting them.

Alder lake is looking really good in leaked benchmarks. I definitely think Intel is down, but not out.

Have to see if they can not only catch up, but keep up.

Where can I find these benchmarks?

https://videocardz.com/newz/intel-core-i7-12700-non-k-alder-...

30x efficiency is specifically for 16 bit FP DGEMM operations, and it is in the context of an HPC compute node including GPUs and any other accelerators or fixed function units.

For general purpose compute, so such luck unfortunately. Performance and efficiency follows process technology to a first order approximation.

https://www.tomshardware.com/news/amd-increase-efficiency-of...

also bear in mind that AMD's standards for these "challenges" have always involved some "funny math", like their previous 25x20 goal, where they considered a 5.02x average gain in performance (10x in CB R15 and 2.5x in 3D Mark 11) at iso-power (same TDP) to be a "32x efficiency gain" because they divided it by idle power or some shit like that.

But 5x average performance gain at the same TDP doesn't mean you do 32x as much computation for the same amount of power. Except in AMD marketing world. But it sounds good!

https://www.anandtech.com/show/15881/amd-succeeds-in-its-25x...

Like, even bearing in mind that that's coming from a Bulldozer derivative on GF 32nm (which is probably more like Intel 40nm) 5x gain in actual computation efficiency is still a lot, and it's actually even more in CPU-based workloads, but AMD marketing can't help but stretch the truth with these "challenges".

Wow that's stupid, I didn't look that closely. So it's really a 5x perf/watt improvement. I assume it will be the same deal for this, around 5-6x perf/watt improvement. Which does make more sense, FP16 should already be pretty well optimized on GPUs today so 30x would be a huge stretch or else require specific fixed function units.

it's an odd coincidence (there's no reason this number would be related, there's no idle power factor here or anything) but 5x also happens to be about the expected gain from NVIDIA's tensor core implementation in real-world code afaik. Sure they advertise a much higher number but that's a microbenchmark looking at just that specific bit of the code and not the program as a whole.

it's possible that the implication here is similar, that AMD does a tensor accelerator or something and they hit "30x" but you end up with similar speedups to NVIDIA's tensor accelerator implementation.

I've seen tensor cores really shining in... tensor operations. If your workload can be expressed in convolutions, and are matching the dimensions and batching needs of tensor cores, there's a world of wild performance out there...

To be fair idle power is really important for a lot of use cases.

In a compute focused cloud environment you might be able to have most of your hardware pegged by compute most of the time, but outside of that CPUs spend most of their time either very far under 100% capacity, or totally idle.

In order to actually calculate real efficiency gains you'd probably have to measure power usage under various scenarios though, not just whatever weird math they did here.

That's not really being fair, because the metric is presented to look like traditional perf/watt. And idle is not so important in supercomputers and cloud compute nodes which get optimized to keep them busy at all costs. But even in cases where it is important, averaging between the two might be reasonable but multiplying the loaded efficiency with the idle efficiency increase is ludicrous. A meaningless unit.

I can't see any possible charitable explanation for this stupidity. MBAs and marketing department run amok.

Yep 100% agree with you - see my last sentence. Just trying to clarify that the issue here isn't that idle power consumption isn't important, it's the nonsense math.

> This definitely is a factor. Another thing that people frequently overlook is how competitive Zen 2 is with M1: the 4800u stands toe-to-toe with the M1 in a lot of benchmarks, and consistently beats it in multicore performance.

It's a bit unfair to compare multicore performance of a chip with 8 cores firing full blast against another 8 core chip with half of them being efficiency cores.

The M1 Max (with 8 performance cores) multicore performance score posted on Geekbench is nearly double the top posted multicore performance scores of the 5800U and 4800U (let alone single core, which the original M1 already managed to dominate).

It'll be interesting to see how it goes in terms of performance-per-a-watt. Which is what really matters in this product segment. The graphs Apple presented indicated that this new line up will be less efficient than the original M1 at lower power levels, but they'll be able to hit it out of the park at higher power levels. We'll have to wait to see the results from the likes of Anandtech to get the full story though.

Personally, I'd love to see a MacBook 14 SE with an O.G. M1, 32 GB memory, a crappy 720p webcam, and no notch. I'd buy as many of those as they'd sell me.

I'm curious to see how the M1 Pro compares to the M1 Max. They are both very similar processors with the main differences being the size of the integrated GPU and the memory bandwidth available.

https://browser.geekbench.com/v5/cpu/search?utf8=%E2%9C%93&q...

The M1 Max isn't competing with the 4800u, considering that it's starting price is ~10x that of the Lenovo Ideapad most people will be benching their Ryzen chips with. I don't think it's unfair to compare it with the M1, since it's still more expensive than the aforementioned Ideapad. Oh, and the 4800u came out 18 months before the M1 Air even hit shelves. Seeing as they're both entry-level consumer laptops, what might you have preferred to compare it with? Maybe a Ryzen 9 that would be more commonplace in $1k+ laptops?

It's hard to compare CPUs based on the price of the products they're packaged in. There are obviously a lot of other bits and bobs that go into them. However, it's worth noting that a MacBook Air is $300 cheaper than the list price for the Ideapad 4800U with an equivalent RAM and storage configuration. So by your logic, is it fair to compare the two? Perhaps, a 4700U based laptop would be a fairer comparison?

The gap between the Ideapad 4800U and the base model 14 inch MacBook Pro is a bit wider, but you'll also get a better display panel and LPDDR5-6400[1] memory.

We'll have to see how the lower specced M1 Pros perform, but it's hardly clear cut.

[1] https://www.anandtech.com/show/17019/apple-announced-m1-pro-...

Edit: I just looked up the price of the cheapest MacBook Pro with a M1 Max processor and it's about 70% more expensive than the Ideapad 4800U. However, with double the memory and much better quality memory and a better display and it seems roughly about 70% better multithreaded performance in Geekbench workloads. Furthermore, you may get very similar performance on CPU bound workloads with the 10 core M1 Pro, the cheapest of which is only 52% more expensive than the Ideapad 4800U.

the 4800u stands toe-to-toe with the M1 in a lot of benchmarks, and consistently beats it in multicore performance

I had a Lenovo ThinkPad with a 4750U (which is very close to the 4800U) and the M1 is definitely quite a bit faster in parallel builds. This is supported by GeekBench scores, the 4800U scores 1028/5894, while the M1 scores 1744/7600.

If AMD had access to 5nm, the CPUs would probably be more or less or par. Well, unless you look at things like matrix multiplication, where even a 3700X has trouble keeping up with the Apple AMX co-processor with all 3700X's 8 cores fully loaded.

During my time, the Mac line has switched CPUs three times, why would Apple not switch to RISC-V if that one is really so much better than ARM?

well....M1 Max single core also beating a 5950x

But tbh, it doesn't seem the new hotness in chips is single core CPU, it's about how fancy you spend the die space in custom processors, in which case the M1 will always be tailored to Apple (and presumably Mac users') specific use-cases...

My 5950x compiles the Linux kernel 10 seconds faster than a 40 core xeon box, in like 1/10th the power envelope.

The chances of actually getting a only a single core working on something are slim with multitasking, I had to purpose build stuff - hardware and kernel/etc for CPU mining a half decade ago to eliminate thermal and pre-emption on single threaded miners.

Single thread performance has been stagnant forever because with Firefox/chrome and whatever the new "browser app" hotness is this month your going to be using more than 1 core virtually 100% of the time, so why target that. Smaller die features means less tdp which means less throttling which means faster overall user experience.

I'm glad someone is calling out the M1 performance finally.

You actually get better single core performance out of a 5900x than a 5950x. The more cores the AMD CPU has, generally the more they constrain the speed it can perform at on the top end. In this case, the 5950x is 3.5GHz and the 5900x is 3.7GHz. The 5800x is even slightly faster than that, and there's some Geekbench results that show single core performance faster than the score listed here, but at that point the multi-core performance isn't winning anymore.

Also, I'm not sure what's up with Geekbench results on MacOS, but here's a 5950x in an iMac Pro that trounces all the results we've mentioned here somehow.[1]

1: https://browser.geekbench.com/v5/cpu/6034871

That's not necessarily true, single core boost on a 5950x is higher than the 5900x (4.9GHz vs 4.8GHz).

>that trounces all the results we've mentioned here somehow.[1]

MacOS, being unix based, has a decent thread scheduler - unlike windows 10/11, which is based on windows NT and 32bits, and never cared about anything other than single core performance until very very recently.

If that's true it puts a lot of comparisons into question. That windows multiprocessing isn't as good as MacOS doesn't matter to a lot of people that run neither. There's not a lot of point in using these benchmarks to say something about the hardware if the software above it but below the benchmark can cause it to be off by such a large amount.

Most comparisons have always been questionable. Main reason MacOS gets away with charging so much more for similar hardware and still dominates the productivity market is it squeezes so much more performance (and stability) out of equivalent hardware. Just check the geekbench top multithread scores, windows starts around the 39th _page_ - and thats for windows enterprise.

That's obviously a Hackintosh or a VM result.

no it doesnt, you must have misread the benchmark scores.

Geekbench score for a 5900x (12 core) against the M1 Max (10 core) score already linked:

https://browser.geekbench.com/v5/cpu/compare/10517471?baseli...

They look to be on par to me. Things will be less murky if and when Apple finally scale this thing up to 100+ watt desktop class machines (Mac Pro) and AMD move to the next gen / latest TSMC process.

In my view Intel and AMD have more incentive to regain the performance crown than Apple do at maintaining it. At some point Apple will go back to focusing on others areas in their marketing.

Geekbench.com says m1 max 1700, 5950x 2200

It wouldn't be a surprise that a CPU that can't run most of the software out there (because that software is x86) and has ditched all compatibility and started a design from scratch, can beat last-gen CPUs from competitors. For specific apps and workloads, for which it has accelerators.

But here I am, with a pretty thin and very durable laptop that has a 6 core Xeon in it. It gets hot, it has huge fans, and it completely obliterates any M1 laptop. I don't mean it's twice as fast. I mean things run at 5x or faster vs an M1.

Now, this is a new version of the M1, but it's an incremental 1-year improvement. It'll be very slightly faster than the old gen. By ditching Intel, what apple did is making sure their pro line - which is about power, not mobility, is no longer a competitor, and never will be. Because when you want a super fast chip, you don't design up from a freaking cell phone CPU. You design down from a server CPU. You know, to get actual work done, professionally. But yeah, I do see their pro battery is 11 hours while mine usually dies at 9. Interesting how I got my computer plugged in most of the time though...

Intel tried to ”design down” their uArch.

Also, there is a TON of pro Mac users. If we define ’pro’ as getting paid for work done on Macs..

Not to mention M1 emulates x86 pretty darn well..

Probably not faster than an M1 Pro and definitely not faster than the M1 Max.

Your machine doesn't have a 512-bit wide memory interface running at over 400GB/s.

Does the Xeon processor in your laptop have 192KB of instruction cache and 24MB of L2 cache?

Every ARM instruction is the same size, enabling many instructions to be in flight all at once, unlike the x86-64 architecture where instructions vary in size and you can't have nearly as many instructions in flight at once.

Apples-to-apple: at the same chip frequency, an M1 has higher throughput than a Xeon and most any other x86 chip. This is basic RISC vs. CISC stuff that's been true forever. It's especially true now as increases in clock speeds has dramatically slowed and the only way to get significantly more performance is by adding more cores.

On just raw performance, I'd take the 8 high-performance cores in an M1 Pro vs. the 6 cores in your Xeon any day of the week and twice on Sunday.

And of course, when it comes to performance per watt, there's no comparison and that's really the story here.

Now, this is a new version of the M1, but it's an incremental 1-year improvement.

If you read AnandTech [1] on this, you'll see this is not the case—there have been huge jumps in several areas.

Incremental would have resulted in the same memory bandwidth with faster cores. And 6 high-performance cores vs. the 4 in the original M1.

Except Apple didn't do that—they doubled the number to 8 high-performance cores and doubled the memory width, etc. There were 8 GPU cores on the original M1 and how you can get up to 32!

Apple stated the Pro and the Max have 1.7x of the CPU performance of Intel's 8-core Core i7-11800H with 70% lower power consumption. There's nothing incremental about that.

By ditching Intel, what apple did is making sure their pro line - which is about power, not mobility, is no longer a competitor, and never will be.

Pro can mean different things to different people. For professional content creators, these new laptops are super professional. Someone could take off from NYC and fly all the way to LA while working on 16-inch MacBook Pro with a 120 MHz mini LED 7.7 million pixel screen that can display a billion colors in 4k or 8k video—battery only.

If you were on the same flight working on the same content, you'd be out of power long before you crossed the Mississippi while the Mac guy is still working. At half the weight of your laptop but a dramatically better display and performance when it comes to video editing and rendering multiple streams of HDR video.

The 16-inch model has 21 hours of video playback which probably comes in handy in many use cases.

Here's a video of a person using the first generation, 8 GB RAM M1 Mac to edit 8K video; the new machines are much more capable: https://youtu.be/HxH3RabNWfE.

[1]: https://www.anandtech.com/show/17019/apple-announced-m1-pro-...

Sorry but this entire post reads (skims) like you're playing top trumps with chip specs.

>Because when you want a super fast chip, you don't design up from a freaking cell phone CPU. You design down from a server CPU.

Is that really true? I don't have any intricate chip knowledge, but it rings false. Whether ARM is coming from a phone background or the Xeon from a server background, what matters in the end is the actual chip used. Maybe phone-derived chips even have an advantage because they are designed to conserve power whereas server chips are designed to harvest every little ounce of performance. IDK a lot about power states in server chips, but it would make sense if they aren't as adapted to rapidly step down power use as a phone chip.

Now, you might be happy with a hot leaf-blower and that's fine. But I would say the market is elsewhere: silent, long-running, light notebooks that can throw around performance if need be, you strike me as an outlier.

Pro laptops should have a beefy CPU, great screen, really fast SSD, long battery life, lots of RAM which (presumably) your notebook features, but the new M somethings seemingly as well. But in the end, people buy laptops so they can use them on their lap occasionally. And I know my HP is getting uncomfy hot, the same was said about the intel laptops from Apple I think.

Apple doesn't need to have the one fastest laptop out there, they need a credible claim to punching in the upper performance echelon - and I think with their M* family, they are there.

You actually have it correct. When you start with an instruction set designed to conserve power, you don't get "max power." The server chips were designed with zero power considerations in mind - the solution to "too much power" is simply "slap a house-sized heatsink on it."

>Apple doesn't need to have the one fastest laptop out there

correct. My complaint, which I have reiterated about 50 times to shiny iphone idiots on here who don't do any real number crunching for work, is when the industry calls "mid tier" something that apple calls "pro" - apple is deceiving the consumer with marketing. The new laptops are a competition to Dell's Latitude and XPS lines. Not their pro lines. Those pro laptops weigh 7lb, and have a huge, loud fan exhaust on the back so they can clock at 5GHz for an hour. They have 128GB of RAM - ECC RAM, because if you have that much RAM w/o ECC, you have a high chance of bit errors.

There are many things you can do to speed up your stuff, if you waste electricity. The issue is not that apple doesn't make a good laptop. It's that they're lying to the consumer. As always. Do you remember when they marketed their acrylic little cube mini-desktop? It was "a supercomputer." They do this as a permanent tactic - sell overpriced underperforming things, and lie with marketing. Like using industry standard terms to describe things not up to that standard.

I’ll happily take my quiet, small, and cool MacBook and number crunch in a data center infinitely more powerful then your laptop. Guess that makes me a shiny iPhone idiot.

Relax, no one is forcing you to use Apple products.

the main reason it does it is because apple bought up all the 5nm capacity though. amd is running at 7nm still. so impressive because they could afford to do that I guess.

Reply on news.ycombinator.com