Hermann Hauser: ‘It’s in Nvidia’s Interests to Destroy Arm’
tech.newstatesman.com
tech.newstatesman.com
Actually, SoftBank got ARM at a special Brexit discount. For them, they saw it as the best deal of 2016 and for the UK, you might as well say they got scammed.
So fast forward to 2020, SoftBank didn't get that 10x return they wanted from ARM and is now selling to get a tiny profit, where it is apparently heading for Santa Clara.
The rest of the UK didn't care and didn't know what they had, until it was too late and the same fate was repeated for Imagination Technologies. What a shame.
[0] https://qz.com/734882/the-32-billion-japanese-takeover-of-a-...
[1] http://blog.policy.manchester.ac.uk/posts/2016/08/arm-holdin...
... and probably lots more that I don't know about since I'm not from the UK. Blaming the misled and uneducated masses for this kind of thing just doesn't work.
Not sure how much money they made selling their stake on the company, but if Wikipedia's valuation for Acorn and ARM are correct, can't be more than a couple of million dollars.
Those shares would be worth many billions of dollars today.
---
Why do you think the brits have anything to do with anything?
If ARM ends up being bought by NVIDIA, that would actually be this Hermann Hauser's guy fault for selling ARM for pocket change. Complaining that their buyer's buyer's buyer now wants to make money with those shares by selling them to the highest bidder is sad. If ARM was so important to him, then maybe they shouldn't have sold their stake on it in the first place... Making a show about it now is like complaining that your ex that you dumped now has a new boyfriend.
I’m sure WeWork imploding and other portfolio businesses not meeting expectations are also influencing the sale.
Recognize biases of hindsight, but it seems like private equity markets in Britain aren't interested in technology at the requisite scale.
There's an incredible historical tradition [1], but the majority tend to be either developed with the help of a (1) few angel investors (and the luck and smaller scale that entails) or (2) government contracts (and the politics that entails).
[1] https://en.m.wikipedia.org/wiki/List_of_British_innovations_...
There is also a weird preoccupation with the London startup scene; politicians decided they were going to nurture it and help fund things there. Which is fine, but they ignored the Cambridge area which is where a lot of the actual innovative tech in the UK gets developed, in favour of (effectively) a bunch of cool-sounding app vendors.
1. In terms of sensible investment, what I've seen is that success breeds success - one success creates wealthy individuals with experience that can lead or by part of the next round.
2. UK markets are quite conservative, cautious and by and large dominated by people who don't understand science and technology, compounding the problem. They tend to want to drip feed money rather than go for the big win.
3. Following on from 1 & 2. There is a culture of selling early, rather than building really big, and if you don't have anyone big in the UK to sell to then it goes abroad.
4. UK government appears to welcome, nay have pride in, the acquisition and destruction of some of the UK's biggest companies by opportunists.
5. UK business leaders typically lack global ambition, nay ego/hubris.
6. UK stock market is rather short term, preferring dividends to inward long term investment - the sort of investment typically required to grow.
So what you get is lots of interesting small to medium size companies - often started in concert with the strong UK academic sector, and/or diaspora from previous successes - and very few transitions to global players.
ARM was one of the few - that's why it's so disappointing to see it sold first to Softbank and now possibly NVidia.
The only silver lining is any change is likely to create a diaspora eager for new opportunities - so in the case of ARM - lots of experienced well off engineers ( more able to take a risk ) in one of the worlds best academic centers of excellence.
Haha, me laughing then crying in underpaid German conservatism that dwarfs the British one.
You guys don't know how good you have it. Brexit aside, London is still Europe's biggest tech capital by a long shot and devs there can make way more than anywhere in Europe bar Switzerland.
Typically they need less upfront investment ( unless you go the buy customer route a la Uber etc - but the jury is out on whether they will ever pay back their investors ), and can grow much quicker and so are less constrained by institutional investors.
London is certainly very good for things like Fintech which employs a lot of software people on good salaries in London. But then city investors understand Fintech.
My experience is mostly hard science tech, rather than software per se.
I'd consider quite a lot of the software 'tech' companies to really be business model innovation companies rather than 'software' innovations ( ie software that's really hard to write ). The tech bubble classic example in London was lastminute.com - a holiday clearing house website - who's founders didn't understand the technology side.
One technical area the UK is strong in is ML - Googles deepmind started in London, and machine learning for Apple, Amazon, and Microsoft all worked on up the road in Cambridge. Again seeded by small companies bought by larger foreign companies.
In my view the biggest reason for the difference in success between UK/EU tech companies and US ones is simply the sheer size of the US market. Particularly for the current set of big-tech incumbents which are largely consumer focussed.
The creation EU Single Market (in which ironically the UK played a large part) was an attempt to compete, but while larger and having greater regulatory uniformity in many areas than the US, different EU countries speak different languages and have a less integrated culture than the US. This makes it harder to grow companies to the size they might in the US.
Britain has the problem that too much of the country is focused on finance + the services that surround it, the country bends over backwards to support those industries through deregulation/not enforcing rules (LIBOR scandal et al), intervening with foreign governments to protect banks (see HSBC 'too big to jail') while pretending it is lassez-faire & does not have an industrial policy.
The ironic thing is that the financial industry could help develop these seeds of technology into behemoths but have misguidedly focused on inflating land values + developing dodgy financial instruments.
Another problem is culture - many Brits have the ambition of making $ millions of money and then retiring to a country house. Not many Zuckerbergs have emerged from the UK.
It seems we're yet to learn that there's a possibility to go all the way.
Why is it the US's fault that they consider tech to be valuable to their economy and national security and we don't? What's stopping us from doing the same?
As long as we'll keep treating tech as some cost center instead of a core business while we keep pumping money to keep afloat 19th century industries who's ships have sailed long ago, then we deserve what's coming to us.
We let Germany buy one of the most important companies on the planet in Monsanto. I consider Monsanto to be a national security level company, we should have never let it go. Seed & AG tech is critical for the coming century and Monsanto is one of the very few global giants.
For a long time we've been allowing European pharma giants to catch up in biotech by acquiring several of our important biotech companies, as with Genentech, Genzyme, Baxalta, MedImmune, Bioverativ and so on. We let Germany's Merck buy up Sigma-Aldrich. We let our medtech juggernaut, Medtronic, rebase to Ireland, no doubt taking important IP and jobs with it over time.
When I watch important US companies like Genzyme get consumed by a European company, I fully understand how the Europeans feel about ARM etc.
Rather, that Europeans should place more value on those companies--and either: for competing European firms to outbid US acquisition attempts; or for the owners to not sell them when the offers are too low.
One of the major challenges is "synergies" (gak!). Acquiring, say, ARM adds much more value to a company like Apple or nVidia than it does to SAP or Spotify. There's no way around the fact that right now American tech companies can afford to outbid European ones for ARM because ARM adds more value to American companies.
You seem to imply here that the US acquired this company against the will of the seller, which I don't think the parent thread was getting at (nor is true, as far as I know).
Many things around have "global importance," and it is long past the time for any modern, globalised economy to undo that.
If China wanted to f__k up the West economically on the supply chain side, they have far, far, far more options to do so than the other way around, and this is not limited by companies on the extremes of labour intensity scale. They are plenty of both low, and high tech companies which are one and only in the world in China.
And so can many countries with industrial economies, not event necessarily large one. Take Thailand for example, and how they gave a fright of a lifetime to Internet companies with a single phrase "It may take at least a few years before the world can make new hard drives."
Or Taiwan, a Microchip embargo today from them may well be a worse thing than an oil embargo of 1970s to a modern economy.
Japan, and Korea both have a deadgrip on high material science products, and not so high as well. Both have giants which control chemicals used in 90%+ of all synthetic fabrics. Life without synthetic fabrics today will not be fun to anybody.
India has a generic drugs supremacy, and is relied upon by much of the world other than the West for just any drugs.
The one and only in the world top tier substrate plant went up in flames in China week ago. All major chip vendors been using their substrate for the package for a few years. If you see a chip with HBM memory, or any model CPU, it's their substrate which is used.
On the other end of high techness scale, China has 90%+ of global market for such essential industrial goods as screws, nuts, and bolts.
Neither of these examples avail to being easily substituted, and I can come with many more.
USA has created monstrous monopolies that are very heavily subsidized by the government. Amazon, Apple, etc. do not pay taxes and yet get big sums of money to settle in cities.
That is the money that USA is using to buy European companies. China blocked the USA from buying their tech, and the result is a healthy ever growing tech scene.
I have worked for a couple of European tech companies purchased by USA corporations. First "nothing will change" is the message, two years later the reduction in quality is clear and all the tech is siphoned to the USA side of the business. It makes sense, it is good for the USA.
Europe should follow the Chinese example and ban most of USA tech. No more Facebook radicalizing our citizens, no more Amazon taking over all business without control, ... When TikTok started to grow, the USA banned it. That is the game. Europe should follow, ban Facebook, Twitter, ... let local talent grow. There is much wasted talent in Europe overwhelmed by the current position of USA companies.
"Just don't sell" does not work when your adversary puts all the weight of the state to protect and subsidize their tech industry.
> I have worked for a couple of European tech companies [...] all the tech is siphoned to the USA side of the business. It makes sense, it is good for the USA.
Out of curiosity here: Could you name one/a couple of examples? Not necessarily the ones you worked at, if you are prohibited from disclosing this information.
Email is in my profile, if you do not want to make any public statements.
Cheers!
It was a successful startup but $8.5 Billion is a lot of cash.
I heard that developers were not happy with Microsoft way. Promotions/Pay/etc. became a competition between colleagues instead of a collaboration.
Of course in that case my colleague leaved the company, maybe other people really liked it that way and stayed.
But, my personal experience is similar. USA companies want employees competing for recognition instead of collaborating to compete with other companies.
Amazon taking over all our businesses? That's a bold claim, any stats to back that up. A family member of mine runs a business on Amazon. You'd like to kill her business? There is literally no other platform that does something like fba.
Also the reason Amazon and Google are successful here in the UK (and everywhere they are allowed to operate) is because they are way better than the competition.
Yes. That is the point. The need exists. European companies will offer that services creating high quality jobs in Europe and they will follow European cultural norms instead of American ones.
> There is literally no other platform that does something like fba.
That is the monopoly part promoted by the USA and financed with public money. Amazon is a monopoly protected by a friendly but in the end foreign power that has misaligned interests.
I can assure you that if tomorrow Alibaba grew to challenge that number one position, it will be banned from doing business in the USA.
The services are needed, here, in the USA and in China. USA has its own tech companies, China has its own tech companies, Europe has not anything at that level because USA just buys anything Europe creates backed up by USA tax-payers money. That is not fair.
Are you really trying to claim Amazon is financed by tax payer money in the USA? Can you provide examples of this investment?
Fba is monopoly because Amazon took a huge bet and invented it. Same with prime.
They don't pay taxes and pay workers like shit. The federal government gets no revenue, and still has to increase social spending and policing to avoid workers burning the place down, protecting Bezos' investment. Then there's Blue Origin, which involves direct handouts from Uncle Sam.
You are right there. We are the product, we give our data, we give our attention, it is not free.
> unopposed monopolists that acquire or destroy any business that might pose a risk to them
Good point. That is also true for other ones like Facebook that purchased Whatsapp just to keep the monopoly in personal communications.
Here [1], for example, is one for Apple, showing them paying an income tax rate north of 20%, for tens of billions of dollars. Then they pay payroll taxes, property taxes, import taxes, sales taxes, for many more billions.
All the companies you claim aren’t paying taxes are.
[1] https://www.sec.gov/Archives/edgar/data/320193/0000320193180...
https://fortune.com/2018/01/18/apple-bonuses-money-us-350-bi...
You are taking the statement literally. In that sense you are completely right, that companies pay an amount of taxes above zero. But, if I pay $100 in taxes but I get back $110 then I have effectively not paid taxes. Even that is not literally true as I paid and then I got them back.
> income tax rate north of 20%
That is after all the accounting magic. As a non-tech example, did you knew that Harry Potter movies did not won any money (https://deadline.com/2010/07/studio-shame-even-harry-potter-...). Of course, that is ridiculous, but it is accounting truth.
Should they also pay US Social Security and employment insurance on their foreign workers in foreign countries?
If you want to look at the foreign country equivalents of these SEC filings, they're all public, and you can look through them. When you do, you'll find Apple pays north of 20% of worldwide income, when totaled worldwide. I've been through their biggest several markets looking through financial data. It's all there for you to check.
You're just making my point - Apple paid all necessary US tax, and it's quite a significant rate (just like US law) compared to US profits. They also paid all required foreign taxes, also at a pretty high rate.
Also, if you think somehow Harry Potter accounting implies Apple accounting, then if I point to a company paying the amount of taxes you like, is that evidence Apple did too? Of course not. It's an invalid comparison or implication in either case.
Same for your made up $100 and $110 story. What if I make up the same story with $100 made and $110 paid in taxes - does that imply Apple paid more in taxes than they made? Of course not.
>That is after all the accounting magic.
What SEC number do you dispute, say for 2017: US sales of 229B, operating income of 61B, or 15B in taxes? These are all well documented, quite reasonable numbers. If they're lying on these, there are significant costs for them if caught. And there's thousands to millions of investors that also care a lot about how accurately Apple records numbers. It makes much more sense these are the legally correct tax numbers.
I presented you Apple's legal filings to cut through all this opinion piece nonsense, since for every opinion piece one way there's one the other way.
The truth in the matter are the legal filings: IRS and SEC. And they're pretty clear Apple pays more tax than the pop believer thinks.
If you want to argue Apple doesn't pay the proper taxes on US income, show me a SEC or IRS filing that shows it.
So you can believe opinion pieces that match your current beliefs. or you can look at the legal filings with the actual data, usually left out of opinion pieces. One will give you a much better and more accurate worldview. One is mostly designed to create outrage and pageviews - the more pageviews the more the opinion piece place gets paid.
Otherwise they, like you and pretty much every entity, pays the tax their required to be law. And like you, and pretty much every entity, they do their best to not pay more than the required amount.
Many small and mid-sized business struggle or refuse to scale because it exposes them, their owners and management to taxes which cannot be escaped (by contrast small businesses can pay their staff in cash, and charge personal expenses to business accounts for VAT writeoffs).
Alternatively many European businesses are optimised to receive Government and EU grants and subsidies, not to compete on the international market.
Yes, the USA is unequal compared to Europe, but this might be the price worth paying for innovation and technological dominance: America's poorest are richer than the average Europeans:
https://www.acton.org/publications/transatlantic/2019/08/27/...
The solution is for Europe is to dramatically cut taxes (eliminating payroll taxes and reducing VAT to 16% across Europe should be a priority), increase the pension age to 70, and limit access to bachelor's degrees to the top 20% and Master's degrees to the top 10% (many Europeans spend their entire 20s at university, when they could be working instead).
The alternative is basically slow Demographic decline and descent in complete central planning, which Europe seems to be pursuing.
Having lived on both sides of the Atlantic, I would say it felt much easier to start a business in the UK: Healthcare is covered, believe it or not there is less bureaucracy, there's great digital banking options that the US is only just getting its head around, etc. In Europe I always felt I had a safety net to fall back on if my business failed – not so in the US.
My own hypothesis is that the European stock exchanges aren't a particularly attractive place for an IPO. London's AIM, isn't quite the NASDAQ. The US tech stocks are now worth more than the entire European stock market according to Bank of America Global Research.
And so I think it's always in the back of a Founder's mind – not to mention their investors – that they'll have better liquidity if they're packaged as a US inc. to start with – and if not just sell to one.
Hopefully with Sequoia opening an office in London though, this will begin to change.
It even goes as far to basically blame the poor for their own poverty. It makes zero consideration for the higher health care costs or general cost of living differences between America and Europe (for example: The average European does not need to own a car, does not need expensive heating or extreme cooling, and their healthcare costs are far cheaper) The most notable bit is healthcare, as the hilarious math says 'well the CHIP benefits save them X thousands of $ vs out of pocket costs!' (Because healthcare costs WAY more in the US) but the purchasing power index is not biased in that category so it accounts the same.
Might I add that the author's main book is about proving the bible is true and that the earth is 7000 years old from a 'I think it is true so here are my mental gymnastics ' perspective.
The USA's privatised healthcare system would operate a lot more efficiently if sugar was heavily taxed, fresh fruit and vegetables consistently available (no 'food deserts'), and coal and oil burning for transport and energy minimised.
I think that's too much of a blanket statement. France (the second or third largest economy in Europe depending on the year) is a potent counter argument to that premise.
Bernard Arnault, $89 billion. Francoise Bettencourt Meyers, $66 billion. Francois Pinault, $42 billion. Alain Wertheimer, $32 billion. Gerard Wertheimer, $32 billion.
Those are immense dynastic fortunes and there's no indication those people plan to give it all away in the style of Gates, Buffett, Zuckerberg, Scott, Carnegie, Rockefeller.
The US economy is about eight times larger than the French economy. If you expect that an economy eight times larger might produce something in the ballpark of fortunes eight times larger (one can apply a discount and it'll still be extreme), that wealth scale & concentration is mind-boggling large even by Bezos standards. Arnault's wealth might be like someone in the US being worth $712 billion in terms of how it towers over the economy and other wealth in the nation.
Germany has 34 of the top 500 fortunes in the world. 6.8% of the 500 richest, and Germany is slightly over 1% of the world's population, and about 4% of the global economy. Clearly they like their giant fortunes as much as the next affluent nation. And the Germans are notorious for their dynastic wealth as well, they have no problem passing down immense fortunes to the next generations.
(I'd do national share of global wealth for Germany, however that's a super messy batch of numbers, especially settling on even semi-accurate estimates for global wealth; it's still fairly interesting to see how their economy represents itself as part of the global share, and then how many top 500 fortunes that has produced.)
Sweden, with just ten million people, has 12 of the top 500 fortunes.
Russia - now infamous for their billionaire oligarchs - has 24 of the top 500 fortunes, while simultaneously having tens of millions of the poorest people in Europe.
It turns out that extreme taxation helps aristocrats...
But one the other side, you have tons of small to mid size enterprises that got bought.
For example, Adobe has bought quite a few French companies such as Neolane (marketing solution) Fotolia (image bank) or Alegoritmic (3D texture editor) in the past few years.
In another world, I could imagine the Nintendo Switch running on PowerVR tech. Instead, it's an NVIDIA SoC (not a bad one, to be fair)
There is almost no chance they make a profit on ARM.
>SoftBank got ARM at a special Brexit discount...
ARM was sold for $32B. It is easy to say they got a discount when even AMD is currently valued at $100B+ today. ( Yes.... the market is quite insane ) At the time many thought it was over priced, just like how WeWork was over priced.
ARM make less total revenue than Intel's Net Profits. And at the time, $32B was 100x + its Earning. And I would not be surprised it is still in the three digit times earning in 2020.
>So fast forward to 2020, SoftBank didn't get that 10x return
I am not sure if SB wanted 10x return. But surely no one sane was expecting it to be $320B.
You have RISC-V chipping away the bottom, low end embedded category, You have Qualcomm, ( or Had ) Huawei, Apple, Samsung representing ~80% of total Smartphone SoC Shipment. These four paid a one time license fees. I.e ARM could raise price for the next generation ARMv9, or current v8. But it wouldn't impact the big four. They could choose to stay with their current v8.
It was the same with Imagination Technologies. Yes I am bitter mostly because I think Apple lied. ( They are still using PowerVR technology. ) I guess we may never know the full story. And it was sold off to some Chinese Capital. But IMG was not without its faults. Buying MIPS? Seriously? Had they done it 5 years earlier it might have been a better move. But when they bought it most MIPS vendors ( Mostly in Networking ) already had plan to move platform to ARM counter part. IMG PowerVR was competing with ARM Mali which is much cheaper ( ARM's main focus is licensing their CPU iP ). I dont see IMG fighting for more IP licensing, it seems they were happy Apple is using their IP and wants to simply continue to do what is working. And they could have sold themselves to Apple. Probably for a better price than they did to Chinese VC.
I am not sure what is missing in the UK. But from a Strategic Perspective they are very often lacking vision and big picture. And that is especially true in Tech Industry.
Talents within the UK saw their leaders and management have absolutely dismal understanding and decide to go somewhere else, mainly US.
Yes. For relatively simple in-order cores, the RV32 instruction set is far simpler to implement than ARM, or even MIPS, and should be able to provide better performance.
If the existing SW dev toolchain is sufficient for your needs right now, I think RISC-V can be a great choice for small embedded SoCs.
Faced with the political impact of "that amazing US trade deal we promised you is off the table" versus "a company you probably haven't actually heard of, which was already Japanese-owned, is becoming US-owned"… well, I don't have much confidence.
(Is it still true the UK-US trade deal is awaiting elections in the latter? EU-UK has certainly not been progressing.)
Is it a possibility that an acquisition by Nvidia will be blocked due to anti-trust reasons? Who is the authority who has to decide on that? Is it the UK cause ARM is a UK company? Or Japan because SoftBank is the largest shareholder?
Of course, the larger such a country (group), the more the to-be-merged companies will think pursuing the merger isn’t worth it.
ARM sells licenses for instruction sets and IP cores for processor layouts but not finished processors.
Companies like Apple might have started by buying IP cores but have moved from there to just licensing the ISA and parts of the IP cores but making their own processors (Which is why saying Apple runs a ARM processor is wrong, they run a Apple processors with ARM ISA and some ARM invented components in them).
Furthermore companies like Apple have likely long running binding license contracts where Nvidia has little power to mess with them.
OpenPower, MIPS and RISC-V are also still a thing which have each their own potential to "eat up" ARM if ARM gets "burned" (well OpenPower is more a concurrent to the server/pc space then the tablet/phone or the embedded space).
The ARM ecosystem is fundamentally different to the classical PC ecosystem due to how Android/iOs differ from Windows/Linux and due to a good amount of it being embedded. Which makes it generally easier to switch away from it for new products as no one expects to "carry over binaries" or similar.
So if Nvidia tries to burn ARM I think it would due to all kind of reasons (existing contracts, chips etc.) take long enough so that affected parties can "move on" without to much fallout.
But then Nvidia has good reasons not to burn ARM but keep it like it is and then slowly sprinkle their thinks in.
I mean Nvidia totally lost out on the mobile (phone/tablet) but with controlling ARM they can use that to get back into that marked (i.e. Android phones and tablets with Nvidia graphics) they already shipped chips which are viable for tablets (e.g. the one in the switch) and had some major improvements from what I heard in their coming generations which should allow them to ship pretty good low power GPUs.
But not only do they missed out on Mobile they also are trying to get into "AI at the edge" thinks like self-driving cars and Robots all based on Nvidia chips combined with ARM CPUs. By controlling ARM not only will it be easier for them to do so, but by keeping ARM open they can slowly nudge people to consider/use Nvidia GPUs/Tensor Cores.
So a lot of benefits for Nvidia to buy ARM and keep it open for now.
Through yes on the long run they might slowly close ARM up so that there aren't ARM chips with non Nvidia GPU and similar. But to prevent legal and image problems they should take many years to get there everything else would hurt them on the long term I think.
Which also means there is enough time to switch to alternatives.
But will you switch if ARM keeps being good and now ships with powerful reasonable priced GPUs or TensorCores just because it a bit more vendor lock in which you are likely used to anyway from other HW thinks?? Which is I think what Nvidia is out for.
Substantial anti-trust concerns are avoided, while building exactly the moat you want: no competitor can afford to compete with the level of integration investment.
Having some advantage may be worth the purchase.
Example: Nvidia would prefer certain specifications on an internal bus they'd use, but nobody else uses as a standard.
There doesn't appear to be sufficient profit margin in the industry for too many greenfield, non-reference custom implementations.
So if Nvidia-ARM sticks out a reference implementation or architecture choice that's beneficial to Nvidia, there's limited capability for others (Apple, Cavium/Marvell, Qualcomm, & Samsung?) to strike their own path, and they'll certainly be paying steeply to do so.
Thats if they keep ARM alive and mostly like it is for now at least.
Imagine if it were somehow possible for a company like IBM to acquire the rights to direct Linux development and licensure, including the right to just refuse to let other Linux distribution vendors have access to new kernel features, or to price gouge them over updates, in an effort to squash all competition. Linux would continue to exist, but plenty of people who love what Linux had been would be saying, not without reason, that IBM had destroyed Linux.
Intel makes a lot more money as (a dominant) duopoly supplier of x86 than ARM will ever do licensing its IP.
So when Hauser talks about 'destroying' ARM he really means destroying the existing business model - and that will not necessarily mean flushing money down the drain. Rather it may mean making Nvidia a very profitable dominant supplier of ARM SoCs.
ARM is currently owned by a Japanese conglomerate.
He doesn't consider the Japanese conglomerate owing ARM a blow to the UK's technological sovereignty, but he does for Nvidia.
Follow-up by noting ARMs actual economic value is less than $10B, do it seems unlikely NVidia or anyone else would pay what SoftBank is asking.
Lastly Hauser says NVidia would do this to become the “dominant microprocessor company” in the world, then cut its customers off from new developments.
First, ARM isn’t a dominant microprocessor company, it’s architecture is dominant but it’s revenues make it a flea among elephants. It doesn’t make anything like Intel, AMD, or TSMC, it simply licensed designs. And it’s own designs for that architecture, while popular, aren’t near the best, Apples are.
Apple and others have perpetual licenses allowing them to make their designs and enhancements. No one is getting “cut off” from ARMs mediocre next gen designs, they may never use them to begin with.
All these technologies are intentionally built to make them hard to copy for competitors and easy to protect with patent encumbrance, but on the bright side technologies that actually prove to be useful slowly turn into open industry standards. So basically Nvidia invest a lot in niche features and then recoup the costs using their top-notch marketing.
IMHO what people missing about ARM is that on mobile market there are already companies with worse lock-in who don't tend to share their hardware innovations with others at all, e.g Apple. And high-end ARM SOCs market would only benefit if someone finally gonna compete with Apple on more than just marketing.
Lock-in is not a driver of innovation, it's a brake on it.
There simply very few companies that doing it. Big reason why Intel and AMD have their open source teams is because that's give them edge against Nvidia. All we see from Nvidia competitors is just reimplementation of the same tech Nvidia already had, but once AMD have some good tech they keep it proprietary too.
ARM space is already locked-in and fully-proprietary company (Apple) have leadership. I don't see how Nvidia getting into competition can make it any worse.
Point is, Nvidia can't be trusted not to use that anti-competitively, given their existing reputation and amount of leverage they would get on even more things. So if competition laws would have worked as intended, Nvidia wouldn't have been allowed to buy ARM.
Through you could decide go for SiFive, Inc through I have no idea if they entered or will enter the stock marked.
Don't worry mr. Hauser. The target of the US cornering of the chip market is of course China ...