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

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

#201

Post by j4yav » Tue, Nov 17, 2020, 6:41 PM UTC

reaperducer wrote: post #187

bmurphy1976 wrote: post #69

Earlier quoted context omitted.

Normally this wouldn't feel so ground breaking, but the stars are in alignment and all these improvements are hitting at the same time. We're seeing years of work and investment paying off (AMD, Apple, ARM, Nvidia, Amazon), new process nodes (TSMC), and new tech (Ray Tracing, DLSS, machine learning) all hitting at the same time. And that's the big stuff! There's also the steady incremental improvements such as batter…

Ray Tracing I'm not someone into the inner-working of chips much, but is "ray tracing" a new term used for something in microprocessors now? Or is this the same graphic "ray tracing" we were doing back in the 80's on Amigas and Atari STs?

The latter. Systems have become fast enough recently to do realtime ray tracing.

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

#202

Post by epistasis » Tue, Nov 17, 2020, 6:41 PM UTC

bryanlarsen wrote: post #183

ogre_codes wrote: post #78

Earlier quoted context omitted.

The CPU gains on ARM have been increasing consistently year over year for the past decade. People have posted benchmarks of the A11/ A12/ A13 versus Intel for a while so this has been pretty obvious. It's just surfacing because suddenly we have a desktop CPU with desktop software like compilers and other things where it's more obvious outside of benchmark tools. Apple is just jumping onto their existing ARM track, on…

IMO, this has little to do with it being ARM. 30 years ago ARM had a significant micro architectural advantage in performance per watt, but in this era of 10 billion transistor chips, that advantage has disappeared. x86_64 rationalized the x86 architecture and decode is such a small fraction of the power budget that it really doesn't matter anyways. What does matter, IMO: - assembling a killer team - 5nm process - hi…

I'm no expert, but the only big architectural differences are a massively larger decoder and a reorder buffer that's several times as large as x86 designs.

If these are actually the reasons for the performance difference, and it's difficult to do these on x86 because of the instruction set, it seems to this amateur that ARM64 really does have an advantage over x86.

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

#203

Post by mynegation » Tue, Nov 17, 2020, 6:41 PM UTC

reaperducer wrote: post #187

bmurphy1976 wrote: post #69

Earlier quoted context omitted.

Normally this wouldn't feel so ground breaking, but the stars are in alignment and all these improvements are hitting at the same time. We're seeing years of work and investment paying off (AMD, Apple, ARM, Nvidia, Amazon), new process nodes (TSMC), and new tech (Ray Tracing, DLSS, machine learning) all hitting at the same time. And that's the big stuff! There's also the steady incremental improvements such as batter…

Ray Tracing I'm not someone into the inner-working of chips much, but is "ray tracing" a new term used for something in microprocessors now? Or is this the same graphic "ray tracing" we were doing back in the 80's on Amigas and Atari STs?

I did not read it as a list of technologies used in chip fabrication, but as a list of “killer apps” driving the need for more performance.

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

#204

Post by bartread » Tue, Nov 17, 2020, 6:41 PM UTC

I think Apple have screwed themselves over a bit here by sticking ARM-based Apple silicon in their lower-end/entry level devices first.

I mean what's the point in spending £5k on a fully tricked out 16-inch MBP, as I'd been considering, when an entry level Macbook Air or Mac Mini is going to run rings around it?

The reason I'm not going to buy one of these lower end Macs (the Mini would be the best fit) is that I can't stick enough memory in one, and the Air obviously doesn't really have any ports.

So the upshot is I'm not going to be spending any money with Apple anytime soon.

OTOH, if they'd started at the high end, I'd be looking at spending £5k on a tricked out laptop with absolutely unbelievable performance and as much memory and storage as I want/need, and would be entirely happy to do so because I'd feel like I was getting decent value for money rather than being taken for a mug.

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

#205

Post by wmf » Tue, Nov 17, 2020, 6:42 PM UTC

throwaway894345 wrote: post #195

cercatrova wrote: post #26

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’m guessing the performance improvements derive from integrating the memory onto the same chip (instead of using external memory), not from ARM (although power savings come from ARM). So we will probably see a new era of laptop SoCs, but that also means coupling RAM with CPU (or maybe you can mix and match the on-chip RAM with external RAM?).

I’m guessing the performance improvements derive from integrating the memory onto the same chip

Nope, LPDDR4x-4266 is LPDDR4x-4266. Apple, Intel, and AMD all have access to the same RAM. The Firestorm core is the real advantage.

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

#206

Post by stu2b50 » Tue, Nov 17, 2020, 6:42 PM UTC

harrygeez wrote: post #198

tonyedgecombe wrote: post #74

Earlier quoted context omitted.

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

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?

Uh, unless there's some new Qualcomm chip I haven't heard of, the Qualcomm chips are all being utterly crushed by Apple's offerings in geek bench and specperf.

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

#207

Post by macintux » Tue, Nov 17, 2020, 6:42 PM UTC

egao1980 wrote: post #145

M1 pushes users further to mobile phone mentality - no more universal computing, data interoperability. Welcome to a massive vendor lock. I doubt it will run any modern games.

WoW has been around forever, so perhaps it’s not “modern”, but this seems like an early vote of confidence.

https://appleinsider.com/articles/20/11/17/blizzard-updates-...

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

#208

Post by yjftsjthsd-h » Tue, Nov 17, 2020, 6:42 PM UTC

minimalist wrote: post #140

whalesalad wrote: post #48

I love everything about the move to Apple Silicon with the exception of the decision to put memory on-die or in-package (not sure how it is configured). They call it 'Unified Memory'. It makes a lot of sense but I don't know if they are going to be able to pack enough memory in there. A lot of folks are fixated on CPU performance lately (which is rad) but I think that there is a tendency to ignore memory. I have 32gb…

Perhaps this will encourage programmers to write true native applications again instead of wrapping web scripting languages in a browser and calling it a day.

Why? Unless the browser is proportionately slower, they'll still benefit.

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

#209

Post by deerIRL » Tue, Nov 17, 2020, 6:42 PM UTC

reaperducer wrote: post #187

bmurphy1976 wrote: post #69

Earlier quoted context omitted.

Normally this wouldn't feel so ground breaking, but the stars are in alignment and all these improvements are hitting at the same time. We're seeing years of work and investment paying off (AMD, Apple, ARM, Nvidia, Amazon), new process nodes (TSMC), and new tech (Ray Tracing, DLSS, machine learning) all hitting at the same time. And that's the big stuff! There's also the steady incremental improvements such as batter…

Ray Tracing I'm not someone into the inner-working of chips much, but is "ray tracing" a new term used for something in microprocessors now? Or is this the same graphic "ray tracing" we were doing back in the 80's on Amigas and Atari STs?

The new ray tracing is being done in real time rather than being baked into the scenes.

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

#210

Post by danaris » Tue, Nov 17, 2020, 6:43 PM UTC

ogre_codes wrote: post #174

whalesalad wrote: post #149

Earlier quoted context omitted.

I think if you have a very specific role where your workload is constant it makes sense. I am an independent contractor and work across a lot of different projects. Some of my client projects require running a full Rails/Worker/DB/Elasticsearch/Redis stack. Then add in my dev tools, browser windows, music, Slack, etc... it adds up. If I want to run a migration for one client in a stack like that and then want to swit…

Update: Re-reading your above comment I realized I mis-read your post and though you were suggesting 32GB was table-stakes... which isn't quite right. Likewise much of below is based on that original mis-read. I'm not convinced that going from 16GB to 32GB is going to be a huge instant performance boost for a lot of developers. If I was given the choice right now between getting one of these machines with 16GB and ge…

I've seen others suggest that 32GB is table-stakes in their rush to pooh-pooh the M1.

I, personally, am a developer who has gone from 16GB to 32GB just this past summer, and seen no noticeable performance gains—just a bit less worry about closing my dev work down in the evening when I want to spin up a more resource-intensive game.