Live data from Hacker News

Apple’s M1 Positioning Mocks the Entire x86 Business Model

extremetech.com

581–590 of 942 posts

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#581

Earlier quoted context omitted.

Apple differentiates the majority of their products by generation rather than binning. If you buy a low-end iPad, you get an A12; the iPad Air steps you up to an A14; and the iPad Pro gets you an M1. This is less evident on the Macintosh side of the business right now because they're just trying to get M1 silicon into as many product lines as their fab capacity will allow. They don't actually have an M2 (or even M1X)…

> Apple differentiates the majority of their products by generation rather than binning. This simplifies things for consumers, but how do you make chips without binning? Are they all just that reliable? Do they all have extra cores? Maybe Intel bins more because they can squeeze out a significantly better price for more core and cache, but Apple's margins are already so high they don't care?

So, there are a few instances where Apple does bin their products:

1. "7-core GPU" M1 Macs, which have one of the eight GPU cores disabled for yield

2. The A12X, which also had one GPU core disabled (which was later shipped in an 8-core GPU configuration for the A12Z)

3. iPod Touch, which uses lower-clocked A10 chips

It's not like Apple is massively overbuilding their chips or has a zero defect rate. It's more that Intel is massively overbinning their chips for product segmentation purposes. Defect rates rarely, if ever, fit a nice product demand curve. You'll wind up producing too many good chips and not enough bad ones, and this will get worse as your yields improve. Meanwhile, the actual demand curve means that you'll sell far more cheap CPUs than expensive ones. So in order to meet demand you have to start turning off or limiting perfectly working hardware in order to make a worse product.

Apple doesn't have to do this because they're vertically integrated. The chip design part of the business doesn't have to worry about maximizing profit on individual designs - they just have to make the best chip they can within the cost budget of the business units they serve. So the company as a whole can afford to differentiate products by what CPU design you're getting, rather than what hardware has been turned off. Again, they aren't generating that many defects to begin with, and old chips are going to have better yields and cost less to make anyway. It makes more sense for Apple to leave a bit of money on the table at the middle of the product stack and charge more for other, more obviously understandable upgrades (e.g. more RAM or storage) at the high end instead.

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#582

Earlier quoted context omitted.

So my sister does professional photography, and went from a 16GB MacBook Pro(2016) to an 8GB Air, and reports no decrease of productivity - quite the opposite, in her experience all Adobe programs run much faster, and the machine is quieter and lighter to boot. So yeah, I'm not sure - maybe the ram amount isn't as a big deal as people make it out to be. On the other hand I'm a C++ programer and my workstation has 128…

Web Developer here. Wouldn't accept anything less than 32GB for a work machine.

Anyone remembers the times when "web development" was something you could do on a pretty any computer available?

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#583
post #569
post #552

Earlier quoted context omitted.

It means the usable life is shorter, so even if it’s fine for a while, you need to think of it as a much more expensive device per useful year.

Sadly, with the current trends I think the device will be more likely rendered by an irreparable hardware failure before it reaches end of life due to speed.

I mean that seems inevitable.

I have a 2014 retina iMac. At some point in the next few years, I’ll probably replace it with an M series iMac, and then put Linux on it. It’s hard to imagine that it will ever really reach end of life due to speed at that point.

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#584

Make it as fabulous as much as you want, $1700 for an 8 core, 8 gigs of ram machine is just one plain fabulous joke. This in the time when 16 gigs ram is the baseline if you plan to do anything more than facebook and instagram.

The 8 gigabyte machine is fine for 90% of people. I've recommended just getting that platform to a number of people, and none of them have had issues.

I've got an 8 Gigabyte MBAir, and it never stutters. Meanwhile, on my 16 GByte Dell XPS (Ubuntu 20.04) - I routinely live in fear of exceeding my chrome tab quota because I know it will bring the system to a crashing halt. Somewhere around 45 is the point it all comes down.

Meanwhile, I don't even think about how many safari tabs I have open (hundred+) - and 8-10 applications open at the same time.

Different operating system has different models of swapping and degrading performance. Apple has nailed it.

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#585
post #541

Earlier quoted context omitted.

I have to admit this is right. Even my old 2012 i7 Mac mini is still a perfectly serviceable machine to this day.

Late-2012 ivy bridge i5 Mac Mini here. For a dual core machine it's really not all that bad of a daily driver experience. Makes me sad that I have to spend a crazy amount of money to get 16GB of ram and 1TB of SSD in a M1 Mini to match what I currently use. (although I understand the 1TB of SSD in the M1 is vastly superior)

Just for the fun of it, I tried to play WoW classic on my late-2012 mac mini and it ran like a toaster oven, but it worked just fine. Pretty fun. After putting the SSD in it, of course.

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#586
post #396

Earlier quoted context omitted.

I've been trying to get to the bottom of this here: https://github.com/solvespace/solvespace/issues/972 With my 2400G the benchmark takes 5.5 seconds. With someones M1 it takes 1.3 seconds. We still need to verify that was run at the same settings, but it looks like the M1 is going to be faster than the 5000 series APUs regardless, possibly by a large margin and I'm assuming the 5000s are between 1.4x and 2.0x the sp…

The 2400G is a relatively low end part that is four generations old. In the past AMD did not release high end desktop APUs, preferring customers to buy CPUs + GPUs. This is already announced to change with a 5700G being produced. https://www.amd.com/en/products/apu/amd-ryzen-7-5700g - On paper this is a 5800H in a desktop form factor with a higher max TDP and higher clock speeds. The 5800H gets 2.5x on cpu benchmark'…

>> The 5800H gets 2.5x on cpu benchmark's rating vs the 2400G.

That's in large part because its 8 cores instead of 4. The test I ran makes good use of multicore, but going from 4 to 8 won't cut the time completely in half. I suspect the M1 will still beat it by a bit.

I take back my comment on availability. We ideally need to compare things on the same process node to see which design team has done a better job ;-)

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#587

Earlier quoted context omitted.

Web developer, I had an 8GB M1 MacBook Air and if I ran vscode with my 50kish typescript codebase and dev server for api and frontend + native postgres and redis at the same time I’d be right at the limit of my machine slowing to unusable levels. Switched it for a 16GB one and I’ve never had any noticeable performance issues since.

I think the efficiency of the new M1 is what blows everyone away. When the M1 first came out I was super super skeptical, seemed like an under powered chip compared to what's in the x86 world. Now I'm convinced that it's got some magic inside it. Everyone I've talked to said it chews throw whatever they throw at it. I'm consistently floored by Apple's ability to innovate.

It’s funny you say “innovate.” They certainly innovated with the M1, but with one key difference. They did so in public, step-by-step over the course of a decade.

Perhaps more than any other Apple innovation, we have the greatest visibility into the process with the M1.

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#588
post #377

Earlier quoted context omitted.

Yeah the marketing effort is immense, and it's really surprising to me how many eagerly lap up the angle Apple is pushing without much reflection, even on HN. Like the good old Apple hype days. Now to be fair, the M1 is an impressive iteration on the A* chip line. But the Ryzen 5000 mobile chips really don't look bad at all in comparison, and a 5nm version of those would level the field.

I didn’t buy the hype until I tested an actual M1. It exceeded the hype. It feels like when I was a kid and went from a 386SX to a Pentium at 4X the speed, except the M1 was also a quarter the power consumption. It’s the first time in a over a decade I’ve felt a revolutionary step forward in a chip and the only time it has ever come with less power. Process node is a factor but I find it hard to believe it’s the only…

> It feels like when I was a kid and went from a 386SX to a Pentium at 4X the speed

I remember the jump from the 486 chips to the Pentium being that dramatic. Heck, 486DX2-66 to P100 brought Linux kernel compile times from 45 minutes down to like 5 minutes.

Back then, I think a big part of these jumps wasn't just CPU design. I think a lot of it was that those CPU design changes were often accompanied by massive improvements to overall system architecture. So the I/O buses, memory, peripherals, all of it got a lot faster as part of the same upgrade.

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#589
post #533

Earlier quoted context omitted.

I am continually astounded that the quad core sandy bridge core-i7 I bought in January 2011 is still completely serviceable for pretty much all tasks outside of gaming. The bezel and overall form factor is clownishly large by todays standards, and the screen is starting to look a bit faded, but as a portable but mostly stationary PC it still works great alongside my main laptop I use for work. I knew back in 2011 I w…

I think that’s because 99% of regular computer tasks don’t need 24 cores operating at 5 billion cycles per second. The differences between an i7-4xxx and i7-11xxx are more about hardware acceleration, eg HEVC and SIMD. That’s why they look so much better in benchmarks that are designed with that in mind. My main desktop is an unRAID server that hosts multiple VMs. Only yesterday I discovered that I accidentally only…

Speak for yourself. Me personally? I would like one core per Chrome tab.

Re: Apple’s M1 Positioning Mocks the Entire x86 Business Model

#590
post #500
post #182

Earlier quoted context omitted.

A blazingly fast SSD is excruciatingly slow RAM.

In-memory memory compression is the most meaningful part of what parent comment said. And it helps tremendously on other systems as well, for example zram is quite underhyped on linux.

zram is a game changer. I believed I needed to upgrade 16gb to 32gb, with zram and a decent cpu no I don't. 97% full memory + 20GB zram swap and no noticeable problems.
Post reply on HN