M1 Macs: Truth and Truthiness • Hacker News Board

Re: M1 Macs: Truth and Truthiness

#491

Post by folmar » Thu, Dec 03, 2020, 10:40 PM UTC

ogre_codes wrote: post #254

ekianjo wrote: post #208

Earlier quoted context omitted.

> But I think the overarching point is true: now desktop and laptop makers can't just pretend that having a hot and/or slow, clunky, noisy x86-based architecture is just a fact of life. I hope and expect that it'll help create a new generation of ARM-based laptops and maybe even desktops that will run cool and smooth. Be careful what you wish for. X86 is the most open architecture right now. ARM, on the other hand, h…

> X86 is the most open architecture right now. x86 is not "Open" in any fashion. You can't license x86 even for a fee, let alone free. While ARM isn't open either, it is by comparison far more available to license at reasonable cost. > ARM, on the other hand, has locked bootloaders It also has open/ unlocked boot loaders. One of the advantages of licensable technology like this is you can have a lot of different arch…

The is a free one: https://opencores.org/projects/zet86 (I don't know how complete/production grade is it).

Re: M1 Macs: Truth and Truthiness

#492

Post by PaulDavisThe1st » Thu, Dec 03, 2020, 10:40 PM UTC

joakleaf wrote: post #469

PaulDavisThe1st wrote: post #452

Gruber notes midway: "Perversely, developers, who by nature of their profession best understand exactly what an architecture transition like this entails, might be among the few professions who can’t yet move their primary computing to an Apple Silicon device by nature of the software tools they depend upon. (Some developers can move now, and — because Xcode running on the M1 compiles code so much faster than any Int…

I disagree. I compile a lot. Every day. Almost daily in Java, Objective C, and C++ (Yes, I work actively on all three languages these months). 90% of the time I compile one file (the one I just changed) and then I link. Linking is not parallel and it takes about 4-15 seconds (dep. on the project/lang.). I use an i9 with 8 cores, but I suspect that a CPU with fewer cores, that go really fast for seconds before throttl…

It does depend on project scale, and what you're doing with it.

A full compile of my project on my 2950X (16 core threadripper gen 1) takes 4m20s.

If I'm messing around with a new GUI-visible feature, then yes, often it's 1 or 2 files to compile, then a link step (thank you, llc, for parallel linking). Anything can handle that, even if it is C++ :)

But if I'm doing a change to the core headers, it typically means multi-minute wait even on the threadripper. And that's something I've been doing for a couple of months now, working on basic architectural changes. The idea of doing such work on machine that took 50-100% longer to do a full build would be a serious problem.

I don't see anything on the horizon (yet) that uses Apple silicon that helps that kind of development process.

Re: M1 Macs: Truth and Truthiness

#493

Post by valuearb » Thu, Dec 03, 2020, 10:42 PM UTC

thu2111 wrote: post #198

valuearb wrote: post #193

Earlier quoted context omitted.

Your TR is what, 280 watts TDP? nearly 20 times more than a M1. And is 30% slower in single core. Firestorm Cores use less than 4 watts each. A 16 firestorm core Mx will produce roughly the same multicore performance as your TR, with a less than 70 watt TDP. And a $65 M1 dusts a 4800h that costs over 4x as much and using 1/3 the power. Top end ThreadRippers cost $2,000 to $3,000 each. You can bet an Apple Mx ThreadRi…

How does that matter though? His machine isn't sitting on his lap or running on a battery. Electricity isn't that expensive.

It matters because its not just adding electricity, its adding heat. Pretty clearly Apple can add a whole bunch of Firestorm cores to future M1 derivatives and still fit in smaller power/heat budgets that current generation Ryzen and Threadrippers operate in.

Re: M1 Macs: Truth and Truthiness

#494

Post by compiler-guy » Thu, Dec 03, 2020, 10:42 PM UTC

ogre_codes wrote: post #335

6d65 wrote: post #297

The author is bashing another author for not doing measurements, yet himself declares that m1 beats all the other CPUs, without providing any measurements. With regards to the M1, I'm grateful for a good competition. As this will likely push premium laptop manufactures (Razer, Dell) towards ryzen CPUs, as opposed to using them in cheaper models. Loosening the grip that intel has on them. With Zen3 on mobile, giving a…

> The author is bashing another author for not doing measurements That is a poor characterization of this article. Gruber's critic of Moorhead's article doesn't reference a lack of benchmarks or measurements at all.

Gruber has also cited plenty of benchmarks in other posts. So he didn’t on this one. So what?

Re: M1 Macs: Truth and Truthiness

#495

Post by ravi-delia » Thu, Dec 03, 2020, 10:44 PM UTC

GekkePrutser wrote: post #350

ravi-delia wrote: post #329

Earlier quoted context omitted.

The M1 as a computer is too locked down for me, and it seems like for you. But the M1 as a demonstration of the power of modern ARM chips is absolutely unmatched. I have seen a lot of people use the lockdown argument to justify a disregard of ARM overall without realizing that no, the open version of this technology will be incredible .

Oh I'm not against ARM at all. In fact I have many systems running it :) Raspberries but also a Pinebook. I'm also getting an M1 mini from work. I just don't like the almost-religious overhyping of everything Apple. They're doing a great job. They're not perfect. For example: Moorhead makes a good point IMO. Yes he's talking about software issues. But in the case of Apple, you can't separate those two. The OS doesn't…

> I just don't like the almost-religious overhyping of everything Apple. They're doing a great job. They're not perfect.

That may be so, but again that doesn't really apply to the M1's performance as hardware. If the article was about Apple as an investment, or Apple as the designer of the next computer for me, I'd agree that taking into account their software would make sense. But the whole point of the article is that equating Apple's being Apple with the M1 being a badly designed chip is wrong.

Re: M1 Macs: Truth and Truthiness

#496

Post by ricardobeat » Thu, Dec 03, 2020, 10:46 PM UTC

6d65 wrote: post #297

The author is bashing another author for not doing measurements, yet himself declares that m1 beats all the other CPUs, without providing any measurements. With regards to the M1, I'm grateful for a good competition. As this will likely push premium laptop manufactures (Razer, Dell) towards ryzen CPUs, as opposed to using them in cheaper models. Loosening the grip that intel has on them. With Zen3 on mobile, giving a…

Multiple rounds of 20% improvement will still not give you the performance jump brought by M1. It takes tasks that were simply unthinkable on a light laptop and not only makes them possible but manages to beat something like a 10-core i9 [1], in a machine without a fan.

[1] https://arstechnica.com/gadgets/2020/11/mac-mini-and-apple-s...

Re: M1 Macs: Truth and Truthiness

#497

Post by notreallytrue » Thu, Dec 03, 2020, 10:47 PM UTC

egsmi wrote: post #430

ianhowson wrote: post #418

Earlier quoted context omitted.

M1 is using standard LPDDRx. It's not "very high performance". It uses a different interconnect -- that's it. I guarantee that 64GB does not cost anything near EUR300. You might be thinking of HBM[2] which has a wider I/O path and costs more.

https://news.ycombinator.com/item?id=25050625 showed https://www.cpu-monkey.com/en/cpu-apple_m1-1804 which determined the M1's memory is LPDDR4X-4266 or LPDDR5-5500. If those memories are not high performance, what is?

You can't do what you do on a desktop on a laptop, not even a good one

Who cares if an M1 consumes less energy than a candle if I can buy 64GB of DDR4 3600 for 250 bucks and render the VFX for a 2 hours movie in 4k?

Another 300 bucks buy me a second GPU

When I deliver the job I put aside another 300 bucks and buy a third GPU

Or a better CPU

vertical products are an absolute waste of money when you chase the last bit of performance to save time (for you and your clients) and don't have the budget of Elon Musk

The M1 changes nothing in that space

Which is also a very lucrative space where every hour saved is an hour billed doing a new job instead of waiting to finish the last one to get paid

You can't mount your old gear on a rack and use it as a rendering node, plus you're paying for things you don't need: design, thermal constraints, a very expensive panel (a very good one, but still attached to the laptop body, and small)

So no, M1 is not comparable to a Threadripper, it's not even close, even if it consumes a lot more energy

When I'll see the same performances and freedom to upgrade in 20W chips, I will be the first one to buy them!

https://www.newegg.com/corsair-64gb-288-pin-ddr4-sdram/p/N82...

Then there's the 92% (actually 92.4%) of the remaining market that is not using an Apple computer that will keep buying non Apple hardware

Even if Apple doubled their market share, it would still be 15% Vs 85%

How is it possible that on HN people don't realise that 90 is much bigger than 10 and it's not a new laptop that will overturn the situation in a month is beyond me

Re: M1 Macs: Truth and Truthiness

#498

Post by p1necone » Thu, Dec 03, 2020, 10:47 PM UTC

valuearb wrote: post #487

p1necone wrote: post #298

Earlier quoted context omitted.

"$65 m1" where are you getting this number from? You can't buy an m1 CPU from a store and slap it in a system, you have to buy it in a $1000 laptop.

It's a pretty accurate cost estimate from those tracking Ax series costs. A chip AMD sells for $300 doesn't cost AMD $300 to make, but being a public company their gross margins tell us it costs roughly $150.

But you were comparing this theoretical production cost to the retail cost of other chips, that doesn't make sense.

Hardware companies don't charge cost + some reasonable markup, they charge whatever the market is willing to pay for it. I highly doubt Apple would release some hypothetical "threadripper killer" at a price any lower than they needed to to compete.

Re: M1 Macs: Truth and Truthiness

#499

Post by LeoPanthera » Thu, Dec 03, 2020, 10:50 PM UTC

shawnz wrote: post #21

faitswulff wrote: post #14

Earlier quoted context omitted.

I had a suspicion upon reading Moorhead's review that he was trying to use familiar enterprise Windows software on the M1 instead of trying to use native apps: > So far, I have experienced application crashes in Microsoft Edge, Outlook, WinZip and Logitech Camera Control. I got installation errors with Adobe Reader XI, Adobe Acrobat Reader DC, a Samsung SSD backup application, and Xbox 360 Controller for Mac. What Ma…

Not to judge a book by it's cover, but mind-bogglingly his "preferred browser" for OSX is Edge. https://twitter.com/PatrickMoorhead/status/13296105698261319...

This guy is a tech reporter? But he doesn't know how to take a screenshot.

Says a lot.

Re: M1 Macs: Truth and Truthiness

#500

Post by joakleaf » Thu, Dec 03, 2020, 10:51 PM UTC

valuearb wrote: post #474

PaulDavisThe1st wrote: post #452

Gruber notes midway: "Perversely, developers, who by nature of their profession best understand exactly what an architecture transition like this entails, might be among the few professions who can’t yet move their primary computing to an Apple Silicon device by nature of the software tools they depend upon. (Some developers can move now, and — because Xcode running on the M1 compiles code so much faster than any Int…

The M1 is their entry level CPU and it's compiling software as fast as a fast i7 or i9. They don't have very far to go to catch ThreadRippers, 24 Firestorm cores should more than do it.

The current M1 has 4 firestorms. Geekbench gives a score of ~7400 for the M1s. Assuming the firestorms represent some 80% of the CPU power and multiplying by 6 (6x4=24), we would get a Geekbench score around 35000.

That would place it between a 32- and 64-core threadripper.

I know, that's a lot of handwaving!

I must admit, while I have a lot of doubts that Apple can simply scale up to 24+ core M1s while having acceptable yield, what they have going for them is low amount of heat with current core count and clock speed. So I think, they might also be able to increase the clock speed significantly from here.

But, I have been wondering if their plan is instead to create an m1x (4+8 core), and then combine up-to 3-4 m1xs in the Mac Pro and iMac Pro machines to get up-to 24-32 total cores. That way they might still have high yield and only a few lines of chips (keeping the supply chain tight, which seems like Apple / Tim Cook's MO).