Live data from Hacker News

Intel's disruption is now complete

jamesallworth.medium.com

141–150 of 638 posts

Re: Intel's disruption is now complete

#141

Maybe this is a bit of a side topic, and I'm a layman, but I'm interested to know why the M1 chip is able to perform so surprisingly better? And why Intel or others can't just come up with a similar design, if it's clear from the specs/die images what's being done? Is it years of work to figure out the placement of CPU/GPU/memory units on the die, routings, and come up with new instruction sets? I imagine this wasn't…

Intel's current design are 1-3 years late depending on which roadmap and sector you are counting. Caused by their 10nm delay.

Intel had Tigerlake / WillowCove Design but they cant ship it. They also have Golden Cove. If you have put the design they were suppose to be using now, on a IPC basis Intel might still be in the lead.

Re: Intel's disruption is now complete

#142
post #9

Earlier quoted context omitted.

It would seem really out of character for Apple to make a move in the server space.

https://en.m.wikipedia.org/wiki/Xserve Apple’s Xserves were awesome. Not cheap. Only marginally competitive. But a joy to work with because of all the nice little details like a FireWire port on the front so you could boot it off of an iPod.

Both Xserve and A/UX were only relevant to the extent of Apple shops, nothing else.

Re: Intel's disruption is now complete

#143

Earlier quoted context omitted.

Reminds me of when I went to visit Intel a few years ago. I was shocked at the bare bones amenities and that employees had to pay for everything (snacks/drinks/etc.). That might sound insignificant but in the talent war in SV that stuff mattered a lot back then. It was a stark contrast from nearly every other fast-growing tech company in the area.

> employees had to pay for everything (snacks/drinks/etc.). That’s not entirely true, at least in Hillsboro (I interned there recently). Drinks were free, and so was fruit, but you had to pay for e.g. donuts or chips.

The fruit's hilarious because if you're a contractor the fruit's not free. It's forbidden fruit.

Re: Intel's disruption is now complete

#144
post #81
post #17

There’s just one missing detail here: intel was terribly, terribly bad at executing its plan. Tiger lake was supposed to be here on 10++nm or 7nm or whatever much improved process, which would make it extremely competitive if not the absolute best. Intel very well knew they’re in trouble if the expected process upgrades don’t work and they didn’t. It’s not like folks there don’t know how to plot that core vs Apple si…

What's interesting is that Apple apparently plotted that graph too. They knew exactly when that crossover was going to occur, for years now.

They certainly didn't expect the massive Intel fuckups from the last years. With the normal performance growth curve they would still be ahead.

* Broken 10nm and 7nm fab process. No end in sight. Even IBM could surpass them.

* Even with much stricter memory guarantee orderings they managed to break caching and C3 kernel/ user seperation security (Spectre, Meltdown, ...).

* Temperature management. Eg AVX512!

* The government "mandated" backdoors.

Who would have expected such massive Boeing-like fuckups from the industry leader?

Re: Intel's disruption is now complete

#145

Earlier quoted context omitted.

That's why I commute to the office via horse — so I don't need to deal with those fascists at the DMV. Sure the costs of boarding and hay and horseshoes pile up, but at least I'm not being hassled by The Man.

I think you will find that The Man is especially going to be hassling you with your horse.

Yeah, just like they hassle me with my wire-wrapped homebrew OpenBSD-based SDR mobile phone in my blue Radio Shack project box every time I try to get on an airplane. Why does The Man always pick on me?!

Re: Intel's disruption is now complete

#146
post #9

Earlier quoted context omitted.

It would seem really out of character for Apple to make a move in the server space.

https://en.m.wikipedia.org/wiki/Xserve Apple’s Xserves were awesome. Not cheap. Only marginally competitive. But a joy to work with because of all the nice little details like a FireWire port on the front so you could boot it off of an iPod.

> But a joy to work with

I think you are using the past tense. Those things are still in production in some places. I wish they weren’t.

Also, I’d like one as a chassis for a Pi or NUC cluster.

Re: Intel's disruption is now complete

#147
post #2

What's the real reason Apple can design a higher performance device? Better fab process? ARM ISA is fundamentally faster? Graph conveniently omits AMD who now beat Intel using the same ISA but different process.

Intel dropped the ball on 10nm and 7nm processes constantly for the last 5 years. Skylake was the "tick", then they couldn't "Tock" to 10nm, so they made 14nm+, then 14nm++, then 14nm+++... Intel's been stuck on the same process for half a decade now.

This doesn’t come up enough: intel’s ceo had to abruptly resign in June 2018 due to a relationship with an employee. [1]

That could not have come at a worse time for a company like intel.

[1] https://www.reuters.com/article/us-intel-ceo-idUSKBN1JH1VW

Re: Intel's disruption is now complete

#148

What percentage of pc unit sales do Mac's make up? Anybody seriously see that market share growing because of M1? Mac's have a lot of baggage to overcome, not saying they can't but by both amd and Intel are not work shy lightweights and many of the advantages of the M1 do not translate to the expandable platforms the pc world expects.

I think another thing to consider is AWS Graviton2 and that based off ARM as well.

Totally agree but Graviton, impressive though it is, doesn't have the aggregate advantage of M1, Intel is setup to respond to that challenge, M1, not so much.

Re: Intel's disruption is now complete

#149
post #65

Earlier quoted context omitted.

Apple’s core is very wide. But wide usually means very high power consumption. M1 apparently has much lower power consumption. So something else is going on other than “Apple is willing to spend more to make cores wider”.

That, in part, comes from x86 vs arm. x86, with its variable length instructions, means your logic gets increasingly complex as you widen. Meanwhile with arm, you have constant instruction size (other than thumb, but that’s more straightforward than variable size), that lets Apple do things like an 8-bit wide decode block (vs. 4-bit for Intel) and a humongous 630 OOB (almost double Intel’s deepest cores).

I think this marks the definitive point where RISC dominates CISC. Been a long time coming but M1 spells it out in bold. Sure variable size instructions are great when cache is limited and page faults are expensive. But with clock speeds and node sizes ansymptoting, the only way to scale is horizontal. More chiplets, more cores, bigger caches, bringing DRAM closer to cache. Minimizing losses from cache flushes by more threads, and less context-switching.

Basically computers that start looking and acting more like clusters. Smarter memory and caching, zero-copy/fast-copy-on-mutate IPC as first-class citizens. More system primitives like io_uring to facilitate i/o.

The popularity of modern languages that make concurrency easy means more leveraging of all those cores.

Re: Intel's disruption is now complete

#150
post #67

Earlier quoted context omitted.

While Intel is indeed losing the performance game, even with the shady management engine shenanigans intel CPUs are still much more open than the locked down proprietary nightmare that apparently are the new Apple ARM SoCs.

That's why I commute to the office via horse — so I don't need to deal with those fascists at the DMV. Sure the costs of boarding and hay and horseshoes pile up, but at least I'm not being hassled by The Man.

No private copter yet? Look at Elon. He flies to work with his Rocketman suite and disturbs the LAX flight controllers and pilots all the way. The Man
Post reply on HN