Calling Marvell an "AI chip maker" is bizarre. They're mostly a networking company, with a fair bit of storage stuff, too. Like any tech company these days, they're trying to jump on the AI bandwagon, but that hasn't actually changed what markets they're operating in.
> that hasn't actually changed what markets they're operating in.
Actually, if you look at their quarterly report, the breakdown of their revenue by market has changed dramatically in the last year. Revenue from the data center market went from 36% to 71% of their revenue, a change which they attribute to AI creating demand. Meanwhile networking dropped from 24% to 10% of their revenue, and other categories have gone down as well.
From the report:
> Net revenue in the third quarter of fiscal 2025 was $1.5 billion and was 7% higher than net revenue in the third quarter of fiscal 2024. This was due to a 98% increase in sales from the data center end market compared to the three months ended October 28, 2023. The increase was partially offset by decreases in sales from the carrier infrastructure end market by 73%, from the enterprise networking end market by 44%, from the consumer end market by 43% and from the automotive/industrial end market by 22%.
> We have seen strong revenue growth from our data center end market, driven by robust demand for our electro-optics and custom compute products from AI applications.
They've sprinkled "AI" everywhere above the fold, but all the actual products they're talking about are still networking products.
> Meanwhile networking dropped from 24% to 10% of their revenue, and other categories have gone down as well.
Enterprise networking is a category they report as dropping from 19% to 10% year-over-year for a three month period. I think you got the 24% from an incomparable column reporting on a nine month period. But either way, enterprise networking is obviously just a small subset of all their networking business.
Marvell is clearly still primarily a networking business. AI is driving demand for some of their networking products (or at least, that's what they're attributing the demand to), but that's not the same thing as Marvell actually making products that are doing AI, or selling such products in enough volume to double their datacenter revenue.
> Enterprise networking is a category they report as dropping from 19% to 10% year-over-year for a three month period. I think you got the 24% from an incomparable column reporting on a nine month period.
Oops, you're right, the comparison I meant to make was 24% to 12% over a 9 month period.
> But either way, enterprise networking is obviously just a small subset of all their networking business.
Is it, though? That table [0] puts everything "consumer" into a chunk that's now only 6% of their revenue. Carrier infrastructure is another 6%. Together that adds up to roughly equal to the enterprise networking row, but that's still a tiny amount compared to the data center revenue.
Are you saying that "data center" includes a lot of networking equipment that isn't filed under "enterprise networking"?
> Are you saying that "data center" includes a lot of networking equipment that isn't filed under "enterprise networking"?
Yes, exactly. "Enterprise" and "data center" are treated as disjoint segments by many tech companies, even though both tend to refer to rackmount hardware. That's why I linked to Marvell's own page describing their data center products, so you could see for yourself that they are indeed networking products: switches, PHYs, NICs, etc.
All of the categories under the "Net revenue by end market" table are comprised primarily of networking products: "Data center" is mostly networking stuff, "Enterprise networking" is obviously networking stuff, "Carrier infrastructure" is where you find networking stuff with the now-passé "5G" buzzword, "Consumer" is probably mostly networking but might actually be the segment with the lowest proportion of networking products (though Marvell's SSD controllers aren't too common these days in consumer SSDs. Also, Realtek has recently started eating away at Marvell's Aquantia revenue for 2.5Gb and 5Gb Ethernet in consumer products.) "Automotive/industrial" is primarily about automotive Ethernet.
Every company is broadly free to organize their subdivisions as they like and see fit. That's how the Oracle Cloud became the biggest in the world, they just optimized their reporting to attribute as much revenue as possible to the cloud org. Massive growth market buzzword "cloud" division up, boring old "enterprise DBMS" down.
They are mostly a chip maker though, both directly selling semiconductors and selling products based on them. Sure, they call their processors an AI product, but so do Intel, AMD, and pretty much every other ARM licensee.
The real kicker is that they picked up their ARM license from Intel, when they bought the XScale line (now Armada) in 2006.
It's been a long time since Marvell was seeing any lingering benefits from that XScale acquisition. Their more compute-oriented products owe a lot more to the Cavium acquisition.
AI chip or even networking chip maker is somewhat incorrect. Marvell makes every kind of chip, will design you a chip, sell you IP, whatever you want.
They started with disk drive controllers, recently they’ve made a decent NVME controller. They bought global foundries custom asic devision in 2019 and that’s when they really got into the custom chip business (according to wikipedia).
> Murphy, who has served as chief executive officer of Marvell Technology since 2016, has been among the few names floated as potential replacements for the recently ousted Pat Gelsinger at Intel’s corner office.
The "job-centric" American economy feels...harrowing now
"While Grove supported helping technology startups, he also felt that America was wrong in thinking that those new companies would increase employment. "Startups are a wonderful thing," he wrote in a 2010 article for Bloomberg, "but they cannot by themselves increase tech employment."[40] Although many of those startups and entrepreneurs would achieve tremendous success and wealth, said Grove, he was more concerned with the overall negative effect on America: "What kind of a society are we going to have if it consists of highly paid people doing high-value-added work and masses of unemployed?"
How many jobs have Uber and Amazon destroyed? How well do the jobs pay? How much work do they demand for that pay? What's the net? Do they create any negative externalities?
You can't just look at revenues without looking at expenses.
For Amazon the claim is more obvious - picking up the "killing local business" and "killing US suppliers" torches from Walmart before them - but I would be a bit suprised if there were any cities where there were fewer transportation drivers now than pre-Uber in the US. Taxis in the vast majority of the country were pretty few and far-between.
They are very different types of tech company; not as completely different as something like SaaS with no cost of physical goods at all, but Uber (while hardly an example of a good citizen company) is not really the sort of "some people in one city have jobs and nobody else does as a result" tech company as some others.
I disagree. You'd be correct IMHO if the welfare stopped if the worker quit his Uber or Walmart gig.
If a poor person has two income sources, either source is less essential than if he had just one of the sources. The less essential the pay from the Uber gig, the less Uber can squeeze the worker.
...wow, this is a really great case against welfare work requirements! You're basically just subsidizing the companies that pay their workers the least.
Without the work requirement, you're actually helping people.
The admin overhead for work requirements is also very much not worth it. It's basically just mandating extra paperwork and if you file it wrong you can't eat.
The reason we have it partly that welfare recipients literally don't like getting welfare without strings attached because they think it's embarrassing (or that the other people getting it don't deserve it), and part that states are hoping to save money by kicking people off for doing the paperwork wrong.
EITC is a way to do "work requirements" without the downsides.
No. Giving people money (or food) increases their negotiating power, which lets them ask for better wages from whoever else is paying them. Uber is especially weak to negotiation because you can just drive less or switch to Lyft.
You're actually arguing that welfare is bad - if giving people welfare allowed Uber to lower their wages that means people would be better off without it.
Thank you so much for this comment, detail and link. I was not aware of this aspect of his management style. The current CEO of my company has spoken highly of Andy Grove (I believe he was at Intel during Grove's tenure) and learning these details has helped me to see some similarities in my own company leadership.
You’re paying way too much, man. Who’s your WSJ guy? They’ll give it to you for much less if you go through the motions of cancelling it on the website.
I'm an old man, 76 years old. The time it takes to go through the motions... is way more than I want to pay, considering each day is more likely to be my last. But I agree with your sentiment.
The WSJ pays it journalists. It's not free. It needs money. Bezos' interference may have left a bit of a bad taste, it's still a reputable paper that doesn't need exposure here.
A common mistake to make. If you just remember that paper that's owned by that billionaire who has politics I don't agree with then you've described all of them.
That’s not quite the truth. Parts of Lunar Lake and most of Arrow Lake will be built at TSMC, but Intel still plans to use its own 18a process for Panther Lake
That doesn't actually change the fact that Intel is now significantly relying on TSMC for several important product lines, nor does it change the fact that even if TSMC were destroyed in a war, Intel would still be struggling with a shortage of competence.
Compare https://images.anandtech.com/doci/16823/AnandTechRoadmaps3.p... against what actually happened: Intel 7 was reasonably successful, except for the desktop chips that were burning themselves out because Intel couldn't get their voltage regulation working right. Intel 4 resulted in slight performance regressions for laptops, but battery life improvements because they moved everything except the CPU cores over to TSMC. Intel 3 was late. Intel 20A was cancelled. Intel 18A isn't here, and neither is the CEO in charge of the above strategy.
It was a combination of issues. Intel needed to be able to supply quite high voltages (at the right moments) to reach high enough clock speeds to win benchmarks with their desktop CPUs, so they couldn't implement a simple conservative, safe cap on the maximum voltage that would be requested. Their attempts to build an on-die voltage regulator (DLVR) repeatedly failed, leaving Intel relying exclusively on a different voltage regulator mechanism than they'd originally planned for. And then they seemingly did a sloppy rush job of characterizing how the chip behaved with their fallback voltage regulator strategy, and messed up the code controlling that system.
There were many steps at which Intel could have avoided this outcome, but it started from the fact that Intel had to push voltages and clocks beyond reasonable limits to make up for having a microarchitecture with poor enough performance per clock that Intel needed to hit 6+ GHz to be competitive.
Samsung is much closer second to TSMC, not Intel. And in any case, we have gone so small that quantum effects have started creeping in. My guess is that 5-10 years from now everyone would be converged to 5% of each other, unless something drastically changes in fabrication process.
In an ironic twist, Intel used to make the highest-performance ARM chips until 2006. They were popular in PDAs and early smartphones, but not x86 and therefore strategically embarrassing.
So Intel sold the Xscale family, just before iPhone and Android happened… To Marvell.
Intel sold just one of the XScale product lines to Marvell, who almost immediately dropped the XScale CPU cores in favor of their own CPU cores and later ARM's CPU core IP.
Marvell has a fraction of the revenue, so I assume this is just a verdict on which company has better growth prospects? I mean, that's always true of market valuations, but: could have been the case whether or not Intel was hitting its marks?
"More valuable than" is exactly the mindset that likely has ruined Intel. "Stock must go up" is often a great example of Goodhart's Law in action. Markets, especially speculative markets like the stock market, are known to behave irrationally time and again. Also, it's rational to buy a growing stock, so the stock market actively punishes activities like internal restructuring, which an ailing company that Intel is has been needing for years, and was / is undergoing now.
NVidia's P/E is currently about 120, that is, the market values NVDA as 120 years of its current earnings. I'm not even certain that NVidia will last as long, or will be in an equally good shape. E.g. IBM has lasted 120 years, and it had wild fluctuations of its market position in these years; toda it's hardly comparable to what it was a century ago.
I would say that Intel is now easier to buy in a large acquisition, without using words like "valuable".
Is it worth the risk? Any company buying Intel would have to deal with the internal cultural rot that has taken hold. Would the buyer be able to survive and make things better or would it turn out like the McDonnell Douglas merger with Boeing?
Intel is no longer relevant. Time to move on. Intel is not inside any phones, gaming consoles, Apple laptops or the plethora of other devices we use daily. The only people still using Intel are old legacy companies with a penchant for overspending on a noisy, hot data center. They will soon be out of business themselves because of the rise of AI.
Intel is no longer relevant and should be shut down, and the money invested in other more promising ventures with a future.
