Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Intel's disruption is now complete

jamesallworth.medium.com

Post reply on HN
Print view View original

51–60 of 638 posts

Re: Intel's disruption is now complete

#51

Post by SoSoRoCoCo »

pinewurst wrote: post #38
SoSoRoCoCo wrote: post #28

Earlier quoted context omitted.

I guess you didn't read about their Xeon Phi bi-directional ring architecture that achieved dozens of terrabytes/sec of bandwidth, their 512b vector-processor ISA, or, heck, just for giggles, the industry's first FinFet (and high-k dielectric) on p1270, p1272, and p1274. But, please, do go on.

And it was such an outstanding success! Intel's customers were able to get so much use out of it - it'll always be remembered as a high point in supercomputing. Seriously - it was a resounding failure, even with Intel spending big bucks to convince this class of customers it was the future. There was little uptake and vast dissatisfaction with achieved results. IMHO I expect Xe to follow the same (very shallow) traje…

Oh I completely agree. But success and innovation are orthogonal, and the original question was about innovation.

Re: Intel's disruption is now complete

#52

Post by raverbashing »

jeffbee wrote: post #26
dboreham wrote: 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.

I'm still waiting for my M1 to show up but from studying the details that other people have commented on, it's clear that Apple is slathering on the microarchitectural resources that both Intel and AMD are stingy with. Their core has a gigantic L1 instruction cache, rather a lot of L1 TLB entries (4x more than Intel), large(r) pages by default, and many other aspects. L1 TLB entries in particular are a well-known pro…

Interesting that you mention page size, IMHO 4k was small even 10 years ago. Though of course the next bump is 4M which I think might be too big for most apps

Apparently ARM supports 4KB, 64KB and 1MB (I just googled that)

Also maybe all the x86 crud might be finally taking its toll (and not just the ISA but all the cruft that got on top of it for 40yrs+ like legacy buses and functionalities, ACPI, or even the overengineered mess that's UEFI)

Re: Intel's disruption is now complete

#53

Post by threatofrain »

agumonkey wrote: post #48
SoSoRoCoCo wrote: post #20

As someone who worked on Intel's phone chip: we definitely didn't win it. We fucked it up twelve ways to Sunday. Why: giant egos. There were turf wars between Austin, Santa Clara and Israel over who would design it, and the team that won out had long since lost its best principle engineers and had no clue how to spin the architecture to meet the design win. Otellini's hindsight hedge is pure spin: we knew the landing…

It's somehow incredible how this infiltrates even the most advanced classes of company in terms of technical knowledge. You'd expect a bit more objectivity

You don't use computer science to make sure people are properly incentivized.

Re: Intel's disruption is now complete

#54

Post by _the_inflator »

SoSoRoCoCo wrote: post #20

As someone who worked on Intel's phone chip: we definitely didn't win it. We fucked it up twelve ways to Sunday. Why: giant egos. There were turf wars between Austin, Santa Clara and Israel over who would design it, and the team that won out had long since lost its best principle engineers and had no clue how to spin the architecture to meet the design win. Otellini's hindsight hedge is pure spin: we knew the landing…

Corporate mess. I know it, not Intel, but politics, who gets all the credits?, before the first step has been taken - been there.

Re: Intel's disruption is now complete

#55

Post by samfisher83 »

In general mobile chip margins aren't great unless you are apple. That is why you don't see too many competitors in the mobile chip market. You have Qualcomm, but they have a Radio monopoly and their performance isn't that great. Intel architecture isn't even that bad.

Intel used to have a manufacturing advantage which they longer do. That is where they are trailing: Material science and manufacturing. This stuff cost a lot of money and requires and lot of Research. This is the Kind of stuff MBA types don't necessarily want to invest in.

Re: Intel's disruption is now complete

#56

Post by S_A_P »

whatever1 wrote: post #40

My question is why Qualcomm while using the same arch, has not produced a chip as fast as the M1?

I suspect there was some talent at apple that figured out some details which Qualcomm have not at this time. This guy seems to improve things at every place he has been. He left abruptly from Intel this year, which hopefully is not due to health reasons.

https://www.anandtech.com/show/15846/jim-keller-resigns-from...

Re: Intel's disruption is now complete

#57

Post by Joeri »

baq wrote: 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…

Even if they had seen two process shrinks more, the relative efficiency benefit of that wouldn’t have topped 30%, which still wouldn’t make intel chips competitive with the m1. What intel needs is both a better fab process and a new core architecture. Apple might have waited one more year or maybe two at the most to get the same performance delta, but they would have switched regardless.

Re: Intel's disruption is now complete

#58

Post by lgessler »

agumonkey wrote: post #48
SoSoRoCoCo wrote: post #20

As someone who worked on Intel's phone chip: we definitely didn't win it. We fucked it up twelve ways to Sunday. Why: giant egos. There were turf wars between Austin, Santa Clara and Israel over who would design it, and the team that won out had long since lost its best principle engineers and had no clue how to spin the architecture to meet the design win. Otellini's hindsight hedge is pure spin: we knew the landing…

It's somehow incredible how this infiltrates even the most advanced classes of company in terms of technical knowledge. You'd expect a bit more objectivity

Indeed, for we computer geeks are known for our lack of ego...

Re: Intel's disruption is now complete

#59

Post by rsynnott »

whatever1 wrote: post #40

My question is why Qualcomm while using the same arch, has not produced a chip as fast as the M1?

Qualcomm is a virtual monopoly in the space it serves (mid to high end non-Apple non-Chinese phone chips); arguably it doesn't have that much incentive to do anything much.

Re: Intel's disruption is now complete

#60

Post by pinewurst »

SoSoRoCoCo wrote: post #51
pinewurst wrote: post #38

Earlier quoted context omitted.

And it was such an outstanding success! Intel's customers were able to get so much use out of it - it'll always be remembered as a high point in supercomputing. Seriously - it was a resounding failure, even with Intel spending big bucks to convince this class of customers it was the future. There was little uptake and vast dissatisfaction with achieved results. IMHO I expect Xe to follow the same (very shallow) traje…

Oh I completely agree. But success and innovation are orthogonal, and the original question was about innovation.

But whatever real innovation underlay Xeon Phi/KNx was performed two decades ago. Current Intel is only capable of SKUsmanship - minutely partitioning function for projected max margins.

Edit: Also too, remember that KNx was really an attempt to salvage Intel's failed Larrabee GPU effort.

Post reply on HN