Apple M1 ARM Performance With A 2020 Mac Mini

[carewolf]

It outperforms the fastest x86 desktops including Zen 3 at a fraction of the power. The only conclusion: game over.

I wonder what on Earth you are smoking. It loses most benchmarks against a 3 year old low power ultrabook chip in these tests.

[ultimA]

The results on Openbenchmarks with modern CPUs tell a different story

https://openbenchmarking.org/result/...MERGE68890&sor

Again, the M1 performance are good -- this is not a weak CPU
But it is not (yet?) a revolution

And yes, I agree with you that this is a wake up call for AMD and Intel.
But I do not want that the wake up call translates into RAM-on-chip for all.

I personally believe that the major advantage of he M1 is the RAM in the same package, but this has serious limitation for i.e. ppl that build their own PC from individual parts, ect

The results on that page you linked are very mixed. In some (many) cases the Intel CPUs have the advantage, on other some (many) cases the M1. In both cases, perf differences range from small to great. Which again is very impressive for the M1 given its much better heat and power characteristics.

As for the concept of RAM in same package as CPU, it is not actually new. It has a long history on embedded and SBCs. It is only new in the desktop(/laptop) segment.

[JackLilhammers]

7% isn't that much for company specializing in desktop computers. Linux was even over 3% not so long ago.

Yes, and then it came back to the normal 1-2%, because it was just a hiccup. Linux adoption has been quite stable (and stagnant) over the last few years

[BillBroadley]

AnandTech has some memory system results, it's 68GB/s max. But with those humongous caches you can run most code without relying on fast DRAM.

Ah, that does look good. But I expect the dig differences in parallel random lookups.

[edwaleni]

Looks like somebody had too many cups of "Joe Bitch" coffee.

Uhhh...you do know that this is x86 binaries running in a translation similar in vein to Wine...right? And the M1 STILL came in second in WebP encoding. Right?

And on FLAC it only ran bad on native ARM because, as Michael pointed out, FLAC has NO ARM optimizations to speak of.

Which is why i have been saying ever since Apple revealed months ago they were going to release ARM only Macs from here on out...and even BEFORE Apple revealed this and it was just rumor for years, that the FLOSS world and Linux had better get their asses in gear and port EVERYTHING over to ARM like yesterday. Every FLOSS project should have a dual track of x86 AND ARM development.

The Age of ARM is here.

By the end of the decade, 2029 + 1, the majority of laptops and desktops will be ARM based. How do I calculate that? Simple. Every Apple laptop and desktop by the end of 2021 will be ARM based. More than half of EVERY Chromebook and Chrome desktop will be ARM based by 2030. And at least a quarter if not half of EVERY Windows based laptop and desktop will be ARM based.

Of course, that's not to include every mobile device in the world, which is 99.5% ARM.

And not to mention that the world's fastest in Raw Compute AND the world's most power efficient Supercomputer is ARM based.

Tick Tock desktop Linux. We are becoming increasingly marginalized and irrelevant.

starshipeleven should be landing any moment now.

[Jumbotron]

This from Ars Technica.....Windows CAN run on Apple Silicon M1.

What about Windows?


But while Apple Silicon Macs can run existing Mac, iPhone, and iPad software, the new architecture cannot immediately run applications built for x86 operating systems besides macOS, and Rosetta 2 doesn't offer any help on this front. The ability to run Windows software was a small part of the success of Intel-based Macs, so some users—particularly those with certain professional workflows—will see that as a loss.

We asked what an Apple Silicon workflow will look like for a technologist who lives in multiple operating systems simultaneously. Federighi pointed out that the M1 Macs do use a virtualization framework that supports products like Parallels or VMWare, but he acknowledged that these would typically virtualize other ARM operating systems.

"For instance, running ARM Linux of many vintages runs great in virtualization on these Macs. Those in turn often have a user mode x86 emulation in the same way that Rosetta does, running on our kernel in macOS," he explained.

As for Windows running natively on the machine, "that's really up to Microsoft," he said. "We have the core technologies for them to do that, to run their ARM version of Windows, which in turn of course supports x86 user mode applications. But that's a decision Microsoft has to make, to bring to license that technology for users to run on these Macs. But the Macs are certainly very capable of it."

https://arstechnica.com/gadgets/2020/11/we-are-giddy-interviewing-apple-about-its-mac-silicon-revolution/

[Slartifartblast]

Zen 4? That's due for 2022, by which time there will be an M2, perhaps even M3 depending on the timeline. Either way Apple will be even more ahead then.

I'll leave Apple to the fanboys and their tithe to the Church of Jobs, meanwhile I'll look at value for money and no vendor lock-in. Good luck running Linux on their hardware.

A classic Jobsian reality distortion field.

[Jumbotron]

Those L1 sizes are per-core. So in total there is 2MB L1, 16MB L2 plus 16MB L3. No surprise it's the fastest CPU in the world! The translation performance is also absolutely incredible, I wasn't expecting it to beat native x86 CPUs in so many cases.

Ahhh....yes, of course you are absolutely correct. The L1 numbers do get fractal after a while. I overlooked that.

As I said at the earlier thread when Michael first posted the news of the M1 launch in the new Macbook Pro, I mentioned that Apple took a page from the world of Supercomputing and loaded up the M1 with scads and scads of high speed memory to keep the cores loaded with data at the ready. Memory pools are a thing now in Supercomputing to do just that because the data sets are so stupidly massive for EVERYTHING now from Nuclear Simulations to Meteorology to A.I., you can't just have enough RAM sitting on the boards.

Apple on the other hand, fundamentally understands the difference between an ARM core and a hot and heavy x86 core and just loaded a memory pool right on die and on the same package as the SoC. Now to be sure. L1, L2, L3 and RAM have been around forever. Nothing new. But what IS new is Apple going MASSIVE on the L1, keeping L2 sufficiently big, using built in storage for composable L3, according to Anandtech, AND using low latency, HBM RAM on package tied into the SoC with what is probably the CCIX ARM zero copy cache coherent interconnect to all cores simultaneously. You could look at Apple's RAM as a "Data Pool" but not dissagregated out to another rack as it would be in a Supercomputer setup.

The M1 is quite literally a Supercomputer cabinet architecture that has been shrunk down to the wafer.

[PerformanceExpert]

I wonder what on Earth you are smoking. It loses most benchmarks against a 3 year old low power ultrabook chip in these tests.

Just look at native results for SPEC on AnandTech. Beware, it's a massacre...

[Jumbotron]

Ohhh...and for all you Apple haters who poo poo the M1 as a "phone chip", wrong again.

" The M1 is essentially a superset, if you want to think of it relative to A14. Because as we set out to build a Mac chip, there were many differences from what we otherwise would have had in a corresponding, say, A14X or something.

We had done lots of analysis of Mac application workloads, the kinds of graphic/GPU capabilities that were required to run a typical Mac workload, the kinds of texture formats that were required, support for different kinds of GPU compute and things that were available on the Mac… just even the number of cores, the ability to drive Mac-sized displays, support for virtualization and Thunderbolt.

There are many, many capabilities we engineered into M1 that were requirements for the Mac, but those are all superset capabilities relative to what an app that was compiled for the iPhone would expect.

The foundation of many of the IPs that we have built and that became foundations for M1 to go build on top of it… started over a decade ago. As you may know, we started with our own CPU, then graphics and ISP and Neural Engine.

So we've been building these great technologies over a decade, and then several years back, we said, "Now it's time to use what we call the scalable architecture." Because we had the foundation of these great IPs, and the architecture is scalable with UMA.

Then we said, "Now it's time to go build a custom chip for the Mac," which is M1. It's not like some iPhone chip that is on steroids. It's a whole different custom chip, but we do use the foundation of many of these great IPs. "

https://arstechnica.com/gadgets/2020...on-revolution/