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

561–570 of 638 posts

Re: Intel's disruption is now complete

#561

Post by thu2111 »

cptnapalm wrote: post #508
Balgair wrote: post #468

Earlier quoted context omitted.

I had always heard that there are two 'Iron Laws of Bureaucracy': 1. The only thing that matters is the budget. 2. Bureaucracies only grow, never shrink, and the only thing you can control is the rate of growth. Thanks for sharing another one with us!

This reminds me of an observation about the British Empire that the bureaucracy was at its most vast the day it was abolished.

Yes, that comes from an essay written in 1955 by Parkinson:

http://doc.cat-v.org/economics/parkinsons-law/the-economist-...

What we have to note is that the 2,000 Admiralty officials of 1914 had become the 3,569 of 1928; and that this growth was unrelated to any possible increase in their work. The Navy during that period had diminished, in point of fact, by a third in men and two-thirds in ships. Nor, from 1922 onwards, was its strength even expected to increase, for its total of ships (unlike its total of officials) was limited by the Washington Naval Agreement of that year.

Re: Intel's disruption is now complete

#562

Post by thu2111 »

whatifitoldyou wrote: post #536
dmix wrote: post #371

Earlier quoted context omitted.

As I've linked to many times on HN: The Iron Law of Bureaucracy > Pournelle's Iron Law of Bureaucracy states that in any bureaucratic organization there will be two kinds of people": > First, there will be those who are devoted to the goals of the organization. Examples are dedicated classroom teachers in an educational bureaucracy, many of the engineers and launch technicians and scientists at NASA, even some agricu…

But isn't apple that's overpassing Intel an equally big organization? How is it exempt from the rules of bureaucracy?

Apple is actually bigger by employee count, however about half (I think) are retail employees in the stores.

So yes - a very large organisation. However Apple's chip design teams are much, much smaller than that. And because Apple mostly doesn't do chip design those teams are likely able to work like a startup, without much internal politics or in-fighting. Where as all of Intel's 100,000+ employees are basically devoted to chip design and manufacture.

Apple as a whole, though, is starting to look rather sick from the same sort of problems that Intel has. Under Jobs there was an energy and clarity of purpose that's been lacking for some time. The drift in Apple's core businesses are obvious. The iPhone has been stuck in turgid incrementalism for a long time now, and has already been largely "disrupted" by Android which sold into cheaper markets and used that to fund R&D budgets that match Apple's. We don't think of it as disruption because Android arrive so soon after the iPhone and thus there was no obvious delayed "disruption event", Apple just had their market share capped by the refusal to compete on price, which is why Apple fans have for years been forced to make arguments about why Apple is successful because they take a larger profit margin than others.

Their core Mac business has also been adrift for many years now. I and many others actively avoid trying to upgrade because things frequently get worse rather than better. The first decade of the century saw constant innovation in the Mac business, after all their best people were reallocated to the iPhone and iPad, quality entered a long period of decline. macOS releases struggled with regressions and rarely introduced new innovations worth caring about. Their apps and hardware have also stalled with own goals and unforced errors like the keyboard fiasco .

Developer experience and docs are another area of problems, there was a good rant about it posted here the other day.

All these are due to organisational rot, and most obviously, a form of in-fighting between iOS / macOS in which the iWorld got the best people and resources, leaving macWorld to pathetically try and steal things from the other group occasionally, regardless of whether it made sense for a laptop context or not.

Re: Intel's disruption is now complete

#563

Post by supernova87a »

ksec wrote: post #141
supernova87a wrote: post #127

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.

What takes time to bring a chip to market? As in the question, if you know exactly what needed to be designed (in concept), how long would it take you to have a fab turning out units?

Is it the work to figure out how to lay down the component blocks on the die and all the circuit routing? Is it figuring out how to assemble the power-efficient components from the Arm library? Does that take years? Is it the instruction set? Is it the planning for manufacturing capacity?

I'm just curious to know what the actual work and division of labor / time is to "make a chip".

I would love to know for example:

-- Initial design, 10 people, 1 year

-- Prototype runs, 5 people, 6 months

-- Final design, 5 people, 1 year

-- etc etc.

-- Acquiring fab capacity / manufacturing hardware, 10 people, 1 year

-- etc etc.

But sounds like there's no way Intel or others can just "leapfrog" even if they knew what needed to be done and could abandon their prior choices.

Re: Intel's disruption is now complete

#564

Post by baybal2 »

SoSoRoCoCo wrote: post #216
baybal2 wrote: post #62

Earlier quoted context omitted.

> And after 30+ years of this, I also don' think there's anything new or revolutionary in silicon fabrication on the horizon. Why so? Few more material, and device generations are on the way. Litho has went through more generation changes in the last 1 decade, than in any previous one. Whole new device classes are on the way: optics on silicon, new memory, logic in the backend, and logic on package. Semiconductor eng…

This is my "640K should be enough comment", but we've reached the limits of optical fabrication with ~2nm. And really, it is just smaller transistors. Yes, that glibly does a disservice to the process engineers making this miracle happen, but I can get excited about "look, smaller transistors!" only so many times in my lifetime. :) > optics on silicon, Yep! That and quantum and post VNeumann architectures. There's so…

Optics may come earlier than people expect.

And not even for I/O, but for on-die communications, and busses.

High speed SERDES may well be replaced by same-die optics, as a lower transistor count, dumber, and cooler solutions.

Re: Intel's disruption is now complete

#565

Post by AdrianB1 »

nicoburns wrote: post #559
airstrike wrote: post #448

Earlier quoted context omitted.

> But is there empirical evidence for it? And a rational explanation behind it? I don't have empirical evidence for it, but I can take a stab at a rational explanation. The second group faces no opposition in their pursuit for control of the organization and its processes. At best, members of that group will fight among themselves, but certainly won't find resistance from the first group. Control over various parts o…

Shouldn't the senior leadership team be providing resistance against the bureaucrats? This doesn't always happen, but I think it does sometimes.

Eventually the bureaucrats become the senior leadership and then they protect themselves.

This grows like a cancer: it starts from the bottom of the pool with people with no much skills, but looking for promotions while people with skills are looking for work results. In some very large companies I worked in (or with), techies focus on work while less competent people become managers; techies regard managers initially as admins that do work techies don't want to, but then the managers take over the organization and make it rot. This is initially far from senior leadership's sight, when it becomes visible it means most of the organization is already affected and it is too late. In most cases HR is helping with this because HR is completely disconnected and not understanding techies and the power dynamics ("those geeks nobody understands") and side with the bureaucrats because HR is also a form of bureaucracy (you don't get in HR is you are a brilliant STEM graduate).

Re: Intel's disruption is now complete

#566

Post by mrtksn »

eternauta3k wrote: post #558
mrtksn wrote: post #527

Earlier quoted context omitted.

Sounds like US screw up to prevent a monopoly which is being broken by an outsider, now complaining that there are no western alternatives to the outsider. Anyway, the machinery that TSMC is using is produced in Netherlands AFAIK. US should be fine, there's a lot of capital available to buy the machinery, recruit the talent and build the Fabs if it comes to TSMC not being reliable partner at some point.

Critical technologies like optical proximity correction are fab proprietary

That's why your recruit the talent too.

Re: Intel's disruption is now complete

#567

Post by vagrantJin »

PostOnce wrote: post #500
vslira wrote: post #387

Earlier quoted context omitted.

The Boeing of semiconductors

War must always be considered, if you want to keep existing as a nation. People have been repeating "Si vis pacem, para bellum" for thousands of years for a reason. Semiconductors and aerospace are important to war efforts, and having domestic manufacturing capability is a big deal if you have serious, wealthy, technically advanced adversaries. Reality is complicated and many-faceted, and there are a number of reason…

Your praise of war comes off quite naive. Perhaps you figure "war" or violence as viable solutions despite being a much more connected world where diplomacy and fairness can be judges more democratically. Such talk of war is neccesary is an utterance of someone born on the right side of the fence and quite clearly no clue as what they are talking about.

Re: Intel's disruption is now complete

#568

Post by macintux »

vagrantJin wrote: post #567
PostOnce wrote: post #500

Earlier quoted context omitted.

War must always be considered, if you want to keep existing as a nation. People have been repeating "Si vis pacem, para bellum" for thousands of years for a reason. Semiconductors and aerospace are important to war efforts, and having domestic manufacturing capability is a big deal if you have serious, wealthy, technically advanced adversaries. Reality is complicated and many-faceted, and there are a number of reason…

Your praise of war comes off quite naive. Perhaps you figure "war" or violence as viable solutions despite being a much more connected world where diplomacy and fairness can be judges more democratically. Such talk of war is neccesary is an utterance of someone born on the right side of the fence and quite clearly no clue as what they are talking about.

Ask Ukraine about the practicality of disregarding defense as a necessary evil.

It’s hardly naive to assume that, since there has always been conflict, that conflict will continue for the foreseeable future.

Re: Intel's disruption is now complete

#569

Post by AnthonyMouse »

andrewjl wrote: post #502
AnthonyMouse wrote: post #408

Earlier quoted context omitted.

> If you can’t get rich by playing office politics you might as well enjoy making a good product. It seems to me this would go the other way. Making a good product (in the sense that it's successful in the market) is what you want a company to do, but that's the thing mostly aligned with the profit incentive. If profits are taxed heavily then spending resources on making a better product so that more people buy it do…

Corporate tax rates can be lower, ideally at 0, but I doubt our current system incentivizes corporations retaining profits. On the contrary, recently many needed to be bailed out because they paid out too much to shareholders instead of saving enough for a rainy day. Making a better product also doesn't necessarily lead to profits. That's an oversimplified perspective. There's a far greater correlation between monopo…

> Corporate tax rates can be lower, ideally at 0, but I doubt our current system incentivizes corporations retaining profits.

Multinational corporations commonly avoid taxes by keeping profits offshore in a low tax jurisdiction. In order to pay the money to shareholders they have to repatriate them and pay corporate income tax on the money, and then the shareholder has to pay tax again on the dividend or the capital gain from the buyback.

Or they can leave the money where it is, not pay any of those taxes and invest it in some index fund from there. But then the money is stuck inside the corporation, and gets invested in some index fund instead of potentially going to some higher risk/reward investments that some of the original shareholders would have chosen.

> On the contrary, recently many needed to be bailed out because they paid out too much to shareholders instead of saving enough for a rainy day.

COVID-19 isn't a rainy day, it's a once in a century pandemic. Most companies have never made enough money to be able to survive it, and the ones that did still shouldn't be hoarding it just in case. Systemic problems get solved through systemic actions, e.g. lowering interest rates or sending stimulus checks. To expect companies to have saved enough money to survive rare systemic events without that is to penalize and destroy any company lean enough to have been unable to save it to begin with.

> Making a better product also doesn't necessarily lead to profits.

Making a worse product in a competitive market has a strong relationship with going out of business.

> There's a far greater correlation between monopoly power and profit.

But that's an independent problem. The solution there isn't to raise taxes, it's to break up monopolies.

Re: Intel's disruption is now complete

#570

Post by hajile »

systemvoltage wrote: post #231
hapless wrote: post #227

Earlier quoted context omitted.

Fabrication is where Intel is falling down. They have fresh new chip designs that would be perfectly adequate, but they are still hobbled by 14nm. They have now released three? four? iterations of x86 design on the same 14 nm process, because 10nm is that troubled If you spun off the fabrication, as AMD did, you would have a much smaller, but more-competitive fabless business, and the fabricator would implode. (Which…

I am afraid, they're going to be using US gov as their wheelchair to survive a few more decades: https://newsroom.intel.com/news/intel-wins-us-government-adv...

That shouldn't be surprising.

The Military Industrial Complex is a huge beast. Almost all the big tech companies started before 1970 or so were funded by DoD research money. The only reason silicon valley is where it is lies in DoD radar research constraints. Computer tech started as stuff like solid state transistors to improve radio efficiency and size.

Since the MIC is a good ol' boys club with the tax dollars to pick winners and losers, of course they'll help their friends first "for the good of the country".

I'd also like to note my displeasure that my tax dollars pay must of the research bill, but the various companies walk away with the patents, copyrights, and profits. There's no reason for someone else to get rich off my investment.

Want to solve the national debt? Require royalties for government investment.

Post reply on HN