News Intel might be too big to fail — Washington policymakers are already discussing potential solutions if the chipmaker cannot recover

[evilpaul]

Intel spent like fifteen years buying their own stock back. With how much they improved shareholder value, they couldn't really be having problems.

Also, according to the US State Department, since the Nixon presidency, Taiwan is part of China. It's a pretty recent thing. https://www.state.gov/u-s-relations-with-taiwan/ It's why the US has no official relations with Taiwan.

 

[Jagwired]

If I'm the Intel CEO, I spin off the foundry business. It doesn't seem like much of an advantage to own the design and production end to end at this point. In fact, its held Intel back while AMD has been able to use smaller process nodes.

Intel also cannot give up on the GPU business again. They need to keep pushing forward with the Arc GPU's because the CPU business is going to continue to decline.

I also don't understand why Intel (and AMD) haven't been bigger promoters of RISC-V. If I were them, I'd be manufacturing some RISC-V CPU's even if there wasn't a big market. They need RISC-V to gain some footing so they don't have to pay royalty fees to ARM for the next 50 years.

 

[SonoraTechnical]

Intel financials are no longer included in the DOW Industrial Average in the US ... It was replaced by nVidia last week.

 

[ex_bubblehead]

Just a reminder to all that politics of any kind are prohibited in these forums. There have already been some removals and any further violations of the rules will result in the immediate closure of the thread and possible sanctions for the violators.

 

[pug_s]

AMD had been on this road before. They ended up splitting its fabs and its cpu business and came back stronger. Intel is not doing the same thing.

 

[bit_user]

AMD had been on this road before. They ended up splitting its fabs and its cpu business and came back stronger.

AMD came back stronger, but just look at what happened to Global Foundries!

Certainly, there were a lot of factors at play and I'm not saying history will definitely repeat itself. However, care should be taken specifically to avoid the same outcome, or else we'd lose another competitor for leading edge semiconductor fabrication.

 

[bit_user]

A crisis in a company is always seen as bad thing, which is not. It shows there was something wrong to begin with and it's up to the company concerned to work things out

This presumes companies exist in a static world, which they obviously do not. A company might be optimized for one set of conditions among their suppliers, customers, competitors, and technologies. As those conditions change, the company's structure might no longer be optimal, and I think that best describes what's happened with Intel (not to say there weren't other issues, as well).

IMO, the string of execution failures were the first sign of serious trouble at Intel. It's probably unfortunate they had such overwhelming demand for their 14nm products, because I think that masked some deeper problems. Investors only seem willing to take action when these problems are reflected in the bottom line.

Had Intel executed perfectly on Gelsinger's strategy, maybe they could've successfully gotten the fabs to a point of self-sufficiency. However, they have been faltering in execution at a critical time, and now are being forced to deal with the structural problems ahead of when they planned.

 

[bit_user]

In an ideal world we wouldn't even be worried about it yet because investors would have understood the business they invested in and allowed Pat's plan they supposedly believed in play out before panicking.

I supported Pat for a long time, but mounting evidence about his dishonesty, misrepresentations, and lack of forthrightness has destroyed my faith in him. I think he needs to go, but I hope he's replaced with someone who will continue to carry out his plan.

 

[bit_user]

If a company (or bank) is going to fail and needs GOV cash then the Gov should obtain half of the company as they prvent it from goign entirely away.

Gov should not be responsible for other companies bad business choices w/o gettign a very lucrative deal out of it.

The 2008 auto bailouts were loans with (as far as I can tell) an interest rate of 5%. That sounds pretty reasonable, to me.

As the lender of last resort, the government should not be trying to make substantial profits from these activities. If an industry or company is systemically vital to the economy, then it already has enough intrinsic value to justify the loan. If not, then there's no real reason for the government to intervene and loans shouldn't be issued at all.

The idea of the government treating private sector loans as a profit center is hugely problematic. It's almost like a referee betting on one of the teams in the game they're refereeing. Huge conflicts of interest, and incentives for lawmakers & regulators to tip the scales in favor of the companies in which the government is invested.

 

[bit_user]

This is typical institutional decline. Blame the managers of the company, and the end of the day, the Board of Directors. CEOs are not kings of their companies. The Board is.

I think if they just ate into their profit margins and launched on not-ideal yields, they would probably have just worked it out and been as competitive as TSMC. But the Board wanted certain profit margins, damn the future of the company. This is why public companies can be a bad thing.

Public ownership has worked pretty well. It also helps distribute the proceeds, when companies are successful. I think most of the problems with public ownership can be addressed by tweaking the SEC rules of corporate governance and reporting, not by doing away with it.

In Intel's case, I think dividends and share buybacks deprived the company of too many resources that it needed to remain competitive. These are done according to rules/laws, so we could theoretically tweak them to mitigate against such short-sighted behavior.

Steam is private, has FU money, and an FU owner, Gaben Claus. They are still on the cutting edge of content delivery.

Eh, there are plenty of negatives to having private ownership, as well. For instance, plenty of private equity firms that engage in practices like asset-stripping. There's basically zero visibility into such companies. Also, no opportunity for the public to share in their success.

I'm not trying to be prescriptive, here. I'm just saying that it's not as if private ownership is the ideal solution. Intel did quite well, as a publicly-traded company, as are AMD, Nvidia, and the other tech giants. Even ARM is (again) publicly-traded.

 

[bit_user]

Ditto for DEC. Ditto for Compaq. I'm sure the PRC is sponsoring their CPU manufacturer and turning a blind eye to any stolen IP it uses. I don't much remember anyone being up in arms when DEC finally died.

What's even your point? Compaq bought DEC, which then sold off the CPU division to Intel. HP bought Compaq. I'm sure existing DEC customers had their support contracts honored, even as DEC products were phased out.

DEC's Alpha was indeed ground-breaking, but most of DEC's business was geared towards servers and mainframes at a time when x86 was ascendant and mainframes were being squeezed out by commodity servers. Alpha also suffered from being difficult to program, which might've been another reason it saw limited adoption. Jim Keller (one of Alpha's architects) recently said he thinks it was probably a mistake for them to use such a weak memory model. He touches on this at about 14:45, in this presentation he gave at TSMC:

View: https://youtu.be/o70yKYWgtVI?t=887
​

Anyway, I don't see how it's comparable to anything we're discussing here. DEC wasn't the last US-based maker of any particular type of product and they didn't simply disappear overnight or need any kind of bailout. It's an age old story of the industry evolving, DEC failed to keep up, and so they got eaten and digested by their more modern competitors.

BTW, some of the Alpha team quit Intel and founded P.A.Semi, which Apple later purchased as the core design team to create its ARM cores and mobile SoCs. Those processors would later displace Intel from Macs. So, in a way, you could view that as blowback from Intel's decision to kill off Alpha. In another turn of karma, some of the lead designers left Apple to found Nuvia, which Qualcomm bought to help it fend off Apple and make inroads into the laptop market currently dominated by x86.

 

[hotaru251]

If an industry or company is systemically vital to the economy, then it already has enough intrinsic value to justify the loan.

except when it keeps needing help & already getting money to expand (which if it has trouble means wont get expected results) as we las fact that even with aid there is no guarantee it wont fail and become unable to pay back the loans.

The idea of the government treating private sector loans as a profit center is hugely problematic.

except its not because if gov keeps stepping in then large important business know "even if we frick up its ok as the gov will step in to help us" making mistakes mean less.

A gov shouldnt be bailing out mega corps for bad choices and should instead use that money for the people who are in more need of the aid.

Also no company should be "too big to fail" as if thats the case then Gov failed to keep mergers/buyouts from happening that removed alternatives should one company fail.

 

[bit_user]

I was under the impression that TSMC's success is partly attributable to it being a dedicated foundry, i.e. a fabless designer perceives less risk of IP/trade secrets leaking to rival designers if they use TSMC instead of IFS, Samsung, or GlobalFoundries (before AMD spun it off).

To be clear, I'm not saying you're wrong—I don't know anywhere near enough about the semiconductor industry to make that judgement. But I'm having trouble squaring your comment with the above. Is my impression incorrect? e.g. wikipedia lists other advantages of the dedicated foundry model, but doesn't explicitly say whether mitigating IP theft is one of them.

The way I view the situation of IFS' conflicts of interest is like this: a fab has early access to production schedules, node selections, pricing, product code names, die sizes, and other high-level details. If this business-level information leaked to a competitor, they could respond in tactical and strategic ways to better position their own competitive offerings (or even cancel products that would be uncompetitive). That's why I think AMD is reluctant to use Intel Foundry Services, especially since so many people at IFS would be privy to such information that still have contacts in the design side of Intel.

I'm sure IFS has NDAs, but those are only enforceable to the extent you can prove they were broken. If I were AMD, I wouldn't trust IFS with any products that compete against anything from Intel, which is nearly all of their portfolio except perhaps the datacenter network products (which I think Intel has largely divested).

I believe none of this is what Gururu was talking about. I think the concern was specifically that having a chip made at TSMC would potentially give the Chinese an opportunity to reverse engineer the design of the chips. As I said, that's not feasible.

Gee wiz, it's almost like they have what amounts to a monopoly in the modern era. Imagine if we just split them up

FWIW, I wanted Intel to be forced to spin off its fabs about 10 years ago, when they actually had a dominant lead in chip fabrication. That gave their CPUs an unfair advantage. If they had reinvested in their business as they should've, they'd probably still be well ahead of TSMC.

 

[bit_user]

Intel was sanctioned out of China and Russia and many other countries.

The China sanctions imposed by the US were on a limited number of products. For Intel's bottom line, I'm sure the issue is much more China's decision to become self-sufficient for their computing infrastructure, which was destined to happen anyway. At worst, sanctions on which CPUs and GPUs they could buy merely moved up the time table, but then cutting off access to advanced silicon fabrication actually slowed it down by even more.

I'm sure Russia was a much smaller market, for Intel, and likely to (eventually) start sourcing from China anyway, for various reasons. What "many other countries" do you mean? AFAIK, it should be only a few other countries like Iran and North Korea, neither of which would be big markets anyway.

This was meant as a penalty for the both, but the people paying for it are western companies, that loose the market share and sanctions allow competition to thrive in those markets. Nobody is suppressing the competitors anymore in those markets. The western sanctions are a self-sanctioning policy, if you want to apply this approach on a too global scale.

China has long been on a path of self-sufficiency in all things. Sure, the sanctions might've pushed more of the Chinese market to switch over to Loongson a bit sooner, but it's just facing an issue that had been looming. The timing isn't great for Intel, but it's only one of many factors now working against them.

 

[bit_user]

Intel just need keep pushing the 7nm or regbadge to another name...

It's not quite what you mean, but Bartlett Lake will be made on the latest iteration of Intel 7, due out in early 2025.

It will offer up to 12 P-cores (no E-cores) in LGA1700.

 

[mac_angel]

I'd say NVidia should buy Intel and make CPUs and GPUs, the same as AMD. But I can't afford NVidia's <Mod Edit> already; I can't imagine the price gouging they'd get up to if they bought Intel and took over that segment of the computer industry.

 

[bit_user]

But what about those spending their nights under homeless shelters in bad weather?

Lawmakers: WHO CARES? We gotta save this multi billion dollar corp.

There are various justifications for this:

• Investing in domestic industries stimulates the economy, which gives such spending a multiplier effect.
• Bailouts are usually in the form of loans, which are usually paid back.
• The economic impact of having no domestic leading-edge fabs would be quite severe, if China asserts control over Taiwan's fabs.

Modern society literally cannot function without modern technology. It's infused throughout many industries, enabling much higher productivity and efficiency than we had in like the 1970's. This is not just about jobs and the economy, but ultimately an issue of societal importance.

who sends all the revenue offshore and hardly pay any tax. Not to mention, we know exactly how they do all this with corrupt bookkeepers but we not gonna do anything about that because we also get paid to look away.

There are different potential solutions to those problems. I think it would be a mistake to conflate them, like the proverbial "cutting off your nose to spite your face".

 

[bit_user]

If I'm the Intel CEO, I spin off the foundry business. It doesn't seem like much of an advantage to own the design and production end to end at this point. In fact, its held Intel back while AMD has been able to use smaller process nodes.

Wall St. knows this. That's why a lot of people are concerned the US could lose its advanced fab capability, if the fabs are spun out and left to fend for themselves.

Also, being vertically integrated did give them a competitive advantage, back when Intel was ahead and the fabs weren't so capital-intensive. It probably even gave Intel an advantage as recently as Raptor Lake. AMD made its cores smaller, possibly be offset the added cost of having to make them at TSMC.

Intel also cannot give up on the GPU business again. They need to keep pushing forward with the Arc GPU's because the CPU business is going to continue to decline.

I certainly hope they stay in the fight!

I also don't understand why Intel (and AMD) haven't been bigger promoters of RISC-V.

Because it would undermine their cash cow x86 CPUs. They will drag their feet on ARM and RISC-V until the industry hits an inflection point. AMD is rumored to start selling ARM CPUs next year, which is probably necessary for them to avoid hitting a ceiling on their cloud market penetration.

If I were them, I'd be manufacturing some RISC-V CPU's even if there wasn't a big market. They need RISC-V to gain some footing so they don't have to pay royalty fees to ARM for the next 50 years.

Intel took an active role in helping get the RISC-V SoC market bootstrapped, several years ago. Horse Creek was a limited run RISC-V product made by Intel's fabs that integrated cores from SiFive with other IP from Intel, such as memory and PCIe controllers. Intel was even rumored to have put in an offer on SiFive, that was rejected.

 

[80251]

What's even your point? Compaq bought DEC, which then sold off the CPU division to Intel. HP bought Compaq. I'm sure existing DEC customers had their support contracts honored, even as DEC products were phased out.

DEC's Alpha was indeed ground-breaking, but most of DEC's business was geared towards servers and mainframes at a time when x86 was ascendant and mainframes were being squeezed out by commodity servers.

DEC didn't manufacture mainframes. They weren't business computers they were geared towards scientific computing. The whole idea of DEC competing w/Big Blue in the mainframe market in the 20th century is frankly ludicrous.

 

[bit_user]

DEC didn't manufacture mainframes. They weren't business computers they were geared towards scientific computing.

Sorry, I suppose I mixed up mainframes with minicomputers.

I once knew some people who worked at DEC, on their Alpha CPUs. In the 90's, I followed some developments in their workstation line. I don't know too much about their other products, except bits of lore I've heard about VMS and their PDP systems. I've heard of other things, like DECnet.

Bell Telephone was presumably using them for various things, back when the folks at Bell Labs developed UNIX on DEC PDP-11's.

 

[bit_user]

except when it keeps needing help & already getting money to expand (which if it has trouble means wont get expected results) as we las fact that even with aid there is no guarantee it wont fail and become unable to pay back the loans.

I think you're conflating CHIPS with a hypothetical bailout of some kind. CHIPS is actually a grant and it's not Intel-specific. In Intel's case, the point was mostly to help them with capacity build-out, because they supposedly have technology on par with TSMC's. The reasons we expect to need more leading-edge chip production capacity should hopefully be obvious.

The bailout, on the other hand, is to help Intel through a bit of a crisis, which they now find themselves in. This wasn't the point of the CHIPS money, so it shouldn't be seen as a failure of that funding that it didn't avert them getting into this predicament.

except its not because if gov keeps stepping in then large important business know "even if we frick up its ok as the gov will step in to help us" making mistakes mean less.

There are ways to mitigate against the moral hazard of bailouts, such as by restricting activities which reward executives and investors. In most cases, these folks are already hurting, by the time it comes to needing a bailout, because the share price will already have tanked and most executives get a large amount of their compensation in the form of stock options, which would then be worthless.

A gov shouldnt be bailing out mega corps for bad choices and should instead use that money for the people who are in more need of the aid.

Government spending should either benefit citizens or protect them from harm, and only when private sector finance can't or won't step up. If the US got cut off from some or all Asian fabs, the economy would go into a full-blown depression, as the impacts ripple throughout countless industries. That would put a lot more people in desperate financial conditions and take a long time to unwind. You forget too quickly what kind of pain was felt during the pandemic-era chip supply crunch, and that was but a mere taste of what could happen.

Also no company should be "too big to fail" as if thats the case then Gov failed to keep mergers/buyouts from happening that removed alternatives should one company fail.

Okay, so tell us what the US government could've blocked that would've prevented Intel from having a lock on leading-edge domestic semiconductor production. Should the US government have forced Global Foundries to keep working on its 7nm node & beyond? What legal basis would they even have for doing that? What about other domestic fabs, like Texas Instruments? At what point do you draw the line?

I'd point out that other governments don't apparently have such qualms about helping their industries become & remain competitive. China, South Korea, Taiwan, and Japan all spend lots of money to develop their semiconductor production capacity. This requires big and long-range spending of the sort that private finance isn't really geared to do.

 

[tommo1982]

AMD came back stronger, but just look at what happened to Global Foundries!

Certainly, there were a lot of factors at play and I'm not saying history will definitely repeat itself. However, care should be taken specifically to avoid the same outcome, or else we'd lose another competitor for leading edge semiconductor fabrication.

Nope, it did not. It almost went bankrupt with Bulldozer. It took them a few yers of bleeding cash to become profitable and ZEN was pretty much a gamble, all or nothing.
It's even worse for Intel. I really hope it can resolve its internal issues. Intel made many interesting tech. I wish it'd continue to do so in the future.

 

[bit_user]

Nope, it did not. It almost went bankrupt with Bulldozer. It took them a few yers of bleeding cash to become profitable and ZEN was pretty much a gamble, all or nothing.

pug_s was the one who originally said it "came back stronger", which is true if you don't stipulate the time period involved. Sure, I could've nit-picked that part, but I think the more important point was what happened to the fabs it spun off. I decided to place the focus solely there.

I'm not sure we can say, for certain, what would've happened to AMD had it not spun off its fabs. It's certainly conceivable they would've simply gone bankrupt.

 

[UniFan]

can somebody give me a rundown on why Intel is in financial trouble? I haven't really been paying attention as of late. I don't mind a very long rundown, just a rundown is all I need. Thanks in advance

 

[bit_user]

can somebody give me a rundown on why Intel is in financial trouble? I haven't really been paying attention as of late. I don't mind a very long rundown, just a rundown is all I need.

Hopefully, someone can provide you with a better answer than this, which actually ranks the various factors in terms of impact, but I can list some of the bigger problems they're facing.

Their cash cow has long been the datacenter market. Most of their biggest customers have turned to using custom ARM-based CPUs, and AMD has now taken a huge chunk of the x86 server market that remains. This not only reduces their volume, but also their negotiating power on server CPU prices. Plus, China is trying to make its own server CPUs, though I don't know how much real decline they've seen in that market.

Meanwhile, newer manufacturing processes are more expensive and they're having to include more CPU die area to compete, which squeezes them on the cost side.

As if that weren't bad enough, they've been investing in fabs for future manufacturing nodes, which is getting ever more expensive. I think a new fab takes something like 5 years from start to finish, before it can start churning out product and recouping the investment. If you stop building fabs, you risk falling behind in the future, so options are limited on that end.

On the desktop products front, they had to cancel Meteor Lake, which left them selling defective Raptor Lake CPUs for two years. They claim now to have fixed the issue via firmware/microcode, but are still on the hook for an unspecified number of warranty claims.

It gets worse: their latest desktop and laptop CPUs are now made almost entirely by TSMC, which has made them less profitable at a time when Intel could really use the money.

One of their biggest customers, Apple, switched to making its own CPUs starting about 4 years ago. The ARM crowd is also gunning for the lucrative Windows laptop market, with Qualcomm leading the charge.

Oh, and their GPU and AI businesses have yet to turn a profit. Their GPUs missed the crypto boom, by a couple years, and had middling performance (at best) and could only be sold at bargain prices. Their AI solution, Gaudi, has yet to find a good market niche.

Here's a historical plot I found of their revenue, though it only includes data through 2023:

Intel revenue 1999-2023 | Statista

In 2023, Intel's revenue amounted to approximately 54.2 billion U.S.

www.statista.com

Here's the article discussing their latest quarterly report, which goes into some of these areas in a little more detail:

https://www.tomshardware.com/pc-com...lion-in-q3-reports-usd13-3-billion-in-revenue
​