The arguments made by the Plaintiff are thoroughly convincing to me. The fact that those 8 points are not enough to convict indicate the industry is fucked. Of course the defendants didn't leave a paper trail - they've already been convicted of collusion before.
It's the people and country that suffer when our government fails to ensure markets are free and fair.
If there is no agreement, then the magic term is "tacit collusion."
Why not ask the same ridiculous amount of money your competitors do? People seem to be paying for it. Their fault. If suppliers have sufficiently different products, they can make some more expensive, others cheaper; on average, everybody pays more. A high barrier to entry might help such practices.
That doesn't mean I'm saying this is what is happening. Sometimes things just suck, and somebody bought the world's supply of RAM wafers to use as frisbees.
Yeah your qualifier is important. If I have zyngots and suddenly the price goes through the roof, it is not collusion or illegal for me to raise my prices.
Tacit collusion requires at least signaling. There has to be some intent. Like in the US rental market where Realpage coordinates tacit collusion.
But if there is legitimately demand for far more memory than can be produced, it seems silly to claim collusion to raise prices.
> Why not ask the same ridiculous amount of money your competitors do? People seem to be paying for it. Their fault.
That's not how commodity markets work. This stuff is essentially sold at auction with the price set by supply and demand. The way they would fix the price is by constraining supply so that people have to outbid each other on a smaller amount of inventory.
But that's not that hard to measure -- are they producing less than they were before prices went up? The answer is actually that they're producing more. The reason prices went up anyway is the huge increase in demand.
You would then have to make the case that it's not just that they're reducing supply but that they're not increasing it fast enough. That's theoretically possible but it's also very plausible that building new fabs just takes time, so if someone's theory is that they're colluding then they need to present some evidence.
That also isn't illegal. A business can choose to artificially restrict supply if they want, there's no mandate that they must meet demand.
It only crosses into illegal territory if multiple companies get together and secretly agree to cap production to keep prices high. Then it becomes collusion. It also becomes illegal if there's a monopoly power that is intentionally constricting supply to specifically stop a smaller competitor or lock them out of the market.
The hard part is how do you prove Samsung, SK Hynix, Micron are acting as a unified cartel when there obviously isn't going to be a paper trail for secret meetings.
> A business can choose to artificially restrict supply if they want, there's no mandate that they must meet demand.
There is, however, no incentive to do this when prices are high unless you expect competitors to do likewise, since otherwise you're just handing the business to the others when they increase production. Which strongly implies that if that's what everybody is doing, they're colluding.
> The hard part is how do you prove Samsung, SK Hynix, Micron are acting as a unified cartel when there obviously isn't going to be a paper trail for secret meetings.
Which is the problem when that's what's happening, and why we should maybe change the law to infer the collusion from the outcome in cases where prices are high yet nobody is responding by increasing their market share.
> There is, however, no incentive to do this when prices are high unless you expect competitors to do likewise
This imagines a world where water production can be dialed up and down instantly, at no cost, with no risk.
It doesn’t work that way though. Hiring people, scaling up factories, converting to 24/7 all have costs and risks.
Just google (or Claude, or whatever) memory companies that went backrupt overbuilding to meet a spike in demand. There are many. You probably don’t even know their names. Because Micron/etc stood pat and survived.
> This imagines a world where water production can be dialed up and down instantly, at no cost, with no risk.
It only images a world in which the profit-maximizing amount of wafer production is higher when demand is higher, which is a completely reasonable premise to take as the default.
> Just google (or Claude, or whatever) memory companies that went backrupt overbuilding to meet a spike in demand. There are many. You probably don’t even know their names. Because Micron/etc stood pat and survived.
Those long forgotten companies like Intel, Toshiba, Texas Instruments, IBM, etc. that are all now bankrupt and no longer exist. Except that they still do exist, they just sold their DRAM divisions to the current incumbents. Even the divisions that went bankrupt -- for Qimonda (Infineon) it was the 2008 housing crash that did them in.
Predicting events like that isn't reasonably possible. What Micron actually did to survive it was keep a large reserve of cash to survive the lean years, not refuse to build fabs. The South Korean companies did the equivalent by being subsidiares of giant conglomerates.
I mean, if they're dumb enough to try to game the system blatantly - and there are plenty of execs that are - they're dumb enough to write down their super secret criminal collusion plans in a group text, or a lengthy email chain. That level of stupidity among white collar criminals is shockingly common.
If there's no evidence to be found, that might be a good indicator that either A.) nothing illegal occured
or
B.) the senior leadership involved are competent, intelligent, and discrete
Given the number of companies involved, all it takes is one idiot, so if it's collusion, I expect we'll see evidence. If not, number go up.
We have to decide what part is damaging to society: the actual physical agreement, or the effects of the agreement?
If it's the actual physical agreement that's the problem - the system is working as intended.
But if we are looking to prevent the negative outcomes associated with price fixing and collusion, our system is failing us.
They are never going to find proof of conspiracy. The people involved covered their tracks, and doing so is trivial. So the best we can do is punish the appearance of collusion. And if the goal is to actually prevent harm to customers, that's a better solution anyway, since it encourages leaders of companies to behave in a manner that's the opposite of collusion.
Not directly addressing your points, but I fail to see how this whole situation can be can be viewed as cartel price fixing. There is no oversupply. The strange thing is that we are at the mercy of 3 corps.
You'll have to create a case that harm is taking place. Harm does not mean a PlayStation 5 is now $200 more expensive or that inflation exists.
I would look at questions regarding what harm is created: Is it discriminatory? Are parts of society shutting down, and is that unreasonable? Are groups of people now unable to afford a living? Does it move the poverty line? Is that permanent? And how do you prove this is exclusively due to the price increase of tech components, and RAM specifically?
It needs to be unfuzzy in some way in order to make sense, but that's just my opinion.
I do agree prices are insane and wish for them to come down today. I liked the ubiquitous amounts of RAM any system could have. In those days, forums were also filled with how insanely expensive 32 gigabytes of RAM was, about $100 :)
> You'll have to create a case that harm is taking place. Harm does not mean a PlayStation 5 is now $200 more expensive or that inflation exists.
"Things cost more because of collusion" is always a harm. It doesn't matter if the product is maize or gold-plated haute couture, competitors are supposed to compete.
The question is, what's the best way to tell the difference between tacit collusion and just normal supply and demand?
It's not that easy, but here's a decent test: It's tacit conclusion if 1) net margins have been high for e.g. 3 years and 2) no new companies have entered the market in that period of time, or were acquired by an incumbent if they did.
Notice that this works for everything. Even if you're making luxury goods, the price may be high, but so are production costs, and there is a lower volume to amortize fixed costs over, so long-term net margins should be the same as they are anywhere else or you should see new entrants. If you don't, it's reasonable to infer collusion and leave it on the companies to prove otherwise.
>Harm does not mean a PlayStation 5 is now $200 more expensive or that inflation exists.
Yes it does lmao. If consumers can't get access to devices then they cannot be used for work or education. It's counterproductive.
How many "attention is all you need" papers aren't being written because as soon as there's a sniff of money the rabid suit and tie MBA fucks leap onto any opportunity like a dog in heat and fuck it to death.
HBM is also DRAM. I also think it's kind of a weak argument to say that them discontinuing ddr3 (which while in use still today in industrial/embedded was on the way out for consumers 10 years ago) and ddr4 which last had consumer CPUs for it 3 years ago is meaningful. What we need now is ddr5. Turning off the old fabs and moving those resources including people to ddr5 is a good thing. That's not price fixing. It's possible price fixing is in play, but discontinuing products people objectively don't use as much anymore isn't it.
Switching off DDR3 manufacturing I can understand, but DDR4 machines are still quite relevant and usable… Ryzen 5000 series boxes for example don’t feel meaningfully weaker than they did when new. My 5950X tower certainly doesn’t, and it’d really be nice to be able to upgrade its RAM should I need to because it will continue to be useful for quite some time.
AMD just re-released their 5800X3D for AM4 board users who wish to upgrade which is further evidence that shutting off DDR4 production is premature.
They're running a business not a charity. Their job is to manufacture what the market as a whole demands. If they can make more money making HBM than DDR4 then they have to make HBM. Why would a business go out of its way to make less money?
1. But it might last for at least few more years, see Nvidia 1 trillion backlog.
2. Semiconductor manufacturing is the most complex industrial process in the world. You need billions of capex and decades of experience. Even existing semi players like intel cannot switch production to memory.
China CXMT is gaing traction in DDR market. New fabs from all players wil come online in the next two years.
That's incorrect. Apple purchases RAM from all of these providers to produce their unified memory. They also rely on TSMC fabs for all of the chips their memory relies on. If they haven't doubled the prices of their machines, that just goes to show how fat of a profit margin they take on everyone that buys from them...
You're most welcome! I'll admit I did search before answering because you said Apple produced their memories so confidently LOL
I guess they would love to produce it themselves, but for the average scenario the production reserves they have with Samsung already work well enough and prevent them from having to get into such a complicated industry.
DDR4 machines are relevant and usable, but it's pretty unusual that new systems on older ram are still big sellers. The 5800x3d was originally released in 2022, AM5 processors were also originally released in 2022. We'll probably see Zen6 soon, the third generation on AM5... I don't think AMD released new AM3 processors after the introduction of AM4, certainly not when Zen3 was coming soon.
When making long term plans in 2022, I don't think anyone expected DDR4 to need significant production in 2026. Since ram makers can pretty much sell whatever they make in today's marketplace, it makes sense (for those fabs that can) to stop making DDR4 and repurpose those fabs to make newer generation ram.
That's what I have in my gaming tower, and yeah I feel zero pressing need to upgrade. I did manage to put 64GB of DDR4 in it just before prices went totally bonkers, thankfully. Where I'm falling behind is my GPU I'm still on an nvidia 1660 super, but I just can't justify paying what they cost right now.
I would gain pretty much nothing moving to a newer board w/ DDR5.
Exactly what I'm feeling. The CPU is great for what I'm using it for and I have a killer motherboard. Have a nice GPU capable of current gen games, too. The equivalent current gen board and CPU cost a small fortune for marginal readily visible gain.
Unfortunately, I opted for only 32GB of RAM because at that point more felt like overkill, which as a decision has aged poorly. I should've gotten more while it was still cheap.
Not to be a bootlicker but AMD releasing a product doesn't mean another company should make more DDR4. That's not price fixing. In the embedded space it's sadly very common for a part to be compatible with a very low number of options (shout out to cellular/admux RAM on the STM32H745 nucleo). That's just the way the cookie crumbles.
Dropping DDR-4 is anything but meaningful. It'll easily last 10 more years, machines from this gen are still much more affordable and quite powerful. In fact for most dev and gaming workflows the difference between the DDR-4 and DDR-5 generation of hardware is more or less negligible. I am exaggerating a bit -- but really, not too much.
Of course it might be a ploy to sheep-herd consumers and companies towards the expensive DDR-5. I would not put that below the ring of RAM producers.
>Dropping DDR-4 is anything but meaningful. It'll easily last 10 more years, machines from this gen are still much more affordable and quite powerful. In fact for most dev and gaming workflows the difference between the DDR-4 and DDR-5 generation of hardware is more or less negligible. I am exaggerating a bit -- but really, not too much.
How much % of the DRAM market do you think is made from computer enthusiasts upgrading their Zen 1/2 CPUs to Zen 3? Note intel and AMD both switched to DDR5 well before the exit from DDR3/DDR4 ("2024-2025", according to the complaint).
Note though that for Intel, the first gen of DDR5 CPUs also supported DDR4, and many buyers bought the DDR4 versions of their boards because at that point DDR5 RAM was much more expensive for gains that were marginal at best, which effectively makes Intel’s following generation of DDR5 CPUs the actual transition point.
The point is that just because there's a handful of people (relatively speaking) looking to buy more RAM to upgrade their last gen systems, doesn't mean there's robust demand for DRAM manufacturers to keep DDR4 manufacturing lines going. It's like arguing Sony shouldn't have exited the CRT business because there's retro enthusiasts on youtube scouring the earth for CRT monitors.
For the reasons stated above -- that DDR5-gen hardware is expensive right now -- I'd think the DDR4-gen market will remain alive for quite a big longer. Though that's likely much more on the second-hand market side of things.
While I wouldn't necessarily agree with "a handful of people", the fact is that neither of us can prove their lean -- so no point pursuing that argument thread.
So you might be right that it's a pure numbers/statistics decision. Or I might be right that they want to herd people into the more expensive hardware while forcing them to do so by phasing out production of the cheaper hardware.
No way to truly know IMO. We are exchanging hypotheses.
I’m not sure that comparison makes much sense. By the time CRTs were phased out, demand was down to almost nothing and what little existed was confined to the extreme budget market. While I don’t have industry insights or anything I don’t think demand for DDR4 is anywhere near the bottom yet, and the remaining demand is centered on premium product (nobody running cheap DDR4 is upgrading). In a more normal market would be more than enough to justify continued production for several more years.
DDR4 production is likely still quite profitable, just not drowning-in-money AI-bubble profitable. If smaller foundries existed they’d be happy to take up the business.
Maybe really what needs to happen is some busting up of the giants…
There is a minimum amount of production volume for it to fit the price equation. If the market doesn't have that demand, it is fundamentally no different to CRT.
Otherwise they could continue to make DDR4 at a higher cost and sold at a higher price to which people will complain price fixing again.
Consumers use these every single day in embedded devices without knowing it.
I wouldn't be surprised in the slightest if the embedded DDR3/DDR4 market greatly exceeds the number of consumer desktop computing devices in terms of "devices with memory" (not in sheer IC count or nominal size though.)
The level of design effort and PCB expense to go from DDR3 to DDR5 is enormous.
Especially industrially, ddr3l is just fine. At a certain point you don't need the speed of newer generations, and buying new idk lpddr5x/t controllers for 30 year old process nodes just isn't worth it. Until ddr3/l to EOL from everybody. The big 3 aren't the only memory fabs.
It's easy to see why your argument is wrong with a simple hypothetical: what if they were still making DDR4 today? Would people still buy it?
The answer is an obvious "fuck yeah", even if you ignore the DDR5 price gouging. People will buy it because people still have DDR4 hardware, and that hardware is still extremely relevant.
So if there's a market for it, but none of the suppliers are trying to sell to it... Wtf is happening? Basic capitalism logic says any rational supplier would sell DDR4 for easy profits, meeting an unmet demand. That it isn't happen points to some kind of collusion, IMO.
Because the market pays less for DDR4 than for HBM (or DDR5), and since HBM is heavily modified, vertically stacked DRAM, it competes for the same raw inputs and fab space than DDR4 used.
If I can produce DDR4 for modest profit or HBM for a lot more profit I will obviously produce HBM. And given physical realities producing HBM takes from existing DDR4 production capacity. Worse still, it takes roughly 3GB of ram to produce 1GB of hbm iirc.
> People will buy it because people still have DDR4 hardware
The question is whether there’s enough meaningful demand for aftermarket DDR4 upgrades to make it worthwhile to a manufacturer to keep producing DDR4 instead of switching to HBM and DDR5.
Micron claimed retail is a rounding error, a market not worth serving. So you’d presumably need to find industrial buyers who would be willing to buy DDR4.
There's "fat margins" and then there's the kind of margins that tech industry shareholders expect.
OpenAI's original corporate agreements capped its returns at 100x, which is seen as too paltry for its current holders, so they scrapped those to prepare for an IPO [0]
>But god forbid if someone else captures the margins and squeeze them out of easy profits.
Yeah that's how law works. Everyone likes money, but that doesn't mean it's fine to steal money. Yes, even from maligned entities like "big tech" or "private equity"
Micron is forced to stop under-investing in plants and will increase production. This will trigger everyone else to expand production and lower prices.
The whole point of the collusion is to ensure everyone is producing the same volumes and keeping prices high. The company that expands is the company that "wins" because memory is a volume game and it's all about hanging on the longest during the glut. So once one company expands, the rest have a choice of expanding or planning their exit.
If Samsung and SK lose access to the US market, they'd be fucked long term. Micron would kill them selling at higher margins and higher volumes in the USDM, while the rest are stuck competing for the international scraps - markets Micron is also allowed to compete in, if they wanted to.
Oddly enough, Samsung and SK Hynix decided that they needed to expand production on the same day another lawsuit accused them of colluding to keep production low and prices high.
> South Korea announces $520bn chip plant project with Samsung, SK Hynix
Everyone is using something from someone, you can even argue that US owes India and Europe huge compensation because pretty much everything US did in the last half century was made using technology or people funded by those people. Johny Ive is British, Almost all the AI stuff is created by Europeans, Israelis, and Canadians - thus funded by their respective taxpayers.
The thing about the US losing its grip on the world and the collapse of the global world order means that the words on the paper don't mean much. Embargoes on Russia didn't mean much so Europeans are physically taking over their ships and Ukrainians are physically sinking the rest of their ships. In Iran nothing other than physically sinking ships and blowing up places meant anything.
Europeans can ship EUV machines because they are physically building them for people who will use these to physically build the most valuable products currently there is. US wasn't able to enforce its will to Iran, what if Koreans, Europeans and the Chinese decide that its not into their interest to act according to US courts?
The collapse of global trade would greatly reduce economic efficiency, output, and investment. It has been coming for while, though greatly accelerated by the orange pdf file. It takes a lot longer to build systems of trust and belief in enforcements of global order than to disrupt them. I suppose we'll move closer to the fear side of the financial/political axis from the greed side.
If history is any guidelines that would be how World War III starts.
There is nothing that stops US from building their own Memory Fabs, or asking / funding Micron building more US Fabs. It will cost a more, but the complexity is certainly no where near replicating TSMC.
US is in a very advantageous position regarding geography and resources but its problem is that its geared towards having access to the whole worlds markets. Apple, Google etc. are all possible because they server billions of people, not just 350M. IMHO US will have serious internal trouble for years, eventually stabilizing and being a nice place again.
Steve Jobs’s dad, his biological dad, is Syrian, and he may still be alive because he outlived his son lived in northern Nevada, Steve’s circumstances somewhat similar to Barack Obama, where two college students in the fifties early sixties, who were unmarried had an unexpected pregnancy.
Note Steve’s biological mom and biological dad, the Syrian, kept their second child, a girl, and Steve even met her later on in life.
Then Samsung, Hyundai, LG, Kia would go bye-bye in the United States. I think the current memory fiasco will be their last big payday, so they should enjoy it while they can. There will be many companies that will rethink how they approach memory going into the future.
The market is the AI boom and the US is the host, they can sell the exact same stuff to someone else. What are the capitalists who fund the AI build up do? Invest in SaaS when they can't buy chips? I bet if something like that happens the chip manufacturers wouldn't end up with product they have no one to sell to.
> The market is the AI boom and the US is the host, they can sell the exact same stuff to someone else.
In theory yeah, but not at the levels that the US companies are buying.. Last I heard OpenAI is sitting on 50% of the world useless wafers on their own for some reason, and Microsoft cant find a datacenter fast enough to plug their stock in.
I'm a customer, looking for a simple 32GB ddr5 6000 ram, I'm now competing with Microsoft because there isn't enough supply? I'm not sure about that.
So in other words, if Koreans and Europeans decide not to sell their stuff to America,quits the AI race and the capitalist do something else instead? I don't think so, in the hypothetical world where Korea and Europe don't sell to US, the American money that is invested in AI will go wherever they can actually build it, the people who are using these machines to build those models are mostly immigrants anyway.
If Koreans/Europeans decide not to sell their stuff, some of the money would go into circumventing the ban:
1) This is not good for Korean/Europen sellers, because it negatively affects sales volume, and is unlikely to be compensated by margins, because a good chunk of those will go towards circumvention instead of the original seller.
2) Some more money will go towards replacing those sellers completely. This is extra not good from the sellers perspective.
In this new world there are more important things like nationalism, the ego of the politicians and the personal ambitions of the super rich. Bot being good idea is irrelevant.
I don't understand what this hypothetical world is supposed to be. Samsung and SK Hynix are run by Korean capitalists who want to sell their stuff to America; the reasons you're describing are precisely why they would not exit the American market just to dodge an inconvenient antitrust investigation.
memory is a commodity is laughable. Then software engineering is even more a commodity, the amount of engineering going into making memory chips the vast majority of people don't understand. There are a lot of software engineers getting this field after leetcoding and copy from hellointerview. Claude can write you an app in 30 minutes. Try build a lpddr5 dram chip in 30 minutes. Manufacturing know how itself is a specialty and barrier to entry.
It’s a barrier, but not an insurmountable barrier. Most of the companies are somewhat lazy and somewhat cheap, but if circumstances change in the market, that make it untenable, they will take on making their own memory or making arrangements with someone else.
It’s been done with processors, modems, SSDs and many other specialized chips. The design and engineering part can already be done at several companies, they just elect not to do anything on the fab side. But they will, if they have to in the long run.
It takes two to four years and Five to twenty billion dollars. Some of the bigger companies, if pushed, particularly, if the alternative is getting chips from China, will do something on the fab side if they absolutely have to. They probably won’t live with the current conditions forever.
There was a company that had three companies say no over about 10 years, eventually, they made a decision to build their own processor known as the M-series…
I mean your view isn't flawless but overall I agree. Too many people think building things amount to spending money and completely overlook the thousands of people required who are not just unskilled labor hired off the street.
I also feel like there should be FTC or other antitrust actions against OpenAI and other hyperscalers, with nvidia being complicit, if they are cornering the market that badly on consumer RAM, SSD and other components, especially if these volume purchases are for projected datacenter that haven't even broken ground (or have been paid for) for many months.
Furthermore, I think there should be a tax on algorithmic inefficiency, in that if a LLM, frontier or not, consumes more than a certain amount of KWH per token, it should be taxed such as to put emphasis on models than can run locally, on a normal PC
Honestly, the knock on effects of this cartel behavior should concern all countries, and be remediated with expediency.
From consumer electronics to the data-center, the rising real mfg costs and lack of supply is putting huge pressure on pricing, and may just drive anyone who cant negotiate with these suppliers out of business.
Once the dominoes start, I fail to see how things recover in less than 3-5 years, not counting all the businesses wiped out in the meanwhile.
Just a reminder that anyone can file a lawsuit over anything, and the initial complaint is written by lawyers and reads with tabloid levels of sensationalism and allegation. The goal being to maximize the appearance of harm as much as possible so the suit has the greatest chance of sticking.
It is not in any way, shape, or form a ruling much less even a piece of well researched work. It's "my side of the story that makes me look perfect, with lawyers turning the heat up to 11"
In reality a lawsuit needs to have some basis. The bar might be low but judges frown on having their time wasted with a truly frivolous claim or a claim with no clear damages or harm shown.
Honestly a lot of lawsuits feel like they just want the other party to pay them (settle) to make the lawsuit go away, even if there is no case; what are the repercussions for a frivolous lawsuit?
Realistically very few. The plaintiff carries the burden of proving the claim, but they are afforded the power of the law to do so via discovery. The defendant then has to spend a large number of legal resources refusing the claims and also providing the evidence required.
If the plaintiff loses the lawsuit a countersuit is pretty unlikely to succeed unless the lawyer participates in gross misconduct. Generally, countersuits are filed more to put the original plaintiff on the defense and don't result in a large judgement.
If you are operating in good faith, then you are pretty insular from kickback as a plaintiff
>what are the repercussions for a frivolous lawsuit?
None, hence the high price of liability baked into basically everything in America. And not just in nominal prices, but in terms of things like restricting access to spaces, restricting access to information, etc.
Maybe software people can get around this issue by not making every app a electron bloat? This is now more than doable now you got AI right? And it will save your job
No, they suck. No swiping, no pull-to-refresh, no multitouch zoom, the calendars dropdowns are horrible to use, the buttons look like something out of Windows 3.1
,,The plaintiffs claimed the three companies reduced D-RAM supply under the pretext of transitioning to high-bandwidth memory (HBM). "The D-RAM oligopoly companies systematically coordinated the shift to HBM and the discontinuation of DDR3 and DDR4," they said. They added that Apple's recent sweeping product price increases were the trigger for the lawsuit.''
How can they do price fixing and discontinuing a product at the same time?
It just looks like some companies are angry that AI / VC industry is outpricing them.
> An agreement to restrict production, sales, or output is just as illegal as direct price fixing, because reducing the supply of a product or service drives up its price. For example, the FTC challenged an agreement among competing oil importers to restrict the supply of lubricants by refusing to import or sell those products in Puerto Rico. The competitors were seeking to pressure the legislature to repeal an environmental deposit fee on lubricants, and warned of lubricant shortages and higher prices.
People have a tendency to get upset when someone waves a future IOU intent order from another buyer in front of you, one that isn’t taking delivery anytime soon and then proceeds to tell you you must pay more…
Antitrust and anticompetitive behavior continue to evolve. Back when the Sherman Act was passed, we were talking about backroom deals in smoke-filled rooms. One of the more important ways of recent years is where all or most competitors in a market use the same software that outputs the same value as to what they should charge. This is effective collusion even if it isn't explicit. The goal of such software is to raise prices. You raise prices by effectively or actually colluding with other market participants.
The posterchild for this is RealPage [1]. But you're going to see this pop up in every aspect of life, such as gas prices [2] and meat processing [3].
The whole thing is kind of depressing because this is what "innovation" is now: fancy ways to collude on prices.
It used to be possible to bust up a company based (more or less) simply on their having too much of the market.
Right wing interests built up think-tanks after the Second World War, having seen the success of especially the Brookings Institution during the New Deal and world war eras in pushing technocratic-liberalism and wanting to fight fire with fire by creating their own credibility-laundering offices for propaganda advancing the desires of the rich. This effort bore its first great fruit in the mid '70s as Chicago School economists and lobbyists successfully overthrew that enforcement regime and replaced it with one in which specific harm must be proven to win an antitrust case, which de facto ended meaningful anti-trust enforcement in the US and led to the consolidation of money, power, and media over the following decades.
Aaaand... here we are, in the endgame of that movement.
Looking forward to getting my $10 settlement 15 years from now like what happened with the DVD drive price fixing lawsuit (paid as a digital Visa gift card code so it can't actually be spent anywhere).
dumb peripheral question: memory as a circuit would seems easy to add directly on-chip for a company like Apple. Is that blocked by time-frame, IP, technology, cost?
Logic gates and memory bits have very different fabrication processes (mostly because DRAM is optimized for a high density of big capacitors doing the storage).
You can put some memory on the logic wafer (SRAM) but it's area inefficient, which is wasteful on your expensive N2 wafer. So a dedicated DRAM process is vastly cheaper per bit, even at current elevated prices.
My guess is area becomes problematic. Chip becomes bigger, harder to manufacture without defects, costs rise as a result of those defects.
Bonus though, Apple is using "defective" chips in other products now such as the Neo which will help some with costs, but overall, you can't just add memory to the chip because the chip can become very yield sensitive, the process has to be there to produce the yield effectively.
I mean there was an attempt before, and maybe I'm wrong but I'm sure they were fined in 2010 at least in the EU, this seems to be round 2 of their 'Memory Cartel' after learning some lessons, realising all they need to do is keep supply as low as they want and allow the AI companies to spend hand over fist.
Oh, maybe we're mixing terms here. OP was proposing CXMT as a "solution" to the problem of these companies engaging in price fixing. Now you're saying there needs to be a "solution" to the problem of these companies no longer being able to "compete" (engage in price fixing), because CXMT exists.
The latter is not a problem I or almost anybody outside those companies care to fix. It's not the public's job to bail out a company because they forced states to intervene to break up their monopoly after taking the highest profits in the industry's history.
cxmt is also selling their memory similar to the big three. No one hardware company in their right mind will sell their products not as high as possible after you learn how much harder hardware engineer and fab people work for.
"As high as possible" is a function of scale though. There is only so much global demand for memory at a given price point. If CXMT determine that by increasing supply they can increase profit even if it lowers prices, then they will do so within their ability.
EDIT: This is also semiconductor-specific. Because there's little additional cost to doubling the output, ramping up production is the hard part
There is no reason for there not to be a price fixing. But the OpenAI's announcement from 2025-10-01 about buying 40% of supply, removes any need for collusion. It was a public signal for everyone to rise prices, one that each company could figure on their own. And it will be very hard to prove otherwise.
Two things can be true; they could theoretically make agreements to limit supply, like the OPEC does to exert control over international oil prices. But the OPEC is all above board and there's plenty of international competition on the oil market.
Prices skyrocketing create a perfect opportunity to slip in a little artificial bump and hope everyone blames the market. See also: egg prices in 2024.
if demand is too high, the market will adjust by adding supply. If Apple can't take the price of memory, why don't they make memory themselves. Oh no, the technical and manufacturing know-how is a barrier to entry and there is no talent available to make it. That is memory or semi companies' competitive advantages and their pricing power. It also takes Phd degrees to work at hardware companies. You can't have people doing leetcode for 2 months and todo apps and getting into the field and make 300k a year and hardware engineers wih Phds making 100k before this boom. It is long overdue for a rebalance of pricing power between hardware and software companies
Between TSMC, Intel and Apple memory can be designed, engineered and made, and that doesn’t even include the Chinese who will probably use this to take over a large part of the memory market worldwide?, the three-headed memory cartel is unsustainable, they have two to four years maximum and that’s it.
Price fixing is legal as long as your are doing it in the open. In the UK it is called "price match" and eg. if supermarket says they keep prices matched to their competitor. No regulator raises an eyebrow.
So here Samsung and SK Hynix could say they price match to Micron and they are in the clear.
> Price fixing is legal as long as your are doing it in the open.
In the U.S., competitors are allowed to act in similar ways in response to economic realities, as long as they each arrive at that decision independently. But publicly anchoring your price to a competitor’s is potentially illegal.
> Price fixing is an agreement (written, verbal, orinferred from conduct) among competitors to raise, lower, maintain, or stabilize prices or price levels.
Identical prices are often a sign of intense competition. Every gas station on a corner has the same price because it's a highly competitive market not because of collusion. The prices of the much more lucrative chocolate bars inside the gas stations are less likely to be identical.
Too many consumers think that there are enough differences in additives and detergents between the brands for it to make a significant different in mpg and in maintenance costs, so buy from "their" brand even if there is a cheaper brand nearby.
At one time those consumers would have been right, but first the EPA established certain minimal standards that all brands had to meet. The better brands were still better for your car than the ones just meeting the minimum, but it would take longer for it to make a significant difference.
Then BMW, GM, Honda, and Toyota made a spec for what makes a good detergents and additive mix, and most other major automakers joined. They created a certification, "TOP TIER", that gas companies that met that standard could use.
Nowadays most gas in the US is certified TOP TIER. Generally you will only find non-TOP TIER gas at stations not part of a major chain, at some gas brands owned by grocery chains (Safeway for example), and at some gas brands owned by convenience stores (7-Eleven for example).
Most independent tests have found that and detergents and additives beyond those required for TOP TIER certification do not make a significant difference, even cumulative over 300+k miles.
Nowadays then in the US just buy from any TOP TIER brand and that will be good enough. But consumers don't seem to know that and so for a large number of consumers different brands on the same corner aren't actually in competition.
This is a good signal of how competitive your gas market is. Your examples are Seattle and the valley, which are very difficult markets for gas stations, so unsurprisingly are uncompetitive.
The whole US gas station market has gotten less competitive over the last 40 years I've been observing prices and big coastal cities are a lot worse than anywhere else.
Yes, the gas station example is directly cited in the article I linked to. It’s legal for a gas station owner, with knowledge and consideration of a competitor’s price, to reduce their price to the same or just below. What is illegal is for nominally-competing gas station owners in an area to conspire to keep their prices within a range of each other’s, even without explicit agreement.
The article you linked seems to indicate that there has to be active communication between the gas station owners for there to be collusion.
When I worked at a gas station as a teenager there was definitely an unspoken implicit agreement that the price of gas would be 6 cents/liter above wholesale IIRC. Which was highly competitive and didn't completely cover costs.
"Every gas station on a corner has the same price because it's a highly competitive market not because of collusion."
Huh? I can go to most any gas station-occupied intersection and you will always find two that match and one (usually a Persian-owned Chevron) which is consistently a dollar or more higher per gallon across all grades of fuel.
>Price fixing is legal as long as your are doing it in the open. In the UK it is called "price match" and eg. if supermarket says they keep prices matched to their competitor. No regulator raises an eyebrow.
No, the key term is "collusion", which could be done in the open or not. If a competitor told you they were unilaterally raising prices in secret, that would still be legal. Where you get into trouble is if you are cooperating to set prices. And no, this is all determined by a judge so cute workarounds like "I'm telling my competitors that I'm raising prices then gauging his body language" won't work.
I worked in a small town gas station as a kid. We always phoned our competitor to let them know when we were changing our prices, and they extended us the same courtesy. Usually we followed them, but sometimes we didn't.
Gas prices are posted on massive highly visible signs and are public information. This wasn't collusion, it was a sign of intense but friendly competition.
That is not what price fixing actually is. In the UK "price matching" is a one-to-many relationship meaning that the price of goods is set to the "lowest available"
Price fixing is a many-to-one all the manufactures agree to the highest prices they all agree on and set it there.
That is your assumption without evidence. Once the targeted market knows others are tracking the price, it can up the price without fear competing market will undercut.
This was attempted before in 2022 but fell apart because plaintiff couldn't show an agreement took place https://cdn.ca9.uscourts.gov/datastore/opinions/2022/03/07/2...
The arguments made by the Plaintiff are thoroughly convincing to me. The fact that those 8 points are not enough to convict indicate the industry is fucked. Of course the defendants didn't leave a paper trail - they've already been convicted of collusion before.
It's the people and country that suffer when our government fails to ensure markets are free and fair.
If there is no agreement, then the magic term is "tacit collusion."
Why not ask the same ridiculous amount of money your competitors do? People seem to be paying for it. Their fault. If suppliers have sufficiently different products, they can make some more expensive, others cheaper; on average, everybody pays more. A high barrier to entry might help such practices.
That doesn't mean I'm saying this is what is happening. Sometimes things just suck, and somebody bought the world's supply of RAM wafers to use as frisbees.
https://en.wikipedia.org/wiki/Tacit_collusion
Yeah your qualifier is important. If I have zyngots and suddenly the price goes through the roof, it is not collusion or illegal for me to raise my prices.
Tacit collusion requires at least signaling. There has to be some intent. Like in the US rental market where Realpage coordinates tacit collusion.
But if there is legitimately demand for far more memory than can be produced, it seems silly to claim collusion to raise prices.
> Why not ask the same ridiculous amount of money your competitors do? People seem to be paying for it. Their fault.
That's not how commodity markets work. This stuff is essentially sold at auction with the price set by supply and demand. The way they would fix the price is by constraining supply so that people have to outbid each other on a smaller amount of inventory.
But that's not that hard to measure -- are they producing less than they were before prices went up? The answer is actually that they're producing more. The reason prices went up anyway is the huge increase in demand.
You would then have to make the case that it's not just that they're reducing supply but that they're not increasing it fast enough. That's theoretically possible but it's also very plausible that building new fabs just takes time, so if someone's theory is that they're colluding then they need to present some evidence.
> but that they're not increasing it fast enough
That also isn't illegal. A business can choose to artificially restrict supply if they want, there's no mandate that they must meet demand.
It only crosses into illegal territory if multiple companies get together and secretly agree to cap production to keep prices high. Then it becomes collusion. It also becomes illegal if there's a monopoly power that is intentionally constricting supply to specifically stop a smaller competitor or lock them out of the market.
The hard part is how do you prove Samsung, SK Hynix, Micron are acting as a unified cartel when there obviously isn't going to be a paper trail for secret meetings.
> A business can choose to artificially restrict supply if they want, there's no mandate that they must meet demand.
There is, however, no incentive to do this when prices are high unless you expect competitors to do likewise, since otherwise you're just handing the business to the others when they increase production. Which strongly implies that if that's what everybody is doing, they're colluding.
> The hard part is how do you prove Samsung, SK Hynix, Micron are acting as a unified cartel when there obviously isn't going to be a paper trail for secret meetings.
Which is the problem when that's what's happening, and why we should maybe change the law to infer the collusion from the outcome in cases where prices are high yet nobody is responding by increasing their market share.
> There is, however, no incentive to do this when prices are high unless you expect competitors to do likewise
This imagines a world where water production can be dialed up and down instantly, at no cost, with no risk.
It doesn’t work that way though. Hiring people, scaling up factories, converting to 24/7 all have costs and risks.
Just google (or Claude, or whatever) memory companies that went backrupt overbuilding to meet a spike in demand. There are many. You probably don’t even know their names. Because Micron/etc stood pat and survived.
> This imagines a world where water production can be dialed up and down instantly, at no cost, with no risk.
It only images a world in which the profit-maximizing amount of wafer production is higher when demand is higher, which is a completely reasonable premise to take as the default.
> Just google (or Claude, or whatever) memory companies that went backrupt overbuilding to meet a spike in demand. There are many. You probably don’t even know their names. Because Micron/etc stood pat and survived.
Those long forgotten companies like Intel, Toshiba, Texas Instruments, IBM, etc. that are all now bankrupt and no longer exist. Except that they still do exist, they just sold their DRAM divisions to the current incumbents. Even the divisions that went bankrupt -- for Qimonda (Infineon) it was the 2008 housing crash that did them in.
Predicting events like that isn't reasonably possible. What Micron actually did to survive it was keep a large reserve of cash to survive the lean years, not refuse to build fabs. The South Korean companies did the equivalent by being subsidiares of giant conglomerates.
I mean, if they're dumb enough to try to game the system blatantly - and there are plenty of execs that are - they're dumb enough to write down their super secret criminal collusion plans in a group text, or a lengthy email chain. That level of stupidity among white collar criminals is shockingly common.
If there's no evidence to be found, that might be a good indicator that either A.) nothing illegal occured or B.) the senior leadership involved are competent, intelligent, and discrete
Given the number of companies involved, all it takes is one idiot, so if it's collusion, I expect we'll see evidence. If not, number go up.
Isn't it difficult to win on tacit collusion if you can't find some kind of document or email to show they were deliberately taking advantage?
We have to decide what part is damaging to society: the actual physical agreement, or the effects of the agreement?
If it's the actual physical agreement that's the problem - the system is working as intended.
But if we are looking to prevent the negative outcomes associated with price fixing and collusion, our system is failing us.
They are never going to find proof of conspiracy. The people involved covered their tracks, and doing so is trivial. So the best we can do is punish the appearance of collusion. And if the goal is to actually prevent harm to customers, that's a better solution anyway, since it encourages leaders of companies to behave in a manner that's the opposite of collusion.
Not directly addressing your points, but I fail to see how this whole situation can be can be viewed as cartel price fixing. There is no oversupply. The strange thing is that we are at the mercy of 3 corps.
You'll have to create a case that harm is taking place. Harm does not mean a PlayStation 5 is now $200 more expensive or that inflation exists.
I would look at questions regarding what harm is created: Is it discriminatory? Are parts of society shutting down, and is that unreasonable? Are groups of people now unable to afford a living? Does it move the poverty line? Is that permanent? And how do you prove this is exclusively due to the price increase of tech components, and RAM specifically?
It needs to be unfuzzy in some way in order to make sense, but that's just my opinion.
I do agree prices are insane and wish for them to come down today. I liked the ubiquitous amounts of RAM any system could have. In those days, forums were also filled with how insanely expensive 32 gigabytes of RAM was, about $100 :)
> You'll have to create a case that harm is taking place. Harm does not mean a PlayStation 5 is now $200 more expensive or that inflation exists.
"Things cost more because of collusion" is always a harm. It doesn't matter if the product is maize or gold-plated haute couture, competitors are supposed to compete.
The question is, what's the best way to tell the difference between tacit collusion and just normal supply and demand?
It's not that easy, but here's a decent test: It's tacit conclusion if 1) net margins have been high for e.g. 3 years and 2) no new companies have entered the market in that period of time, or were acquired by an incumbent if they did.
Notice that this works for everything. Even if you're making luxury goods, the price may be high, but so are production costs, and there is a lower volume to amortize fixed costs over, so long-term net margins should be the same as they are anywhere else or you should see new entrants. If you don't, it's reasonable to infer collusion and leave it on the companies to prove otherwise.
I see the logic in your reply, thanks :)
>Harm does not mean a PlayStation 5 is now $200 more expensive or that inflation exists.
Yes it does lmao. If consumers can't get access to devices then they cannot be used for work or education. It's counterproductive.
How many "attention is all you need" papers aren't being written because as soon as there's a sniff of money the rabid suit and tie MBA fucks leap onto any opportunity like a dog in heat and fuck it to death.
It wouldn't be a first: https://en.wikipedia.org/wiki/DRAM_price_fixing_scandal
I thought the title missed the year (in parenthesis)
It's been happening once every decade or so. I'm not surprised that they are being sued again.
Total fines of about half a billion. Thats not nothing but it also does not seem like a real disincentive.
Samsung is about to hand out ~$26B in bonuses. SK Hynix something similar-ish.
Yes, but for context Samsung is a massive chaebol. You need to compare it to the business unit in question.
It’s different to, say, Google’s vertical monopoly in advertising where that’s most of their revenue.
This! Samsung is one of the world's biggest ship maker for instance.
HBM is also DRAM. I also think it's kind of a weak argument to say that them discontinuing ddr3 (which while in use still today in industrial/embedded was on the way out for consumers 10 years ago) and ddr4 which last had consumer CPUs for it 3 years ago is meaningful. What we need now is ddr5. Turning off the old fabs and moving those resources including people to ddr5 is a good thing. That's not price fixing. It's possible price fixing is in play, but discontinuing products people objectively don't use as much anymore isn't it.
Switching off DDR3 manufacturing I can understand, but DDR4 machines are still quite relevant and usable… Ryzen 5000 series boxes for example don’t feel meaningfully weaker than they did when new. My 5950X tower certainly doesn’t, and it’d really be nice to be able to upgrade its RAM should I need to because it will continue to be useful for quite some time.
AMD just re-released their 5800X3D for AM4 board users who wish to upgrade which is further evidence that shutting off DDR4 production is premature.
They're running a business not a charity. Their job is to manufacture what the market as a whole demands. If they can make more money making HBM than DDR4 then they have to make HBM. Why would a business go out of its way to make less money?
1. The AI bubble is an insane distortion and the gravy train isn’t going to last forever. Betting the farm on everlasting datacenter demand is myopic.
2. In a healthy, competitive market there would be smaller manufacturers that’d be happy to take up the big guys’ discarded business.
1. But it might last for at least few more years, see Nvidia 1 trillion backlog.
2. Semiconductor manufacturing is the most complex industrial process in the world. You need billions of capex and decades of experience. Even existing semi players like intel cannot switch production to memory.
China CXMT is gaing traction in DDR market. New fabs from all players wil come online in the next two years.
>the gravy train isn’t going to last forever.
Have you seen how the modern stock markets works lmao? It hasn't been based on reality in a long time.
Hell just look at Trump, should've run outta money from all his bad deals ages ago but the grift continues.
It's true that markets have no need to be rational, but everything has an end and it's usually sooner than anybody ever expects.
Deleted
Not sure I understand this statement. Apple is affected by DRAM price fixing the same as any other PC manufacturer.
Ahh you are right, not sure where I got that from…
That's incorrect. Apple purchases RAM from all of these providers to produce their unified memory. They also rely on TSMC fabs for all of the chips their memory relies on. If they haven't doubled the prices of their machines, that just goes to show how fat of a profit margin they take on everyone that buys from them...
I stand corrected, thank you!
You're most welcome! I'll admit I did search before answering because you said Apple produced their memories so confidently LOL
I guess they would love to produce it themselves, but for the average scenario the production reserves they have with Samsung already work well enough and prevent them from having to get into such a complicated industry.
Apple just increased their prices a few days ago, citing high memory costs.
I am an idiot, sorry for wasting your time with confident bullshit.
DDR4 machines are relevant and usable, but it's pretty unusual that new systems on older ram are still big sellers. The 5800x3d was originally released in 2022, AM5 processors were also originally released in 2022. We'll probably see Zen6 soon, the third generation on AM5... I don't think AMD released new AM3 processors after the introduction of AM4, certainly not when Zen3 was coming soon.
When making long term plans in 2022, I don't think anyone expected DDR4 to need significant production in 2026. Since ram makers can pretty much sell whatever they make in today's marketplace, it makes sense (for those fabs that can) to stop making DDR4 and repurpose those fabs to make newer generation ram.
> My 5950X tower certainly doesn’t
That's what I have in my gaming tower, and yeah I feel zero pressing need to upgrade. I did manage to put 64GB of DDR4 in it just before prices went totally bonkers, thankfully. Where I'm falling behind is my GPU I'm still on an nvidia 1660 super, but I just can't justify paying what they cost right now.
I would gain pretty much nothing moving to a newer board w/ DDR5.
Exactly what I'm feeling. The CPU is great for what I'm using it for and I have a killer motherboard. Have a nice GPU capable of current gen games, too. The equivalent current gen board and CPU cost a small fortune for marginal readily visible gain.
Unfortunately, I opted for only 32GB of RAM because at that point more felt like overkill, which as a decision has aged poorly. I should've gotten more while it was still cheap.
Not to be a bootlicker but AMD releasing a product doesn't mean another company should make more DDR4. That's not price fixing. In the embedded space it's sadly very common for a part to be compatible with a very low number of options (shout out to cellular/admux RAM on the STM32H745 nucleo). That's just the way the cookie crumbles.
Dropping DDR-4 is anything but meaningful. It'll easily last 10 more years, machines from this gen are still much more affordable and quite powerful. In fact for most dev and gaming workflows the difference between the DDR-4 and DDR-5 generation of hardware is more or less negligible. I am exaggerating a bit -- but really, not too much.
Of course it might be a ploy to sheep-herd consumers and companies towards the expensive DDR-5. I would not put that below the ring of RAM producers.
>Dropping DDR-4 is anything but meaningful. It'll easily last 10 more years, machines from this gen are still much more affordable and quite powerful. In fact for most dev and gaming workflows the difference between the DDR-4 and DDR-5 generation of hardware is more or less negligible. I am exaggerating a bit -- but really, not too much.
How much % of the DRAM market do you think is made from computer enthusiasts upgrading their Zen 1/2 CPUs to Zen 3? Note intel and AMD both switched to DDR5 well before the exit from DDR3/DDR4 ("2024-2025", according to the complaint).
Note though that for Intel, the first gen of DDR5 CPUs also supported DDR4, and many buyers bought the DDR4 versions of their boards because at that point DDR5 RAM was much more expensive for gains that were marginal at best, which effectively makes Intel’s following generation of DDR5 CPUs the actual transition point.
The mere "enthusiast" word in your question suggests the percent is not too big. But I am not sure I get your point -- elaborate, please?
The point is that just because there's a handful of people (relatively speaking) looking to buy more RAM to upgrade their last gen systems, doesn't mean there's robust demand for DRAM manufacturers to keep DDR4 manufacturing lines going. It's like arguing Sony shouldn't have exited the CRT business because there's retro enthusiasts on youtube scouring the earth for CRT monitors.
For the reasons stated above -- that DDR5-gen hardware is expensive right now -- I'd think the DDR4-gen market will remain alive for quite a big longer. Though that's likely much more on the second-hand market side of things.
While I wouldn't necessarily agree with "a handful of people", the fact is that neither of us can prove their lean -- so no point pursuing that argument thread.
So you might be right that it's a pure numbers/statistics decision. Or I might be right that they want to herd people into the more expensive hardware while forcing them to do so by phasing out production of the cheaper hardware.
No way to truly know IMO. We are exchanging hypotheses.
I’m not sure that comparison makes much sense. By the time CRTs were phased out, demand was down to almost nothing and what little existed was confined to the extreme budget market. While I don’t have industry insights or anything I don’t think demand for DDR4 is anywhere near the bottom yet, and the remaining demand is centered on premium product (nobody running cheap DDR4 is upgrading). In a more normal market would be more than enough to justify continued production for several more years.
DDR4 production is likely still quite profitable, just not drowning-in-money AI-bubble profitable. If smaller foundries existed they’d be happy to take up the business.
Maybe really what needs to happen is some busting up of the giants…
There is a minimum amount of production volume for it to fit the price equation. If the market doesn't have that demand, it is fundamentally no different to CRT.
Otherwise they could continue to make DDR4 at a higher cost and sold at a higher price to which people will complain price fixing again.
> objectively don't use as much anymore
Consumers use these every single day in embedded devices without knowing it.
I wouldn't be surprised in the slightest if the embedded DDR3/DDR4 market greatly exceeds the number of consumer desktop computing devices in terms of "devices with memory" (not in sheer IC count or nominal size though.)
The level of design effort and PCB expense to go from DDR3 to DDR5 is enormous.
I was hoping it was a typo, and they actually meant 4 and 5... I didn't even know people made products that still use ddr3.
Especially industrially, ddr3l is just fine. At a certain point you don't need the speed of newer generations, and buying new idk lpddr5x/t controllers for 30 year old process nodes just isn't worth it. Until ddr3/l to EOL from everybody. The big 3 aren't the only memory fabs.
It's easy to see why your argument is wrong with a simple hypothetical: what if they were still making DDR4 today? Would people still buy it?
The answer is an obvious "fuck yeah", even if you ignore the DDR5 price gouging. People will buy it because people still have DDR4 hardware, and that hardware is still extremely relevant.
So if there's a market for it, but none of the suppliers are trying to sell to it... Wtf is happening? Basic capitalism logic says any rational supplier would sell DDR4 for easy profits, meeting an unmet demand. That it isn't happen points to some kind of collusion, IMO.
Because the market pays less for DDR4 than for HBM (or DDR5), and since HBM is heavily modified, vertically stacked DRAM, it competes for the same raw inputs and fab space than DDR4 used.
If I can produce DDR4 for modest profit or HBM for a lot more profit I will obviously produce HBM. And given physical realities producing HBM takes from existing DDR4 production capacity. Worse still, it takes roughly 3GB of ram to produce 1GB of hbm iirc.
> People will buy it because people still have DDR4 hardware
The question is whether there’s enough meaningful demand for aftermarket DDR4 upgrades to make it worthwhile to a manufacturer to keep producing DDR4 instead of switching to HBM and DDR5.
Micron claimed retail is a rounding error, a market not worth serving. So you’d presumably need to find industrial buyers who would be willing to buy DDR4.
Basic capitalism logic is that if you think it's stupid, you put your money where your mouth is, set up a DRAM fab, and get rich.
I'll do it. Will you give me the seed money?
No, basic capitalism logic is that you already have enough money because barriers to competition are low.
lol :D
Everybody in our industry loves fat margins. But god forbid if someone else captures the margins and squeeze them out of easy profits.
A fat margin is fine. What’s not fine is saying you don’t have any and you don’t plan to make any more.
There's "fat margins" and then there's the kind of margins that tech industry shareholders expect.
OpenAI's original corporate agreements capped its returns at 100x, which is seen as too paltry for its current holders, so they scrapped those to prepare for an IPO [0]
That is, in a word, insane.
[0] https://abhs.in/blog/openai-for-profit-conversion-ipo-develo...
>But god forbid if someone else captures the margins and squeeze them out of easy profits.
Yeah that's how law works. Everyone likes money, but that doesn't mean it's fine to steal money. Yes, even from maligned entities like "big tech" or "private equity"
What happens if Samsung and SK Hynix simply stop selling to US at all? Micron is in US but are the rest still in the US jurisdiction?
They are selling the hottest commodity of the day. It’s made outside of the US using non-American tooling.
Micron is forced to stop under-investing in plants and will increase production. This will trigger everyone else to expand production and lower prices.
The whole point of the collusion is to ensure everyone is producing the same volumes and keeping prices high. The company that expands is the company that "wins" because memory is a volume game and it's all about hanging on the longest during the glut. So once one company expands, the rest have a choice of expanding or planning their exit.
If Samsung and SK lose access to the US market, they'd be fucked long term. Micron would kill them selling at higher margins and higher volumes in the USDM, while the rest are stuck competing for the international scraps - markets Micron is also allowed to compete in, if they wanted to.
Oddly enough, Samsung and SK Hynix decided that they needed to expand production on the same day another lawsuit accused them of colluding to keep production low and prices high.
> South Korea announces $520bn chip plant project with Samsung, SK Hynix
https://asia.nikkei.com/business/tech/semiconductors/south-k...
>It’s made outside of the US using non-American tooling.
Depends if US can demand ASML which uses plenty of US tech inside. In reality even the DRAM and NAND supply chain has plenty of US technologies.
And you say Micron are US but they have lots of Fabs in Japan as well since they acquired Elpida.
Everyone is using something from someone, you can even argue that US owes India and Europe huge compensation because pretty much everything US did in the last half century was made using technology or people funded by those people. Johny Ive is British, Almost all the AI stuff is created by Europeans, Israelis, and Canadians - thus funded by their respective taxpayers.
The thing about the US losing its grip on the world and the collapse of the global world order means that the words on the paper don't mean much. Embargoes on Russia didn't mean much so Europeans are physically taking over their ships and Ukrainians are physically sinking the rest of their ships. In Iran nothing other than physically sinking ships and blowing up places meant anything.
Europeans can ship EUV machines because they are physically building them for people who will use these to physically build the most valuable products currently there is. US wasn't able to enforce its will to Iran, what if Koreans, Europeans and the Chinese decide that its not into their interest to act according to US courts?
The collapse of global trade would greatly reduce economic efficiency, output, and investment. It has been coming for while, though greatly accelerated by the orange pdf file. It takes a lot longer to build systems of trust and belief in enforcements of global order than to disrupt them. I suppose we'll move closer to the fear side of the financial/political axis from the greed side.
If history is any guidelines that would be how World War III starts.
There is nothing that stops US from building their own Memory Fabs, or asking / funding Micron building more US Fabs. It will cost a more, but the complexity is certainly no where near replicating TSMC.
US is in a very advantageous position regarding geography and resources but its problem is that its geared towards having access to the whole worlds markets. Apple, Google etc. are all possible because they server billions of people, not just 350M. IMHO US will have serious internal trouble for years, eventually stabilizing and being a nice place again.
> Steve Jobs is Syrian
lol he is not. at no point was he a syrian. his mom was from ohio or something.
Steve Jobs’s dad, his biological dad, is Syrian, and he may still be alive because he outlived his son lived in northern Nevada, Steve’s circumstances somewhat similar to Barack Obama, where two college students in the fifties early sixties, who were unmarried had an unexpected pregnancy.
Note Steve’s biological mom and biological dad, the Syrian, kept their second child, a girl, and Steve even met her later on in life.
You are going to have to cite more for those claims. Steve Jobs was born in San Francisco. He is as Syrian as Trump is.
This is as biased an Anti-US take as any. Will not grace the rest of the claims with a response.
Then Samsung, Hyundai, LG, Kia would go bye-bye in the United States. I think the current memory fiasco will be their last big payday, so they should enjoy it while they can. There will be many companies that will rethink how they approach memory going into the future.
> What happens if Samsung and SK Hynix simply stop selling to US at all? Micron is in US but are the rest still in the US jurisdiction?
They would lose access to their largest market, I'm sure shareholders would havesomething to say about that ?
The market is the AI boom and the US is the host, they can sell the exact same stuff to someone else. What are the capitalists who fund the AI build up do? Invest in SaaS when they can't buy chips? I bet if something like that happens the chip manufacturers wouldn't end up with product they have no one to sell to.
> The market is the AI boom and the US is the host, they can sell the exact same stuff to someone else.
In theory yeah, but not at the levels that the US companies are buying.. Last I heard OpenAI is sitting on 50% of the world useless wafers on their own for some reason, and Microsoft cant find a datacenter fast enough to plug their stock in.
I'm a customer, looking for a simple 32GB ddr5 6000 ram, I'm now competing with Microsoft because there isn't enough supply? I'm not sure about that.
US is the vast majority of their market - Apple, hyper-scalers, AI labs
Why can't they just buy the exact same product and install it in Kazakhstan or somewhere else?
The RAM buyers have no interest in entertaining something like that because their revenue comes mainly from the US, too.
Corporations avoid picking fights with large nations where lots of revenue comes from for very obvious reasons.
So in other words, if Koreans and Europeans decide not to sell their stuff to America,quits the AI race and the capitalist do something else instead? I don't think so, in the hypothetical world where Korea and Europe don't sell to US, the American money that is invested in AI will go wherever they can actually build it, the people who are using these machines to build those models are mostly immigrants anyway.
If Koreans/Europeans decide not to sell their stuff, some of the money would go into circumventing the ban:
1) This is not good for Korean/Europen sellers, because it negatively affects sales volume, and is unlikely to be compensated by margins, because a good chunk of those will go towards circumvention instead of the original seller.
2) Some more money will go towards replacing those sellers completely. This is extra not good from the sellers perspective.
In this new world there are more important things like nationalism, the ego of the politicians and the personal ambitions of the super rich. Bot being good idea is irrelevant.
I don't understand what this hypothetical world is supposed to be. Samsung and SK Hynix are run by Korean capitalists who want to sell their stuff to America; the reasons you're describing are precisely why they would not exit the American market just to dodge an inconvenient antitrust investigation.
And hyper-scalers, Apple, AI labs all will die if memory makers can't sell to them?
regime change in South Korea. President Lee Jae Myung isn't exactly popular among Washington circles
memory is a commodity is laughable. Then software engineering is even more a commodity, the amount of engineering going into making memory chips the vast majority of people don't understand. There are a lot of software engineers getting this field after leetcoding and copy from hellointerview. Claude can write you an app in 30 minutes. Try build a lpddr5 dram chip in 30 minutes. Manufacturing know how itself is a specialty and barrier to entry.
It’s a barrier, but not an insurmountable barrier. Most of the companies are somewhat lazy and somewhat cheap, but if circumstances change in the market, that make it untenable, they will take on making their own memory or making arrangements with someone else.
It’s been done with processors, modems, SSDs and many other specialized chips. The design and engineering part can already be done at several companies, they just elect not to do anything on the fab side. But they will, if they have to in the long run.
It takes two to four years and Five to twenty billion dollars. Some of the bigger companies, if pushed, particularly, if the alternative is getting chips from China, will do something on the fab side if they absolutely have to. They probably won’t live with the current conditions forever.
There was a company that had three companies say no over about 10 years, eventually, they made a decision to build their own processor known as the M-series…
I mean your view isn't flawless but overall I agree. Too many people think building things amount to spending money and completely overlook the thousands of people required who are not just unskilled labor hired off the street.
I also feel like there should be FTC or other antitrust actions against OpenAI and other hyperscalers, with nvidia being complicit, if they are cornering the market that badly on consumer RAM, SSD and other components, especially if these volume purchases are for projected datacenter that haven't even broken ground (or have been paid for) for many months.
Furthermore, I think there should be a tax on algorithmic inefficiency, in that if a LLM, frontier or not, consumes more than a certain amount of KWH per token, it should be taxed such as to put emphasis on models than can run locally, on a normal PC
Anyone know what will lawfirms cut if they win from the lawsuit?
How many millions are we talking.
Honestly, the knock on effects of this cartel behavior should concern all countries, and be remediated with expediency.
From consumer electronics to the data-center, the rising real mfg costs and lack of supply is putting huge pressure on pricing, and may just drive anyone who cant negotiate with these suppliers out of business.
Once the dominoes start, I fail to see how things recover in less than 3-5 years, not counting all the businesses wiped out in the meanwhile.
How do we know it's cartel behavior, not just free market competition?
Just a reminder that anyone can file a lawsuit over anything, and the initial complaint is written by lawyers and reads with tabloid levels of sensationalism and allegation. The goal being to maximize the appearance of harm as much as possible so the suit has the greatest chance of sticking.
It is not in any way, shape, or form a ruling much less even a piece of well researched work. It's "my side of the story that makes me look perfect, with lawyers turning the heat up to 11"
In reality a lawsuit needs to have some basis. The bar might be low but judges frown on having their time wasted with a truly frivolous claim or a claim with no clear damages or harm shown.
That might have been true at some point in the past and the judiciary might even feel that way still.
However the practical penalty for filling absurd lawsuits is zero unless you do it repeatedly to random people for a decade.
Absolutely nothing bad will happen to the plaintiff or the lawyers representing them in this case.
Western legal systems are broken.
Honestly a lot of lawsuits feel like they just want the other party to pay them (settle) to make the lawsuit go away, even if there is no case; what are the repercussions for a frivolous lawsuit?
Realistically very few. The plaintiff carries the burden of proving the claim, but they are afforded the power of the law to do so via discovery. The defendant then has to spend a large number of legal resources refusing the claims and also providing the evidence required.
If the plaintiff loses the lawsuit a countersuit is pretty unlikely to succeed unless the lawyer participates in gross misconduct. Generally, countersuits are filed more to put the original plaintiff on the defense and don't result in a large judgement.
If you are operating in good faith, then you are pretty insular from kickback as a plaintiff
>what are the repercussions for a frivolous lawsuit?
None, hence the high price of liability baked into basically everything in America. And not just in nominal prices, but in terms of things like restricting access to spaces, restricting access to information, etc.
Maybe software people can get around this issue by not making every app a electron bloat? This is now more than doable now you got AI right? And it will save your job
And replace all your apps with some horrible Tcl/Tk and .NET interface?
Nobody ever made good native widgets.
Qt Widgets are good.
No, they suck. No swiping, no pull-to-refresh, no multitouch zoom, the calendars dropdowns are horrible to use, the buttons look like something out of Windows 3.1
Those are touchscreen actions, uncommon in desktop apps. Did you forget that this thread is about avoiding Electron bloat?
(Also, it has a theme engine. The buttons don't look like something out of Windows 3.1 unless you use a Windows 3.1 theme.)
,,The plaintiffs claimed the three companies reduced D-RAM supply under the pretext of transitioning to high-bandwidth memory (HBM). "The D-RAM oligopoly companies systematically coordinated the shift to HBM and the discontinuation of DDR3 and DDR4," they said. They added that Apple's recent sweeping product price increases were the trigger for the lawsuit.''
How can they do price fixing and discontinuing a product at the same time? It just looks like some companies are angry that AI / VC industry is outpricing them.
> An agreement to restrict production, sales, or output is just as illegal as direct price fixing, because reducing the supply of a product or service drives up its price. For example, the FTC challenged an agreement among competing oil importers to restrict the supply of lubricants by refusing to import or sell those products in Puerto Rico. The competitors were seeking to pressure the legislature to repeal an environmental deposit fee on lubricants, and warned of lubricant shortages and higher prices.
https://www.ftc.gov/advice-guidance/competition-guidance/gui...
But Micron didn't restrict the output: it stopped producing DDR2-3-4 completely, so it's not profiting from DDR customers at all.
So they restricted it to zero?
Not saying I agree with the plaintiffs.
People have a tendency to get upset when someone waves a future IOU intent order from another buyer in front of you, one that isn’t taking delivery anytime soon and then proceeds to tell you you must pay more…
hmm maybe the plaintiff should sue nvidia?
Antitrust and anticompetitive behavior continue to evolve. Back when the Sherman Act was passed, we were talking about backroom deals in smoke-filled rooms. One of the more important ways of recent years is where all or most competitors in a market use the same software that outputs the same value as to what they should charge. This is effective collusion even if it isn't explicit. The goal of such software is to raise prices. You raise prices by effectively or actually colluding with other market participants.
The posterchild for this is RealPage [1]. But you're going to see this pop up in every aspect of life, such as gas prices [2] and meat processing [3].
The whole thing is kind of depressing because this is what "innovation" is now: fancy ways to collude on prices.
[1]: https://www.justice.gov/archives/opa/pr/justice-department-s...
[2]: https://abcnews.com/US/wireStory/ai-helping-gas-stations-col...
[3]: https://www.justice.gov/opa/pr/justice-department-requires-a...
It used to be possible to bust up a company based (more or less) simply on their having too much of the market.
Right wing interests built up think-tanks after the Second World War, having seen the success of especially the Brookings Institution during the New Deal and world war eras in pushing technocratic-liberalism and wanting to fight fire with fire by creating their own credibility-laundering offices for propaganda advancing the desires of the rich. This effort bore its first great fruit in the mid '70s as Chicago School economists and lobbyists successfully overthrew that enforcement regime and replaced it with one in which specific harm must be proven to win an antitrust case, which de facto ended meaningful anti-trust enforcement in the US and led to the consolidation of money, power, and media over the following decades.
Aaaand... here we are, in the endgame of that movement.
Looking forward to getting my $10 settlement 15 years from now like what happened with the DVD drive price fixing lawsuit (paid as a digital Visa gift card code so it can't actually be spent anywhere).
I don't see how this can really go anywhere, but it'd be nice if it does.
Not like it'd be the first time someone shook a stick at them.
At this point the only hope for change is if China finally decides to get in the game rather than just threatening to.
The Chinese will get into the game, all right worldwide where most other places aren’t going to care about anything other than the price.
And our illustrious leader doesn't tariff Chinese RAM and allows its sale in the US.
dumb peripheral question: memory as a circuit would seems easy to add directly on-chip for a company like Apple. Is that blocked by time-frame, IP, technology, cost?
Logic gates and memory bits have very different fabrication processes (mostly because DRAM is optimized for a high density of big capacitors doing the storage).
You can put some memory on the logic wafer (SRAM) but it's area inefficient, which is wasteful on your expensive N2 wafer. So a dedicated DRAM process is vastly cheaper per bit, even at current elevated prices.
My guess is area becomes problematic. Chip becomes bigger, harder to manufacture without defects, costs rise as a result of those defects.
Bonus though, Apple is using "defective" chips in other products now such as the Neo which will help some with costs, but overall, you can't just add memory to the chip because the chip can become very yield sensitive, the process has to be there to produce the yield effectively.
I mean there was an attempt before, and maybe I'm wrong but I'm sure they were fined in 2010 at least in the EU, this seems to be round 2 of their 'Memory Cartel' after learning some lessons, realising all they need to do is keep supply as low as they want and allow the AI companies to spend hand over fist.
https://ec.europa.eu/commission/presscorner/detail/en/ip_10_...
>keep supply as low as they want and allow the AI companies to spend hand over fist
That's indistinguishable from not wanting to get burned by another semiconductor boom/bust cycle.
I think they already know it's gonna take China a year or two to ramp up, so they're banking on that ?
everything is political
I think the only solution to this issue is China. If CXMT can supply, it will put all these monopolies in check.
That’s a nice short-term solution. How do you envision adding an infinitely-deep-pocketed state-sponsored supplier to the mix over the longer term?
Even if CXMT can only sell in China, that reduces the amount of memory China needs to buy from other manufacturers, thus benefiting everyone.
What would the issue be, in the long term? I'm struggling to see a downside to more manufacturing capacity
If you want to realistically compete in a market with state-backed firms, you need state backing. Aerospace is a prime example.
Oh, maybe we're mixing terms here. OP was proposing CXMT as a "solution" to the problem of these companies engaging in price fixing. Now you're saying there needs to be a "solution" to the problem of these companies no longer being able to "compete" (engage in price fixing), because CXMT exists.
The latter is not a problem I or almost anybody outside those companies care to fix. It's not the public's job to bail out a company because they forced states to intervene to break up their monopoly after taking the highest profits in the industry's history.
cxmt is also selling their memory similar to the big three. No one hardware company in their right mind will sell their products not as high as possible after you learn how much harder hardware engineer and fab people work for.
"As high as possible" is a function of scale though. There is only so much global demand for memory at a given price point. If CXMT determine that by increasing supply they can increase profit even if it lowers prices, then they will do so within their ability.
EDIT: This is also semiconductor-specific. Because there's little additional cost to doubling the output, ramping up production is the hard part
I think it's incredibly unlikely to be deliberate price fixing this time. Demand is too high.
There is no reason for there not to be a price fixing. But the OpenAI's announcement from 2025-10-01 about buying 40% of supply, removes any need for collusion. It was a public signal for everyone to rise prices, one that each company could figure on their own. And it will be very hard to prove otherwise.
Two things can be true; they could theoretically make agreements to limit supply, like the OPEC does to exert control over international oil prices. But the OPEC is all above board and there's plenty of international competition on the oil market.
But the challenge is in proving it.
It isn't like OPEC and oil where you can turn on and off bumping out from the ground.
The most expensive foundry / fab isn't the leading edge, it is the "empty" ones. You can't have a fab sitting idle.
Prices skyrocketing create a perfect opportunity to slip in a little artificial bump and hope everyone blames the market. See also: egg prices in 2024.
if demand is too high, the market will adjust by adding supply. If Apple can't take the price of memory, why don't they make memory themselves. Oh no, the technical and manufacturing know-how is a barrier to entry and there is no talent available to make it. That is memory or semi companies' competitive advantages and their pricing power. It also takes Phd degrees to work at hardware companies. You can't have people doing leetcode for 2 months and todo apps and getting into the field and make 300k a year and hardware engineers wih Phds making 100k before this boom. It is long overdue for a rebalance of pricing power between hardware and software companies
Between TSMC, Intel and Apple memory can be designed, engineered and made, and that doesn’t even include the Chinese who will probably use this to take over a large part of the memory market worldwide?, the three-headed memory cartel is unsustainable, they have two to four years maximum and that’s it.
Price fixing is legal as long as your are doing it in the open. In the UK it is called "price match" and eg. if supermarket says they keep prices matched to their competitor. No regulator raises an eyebrow.
So here Samsung and SK Hynix could say they price match to Micron and they are in the clear.
> Price fixing is legal as long as your are doing it in the open.
In the U.S., competitors are allowed to act in similar ways in response to economic realities, as long as they each arrive at that decision independently. But publicly anchoring your price to a competitor’s is potentially illegal.
> Price fixing is an agreement (written, verbal, orinferred from conduct) among competitors to raise, lower, maintain, or stabilize prices or price levels.
[Emphasis added]
https://www.ftc.gov/advice-guidance/competition-guidance/gui...
Identical prices are often a sign of intense competition. Every gas station on a corner has the same price because it's a highly competitive market not because of collusion. The prices of the much more lucrative chocolate bars inside the gas stations are less likely to be identical.
Gas stations on the same corner (or almost the same) in fact frequently have different prices, often quite different. A few examples.
https://maps.app.goo.gl/AeRZsAeGQAHoYvd36
https://maps.app.goo.gl/cdcbxQuEYLBFVECc7
https://maps.app.goo.gl/k8zAVMrwxkKYsoA8A
https://maps.app.goo.gl/nrQmyShTk9P8YEmo9
Too many consumers think that there are enough differences in additives and detergents between the brands for it to make a significant different in mpg and in maintenance costs, so buy from "their" brand even if there is a cheaper brand nearby.
At one time those consumers would have been right, but first the EPA established certain minimal standards that all brands had to meet. The better brands were still better for your car than the ones just meeting the minimum, but it would take longer for it to make a significant difference.
Then BMW, GM, Honda, and Toyota made a spec for what makes a good detergents and additive mix, and most other major automakers joined. They created a certification, "TOP TIER", that gas companies that met that standard could use.
Nowadays most gas in the US is certified TOP TIER. Generally you will only find non-TOP TIER gas at stations not part of a major chain, at some gas brands owned by grocery chains (Safeway for example), and at some gas brands owned by convenience stores (7-Eleven for example).
Most independent tests have found that and detergents and additives beyond those required for TOP TIER certification do not make a significant difference, even cumulative over 300+k miles.
Nowadays then in the US just buy from any TOP TIER brand and that will be good enough. But consumers don't seem to know that and so for a large number of consumers different brands on the same corner aren't actually in competition.
This is a good signal of how competitive your gas market is. Your examples are Seattle and the valley, which are very difficult markets for gas stations, so unsurprisingly are uncompetitive.
The whole US gas station market has gotten less competitive over the last 40 years I've been observing prices and big coastal cities are a lot worse than anywhere else.
Yes, the gas station example is directly cited in the article I linked to. It’s legal for a gas station owner, with knowledge and consideration of a competitor’s price, to reduce their price to the same or just below. What is illegal is for nominally-competing gas station owners in an area to conspire to keep their prices within a range of each other’s, even without explicit agreement.
The article you linked seems to indicate that there has to be active communication between the gas station owners for there to be collusion.
When I worked at a gas station as a teenager there was definitely an unspoken implicit agreement that the price of gas would be 6 cents/liter above wholesale IIRC. Which was highly competitive and didn't completely cover costs.
"Every gas station on a corner has the same price because it's a highly competitive market not because of collusion."
Huh? I can go to most any gas station-occupied intersection and you will always find two that match and one (usually a Persian-owned Chevron) which is consistently a dollar or more higher per gallon across all grades of fuel.
Same here. I've never seen 2 gas stations near each other offer the same price. And large price disparities are common.
>Price fixing is legal as long as your are doing it in the open. In the UK it is called "price match" and eg. if supermarket says they keep prices matched to their competitor. No regulator raises an eyebrow.
No, the key term is "collusion", which could be done in the open or not. If a competitor told you they were unilaterally raising prices in secret, that would still be legal. Where you get into trouble is if you are cooperating to set prices. And no, this is all determined by a judge so cute workarounds like "I'm telling my competitors that I'm raising prices then gauging his body language" won't work.
I worked in a small town gas station as a kid. We always phoned our competitor to let them know when we were changing our prices, and they extended us the same courtesy. Usually we followed them, but sometimes we didn't.
Gas prices are posted on massive highly visible signs and are public information. This wasn't collusion, it was a sign of intense but friendly competition.
That is not what price fixing actually is. In the UK "price matching" is a one-to-many relationship meaning that the price of goods is set to the "lowest available"
Price fixing is a many-to-one all the manufactures agree to the highest prices they all agree on and set it there.
You are arguing semantics. Effect is the same.
Price match creates downward pressure, nobody is gonna sue over lower prices.
This is like if you showed a supermarket that their competitor's oranges were more expensive, and they "matched" by raising their prices for everyone.
That is your assumption without evidence. Once the targeted market knows others are tracking the price, it can up the price without fear competing market will undercut.