Apple’s M1 chip fastest laptop CPU in the world
-
@Klaus said in Apple’s M1 chip fastest laptop CPU in the world:
@George-K said in Apple’s M1 chip fastest laptop CPU in the world:
Perhaps the "Fastest CONSUMER (except Mark’s) CPU ever?"
FIFY.
LOL
-
@Klaus said in Apple’s M1 chip fastest laptop CPU in the world:
@xenon said in Apple’s M1 chip fastest laptop CPU in the world:
So in short - intel has a lot of design legacy to deal with and can’t just optimize for today’s common workloads
I don’t think that this is the main factor. CISC vs RISC, that’s the big difference.
Not necessary a conflict between @xenon and @Klaus' points. Not hard to imagine, with the smaller transistor feature sizes today, that you can pack legacy Intel x86 CISC cores and AArchx64 RISC cores into the same die and have one chip that runs both instruction sets while sharing on-chip common caches and I/Os. Coming up with an OS that can simultaneously take manage both will take more work. The business/legal folks have to remove commercial and legal/cross-licensing impediments, but technically something like this can be done.
As a guy who keeps a Windows VM on my x86 Mac to run Windows, I really want to have a Mac option let me run a Windows VM on my future Mac. Not that it's urgent, I can wait a few years. I am just not confident that Apple will find sufficient commercial incentives to build a M-series chip Mac that can also run Windows and craft a version of macOS that can simultaneously manage AArch-64 and x86 cores. (Though, seeing how macOS can make use of an external GPU, maybe extending that framework to make use of an external x86 CPU isn't all that far fetched.) Else we'll have to resort of using software to emulate an x86 PC inside a macOS. The performance will take a hit and the CPU may be made unnecessarily busy then, but I suppose that will still be better than me having to maintain a separate machine just to run a Windows application.
-
https://www.macrumors.com/2020/11/15/m1-chip-emulating-x86-benchmark/
Got a chart there that shows the benchmark of M1-based MacBook Air running Geekbench in Rosetta 2:
-
I saw that too. As I said earlier, I had no problems with the PowerPC/Intel transition. Rosetta worked just fine. Rmarkably well, as a matter of fact.
I have little doubt that Rosetta 2 will not be as good.
But, when your software runs faster in emulation than on native hardware. that's amazing.
-
@George-K said in Apple’s M1 chip fastest laptop CPU in the world:
But, when your software runs faster in emulation than on native hardware. that's amazing.
Yes, that's bound to happen given enough time, like MS-DOS can run faster in an emulator on a modern computer than it could natively on an Intel 8088 PC XT from the 80s. I just was not sure how long that would take, so it's good to see these benchmarks.
-
@Axtremus said in Apple’s M1 chip fastest laptop CPU in the world:
Posting from from a MacBook Air with M1 chip.
So far so good.How much RAM does it have, and are you bumping into any problems because of the RAM?
I assume Mac OS 11, right?
-
Apple's working on its own cellular modem chip:
https://www.theverge.com/2020/12/10/22168779/apple-leak-cellular-modem-johny-srouji-town-hall
-
@George-K said in Apple’s M1 chip fastest laptop CPU in the world:
@Axtremus said in Apple’s M1 chip fastest laptop CPU in the world:
Posting from from a MacBook Air with M1 chip.
So far so good.How much RAM does it have, and are you bumping into any problems because of the RAM?
I assume Mac OS 11, right?
16GB RAM in this case.
It came with macOS 11 preinstalled. -
@George-K said in Apple’s M1 chip fastest laptop CPU in the world:
Apple's working on its own cellular modem chip:
https://www.theverge.com/2020/12/10/22168779/apple-leak-cellular-modem-johny-srouji-town-hall
Makes sense for Apple to do this, technologically as well as business wise. From a integration perspective, you always gain performance and/or save power when you combine multiple chips into one chip. Qualcomm offers cellular modem chips that come with integrated general purpose CPU cores. Apple can similarly integrate cellular modem circuitries into their A-series CPU chips. As for business case, it’s also not hard to see savings and risk reduction for Apple to have one fewer critical chip supplier.
The only caveat is execution. High performance radio is a bit of a dark art and Qualcomm does it better that just about everyone else. So not easy to see if Apple’s own chip can match Qualcomm’s performance in one metric or another.
-
Application software upgrades take a LOT longer to complete.
Upgrading Apple's own applications like Pages, Numbers, Keynote, GarageBand, iMovie. If I read the descriptions right, it seems only iMovie has been rebuilt with native M1 support. The other applications still have to work through Rosetta 2. Since Apple says a Rosetta 2 translation happens upon installation (and "upgrade" is just a form of "install"), I suppose it makes sense that upgrading takes a longer. -
Waking from "sleep" is practically instantaneous. There was an older version of OS X that could wake a laptop from "sleep" instantaneously, but more recent iterations of macOS 10.x have taken longer to wake from "sleep." Glad to have the "instantaneous wake" back with M1/macOS 11.
-
As pointed out by PCWorld, a huge limitation of Windows on ARM machines is that the operating system was restricted to running emulated 32-bit X86 software. In other words, the system wasn’t able to emulate and run 64-bit apps created for AMD and Intel processors.
32-bit software runs at significantly lower performance — and Apple got rid of them in 2019 with macOS Catalina. At the same time, Apple has introduced the Rosetta 2 technology for the new M1 Macs, which basically translates every software created for Intel Macs into an ARM binary that runs better on Apple Silicon-based computers.
Microsoft recently released a beta version of Windows that features emulation for 64-bit X86 software, but still the performance isn’t even close to the new Macs with the M1 chip. In a Geekbench 5 test, Surface Pro X was outperformed by the new M1 MacBook Air and also ranked behind a cheap HP Pavilion laptop running with an Intel Core i5 processor.
Conclusion: Windows on Arm needs a miracle
Two years ago, the future of Windows on Arm looked bright. With what we hoped was a 64-bit emulator waiting in the wings, the Snapdragon’s “good-enough” performance could hold its own, especially with the perks of all-day battery life and LTE connectivity. Today, Project Athena/Evo laptops from Intel’s partners have caught up in all these area. Qualcomm hasn’t launched a significant Windows on Arm chip in about two years, and during its recent Snapdragon Tech Summit the company had basically nothing to say about its future PC plans.
Microsoft’s 64-bit X86 emulator is still in beta, so we can’t make definitive statements about its success. But it’s hard to believe that further development will bridge the vast gulf of performance between Windows on Arm and Apple’s M1-based Macs. In six months, Microsoft may be able to boast that its emulation performance has improved by a significant amount. But without the combined miracle of a much better CPU from Qualcomm or another Arm chipmaker and continued improvements from Microsoft, the future of Windows on Arm looks grim.
-
Hm, interesting, but I wonder why they don't simply recompile Windows and its applications with an ARM compiler and create native ARM binaries? Why is binary compatibility so important? Should be simple to recompile Windows and the biggest standard applications (Office, Adobe stuff, ...) and then live with slower performance for those applications that aren't available as native ARM binaries.
-
Got my simple performance comparison between my Intel-based MacBook Air (2018, 1.6 GHz Dual-Core Intel Core i5) and my Apple M1-based MacBook Air (2020).
Got both systems to do exactly the same task and timed them using the Unix "time" function, I got:
- Core i5: 575.74 real. 571.99 user. 0.91 sys
- Apple M1: 202.37 real. 202.00 user. 0.41 sys
Think of it as a single thread, single core comparison. I am fairly certain my code's core loop is small enough to fit entirely in cache, the problem does not lend itself to parallelism, and there is barely any I/O, so this is a CPU-centric comparison. Doing exactly the same task, the Apple M1 CPU takes only 35% of the time taken by the Intel Core i5 CPU. Or put another way, a single core of the Apple M1 CPU is roughly 2.8 times faster than a single core of the Intel Core i5 CPU.
Back when playing with the circle of light bulbs puzzle, I wrote a simple program to simulate that puzzle. That's what I used to compare the Intel Core i5 MacBook Air and the Apple M1 MacBook Air -- the comparison was to simulate that puzzle for a ring of 36 light bulbs.
The reason I waited this long to do this comparison is to wait for Go lang to support the M1 silicon natively, and that support was finally released on 2021-02-16.
The M1 is indeed very efficient. Running that simulation of 36 bulbs, the Intel system got really hot to the touch and I could hear the fan kicking into high gear. The M1 system stayed cool to the touch through out and it does not even have a fan.
-
Another comparison, these are for a floating-point heavy computation task, exponentiating a close-to-one floating point number to some ridiculously high degree :
- 1.7 GHz Intel Core i7 (2013 MacBook Air): 206.06s
- Apple M1 in Rosetta 2 emulation (2020 MacBook Air) : 116.02s
-
I encountered a very typical M1 compatibility issue today: I have a Canon camera that can be used as a webcam when installing the "EOS Webcam Utility", but it turns out that it doesn't work on the M1. Most of the big software packages run without major problems, but it's always the little things that are problematic. All in all, I should have bought an Intel Mac. The performance benefits are rarely significant in practice, but the sum of all the minor and major compatibility issues are.