Live data from Hacker News

16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

twitter.com

311–320 of 529 posts

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#311

Is anyone else thinking, what the f*ck? Are we in a new era of computing? It certainly feels that way when looking at these desktop class ARM chips, where performance doubled every year or so, just like back in the 80s and 90s.

Completely! Worth the wait given it took about a decade for Arm chips to suck less ;)

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#312
post #18

Nearly universal praise for the M1 versus it's predecessor but I don't see how this really changes Apple's position the way he implies. Their machines are still far more expensive than Windows and can't use high-powered GPUs like AMD/Nvidia. The people who choose PC, still have the same incentives to do so. For most users, the faster performance will probably not be very noticeable. The reduced power consumption is t…

> Their machines are still far more expensive than Windows Which $1000 Windows laptop has this performance? > and can't use high-powered GPUs like AMD/Nvidia. Not relevant to all markets (particularly ultra bookish laptops, which never have discrete GPUs, but more broadly many people just don't need one). I don't think it changes anything much immediately, but it's extremely bad news for Intel in the long run.

Do people actually care about GPUs in their laptops? It seems like any semi-serious gamer would have strong incentives to have a desktop (modular, the prestige of building you own rig). And anyone using GPUs for compute is just going to remote into some server.

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#313
Now can we get nVidia drivers back for MacOS? Give me the ability to run an nVidia eGPU natively in Mac OS and I'll keep buying apple for the foreseeable future! Sure, my remote box with a "proper" gpu is fine, but Mac OS will always be the best medium between Linux and something that's actually useable day to day. I love linux, but god could I not use it as a daily driver.

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#314

Earlier quoted context omitted.

Intel in 2020 is IBM in 1980.

So in 2060 it will still be a huge company with $20B in revenue?

Yeah. When I'm guessing how long something will last, I guess it's about halfway through its life unless I have really good information to the contrary. Intel was founded in 1968, so let's say it will shut down in 2072 (give or take...many...years). And Fortune 500 companies don't seem to totally shrink back to startup head count and revenue even when they go bankrupt—they might have significant revenue but even more expenses. So it seems pretty reasonable to guess Intel will be around with (at least) $20B of revenue in 2060, in spite of their rough patch today.

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#315

Earlier quoted context omitted.

For some projects, linking non-stripped binaries can take huge amounts of virtual memory. At least on ELF/DWARF platforms with the Gnu linker. Not sure if that applies here, since I assume the Mac will be using the LLVM linker, and Mach-O != ELF.

Virtual memory, or real memory? You're never going to run out of virtual memory on a 64-bit platform.

Virtual memory. I mentioned it in this context because running out of physical memory means the machine starts swapping, which often results in a very undesirable slowdown.

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#316

Is anyone else thinking, what the f*ck? Are we in a new era of computing? It certainly feels that way when looking at these desktop class ARM chips, where performance doubled every year or so, just like back in the 80s and 90s.

I suspect HP/Lenovo/Microsoft are squirming in their chairs right now.

I was waiting for AMD Zen3 to build a new desktop. But now that these tests are out, I am very, very tempted to go with Apple again. The Mac Mini numbers look comparable enough and it will be cheaper than an equivalent Windows machine (never thought I'd see this timeline).

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#318

Earlier quoted context omitted.

Right, and I think some benchmarks I've seen this morning indicate that it throttles during more intensive tasks. The Pro benchmarks better, the only difference being the fan.

All CPUs throttle, all of the time. It's been years since anyone shipped a high-performance processor without a closed-loop dynamic thermal control system.

Having a throttling mechanism != the throttling mechanism is engaged

With enough cooling, you can operate a CPU at full tilt and never engage the throttle.

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#319

Earlier quoted context omitted.

Qualcomm chips are not that far behind Apple's, and Microsoft has Windows on ARM ready for a while now. Those performance advantage won't be there for long. I think the bigger question is what does this spell for x86-64?

> I think the bigger question is what does this spell for x86-64? Not much? People are acting like M1 destroys x86, but as AnandTech showed in the recent benchmark, M1 is trading blows with Zen 3 in single thread performance while having much larger core and process advantage (5nm vs. 7nm) thus being actually more expensive to produce.

The only Zen 3 chips are desktop class, so Apple’s first generation ultra book oriented processor is trading blows with the best line of desktop processors. That’s a pretty big deal!

Re: 16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

#320

Is anyone else thinking, what the f*ck? Are we in a new era of computing? It certainly feels that way when looking at these desktop class ARM chips, where performance doubled every year or so, just like back in the 80s and 90s.

So though it's a new age in some ways, I don't think it will benefit many people. It will just increase the churn. Witnessing all the cries of "16GB of RAM isn't enough!" makes it quite apparent that software behaves like a gas rather than a liquid, and will expand to consume whatever hardware is available to developers. I think that for consumer products, which these are targeting, software responsiveness, usability…

"What Andy giveth, Bill taketh away".

https://en.wikipedia.org/wiki/Andy_and_Bill%27s_law

Post reply on HN