Ok so Samsung, SK Hynix and Micron do not have the capacity to meet demand. Also, what little capacity they do have they are allocating to HBM over DRAM. Based on my limited knowledge HBM can not be easily repurposed for consumer electronics. Translation: main street is cooked for the next 3-4 years.
It doesn't stop there though. OpenAI is currently mired in a capital crunch. Their last round just about sucked all the dry powder out of the private markets. Folks are now starting to ask difficult questions about their burn rate and revenue. It is increasingly looking like they might not commit to the purchase order they made which kick-started this whole panic over RAM.
Soo ... how sure are we that the memory makers themselves are not going to be the ones holding the bag?
The Radeon VII came out in 2019 as a $700 consumer GPU with an 1TB/s HBM2 memory subsystem which is more than any consumer GPU you can get today, including the high-end ones afaik. At that point in time, there was a whole lineup of AMD GPUs with HBM going down into the midrange.
If they could make this stuff and sell it to regular people a decade ago for very palatable prices, why do they come up with the idea that this is the technology of the gods, unaffordable by mere mortals?
I have been wondering this recently. It was the convention that if you wanted to keep costs down, try to keep the memory bus size down as low as possible. Still remember the awful Radeon 9200 SE - 64bit data bus that strangled an already slow GPU.
Heck, I have a phone with a 16bit memory bus for instance. The high(ish) clock rate only makes up the difference slightly.
But with general prices on all components going up, it might not be such a big factor any more.
HBM migght make sense for higher end products which can free up space for the lower end that will never use the tech.
Vega was a card with decent perf/$ for the consumer, but from a pure technical point of view (perf/mm2, perf/BW, perf/W) it was a major failure. Both Vega (and Fiji before it) showed that excess memory BW alone is not sufficient to win.
5090 is an overpriced outlier. A typical consumer GPU, like RTX 5070, has a 3-times lower memory throughput.
Even a RTX 5080 has a lower memory throughput than a Radeon VII from 2019, 7 years ago, while being much more expensive.
The memory throughput of GPUs per dollar has regressed greatly during the last 5 years, despite the fact that the widths of the GPU memory interfaces have been reduced, in order to decrease the production costs.
RTX 5080 has a 256-bit memory interface, while the much cheaper Radeon VII had an 1024-bit memory interface. RTX 5080 has almost 4-times faster memories than Radeon VII, but it has not used this to increase the memory throughput, but only to reduce the production costs, while simultaneously increasing the product price.
Modern GPUs like RTX 5080 are much faster for the applications that are limited by computational capabilities, mainly because they have more execution units, whose clock frequencies have also increased.
I suppose that most games are limited by computation, so they are indeed much faster on modern GPUs.
However, there are applications that are limited by memory throughput, not by computation, including AI inference and many scientific/technical computing applications.
For such applications, old GPUs with higher memory throughput are still faster.
This is why I am still using an old Radeon VII and a couple of other ancient AMD GPUs with high memory throughput.
Last year I have bought an Intel GPU, which is still slower than my old GPUs, but it at least had very good performance per dollar, competitive with that of the old GPUs, because it was very cheap, while the current AMD and especially NVIDIA GPUs have poor performance per dollar.
5090s are certainly expensive compared to most other GPUs, but not expensive enough to be unobtanium for nearly any professional who could utilize one as part of their job
FWIW that depends on the stores you're looking at. There are three models from different manufacturers available here in a few shops. The prices are a bit ouchier than what i paid for mine around Christmas 2024 though (i got mine on a sale).
After NVIDIA essentially removed FP64 from consumer GPUs (their 1:64 performance ratio is worse than what you can obtain by software emulation, so it is useless, except for testing programs intended to run on datacenter GPUs), AMD persisted for a few years, but then they also followed NVIDIA.
AMD Hawaii GPUs still had 1:2 FP64:FP32, while the consumer variant of Radeon VII dropped to 1:4. The following AMD consumer GPUs dropped the FP64 performance to levels that are not competitive with CPUs.
Nowadays the only consumer GPUs with decent FP64 performance are the Intel Battlemage GPUs, which have a 1:8 performance ratio, which provides very good performance per dollar.
Only indirectly. They have most of the money, so if they want something that’s in short supply, the price will rise to the point that it becomes unaffordable to everyone else.
Reason number 7,322 why US-style ultracapitalism is self-destructive, anti-social, and dystopian.
To add a more local hurdle as well, the Dutch power grid is at capacity and its managing company is now telling companies that planned to build a datacenter that they can't be connected to the grid until 2030, even though said companies already paid for and got guarantees about that connection.
That is, memory capacity is reserved for datacenters yet to be built, but this will do weird things if said datacenter construction is postponed or cancelled altogether.
That guarantee is not as much of a guarantee as stated in the media. You get a guarantee it will be planned at a certain time (as in looked at), not that it will be build. The cost of doing business is taking risks and mitigating them. There is a reason the nuclear plant in Borsele was build: an aluminium smelter. Maybe you should arrange for something similar as a datacenter (no politician will fall on a sword for that but you can try). The (original) power draw is about the same 80-100MW.
It says that in 2025, Netherlands was a net exporter of electricity (~14,000 GWh). My guess: Where they want to build data centers, the grid cannot handle it, but the overall system has more than enough power to build data centers. Do you think that sounds like a resonable guess?
> the Dutch power grid is at capacity and its managing company is now telling companies that planned to build a datacenter that they can't be connected to the grid until 2030, even though said companies already paid for and got guarantees about that connection.
Are the Netherlands a large proportion of global datacenters?
Amsterdam hosts a major internet exchange. It's not a bad place to build a datacenter and there are many. Northern latitude brings free air cooling, but also additional distance to clients. Lots of peers in AMS-IX, but not a lot of oceananic cable landings (one with two paths to the US, but most of the submarine cables land nearby in Europe)
Yes. Amsterdam has one of the largest IXPs (AMS-IX) in Europe and is also one of the largest European markets for Internet Infrastructure services (i.e. hosting, DNS provision, domain name registration, etc.)
Do AI data centers not need internet connectivity anymore?
The value of an IX isn't just in the IX itself, but also in the presence of hundreds of parties for direct peering, and excellent connectivity to the rest of the world.
It makes a lot of sense to build your DC near one - even if you have no intention of actually participating in the IX itself.
> Do AI data centers not need internet connectivity anymore?
They don't need entire IX worth of connectivity. You're sending mostly text back and forth and any media is in far lower volume than even normal far less dense DC would generate, all the major traffic is inside the AI DC.
High-level, I would agree with you. One thing that blows me away: I think I read that Northern Virginia, USA has the highest data center density in the world. Mostly it is due to demand from US gov't, military, and spy agencies (like NSA). How did they do it? In mainstream media, I don't see any news about a stressed power grid in this area. I guess the US gov't carefully coordinated with local power providers to continuously upgrade their power grid? This is a real question. It makes no sense to me. No shilling/trolling here.
What the grid looks like in different countries is very different. The Dutch power grid is already almost 50% renewables, which is an inconvenience for adding capacity because that's around where you have to start really dealing with storage in order to add more.
In most other places the percentage is significantly less than that and then you can easily add more of the cheap-but-intermittent stuff because a cloudy day only requires you to make up a 10% shortfall instead of a 50% one, which existing hydro or natural gas plants can handle without new storage when there are more of them to begin with.
I don’t think the source of the electricity is particularly relevant to whether or not you have the transport capacity to add tens of megawatts of demand to the grid. The problem is generally not the supply but whether your local transformers have capacity left.
When you're talking about something that draws megawatts existing transformers are pretty irrelevant because you're going to run high voltage lines directly to the site itself and install new dedicated transformers on site.
What's more common is that they don't have the transmission capacity itself, but that one's pretty easy in this case too, because what that means is that you have an existing transmission line which is already near capacity with generation on one end and customers on the other. So then you just build the data center on the end of the transmission line where the generation is rather than the end where the existing customers are, at which point you can add new generation anywhere you want -- and if you put it near the existing customers you've just freed up transmission capacity because you now have new customers closer to the existing generation and new generation closer to the existing customers.
I know that you comment is midly off topic, but I am going through the same out of body experience each time I see a major project announces an opening date of 2030 or later.
This just highlights what an utter failure and self-inflicted wound the green policies of Euro countries have been. Europe has already lost the AI race to the U.S. and China.
Renewables is the only realistic path to energy independence. Today's global situation should show the absolute necessity of that, even if you dont give a sh*t about the environment.
Had we done more 10 years ago we would have been better of. The second best time to start is now.
(We used to build it at a fraction of the cost and less than half of the time that we do with our modern fuckups and fuel can come from just about anywhere if need be. It might be a lot more expensive than the stuff kazachstan and still be a fraction of the cost.)
I think ideally we would've done both to press the cost of nuclear down and given the fact that the renewables rollout turned out to be a lot lot more expensive than proponents claimed it would be whilst still tying us up into gass to cover winter.
It does not. That is not economically mined. Last big hard coal producer in EU - Poland, has extraction cost x2 or x3 of the mountain top removal mining in US. This sector is shrinking rapidly. Poland coal production came back to ~1915 levels (taking into account current PL territory). This sector would be closed already if not for massive subsidies.
Last year, China's coal use decreased, while China installed 300x more renewables than nuclear. Coal and nuclear aren't cost competitive with renewables, either in a free market or a technocratic top-down economy. Coal and gas still maintain a valid niche of firming intermittency. But that niche is temporary and shrinking.
The free market installs a tiny amount of coal, and a lot of renewable energy. Whether you believe this means "coal is/isn't cost competitive with renewables in a free market" is a debate about word definitions that I'm not terribly interested in.
China, like Brussels, is trying to reduce coal for similar reasons. They don't like the air pollution health hazard (fully believable), and they say they don't like global warming (somewhat believable).
China and Texas are both installing silly amounts of renewables. They install very little new fossil fuels or nuclear. They both maintain cheap electricity prices through abundance.
The problem in the EU is not renewables, it's the same problem that Democratic states in the US face. Regulations and permitting hurdles that block private renewable energy developers.
There appears to be zero advantage to having the datacenter actually in your country apart from minor local property tax, in exchange for which it will put up the electricity bills of every single citizen, who already hate how much they're paying.
The problem in this case seems to have sprung from a lack of collusion. Altman reportedly approached Samsung and SK independently to strike deals for a large chunk of both companies' production. Neither party apparently knew he was negotiating with the other.
If they had actually been communicating or colluding with each other, they would have put the screws to him, making it harder for OpenAI to assert control over the vast majority of the DRAM market.
Failing that, you'd like to think a regulatory agency somewhere would step in to keep a single player from hosing everybody else, but...
> Neither party apparently knew he was negotiating with the other.
I don’t buy it that two of the largest manufacturers of DRAM in the world, from the same country, didn’t know this. Even of you ignore each company’s intelligence teams, that’s also the job of the country’s internal intelligence services, to make sure they know what all companies are doing and then make it so they have the best leverage to gain as much as possible. Both companies would have known “somehow” and played hardball.
There are a lot of cyclical businesses that make money every year. It requires careful management. Factories can produce less than full capacity - but you better design for that. you can make money in the worst years without laying anyone off even - but it requires careful attention to details and not over hiring in good times as if they will never end.
Factories working at (significantly) less than full capacity gets a bit harder when you've got one of the most expensive machines on earth working in them, and production lines that'll be out of date in a couple of years
the normal way to do that is by hiring/firing to meet demand, but in the fab business, you have 10s of billions of dollars of capex with relatively little opex. if you're running at <90% capacity, you're losing money.
Last year ai folks are all over wallstreet and articles, decrying how hardware folks are a roadblock to new frontiers in AI. They just couldn't print and pack the new chips faster.
but if you don't collude during times of feast you will have famine, and during times of famine you will have famine, in an economy based on feast/famine you must sometimes feast or die.
All of the capital intensive businesses face this issue. Chemicals, Shipping, Semiconductors etc.
You get market signals that the demand is there, you acquire the necessary capital, you spend 5 years to build capacity, but guess what, 5 other market players did the same thing. So now you are doomed, because the market is flooded and you have low cash flow since you need to drop prices to compete for pennies.
Now you cannot find capital, you don't invest, but guess what, neither your competitors did. So now the demand is higher than the supply. Your price per unit sold skyrocketed, but you don't have enough capacity!
Forecasting demand 5 years into the future is intrinsically highly unreliable. It doesn’t matter if it is capitalism or a command economy. The bet is always going to be risky and someone will have to pay for that risk.
At least with capitalism you have many different people with different perspectives on the risk making independent bets. That mitigates the more extreme negative outcomes.
Also it is government job to regulate. Monopolies should be busted, and behaviour like that should be culled. But US govt is full on AI to mask them cratering their own economy.
To add to that, investors who do make the bet get punished for over-building, which is better than tax payers paying for it. And before someone says it, big corps do get bailed out by gov't, but that's definitely goes against capitalist ideas.
Is the DRAM industry really capitalist? Focusing on just the Korean parties, it functions like a command economy. I would say the same about most high end semi-conductor manufacturing, TSMC, Intel, ASML are being commanded and driven by nation-state level decision making. Right now the command is to focus on high wattage centralized AI systems at the expense of everything else.
No one at high levels is capitalist, in ideology or action. An ideological capitalist would be in favor of competition, but these people disdain it and collude regularly. The only 'capitalist' actions they take are by accident, the real goal is as much power/money as possible as fast as possible.
We don't even expect companies to plan long-term anymore, it's just moving wealth as fast as possible.
That isn't really a change, very few people could ever have been said to be ideological capitalists. (capitalist is not a word with a hard definition, but I'm considering it a different thing than the more modern pure libertarian zero-regulation ideology)
Liberalism (in the traditional economic sens) likes competition. Capitalism is a mode of production, and capitalists notoriously don’t like competition when they are the incumbents
Because that does not happen exactly as you say for all players. The demand signals will be processed and long-term risk is balanced against short-term gain in a distributed fashion, so not everyone will do the same.
It's more optimal than planned economies until we have AI planned economies with realtime feedback, I guess.
Consumers get cheap goods during oversupply and most inefficient companies get elliminated during bust while consolidation leads to economies of scale.
Why is the opposite of capitalist markets automatically assumed to be a command economy? Co-op style businesses aren't really capitalist orientated but are also not reliant on government action.
The sheer fucking blazing ignorance of this comment
> Capitalists claim that this is optimal.
Compared to starving under communism coz someone at top got the number wrong, yes. And it only really happens when there are massive, unpredictable market movements and governments not doing their job. Govt should look at the whole thing and just say "no", blame them.
No market system self regulates well enough, and it's government job to file down the edge cases like this. But the revolution happened in country which has two utterly incompetent parties, both in pockets of billionaires, fighting for power, and the clowns from one that won last battle use AI to smokescreen the economic growth their actions cratered
I am betting the pendulum swings faster to the other side to excess capacity as all the construction lies of Altman fall through with financiers waking up the the fact they can't build the infrasctructure as fast nor make any profits on that infrastructure that will get built.
What if it takes 100 years to get to AGI or we never achieve it? All bets on AGI will just fail over and over again for decades in that case. It seems a bit like saying financiers can't risk not being involved with Faster-than-light travel technology. Yeah, it would change everything if we got it, but betting that we'll get it soon over and over again is probably not going to get you a lot of money.
We've been projecting both FTL and AGI as future possibilities for almost 100 years now. Do LLMs get us a lot closer to AGI? I think they get us a little closer and Moore's "law" making compute faster probably is a much bigger factor, but I think we're still a very very long ways away.
Who has been projecting FTL as a realistic technology ever? FTL is not possible according to the current laws of Physics, while AGI is at least not forbidden by them.
I think this should be something you can answer for yourself by looking at human media and news over the lasy 100 years. I find it hard to belive you haven't ever noticed anyone seriously saying we may possibly have FTL sometime in the future. Incidentially I think I read on HackerNews that Sam Altman has been talking about building Dyson Spheres in the future. I suppose they're not forbidden by the current laws of phyisics either, but I don't know if I would call them a realistic technology.
The IQ of the smartest human, the perfect memory storing and recollection of computers, the fact that it never tires. I don't know if it's AGI but it's already something greater than us.
Anything where there's _at worst_ solder and traces between the compute and the memory. That's why you see it on GPUs (and Apple hardware). DRAMs advantage is modularity.
HBM is just normal DDR RAM that's been packaged with (much) wider-than-usual data buses. That's where the high bandwidth comes from, not from high clock rates or any other innovation or improvement in core specifications.
But wouldn’t you rather hbm prices come down first ? Memory makers will be fine. There is practically infinite demand.
Unless you get china style rationing of compute per person world wide.
The real issue is everyone wanting to upgrade to hbm, ddr5, and nvme5 at the same time.
Not a rec, but just my source: Atrioc (streamer, YouTuber) is good at gathering all the facts for the rest of us. There's many other things in play, like the Strait of Hormuz (helium, bromine). Ultimately it works out that the shortage, and shortage profits, will continue; the chip makers are probably going to continue to see record profits (as Samsung has).
The specific mix of factors could change at any time, but the supply chain is relatively inelastic, it will take some time to show up on price labels.
I think I’m missing something. Financially, what bag would the memory makers be holding here? I don’t think I’m well informed regarding how these deals were structured.
Memory makers make capital investements (build different factories, convert physical production lines, etc.) to meet orders that have been place for the next ~5 years.
OpenAI (or whoever) crashes and can't pay for the order leaving the memory makers in a tough spot.
The amount of money flowing both from the AI bubble and from quite literally scalping both the server and consumer market... They gambled on the opportunity and if they fail - it's their problem.
The market is already stagnated. Even if OpenAI doesn’t buy what they reserved other players will do so. SK Hynix CEO said there is a 20% gap between supply and demand per year. And that doesn’t account the shock effect that will take place the moment prices normalize and everyone and their dog will go out and start buying inventory to avoid the next crisis. I for one would certainly buy more than I currently need just in case.
> OpenAI’s rapid growth, fueled by the success of ChatGPT and other AI products, led to a landmark agreement in October to purchase 900,000 DRAM wafers per month from Samsung and SK Hynix—amounting to roughly 40% of global supply. This surge in demand, coupled with limited manufacturing capacity, sent prices for memory kits skyrocketing. [0]
They ordered 40% of the global RAM production for 2025/26. It was a non-binding agreement that either side could easily withdraw from but they're essentially trying to buy about half of all the RAM.
If I booked half a hotel's rooms then suddenly said "yeah never mind. Half my friends cancelled and we're not staying", basically any hotel would be coming at me for my money because there's no way they can fill their rooms now and they're losing revenue. But OpenAI can really get the whole world to pivot towards it then say "cool but we don't need your product anymore" and RAM makers are just going to let it go.
Whoever decided that was a good idea needs to be fired and publicly shamed.
Well if that hotel was then able to sell the other half of hotel rooms for 10x the old price. Then the hotel might actually be happy as they can now charge 10x for the other half or slowly lower prices back down over years.
> Soo ... how sure are we that the memory makers themselves are not going to be the ones holding the bag?
I hope they do, they did not have to agree to sell so much RAM to one customer. They’ve been caught colluding and price fixing more than once, I hope they take it in the shorts and new competitors arise or they go bankrupt and new management takes over the existing plants.
Don’t put all your eggs in the one basket is how the old saying goes.
Memory makers did get themselves into this situation by selling all wafers for empty promises and alienating everyone but OpenAI tbh. I do hope they end up holding the bag once again, cause after covid and the cartel thing they don't seem to ever learn their lesson on how to have the tiniest amount of integrity.
> Memory makers did get themselves into this situation by selling all wafers for empty promises and alienating everyone but OpenAI tbh.
Wasn't the problem here that OpenAI was negotiating with Samsung and SK Hynix at the same time without the other one knowing about it? People only realized the implications when they announced both deals at once.
Permanent public ownership of (very large stakes in) these companies doesn't seem like such a bad idea anymore, does it? It's what we used to have for most of the 20th century at least in Europe.
That's only sound is it not. If you take away the hype - nothing critical actually depends on LLMs. You can remove them all today, and nothing bad would happen.
> Soo ... how sure are we that the memory makers themselves are not going to be the ones holding the bag?
We aren't. The remaining memory manufacturers fear getting caught in a "pork cycle" yet again - that is why there's only the three large ones left anyway.
If they don't expand capacity much, the only negative consequences I foresee happening for them is that they might lose spending discipline, and that systems will be set up to make do with a little less memory. Apart from that, it's just very high profits followed by more or less regular profits.
They could wind up losing all their business to China though.
China has memory makers who are creeping up through the stages of production maturity, and once they hit then there's no going back.
If the existing makers can't meet supply such that Chinese exports get their foot in the door, they may find they never get ahead again due to volume - that domestic market is huge so they have scale, and the gaming market isn't going to care because they get anything at the moment, which is all you'll need for enterprise to say "are we really afraid of memory in this business?"
Good point, it's a risk but so far the Chinese competition isn't up to par and it's unclear whether they'll be able to exploit the current window of opportunity.
You think this window is short? We've been dealing with this for years and years, and to me it seems more like incumbent manufacturers are too comfortable milking cash cows.
Surely this can be solved with financial engineering. The memory makers build more capacity, but they finance it with something like floating-rate notes linked to an index of memory prices, or even catastrophe bonds or AT1s. Or more crudely, set up special purpose vehicles to build the extra capacity, and issue convertible bonds from those; if the memory market collapses, investors don't get paid, but they do get a memory factory.
Do the memory makers not have a contract in place for an order this large? I assume that they aren't going to take "trust us bro" as good enough for several million dollars in orders, and even if there is a way to cancel the order it won't be free. I would assume so at least, but i would like if anyone knew for certain.
I was interning at a company that made networking gear that was put out of business when their largest customer canceled an order within a week of the delivery date.
The customer ran out of money. In terms of where you are in line of debtors when you haven't even delivered the product to a customer, it's so far back as to be assured you won't get your money.
If the memory makers got a deposit from OpenAI as part of this deal, that is likely to be the only money they will get for any undelivered memory, particularly if OpenAI runs out of capital.
It's not going to last until 2028, it'll last until 'min(AI_bubble_burst, 2028)', which I expect will be a lot smaller than just '2028'. So the real question is, how long will it take to retool for non-HBM, and will there be a fire sale as they scramble to recover?
Which also explains why production is falling behind demand, companies aren't going to sink billions into creating product for a market that could dry up overnight.
This will result in demand destruction which will starve the enterprise which will starve the hyperscaler. theres no situation where people not being able to afford hardware for 4 years results in the bubble not popping
I'd expect unaffordable hardware to drive demand for thin clients connected to cloud services which is something that had already been happening gradually prior to this.
It's a business with huge up-front capital expenses and typically very low margins. Supply is scaling up slowly because it's hard, and if you overshoot, you go out of business.
Nobody is "allowing" this. It's a natural property of being both advanced technology and a commodity at the same time.
The strange deals on the entire future output are what was allowed. Try to do the same thing with onions and the government understands you are a criminal.
That is quite the amusing read but it seems like a poorly constructed law. It wasn't futures themselves that were the problem there. The duo engaged in blatant market manipulation and severely disrupted part of the food supply in the process.
Cornering the market with the intent to flip the goods is not quite the same as cornering the market because you actually want the goods and intend to use them yourself.
It has the makings of a natural monopoly, except its compounded by RAM cartels colluding to shut out the last of the competitors.
Recently they had a second price fixing lawsuit thrown out (in the US).
Now with the state of things I'm sure another lawsuit will arrive and be thrown out because the government will do anything to keep the AI bubble rolling and a price fixing suit will be a threat to national security, somehow. Obviously thats speculative and opinion but to be clear, people are allowing it. There are and more so were things that could be done.
Allowed? We live in a neoliberal world where corporate monopolies / oligopolies aren’t even remotely regulated. If you try to do even the gentlest regulation of companies people scream about communism and totalitarianism. Unless the regulation serves the monopolies by making it harder to enter the market.
It started with raegan, and even parties on the “left” in the west believe in it with very few exceptions.
> We live in a neoliberal world where corporate monopolies / oligopolies aren’t even remotely regulated. If you try to do even the gentlest regulation of companies people scream about communism and totalitarianism. Unless the regulation serves the monopolies by making it harder to enter the market.
The thing that enables this is pretty obvious. The population is divided into two camps, the first of which holds the heuristic that regulations are "communism and totalitarianism" and this camp is used to prevent e.g. antitrust rules/enforcement. The second camp holds the heuristic that companies need to be aggressively "regulated" and this camp is used to create/sustain rules making it harder to enter the market.
The problem is that ordinary people don't have the resources to dive into the details of any given proposal but the companies do. So what we need is a simple heuristic for ordinary people to distinguish them: Make the majority of "regulations" apply only to companies with more than 20% market share. No one is allowed to dump industrial waste in the river but only dominant companies have bureaucratic reporting requirements etc. Allow private lawsuits against dominant companies for certain offenses but only government-initiated prosecutions against smaller ones, the latter preventing incumbents from miring new challengers in litigation and requiring proof beyond a reasonable doubt.
This even makes logical sense, because most of the rules are attempts to mitigate an uncompetitive market, so applying them to new entrants or markets with >5 competitors is more likely to be deleterious, i.e. drive further consolidation. Whereas if the market is already consolidated then the thicket of rules constrains the incumbents from abusing their dominance in the uncompetitive market while encouraging new entrants who are below the threshold.
Arguably a more efficient approach might just be to have a tax that adds on to corporate tax incrementally for every % of market share a company has above say 7-8%. Then dominant companies are incentivised to re-invest in improving their efficiencies rather than just buying/squeezing out competitors. A more evenly spread market would then, as a result, be against regulations that make smaller market participants less competitive, as they'd all be in relatively less table positions.
> Arguably a more efficient approach might just be to have a tax that adds on to corporate tax incrementally for every % of market share a company has above say 7-8%.
How is this more efficient? You'd still be applying all of the inefficient regulatory rules intended to mitigate a lack of competition to the smaller companies trying to sustain a competitive market, and those rules are much more deleterious for smaller entities than higher tax rates.
If you have $100M in fixed regulatory overhead for a larger company with $10B in profit, it's only equivalent to a 1% tax. The same $100M for a smaller company with $50M in profit is a 200% tax. There is no tax rate you can impose on the larger company to make up for it because the overhead destroys the smaller company regardless of what you do to the larger one.
They won't be, prices are high because they are refusing to build capacity for demand that may evaporated by the time they are done. They are holding back and building only enough so when the bubble pops they will be fine.
You can't build capacity overnight, and even with that in mind, it's hard to say if it is sensible to increase capacity now that we are in an AI bubble. For all we know, the bubble might burst.
So the ML hate is weaponized in the form of memory demand collapse FUD, and the public at large has to pay through their nose for it... thanks party poopers!
I don't think its from the ML collapse FUD, its most likely from the multiple time's in the past when they overbuilt and it resulted in a memory oversupply and price collapses. The 1985–1988, 1993–1994, 1998–2002 and the post pandemic oversupply. These were all cases where shortages followed by over corrections caused oversupply, financial losses due to low prices and fewer surviving companies. I think they're taking their time and are cautiously adding more capacity in such a way that prices won't end up collapsing again. Regardless, the result is still that we the consumers have to pay more.
At this point the remaining memory companies are… the ones that didn’t die during an over-supply collapse, right? I guess there’s been a strong evolutionary pressure against giving consumers what we want, haha.
If they gradually increase production capacity then prices stay high for 10+ years (or for as long as it takes for demand to crash) because a gradual increase in production takes that long for them to add enough capacity for current demand.
If they add enough capacity to meet current demand quickly then if demand crashes they still have billions of dollars in loans used to build capacity for demand that no longer exists and then they go bankrupt.
The biggest problem is predicting future demand, because it often declines quickly rather than gradually.
do we have evidence of RAM manufacturers going bankrupt? do we have evidence that the increased capacities after the mentioned past shortages went unused or were operated at a loss?
Machines take up space in buildings (factories); both of which are discrete rather than continuous functions. If your factory is already full of memory-making machines, and want to add one more, it will cost you billions and many months to build another factory.
If you suppose you have cracked the smooth-ramping problem, perhaps you should throw your hat in the ring and soak up all the pent-up demand that SK Hynix, Samsung and Micron are neglecting.
Think of the factory problem from physics first principals instead, as Elon would say. Musk says he will outcompete earth fabs by building them on the moon in just a few years, deploy radiation harden versions of the chips into space, and beat out TCO vs doing this on earth.
If he can do all that that fast, the RAM makers should be able to at least 1000X their fab capacity on earth in one year. One year for scaling up existing tech is an eternity compared to Elon's timeframe for moon-fabs given the relative complexity of the challenge.
I would expect that OpenAI gets as much money as they ask for for the next 10 years.
There’s virtually infinite capital: if needed, more can be reallocated from the federal government (funded with debt), from public companies (funded with people’s retirement funds), from people’s pockets via wealth redistribution upwards, from offshore investment.
They will be allowed to strangle any part of the supply chain they want.
China already has a well developed DRAM industry, as DRAM is somewhat easier than logic, and can tolerate a much higher defect rate. The industry will figure this out.
Another point is I often see the money argument - like country X has more money, so they can afford to do more and better R&D, make more stuff.
This stuff comes out of factories, that need to be built, the machinery procured, engineers trained and hired.
I think the article has a giant blind spot as far as China is concerned , considering they have already a mature enough memory ecosystem via YMTC that Apple was considering sourcing from them. As well as continued expansion in the DRAM and HBM Fabs [1].
It feels like the memory cartel once again trying to incentivise their various govt to cough up some more tax breaks/funding to cushion the AI buildout bet that they made and the bubble seeming about to pop.
In any case if they leave the consumer market underserved it should be no surprise if before that 2030 prediction we are all on cheaper YMTC memory modules.
I think you're massively overestimating how much money is really accessible here. The parent comment's right that all of the easily available VC & private equity investment is basically used up. OpenAI was struggling to sell $600M of private equity, the big multi-billion dollar investment packages had lots of conditions and non-cash in it.
> more can be reallocated from the federal government (funded with debt)
While this is the most reliable funding, it's still not very accessible. OpenAI is a money pit, and their demands are growing quickly. The US government has started a bunch of very expensive spending. If OpenAI were to require yearly bundles of it's recent "$120B" deal, that's 6% of the US' discretionary budget. 12.5% of the non-military discretionary budget. (And the military is going to ask for a lot more money this year) Even the idea of just issuing more debt is dubious because they're going to want to do that to pay for the wars that are rapidly spiralling out of control.
None of this is saying that the US government can't or wouldn't pay for it, but it's non trivial and it's unclear how much Altman can threaten the US government "give me a trillion dollars or the economy explodes" without consequences.
Further deficit-spending isn't without it's risks for the US government either. Interests rates are already creeping up, and a careless explosion of deficit may well trigger a debt crisis.
> from public companies (funded with people’s retirement funds)
This would be at great cost. OpenAI would need to open up about it's financial performance to go public itself. With it's CFO being put on what is effectively Administrative Leave for pushing against going public, we can assume the financials are so catastrophic an IPO might bomb and take the company down with it. Nobody's going to be investing privately in a company that has no public takers.
Getting money through other companies is also running into limits. Big Tech has deep pockets but they've already started slowing down, switching to debt to finance AI investment, and similarly are increasingly pressured by their own shareholders to show results.
> from people’s pockets via wealth redistribution upwards
The practical mechanism of this is "AI companies raise their prices". That might also just crash the bubble if demand evaporates. For all the hype, the productivity benefit hasn't really shown up in economy-wide aggregates. The moment AI becomes "expensive", all the casual users will drop it. And the non-casual users are likely to follow. The idea of "AI tokens" as a job perk is cute, but exceedingly few are going to accept lower salary in order to use AI at their job.
There's simply not much money to take out of people's pockets these days, with how high cost of living has gotten.
> from offshore investment.
This is a pretty good source of money. The wealthy Arabian oil states have very deep slush funds, extensively investing in AI to get ties to US businesses and in the hope of diversifying their resource economies.
The "no food in other countries" is because of failed/corrupt governments, not because people use AI to generate cat pictures in the West. The economy is not a "fixed pie" that needs to be allocated among people of the world.
Just look at Cuba, which could be a very rich country and one of the prime tourist destinations of the world.
Something I haven’t been able to reconcile: If AI makes software easier to create, that will drive the price down. How are software companies going to make enough revenue to pay for AI, when the amount of money being spent on AI is already multiples of the current total global expenditure on software? This demand for RAM is built on a foundation of sand, there will be a glut of capacity when it all shakes out.
The usage of LLMs is continuing to increase ~exponentially. I'm going to bet on that rather than some half-baked scenario analysis that only takes into account one scenario and assigns a 100% probability to it.
> The usage of LLMs is continuing to increase ~exponentially
I would like a source for that statement. Additionally, I want to know by who? Because it certainly isn't end users. Inflating token usage doesn't make it any more economically viable if your user base, b2b or not, hasn't increased with it. On the contrary, that is a worse scenario for providers.
> This demand for RAM is built on a foundation of sand
Not exactly.
LLMs are already quite useful today if you use them as a tool, so they are there to stay. The remaining problem is scalability, a.k.a. how to make LLMs cheap to use.
But scalability is not really a requirement when you look the bigger picture. If smaller software company/projects can't afford to use AI, the bigger ones might just. Eventually they will discover variable use cases for such tech, even if it only serves big firms i.e. defense, resource extraction, war, finance etc.
To the other end, if scalability is achieved, the use of LLM products will be cheaper too, so smaller project can also use them. But of course, if LLM usage is too cheap, then many were-to-be-consumers will just create software projects by themselves at their homes.
> If AI makes software easier to create, that will drive the price down.
Supposedly AI drives down the cost of producing software,not the "price".
> How are software companies going to make enough revenue to pay for AI, when the amount of money being spent on AI is already multiples of the current total global expenditure on software?
Currently, the cost of AI is between $20/month and around $200/month per developer.
I think the huge billions you're seeing in the news are the investment cost on AI companies, who are burning through cash to invest in compute infrastructure to allow both training and serving users.
> This demand for RAM is built on a foundation of sand, there will be a glut of capacity when it all shakes out.
Who knows? What I know is that I need >64GB of RAM to run local models, and that means most people will need to upgrade from their 8Gb/16GB setup to do the same. Graphics cards follow mostly the same pattern.
You can run huge local models slowly with the weights stored on SSDs.
Nowadays there are many computers that can have e.g. 2 PCIe 5.0 SSDs, which allow a reading throughput of 20 to 30 gigabyte per second, depending on the SSDs.
There are still a lot of improvements that can be done to inference back-ends like llama.cpp to reach the inference speed limit determined by the SSD throughput.
It seems that it is possible to reach inference speed in the range from a few seconds per token to a few tokens per second.
That may be too slow for a chat, but it should be good enough for an AI coding assistant, especially if many tasks are batched, so that they can progress simultaneously during a single read pass over the SSD data.
> Who knows? What I know is that I need >64GB of RAM to run local models, and that means most people will need to upgrade from their 8Gb/16GB setup to do the same. Graphics cards follow mostly the same pattern.
Depends how big the models are, how fast you want them to run and how much context you need for your usage. If you're okay with running only smaller models (which are still very capable in general, their main limitation is world knowledge) making very simple inferences at low overall throughput, you can just repurpose the RAM, CPUs/iGPUs and storage in the average setup.
I’m a bit of an optimist. I think this will smack the hands of developers who don’t manage RAM well and future apps will necessarily be more memory-efficient.
Then again, after many, many years of claims that the following year would be the year of the Linux Desktop, there seems to be more and more of a push into that direction. Or at least into a significant increase in market share. We can thank a current head of state for that.
We're not doing Electron because some popular software also using it. We're doing Electron because the ability to create truly cross-platform interfaces with the web stack is more important to us than 300 MB of user memory.
You should check the memory use of that browser tab. You’re not saving much either way running in a browser or in Electron, which is effectively a browser.
The point is being able to write it once with web developers instead of writing it a minimum of twice (Windows and macOS) with much harder to hire native UI developers.
And HTML/CSS/JS are far more powerful for designing than any of SwiftUI/IB on Apple, Jetpack/XML on Android, or WPF/WinUI on Windows, leaving aside that this is what designers, design platforms and AI models already work best with. Even if all the major OSes converged on one solution, it still wouldn't compete on ergonomics or declarative power for designing.
Lol SwiftUI/Jetpack/WPF aren’t design tools, they’re for writing native UI code. They’re simply not the right tool for building mockups.
I don’t see how design workflows matter in the conversation about cross-platform vs native and RAM efficiency since designers can always write their mockups in HTML/CSS/JS in isolation whenever they like and with any tool of their choice. You could even use purely GUI-based approaches like Figma or Sketch or any photo/vector editor, just tapping buttons and not writing a single line of web frontend code.
Who said anything about mockups? Design goes all the way from concept to real-world. If a designer can specify declaratively how that will look, feel, and animate, that's far better than a developer taking a mockup and trying their hardest to approximate some storyboards. Even as a developer working against mockups, I can move much faster with HTML/CSS than I can with native, and I'm well experienced at both (yes, that includes every tech I mentioned). With native, I either have to compromise on the vision, or I have to spend a long time fighting the system to make it happen (...and even then)
The point is you can be lazy and write the app in html and js. Then you dont need to write c, even though c syntax is similar to js syntax and most gui apps wont require needing advanced c features if the gui framework is generous enough.
Now that everyone who cant be bothered, vibe codes, and electron apps are the overevangelized norm… People will probably not even worry about writing js and electron will be here to stay. The only way out is to evangelize something else.
Like how half the websites have giant in your face cookie banners and half have minimalist banners. The experience will still suck for the end user because the dev doesnt care and neither do the business leaders.
But the point isn’t that they’re more different than alike. The point is that learning c is not really that hard it’s just that corporations don’t want you building apps with a stack they don’t control.
If a js dev really wanted to it wouldn’t be a huge uphill climb to code a c app because the syntax and concepts are similar enough.
Who cares about 300Mb, where is that going to move the needle for you? And if the alternative is a memory-unsafe language then 300Mb is a price more than worth paying. Likewise if the alternative is the app never getting started, or being single-platform-only, because the available build systems suck too bad.
There ought to be a short one-liner that anyone can run to get easily installable "binaries" for their PyQt app for all major platforms. But there isn't, you have to dig up some blog post with 3 config files and a 10 argument incantation and follow it (and every blog post has a different one) when you just wanted to spend 10 minutes writing some code to solve your problem (which is how every good program gets started). So we're stuck with Electron.
If the alternative is memory-safe and easy to build, then maybe people will switch. But until it is it's irresponsible to even try to get them to do so.
Like what? Where else (that's a name brand platform and not, like, some obscure blog post's cobbled-together thing) can I start a project, push one button, and get binaries for all major platforms? Until you solve that people will keep using Electron.
There's a world of difference between using a memory safe language and shipping a web browser with your app. I'm pretty sure Avalonia, JavaFX, and Wails would all be much leaner than electron.
The people who hate Electron hate JavaFX just as much if not more, and I'm not sure it would even use less memory. And while the build experience isn't awful, it's still a significant amount of work to package up in "executable" form especially for a platform different from what you're building on, or was until a couple of years ago. And I'm pretty sure Avalonia is even worse.
In practice, you generally see the opposite. The "CPU" is in fact limited by memory throughput. (The exception is intense number crunching or similar compute-heavy code, where thermal and power limits come into play. But much of that code can be shifted to the GPU.)
RAM throughput and RAM footprint are only weakly related. The throughput is governed by the cache locality of access patterns. A program with a 50MB footprint could put more pressure on the RAM bus than one with a 5GB footprint.
Reducing your RAM consumption is not the best approach to reducing your RAM throughput is my point. It could be effective in some specific situations, but I would definitely not say that those situations are more common than the other ones.
I don't understand how this connects to your original claim, which was about trading ram usage for CPU cycles. Could you elaborate?
From what I understand, increasing cache locality is orthogonal to how much RAM an app is using. It just lets the CPU get cache hits more often, so it only relates to throughout.
That might technically offload work to the CPU, but that's work the CPU is actually good at. We want to offload that.
In the case of Electron apps, they use a lot of RAM and that's not to spare the CPU
Only if the software is optimised for either in the first place.
Ton of software out there where optimisation of both memory and cpu has been pushed to the side because development hours is more costly than a bit of extra resource usage.
The tradeoff has almost exclusively been development time vs resource efficiency. Very few devs are graced with enough time to optimize something to the point of dealing with theoretical tradeoff balances of near optimal implementations.
That's fine, but I was responding to a comment that said that RAM prices would put pressure to optimise footprint. Optimising footprint could often lead to wasting more CPU, even if your starting point was optimising for neither.
My response was that I disagree with this conclusion that something like "pressure to optimize RAM implies another hardware tradeoff" is the primary thing which will give, not that I'm changing the premise.
Pressure to optimize can more often imply just setting aside work to make the program be nearer to being limited by algorithmic bounds rather than doing what was quickest to implement and not caring about any of it. Having the same amount of time, replacing bloated abstractions with something more lightweight overall usually nets more memory gains than trying to tune something heavy to use less RAM at the expense of more CPU.
Some of the algorithms are built deep into the runtime. E.g. languages that rely on malloc/free allocators (which require maintaining free lists) are making a pretty significnant tradoff of wasting CPU to save on RAM as opposed to languages using moving collectors.
I'm a bit surprised the article makes no mention of Google's TurboQuant[0] introduced 26 days prior.
Given that TurboQuant results in a 6x reduction in memory usage for KV caches and up to 8x boost in speed, this optimization is already showing up in llama.cpp, enabling significantly bigger contexts without having to run a smaller model to fit it all in memory.
Some people thought it might significantly improve the RAM situation, though I remain a bit skeptical - the demand is probably still larger than the reduction turboquant brings.
TurboQuant is known across the industry to not be state of the art. There are superior schemes for KV quant at every bitrate. Eg, SpectralQuant: https://github.com/Dynamis-Labs/spectralquant among many, many papers.
> Given that TurboQuant results in a 6x reduction in memory usage for KV caches
All depends on baseline. The "6x" is by stylistic comparison to a BF16 KV cache; not a state of the art 8 or 4 bit KV cache scheme.
Current "TurboQuant" implementations are about 3.8X-4.9X on compression (w/ the higher end taking some significant hits of GSM8K performance) and with about 80-100% baseline speed (no improvement, regression): https://github.com/vllm-project/vllm/pull/38479
For those not paying attention, it's probably worth sending this and ongoing discussion for vLLM https://github.com/vllm-project/vllm/issues/38171 and llama.cpp through your summarizer of choice - TurboQuant is fine, but not a magic bullet. Personally, I've been experimenting with DMS and I think it has a lot more promise and can be stacked with various quantization schemes.
The biggest savings in kvcache though is in improved model architecture. Gemma 4's SWA/global hybrid saves up to 10X kvcache, MLA/DSA (the latter that helps solve global attention compute) does as well, and using linear, SSM layers saves even more.
None of these reduce memory demand (Jevon's paradox, etc), though. Looking at my coding tools, I'm using about 10-15B cached tokens/mo currently (was 5-8B a couple months ago) and while I think I'm probably above average on the curve, I don't consider myself doing anything especially crazy and this year, between mainstream developers, and more and more agents, I don't think there's really any limit to the number of tokens that people will want to consume.
Your skepticism is well placed. Every time a new quantization or compression technique drops, the immediate response is to just scale up context length or run a bigger model to fill whatever headroom was freed up. It's Jevons paradox applied to VRAM - efficiency gains get eaten by increased usage almost immediately.
The work going into local models seems to be targeting lower RAM/VRAM which will definately help.
For example Gemma 4 32B, which you can run on an off-the-shelf laptop, is around the same or even higher intelligence level as the SOTA models from 2 years ago (e.g. gpt-4o). Probably by the time memory prices come down we will have something as smart as Opus 4.7 that can be run locally.
Bigger models of course have more embedded knowledge, but just knowing that they should make a tool call to do a web search can bypass a lot of that.
The net effect won’t be a memory use reduction to achieve the same thing. We’ll do more with the same amount of memory. Companies will increase the context windows of their offerings and people will use it.
I am not convinced that more context will be useful, practical use of current models at 1mil context window shows they get less effective as the window grows. Given model progress is slowing as well, perhaps we end up reaching a balance of context size and competency sooner than expected.
Stuff in more code. Stuff in more system prompt. Stuff in raw utf8 characters instead of tokens to fix strawberries. Stuff in WAY more reasoning steps.
Given the current tech, I also doubt there will be practical uses and I hope we’ll see the opposite of what I wrote. But given the current industry, I fully trust them so somehow fill their hardware.
Market history shows us than when the cost of something goes down, we do more with the same amount, not the same thing with less. But I deeply hope to be wrong here and the memory market will relax.
that will only increase the demand for RAM as models will now be usable in scenarios that weren't feasible prior, and the ceiling for model and context size is not even visible at this point
I hate to mention Jevons paradox as it has become cliche by now, but this is a textbook such scenario
As an aside, recently I wanted to refresh my gaming PC, but the price shock and general lack of availability of buying components individually made it seem hardly worth it, so I just kept deferring the project.
Then, mostly by chance, I saw that my local Microcenter had some pre-builts for sale, and I ended up picking one up for <$5k that had "best in slot" components across the board, including a 5090 and even a high-end power supply.
The last time I built a gaming PC was upwards of a decade ago, and at that time the prevailing wisdom was to never buy a pre-built unless you had a massive amount of disposable income and couldn't spare even just one weekend to dedicate to a hobby project that could benefit you for years. Now, it was absolutely a no-brainer.
I did the exact same thing during Covid, the prebuilt ended up being ~20% cheaper than buying the individual components (I needed a full upgrade). Maybe a little less since I could have reused my case.
> and at that time the prevailing wisdom was to never buy a pre-built
That's still the case, and always will be — with a pre-built you're at the very least paying for someone to assemble it for you, so it's always going to be more expensive as a baseline.
Beyond that, the chance they've chosen good components and haven't tried to screw you over on less flashy ones like the motherboard and power supply is low.
That's not to say it's literally impossible to ever find a good deal. You very well might have. Doesn't change anything though.
> with a pre-built you're at the very least paying for someone to assemble it for you, so it's always going to be more expensive as a baseline
Except isn't it possible that pre-built companies actually get better deals on hardware bought in bulk, and therefore could offset the labor costs with cheaper materials?
>CXMT still trails Samsung, SK Hynix, and Micron by approximately three years in advanced DRAM node development, and yield rates on new production lines remain the variable that determines whether capacity targets translate into reliable supply. Liu notes that lines launched in the second half of 2026 are unlikely to change the global supply-demand balance until 2027.
The Verge article talks about demand exceeding supply in 2028. Your article suggests it'll take until 2029 before Chinese production catches up to current technology.
It'll help drive prices down in five yearss, but the Chinese memory production won't be ready and efficient enough to prevent the shortages from continuing to grow.
I wonder if this might motivate to write more memory efficient software. I mean we have so much memory, but even some trivial programs eat hundreds of megabytes of ram.
I'm skeptical - the apps I use either have a) enough lock-in that they don't have the institutional will to optimize or b) a lack of institutional resources to optimize.
Basically, the optimizing that can happen is that I ditch heavy tools in favour of lighter ones, and hopefully enough other people do the same to help lighter tools with finances/dev resources.
I said it for many years that OS developers need to focus on over optimisations. If it wasnt a chip sgortage it would be the ever slowing progress on chip scaling.
But software optimisation helps all hardware and that doesnt drive sales.
Linux however, they dont have to worry about that. Maybe it is finally the era of Haiku OS as the ghost of BeOS rises!
I think RAM shortages would be the least of our problems…
Assuming China takes TSMC in one piece (unlikely without internal sabotage in the best case scenario), it would still probably take years before it produces another high end GPU or CPU.
We would probably be stuck with the existing inventory of equipment for a long time…
I am surprised we consider TSMC like a natural resource: isn't it really a combination of know-how and build-out according to that know-how? If smarts leave the country, perhaps this moves with them.
The risk with China taking over Taiwan is that they mostly expedite their own production research by a couple of years.
It kinda does resemble a natural resource though. The machines and technology in use at TSMC are so insanely complex, that there isn't a single person on earth who knows everything about how it works. TSMC functions only because of all of the pieces of the puzzle being together in the right place and arranged in just the right way. It's a very fragile balance that keeps it all running, and a major disruption could mean we get thrown back by a decade in chip-making technology.
> I am surprised we consider TSMC like a natural resource: isn't it really a combination of know-how and build-out according to that know-how?
Have you seen how many states and countries look enviously at Silicon Valley’s tech companies, China’s manufacturing dominance, or London’s financial sector and try to replicate them?
Turns out it’s way harder than you’d expect.
Hell, Intel can’t match TSMC despite decades of expertise, much greater fame, and regulators happy to change the law and hand out tens of billions in subsidies.
What you say is absolutely true, and is a serious problem—but the way our system operates does not allow us to correct for it.
Anyone trying to spin up a competitor to TSMC would have to first overcome a significant financial hurdle: the capital investment to build all the industrial equipment needed for fabrication.
Then they'd have to convince institutions to choose them over TSMC when they're unproven, and likely objectively worse than TSMC, given that they would not have its decades of experience and process optimization.
This would be mitigated somewhat if our institutions had common-sense rules in place requiring multiple vendors for every part of their supply chain—note, not just "multiple bids, leading to picking a single vendor" but "multiple vendors actively supplying them at all times". But our system prioritizes efficiency over resiliency.
A wealthy nation-state with a sufficiently motivated voter base could certainly build up a meaningful competitor to TSMC over the course of, say, a decade or two (or three...). But it would require sustained investment at all levels—and not just investment in the simple financial sense; it requires people investing their time in education and research. Dedicating their lives to making the best chips in the world. And the only reason that would work is that it defies our system, and chooses to invest in plants that won't be finished for years, and then pay for chips that they know are inferior in quality, because they're our chips, and paying for them when they're lower quality is the only way to get them to be the best chips in the world.
> A wealthy nation-state with a sufficiently motivated voter base could certainly build up a meaningful competitor to TSMC over the course of, say, a decade or two (or three...).
They've been burned before. The DRAM industry has a long history of booms and busts.
Demand increased, everyone built new fabs, then prices dropped and they couldn't pay off their investments. Many went out of business. It happened in the 80s, it happened in the 90s, it happened in the 2000s.
Now there's only three manufacturers left, and they know very well that demand for their product tends to be cyclical.
I've been in the industry for 30 years and I've worked at companies with fabs were demand was high and customers would only get 30% of what they ordered. Then just 2 years later our fab was only running at 50% capacity and losing money. It takes about $20 billion and 3-4 years to make a modern new fab. If you think that AI is a bubble then do you want to be left with a shiny new factory and no products to sell because demand has collapsed?
The same thing everyone who's paying attention to the real world (and not the financial fantasy world) does: that OpenAI's purchase commitments are wildly unrealistic and unsustainable.
What’s the lose scenario for them? They’re basically a cartel, and you need ram irregardless. If they make less it’s still a cost:demand, just not the most optimal for them. They’ve done that math, and figure this is the best risk and reward for them. Your goodwill or opinion doesn’t matter to them, because you need them more than they need you.
It will last forever. After covid, all manufacturers understood the value of limiting supply and extracting profits. Cars used to super cheap before covid, they will never go back to the same levels.
From now on, RAM will always be super costly for consumers, because they can't make massive deals like Apple/OpenAI/etc. We are the bagholders.
When lithium prices decreased over 80% from 2022 to 2025, it was because lithium miners felt altruistic. Car manufacturers were feeling greedy. This is how bad the thinking has gotten.
Covid inflation was because of supply chain disruptions, loose fiscal policy (like Biden's ARP which a Central Bank analysis said added a few % to annual inflation), and money supply expansions. There was less goods and more money. When you go and trade money for goods, it should be obvious what happens.
if a shortage lasts years, it's not a shortage. "The market clearing price of RAM in the face of expected sustained healthy demand should lead to a stable market for years."
even if gaming is and will remain very popular for years, it and the desire to upgrade gaming rigs is still a discretionary activity with more price elasticity of demand than corporate uses for RAM in the dawn of the AI age. gamers live on the margin of this market, where low prices will stimulate upgrades and high prices will lead to holding out. The complaints about price are real, but that segment of the market is some combination of less large and less important.
It’s not merely a “gaming vs data center“. There’s so many other places DRAM and NVM are needed - mobile, automotive, other consumer electronics,… the current situation is that _all_ of that is deprived of the memory that it needs. And much of this is critical to the real economy.
Why are you only talking about gamers? Apple, the most cautious planners in the whole industry have straight up cancelled their 512gb RAM Mac Studio. Don’t ask; they won’t sell you one.
Your instincts are likely right on this one OP. Memory prices surged 80–90% in Q1 2026 compared to Q4 2025, DRAM, NAND, and HBM all at record highs. 3 suppliers for the entire planet?
Of course, alternatively, the AI companies could go bust before finding profitability. Then, there’d be a ton of supply, prices would crash, and one or two of the current memory suppliers would go out of business. After that, the new Chinese memory companies might be producing at volume, and Renesas could be up and running.
At the moment, nothing is certain. Could this last? Sure. Could it not last? Yup.
I've read that the chip manufacturers are looking into high bandwidth flash for on package storage of ai models. That would solve some of the cost issue, flash is significantly cheaper than dram.
There is enough older hardware floating around to last us for decades. You don't need a gaming rig to do 99% of your computing (excluding gaming obviously). Also computers don't really just break. It's mostly the disks that wear out and PSUs that age.
What do you mean, in 5 years? It's not like everyone just bought a new computer. My gut says it's exactly the other way around: most computers are old. They may fail as soon as today.
I just checked my gaming PC I built a few years ago with 64GB of DDR5 RAM, its actually gone up in value, that is unheard of generally.
Think I will scrap my PC and sell its parts.
I wonder if there are any niche companies building decent rigs with DDR3 and 5/6th generation Intel CPUs out there, it is cheap and might be a business opportunity?
I work at an e-waste recycling company. I have several dozen trays of RAM in my inventory, ~90% of it DDR3. DDR3 was selling as of a month ago, but I haven't tried to sell any RAM since. I'm looking forward to doing a huge one this week.
There's a future where RAM makers tool up for this massively increased demand, then the AI companies go broke as the bubble bursts, so RAM is cheap as. So laptop manufacturers get on that and start making laptops with 1TB+ memory so we can run decent LLMs on the local machine. Everyone happy :)
RAM makers are not increasing their capacity. If AI bubble bursts we might see a momentary drop in RAM prices but it won't be dramatic. Return to "normal" is the best scenario I can imagine but my gut tells me we're probably never going back to early 2025 memory prices.
I fear that the real reason we do have a shortage, I mean, the real reason for the demand, is AI companies scooping what they can so that their competitors, whether existing or incumbent, can’t get to it.
I'm personally hoping that one of the AI or data center companies is suddenly unable to pay for their bills and deflate the entire industry. Probably the only hope of things getting better before the 2030s.
This is simple extrapolation from current demand, nothing more. And that's a borderline silly analysis because it assumes the AI bubble won't burst. The great misadventure in the Persian Gulf probably accelerates that because we're almost certainly going to be facing a recession.
Another thing I've been thinking about is what happens when the next generation of NVidia chips comes out? I suspect NVidia is going to delay this to milk the current demand but at some point you'll be able to buy something that's better than the H100 or B200 or whatever the current state-of-the-art for half the price. And what's that going to do to the trillions in AI DC investment?
I'm interested when the next bump in DRAM chip density is coming. That's going to change things although it seems like much of production has moved from consumer DRAM chips to HBM chips. So maybe that won't help at all.
I do think that companies will start seeing little ot no return from billions spent on AI and that's going to be aproblem. I also think that the hudnreds of billions of capital expenditure of OpenAI is going to come crashing down as there just isn't any even theoretical future revenue that can pay for all that.
I fear the author and most commenters are not aware of the law of demand and supply. If there is demand for consumer RAM, there will be supply for consumer RAM. It just takes time and risk-assessment to scale up operations.
We have RAM shortage now, we will have very cheap RAM tomorrow. It’s not like production is bottlenecked by raw materials. Chip companies just need to assess if the demand by AI companies will last so it’s better to scale up, or perhaps they should wait it out instead of oversupplying and cutting into their profits.
We're talking about advanced semiconductor manufacture. It takes years and 100s millions to billions of dollars to scale up operations. That's something you don't do unless you know there's demand to sustain it in future.
>I fear the author and most commenters are not aware of the law of demand and supply
I cannot stand how you and people like you try to justify everything by supply and demand. Also you act like it's some natural law of nature. It's not a law of nature- if you took an economics class you would realize it's try to maximize PROFIT. It's not for the good of the people.
All of these things are a CHOICE that people are making to now completely screw the average person for, again, the needs of big corporations and the top 0.01%.
It doesn't stop there though. OpenAI is currently mired in a capital crunch. Their last round just about sucked all the dry powder out of the private markets. Folks are now starting to ask difficult questions about their burn rate and revenue. It is increasingly looking like they might not commit to the purchase order they made which kick-started this whole panic over RAM.
Soo ... how sure are we that the memory makers themselves are not going to be the ones holding the bag?
If they could make this stuff and sell it to regular people a decade ago for very palatable prices, why do they come up with the idea that this is the technology of the gods, unaffordable by mere mortals?
because the gods want it all and are willing to pay top dollar.
I wonder whether this is some kind of a racket.
No.
Heck, I have a phone with a 16bit memory bus for instance. The high(ish) clock rate only makes up the difference slightly.
But with general prices on all components going up, it might not be such a big factor any more.
HBM migght make sense for higher end products which can free up space for the lower end that will never use the tech.
That's correct if you're targeting gamers, but local AI inference changes this picture substantially.
5090 has 1.8 TB/s?
Even a RTX 5080 has a lower memory throughput than a Radeon VII from 2019, 7 years ago, while being much more expensive.
The memory throughput of GPUs per dollar has regressed greatly during the last 5 years, despite the fact that the widths of the GPU memory interfaces have been reduced, in order to decrease the production costs.
RTX 5080 has a 256-bit memory interface, while the much cheaper Radeon VII had an 1024-bit memory interface. RTX 5080 has almost 4-times faster memories than Radeon VII, but it has not used this to increase the memory throughput, but only to reduce the production costs, while simultaneously increasing the product price.
And it's faster for gaming, I guess? Which is what matters for the typical user.
Anyway you can buy much faster GPUs now than in 2019. They are also much more expensive, yes.
I suppose that most games are limited by computation, so they are indeed much faster on modern GPUs.
However, there are applications that are limited by memory throughput, not by computation, including AI inference and many scientific/technical computing applications.
For such applications, old GPUs with higher memory throughput are still faster.
This is why I am still using an old Radeon VII and a couple of other ancient AMD GPUs with high memory throughput.
Last year I have bought an Intel GPU, which is still slower than my old GPUs, but it at least had very good performance per dollar, competitive with that of the old GPUs, because it was very cheap, while the current AMD and especially NVIDIA GPUs have poor performance per dollar.
5090s are certainly expensive compared to most other GPUs, but not expensive enough to be unobtanium for nearly any professional who could utilize one as part of their job
R9700 has 32GB and is cheaper than most NVidia consumer GPUs, even though it's a "pro".
AMD Hawaii GPUs still had 1:2 FP64:FP32, while the consumer variant of Radeon VII dropped to 1:4. The following AMD consumer GPUs dropped the FP64 performance to levels that are not competitive with CPUs.
Nowadays the only consumer GPUs with decent FP64 performance are the Intel Battlemage GPUs, which have a 1:8 performance ratio, which provides very good performance per dollar.
Reason number 7,322 why US-style ultracapitalism is self-destructive, anti-social, and dystopian.
That is, memory capacity is reserved for datacenters yet to be built, but this will do weird things if said datacenter construction is postponed or cancelled altogether.
It says that in 2025, Netherlands was a net exporter of electricity (~14,000 GWh). My guess: Where they want to build data centers, the grid cannot handle it, but the overall system has more than enough power to build data centers. Do you think that sounds like a resonable guess?
Are the Netherlands a large proportion of global datacenters?
The value of an IX isn't just in the IX itself, but also in the presence of hundreds of parties for direct peering, and excellent connectivity to the rest of the world.
It makes a lot of sense to build your DC near one - even if you have no intention of actually participating in the IX itself.
They don't need entire IX worth of connectivity. You're sending mostly text back and forth and any media is in far lower volume than even normal far less dense DC would generate, all the major traffic is inside the AI DC.
All it needs is fiber to nearest IX
In most other places the percentage is significantly less than that and then you can easily add more of the cheap-but-intermittent stuff because a cloudy day only requires you to make up a 10% shortfall instead of a 50% one, which existing hydro or natural gas plants can handle without new storage when there are more of them to begin with.
I calculate about 43.5% was solar or wind. What is way crazier is the "bend in the curve" of production sources in the last 10 years. Look here at how fast solar and wind is growing! https://en.wikipedia.org/wiki/File:Netherlands_electricity_g...
What's more common is that they don't have the transmission capacity itself, but that one's pretty easy in this case too, because what that means is that you have an existing transmission line which is already near capacity with generation on one end and customers on the other. So then you just build the data center on the end of the transmission line where the generation is rather than the end where the existing customers are, at which point you can add new generation anywhere you want -- and if you put it near the existing customers you've just freed up transmission capacity because you now have new customers closer to the existing generation and new generation closer to the existing customers.
In every country? Citation needed.
Had we done more 10 years ago we would have been better of. The second best time to start is now.
(We used to build it at a fraction of the cost and less than half of the time that we do with our modern fuckups and fuel can come from just about anywhere if need be. It might be a lot more expensive than the stuff kazachstan and still be a fraction of the cost.)
I think ideally we would've done both to press the cost of nuclear down and given the fact that the renewables rollout turned out to be a lot lot more expensive than proponents claimed it would be whilst still tying us up into gass to cover winter.
Then why all the anti-coal mining diktats coming down from Brussels?
Brussels is trying to reduce "tiny" to zero, because of this: https://en.wikipedia.org/wiki/Tragedy_of_the_commons
China, like Brussels, is trying to reduce coal for similar reasons. They don't like the air pollution health hazard (fully believable), and they say they don't like global warming (somewhat believable).
Renewables deployment is happening fast. Grid upgrades are not. Batteries .. it depends.
Even nuclear darling France has set solar records: https://www.pv-magazine.com/2026/04/15/france-germany-set-da...
The problem in the EU is not renewables, it's the same problem that Democratic states in the US face. Regulations and permitting hurdles that block private renewable energy developers.
(Well that and collusion)
If they had actually been communicating or colluding with each other, they would have put the screws to him, making it harder for OpenAI to assert control over the vast majority of the DRAM market.
Failing that, you'd like to think a regulatory agency somewhere would step in to keep a single player from hosing everybody else, but...
Up until AI there weren't really players being able to gobble 40% of the market so nobody was looking.
I don’t buy it that two of the largest manufacturers of DRAM in the world, from the same country, didn’t know this. Even of you ignore each company’s intelligence teams, that’s also the job of the country’s internal intelligence services, to make sure they know what all companies are doing and then make it so they have the best leverage to gain as much as possible. Both companies would have known “somehow” and played hardball.
By spying?
You get market signals that the demand is there, you acquire the necessary capital, you spend 5 years to build capacity, but guess what, 5 other market players did the same thing. So now you are doomed, because the market is flooded and you have low cash flow since you need to drop prices to compete for pennies.
Now you cannot find capital, you don't invest, but guess what, neither your competitors did. So now the demand is higher than the supply. Your price per unit sold skyrocketed, but you don't have enough capacity!
Rinse and repeat.
Capitalists claim that this is optimal.
If anything, it shows it's possible for you to arbitrage this and in doing so help "smooth out the cycle."
At least with capitalism you have many different people with different perspectives on the risk making independent bets. That mitigates the more extreme negative outcomes.
We don't even expect companies to plan long-term anymore, it's just moving wealth as fast as possible.
That isn't really a change, very few people could ever have been said to be ideological capitalists. (capitalist is not a word with a hard definition, but I'm considering it a different thing than the more modern pure libertarian zero-regulation ideology)
Because that does not happen exactly as you say for all players. The demand signals will be processed and long-term risk is balanced against short-term gain in a distributed fashion, so not everyone will do the same.
It's more optimal than planned economies until we have AI planned economies with realtime feedback, I guess.
Consumers get cheap goods during oversupply and most inefficient companies get elliminated during bust while consolidation leads to economies of scale.
There is an alternative where legislation dampens this behavior but the short term profits will be lower. Hence the hawks don’t like it.
Potentially. Well meaning and thought out legislation still distorts the markets, possibly making things objectively worse.
> Capitalists claim that this is optimal.
Compared to starving under communism coz someone at top got the number wrong, yes. And it only really happens when there are massive, unpredictable market movements and governments not doing their job. Govt should look at the whole thing and just say "no", blame them.
No market system self regulates well enough, and it's government job to file down the edge cases like this. But the revolution happened in country which has two utterly incompetent parties, both in pockets of billionaires, fighting for power, and the clowns from one that won last battle use AI to smokescreen the economic growth their actions cratered
The memory makers specifically did not scale up capacity to avoid being left holding the bag.
It's worse. HBM have lower yields so they are essentially making less GB per wafer too
Do recent actions of Open AI give you the impression of a company that believes it is about to attain AGI imminently?
Hell, all it matters to investors is not being left holding the bag in the end so they don't even need to believe it
We've been projecting both FTL and AGI as future possibilities for almost 100 years now. Do LLMs get us a lot closer to AGI? I think they get us a little closer and Moore's "law" making compute faster probably is a much bigger factor, but I think we're still a very very long ways away.
I think Douglas Hofstadter satisfactorily answered this question.
> We can't even answer if we have free will or not.
Sure we can, it's just that most people don't like the answer.
The Fiji XT architecture after it had 512GB/S on a 4096b HBM bus in 2015.
The Vega architecture did have 400GB/s or so in 2017, which was a bit of a downgrade.
At least as I understand it.
Very few applications other than GPUs need HBM.
The real issue is everyone wanting to upgrade to hbm, ddr5, and nvme5 at the same time.
The specific mix of factors could change at any time, but the supply chain is relatively inelastic, it will take some time to show up on price labels.
this view isn't updated correctly post-claude code and codex. there will clearly be sufficient demand.
OpenAI (or whoever) crashes and can't pay for the order leaving the memory makers in a tough spot.
Oh noes! Think of a poor memory makers!
The amount of money flowing both from the AI bubble and from quite literally scalping both the server and consumer market... They gambled on the opportunity and if they fail - it's their problem.
Edit: also, that demand pressure is going to be applied constantly; there isn’t going to be a shock, it’s just going to keep prices high longer.
Are they really such a big RAM buyer?
> OpenAI’s rapid growth, fueled by the success of ChatGPT and other AI products, led to a landmark agreement in October to purchase 900,000 DRAM wafers per month from Samsung and SK Hynix—amounting to roughly 40% of global supply. This surge in demand, coupled with limited manufacturing capacity, sent prices for memory kits skyrocketing. [0]
[0]: https://peq42.com/blog/openai-canceling-many-large-purchase-...
If I booked half a hotel's rooms then suddenly said "yeah never mind. Half my friends cancelled and we're not staying", basically any hotel would be coming at me for my money because there's no way they can fill their rooms now and they're losing revenue. But OpenAI can really get the whole world to pivot towards it then say "cool but we don't need your product anymore" and RAM makers are just going to let it go.
Whoever decided that was a good idea needs to be fired and publicly shamed.
if anything, OpenAi might be in on it
I hope they do, they did not have to agree to sell so much RAM to one customer. They’ve been caught colluding and price fixing more than once, I hope they take it in the shorts and new competitors arise or they go bankrupt and new management takes over the existing plants.
Don’t put all your eggs in the one basket is how the old saying goes.
Wasn't the problem here that OpenAI was negotiating with Samsung and SK Hynix at the same time without the other one knowing about it? People only realized the implications when they announced both deals at once.
We aren't. The remaining memory manufacturers fear getting caught in a "pork cycle" yet again - that is why there's only the three large ones left anyway.
China has memory makers who are creeping up through the stages of production maturity, and once they hit then there's no going back.
If the existing makers can't meet supply such that Chinese exports get their foot in the door, they may find they never get ahead again due to volume - that domestic market is huge so they have scale, and the gaming market isn't going to care because they get anything at the moment, which is all you'll need for enterprise to say "are we really afraid of memory in this business?"
The customer ran out of money. In terms of where you are in line of debtors when you haven't even delivered the product to a customer, it's so far back as to be assured you won't get your money.
If the memory makers got a deposit from OpenAI as part of this deal, that is likely to be the only money they will get for any undelivered memory, particularly if OpenAI runs out of capital.
Which also explains why production is falling behind demand, companies aren't going to sink billions into creating product for a market that could dry up overnight.
They act as a de-facto monopoly and milk us. Why is this allowed?
Nobody is "allowing" this. It's a natural property of being both advanced technology and a commodity at the same time.
https://en.wikipedia.org/wiki/Onion_Futures_Act
Recently they had a second price fixing lawsuit thrown out (in the US).
Now with the state of things I'm sure another lawsuit will arrive and be thrown out because the government will do anything to keep the AI bubble rolling and a price fixing suit will be a threat to national security, somehow. Obviously thats speculative and opinion but to be clear, people are allowing it. There are and more so were things that could be done.
It started with raegan, and even parties on the “left” in the west believe in it with very few exceptions.
The thing that enables this is pretty obvious. The population is divided into two camps, the first of which holds the heuristic that regulations are "communism and totalitarianism" and this camp is used to prevent e.g. antitrust rules/enforcement. The second camp holds the heuristic that companies need to be aggressively "regulated" and this camp is used to create/sustain rules making it harder to enter the market.
The problem is that ordinary people don't have the resources to dive into the details of any given proposal but the companies do. So what we need is a simple heuristic for ordinary people to distinguish them: Make the majority of "regulations" apply only to companies with more than 20% market share. No one is allowed to dump industrial waste in the river but only dominant companies have bureaucratic reporting requirements etc. Allow private lawsuits against dominant companies for certain offenses but only government-initiated prosecutions against smaller ones, the latter preventing incumbents from miring new challengers in litigation and requiring proof beyond a reasonable doubt.
This even makes logical sense, because most of the rules are attempts to mitigate an uncompetitive market, so applying them to new entrants or markets with >5 competitors is more likely to be deleterious, i.e. drive further consolidation. Whereas if the market is already consolidated then the thicket of rules constrains the incumbents from abusing their dominance in the uncompetitive market while encouraging new entrants who are below the threshold.
How is this more efficient? You'd still be applying all of the inefficient regulatory rules intended to mitigate a lack of competition to the smaller companies trying to sustain a competitive market, and those rules are much more deleterious for smaller entities than higher tax rates.
If you have $100M in fixed regulatory overhead for a larger company with $10B in profit, it's only equivalent to a 1% tax. The same $100M for a smaller company with $50M in profit is a 200% tax. There is no tax rate you can impose on the larger company to make up for it because the overhead destroys the smaller company regardless of what you do to the larger one.
Oh no!
If they add enough capacity to meet current demand quickly then if demand crashes they still have billions of dollars in loans used to build capacity for demand that no longer exists and then they go bankrupt.
The biggest problem is predicting future demand, because it often declines quickly rather than gradually.
If you suppose you have cracked the smooth-ramping problem, perhaps you should throw your hat in the ring and soak up all the pent-up demand that SK Hynix, Samsung and Micron are neglecting.
If he can do all that that fast, the RAM makers should be able to at least 1000X their fab capacity on earth in one year. One year for scaling up existing tech is an eternity compared to Elon's timeframe for moon-fabs given the relative complexity of the challenge.
There’s virtually infinite capital: if needed, more can be reallocated from the federal government (funded with debt), from public companies (funded with people’s retirement funds), from people’s pockets via wealth redistribution upwards, from offshore investment.
They will be allowed to strangle any part of the supply chain they want.
Another point is I often see the money argument - like country X has more money, so they can afford to do more and better R&D, make more stuff.
This stuff comes out of factories, that need to be built, the machinery procured, engineers trained and hired.
[1]https://www.tomshardware.com/tech-industry/semiconductors/ym...
> more can be reallocated from the federal government (funded with debt)
While this is the most reliable funding, it's still not very accessible. OpenAI is a money pit, and their demands are growing quickly. The US government has started a bunch of very expensive spending. If OpenAI were to require yearly bundles of it's recent "$120B" deal, that's 6% of the US' discretionary budget. 12.5% of the non-military discretionary budget. (And the military is going to ask for a lot more money this year) Even the idea of just issuing more debt is dubious because they're going to want to do that to pay for the wars that are rapidly spiralling out of control.
None of this is saying that the US government can't or wouldn't pay for it, but it's non trivial and it's unclear how much Altman can threaten the US government "give me a trillion dollars or the economy explodes" without consequences.
Further deficit-spending isn't without it's risks for the US government either. Interests rates are already creeping up, and a careless explosion of deficit may well trigger a debt crisis.
> from public companies (funded with people’s retirement funds)
This would be at great cost. OpenAI would need to open up about it's financial performance to go public itself. With it's CFO being put on what is effectively Administrative Leave for pushing against going public, we can assume the financials are so catastrophic an IPO might bomb and take the company down with it. Nobody's going to be investing privately in a company that has no public takers.
Getting money through other companies is also running into limits. Big Tech has deep pockets but they've already started slowing down, switching to debt to finance AI investment, and similarly are increasingly pressured by their own shareholders to show results.
> from people’s pockets via wealth redistribution upwards
The practical mechanism of this is "AI companies raise their prices". That might also just crash the bubble if demand evaporates. For all the hype, the productivity benefit hasn't really shown up in economy-wide aggregates. The moment AI becomes "expensive", all the casual users will drop it. And the non-casual users are likely to follow. The idea of "AI tokens" as a job perk is cute, but exceedingly few are going to accept lower salary in order to use AI at their job.
There's simply not much money to take out of people's pockets these days, with how high cost of living has gotten.
> from offshore investment.
This is a pretty good source of money. The wealthy Arabian oil states have very deep slush funds, extensively investing in AI to get ties to US businesses and in the hope of diversifying their resource economies.
...
...
"Was". Was a good source of money.
Just look at Cuba, which could be a very rich country and one of the prime tourist destinations of the world.
Claude Max subscriptions have gone up, but do you think every Netflix user will pay for one?..
https://www.tomshardware.com/tech-industry/artificial-intell...
RAM is built on a foundation of sand.
I would like a source for that statement. Additionally, I want to know by who? Because it certainly isn't end users. Inflating token usage doesn't make it any more economically viable if your user base, b2b or not, hasn't increased with it. On the contrary, that is a worse scenario for providers.
the hope is that Ai is "the next semiconductor" and "the next internet"
Not exactly.
LLMs are already quite useful today if you use them as a tool, so they are there to stay. The remaining problem is scalability, a.k.a. how to make LLMs cheap to use.
But scalability is not really a requirement when you look the bigger picture. If smaller software company/projects can't afford to use AI, the bigger ones might just. Eventually they will discover variable use cases for such tech, even if it only serves big firms i.e. defense, resource extraction, war, finance etc.
To the other end, if scalability is achieved, the use of LLM products will be cheaper too, so smaller project can also use them. But of course, if LLM usage is too cheap, then many were-to-be-consumers will just create software projects by themselves at their homes.
Supposedly AI drives down the cost of producing software,not the "price".
> How are software companies going to make enough revenue to pay for AI, when the amount of money being spent on AI is already multiples of the current total global expenditure on software?
Currently, the cost of AI is between $20/month and around $200/month per developer.
I think the huge billions you're seeing in the news are the investment cost on AI companies, who are burning through cash to invest in compute infrastructure to allow both training and serving users.
> This demand for RAM is built on a foundation of sand, there will be a glut of capacity when it all shakes out.
Who knows? What I know is that I need >64GB of RAM to run local models, and that means most people will need to upgrade from their 8Gb/16GB setup to do the same. Graphics cards follow mostly the same pattern.
You can run huge local models slowly with the weights stored on SSDs.
Nowadays there are many computers that can have e.g. 2 PCIe 5.0 SSDs, which allow a reading throughput of 20 to 30 gigabyte per second, depending on the SSDs.
There are still a lot of improvements that can be done to inference back-ends like llama.cpp to reach the inference speed limit determined by the SSD throughput.
It seems that it is possible to reach inference speed in the range from a few seconds per token to a few tokens per second.
That may be too slow for a chat, but it should be good enough for an AI coding assistant, especially if many tasks are batched, so that they can progress simultaneously during a single read pass over the SSD data.
Depends how big the models are, how fast you want them to run and how much context you need for your usage. If you're okay with running only smaller models (which are still very capable in general, their main limitation is world knowledge) making very simple inferences at low overall throughput, you can just repurpose the RAM, CPUs/iGPUs and storage in the average setup.
Then again, after many, many years of claims that the following year would be the year of the Linux Desktop, there seems to be more and more of a push into that direction. Or at least into a significant increase in market share. We can thank a current head of state for that.
At a cost of simplicity and beauty. And two lost decades of mediocre performance. Sigh
And hopefully kill Electron.
I have never seen the point of spinning up a 300+Mb app just to display something that ought to need only 500Kb to paint onto the screen.
We're not doing Electron because some popular software also using it. We're doing Electron because the ability to create truly cross-platform interfaces with the web stack is more important to us than 300 MB of user memory.
May I never have to use or work on your project's software.
I don’t see how design workflows matter in the conversation about cross-platform vs native and RAM efficiency since designers can always write their mockups in HTML/CSS/JS in isolation whenever they like and with any tool of their choice. You could even use purely GUI-based approaches like Figma or Sketch or any photo/vector editor, just tapping buttons and not writing a single line of web frontend code.
It's bad enough having to run one boated browser, now we have to run multiples?
This is not the right path.
Now that everyone who cant be bothered, vibe codes, and electron apps are the overevangelized norm… People will probably not even worry about writing js and electron will be here to stay. The only way out is to evangelize something else.
Like how half the websites have giant in your face cookie banners and half have minimalist banners. The experience will still suck for the end user because the dev doesnt care and neither do the business leaders.
If a js dev really wanted to it wouldn’t be a huge uphill climb to code a c app because the syntax and concepts are similar enough.
About the only thing they share is curly braces.
This comment makes no sense.
There ought to be a short one-liner that anyone can run to get easily installable "binaries" for their PyQt app for all major platforms. But there isn't, you have to dig up some blog post with 3 config files and a 10 argument incantation and follow it (and every blog post has a different one) when you just wanted to spend 10 minutes writing some code to solve your problem (which is how every good program gets started). So we're stuck with Electron.
and if not?
If the alternative is memory-safe and easy to build, then maybe people will switch. But until it is it's irresponsible to even try to get them to do so.
From what I understand, increasing cache locality is orthogonal to how much RAM an app is using. It just lets the CPU get cache hits more often, so it only relates to throughout.
That might technically offload work to the CPU, but that's work the CPU is actually good at. We want to offload that.
In the case of Electron apps, they use a lot of RAM and that's not to spare the CPU
Ton of software out there where optimisation of both memory and cpu has been pushed to the side because development hours is more costly than a bit of extra resource usage.
Pressure to optimize can more often imply just setting aside work to make the program be nearer to being limited by algorithmic bounds rather than doing what was quickest to implement and not caring about any of it. Having the same amount of time, replacing bloated abstractions with something more lightweight overall usually nets more memory gains than trying to tune something heavy to use less RAM at the expense of more CPU.
Of course memory safety has a quality all its own.
Given that TurboQuant results in a 6x reduction in memory usage for KV caches and up to 8x boost in speed, this optimization is already showing up in llama.cpp, enabling significantly bigger contexts without having to run a smaller model to fit it all in memory.
Some people thought it might significantly improve the RAM situation, though I remain a bit skeptical - the demand is probably still larger than the reduction turboquant brings.
[0] https://news.ycombinator.com/item?id=47513475
> Given that TurboQuant results in a 6x reduction in memory usage for KV caches
All depends on baseline. The "6x" is by stylistic comparison to a BF16 KV cache; not a state of the art 8 or 4 bit KV cache scheme.
Current "TurboQuant" implementations are about 3.8X-4.9X on compression (w/ the higher end taking some significant hits of GSM8K performance) and with about 80-100% baseline speed (no improvement, regression): https://github.com/vllm-project/vllm/pull/38479
For those not paying attention, it's probably worth sending this and ongoing discussion for vLLM https://github.com/vllm-project/vllm/issues/38171 and llama.cpp through your summarizer of choice - TurboQuant is fine, but not a magic bullet. Personally, I've been experimenting with DMS and I think it has a lot more promise and can be stacked with various quantization schemes.
The biggest savings in kvcache though is in improved model architecture. Gemma 4's SWA/global hybrid saves up to 10X kvcache, MLA/DSA (the latter that helps solve global attention compute) does as well, and using linear, SSM layers saves even more.
None of these reduce memory demand (Jevon's paradox, etc), though. Looking at my coding tools, I'm using about 10-15B cached tokens/mo currently (was 5-8B a couple months ago) and while I think I'm probably above average on the curve, I don't consider myself doing anything especially crazy and this year, between mainstream developers, and more and more agents, I don't think there's really any limit to the number of tokens that people will want to consume.
mind that you're quoting marketing material that's largely based on unfair baseline testing (like comparing 4 bit vs 32 bit to get "8x speed")
https://www.youtube.com/watch?v=haoAI2lIZ74
For example Gemma 4 32B, which you can run on an off-the-shelf laptop, is around the same or even higher intelligence level as the SOTA models from 2 years ago (e.g. gpt-4o). Probably by the time memory prices come down we will have something as smart as Opus 4.7 that can be run locally.
Bigger models of course have more embedded knowledge, but just knowing that they should make a tool call to do a web search can bypass a lot of that.
That is the sad reality of the future of memory.
Given the current tech, I also doubt there will be practical uses and I hope we’ll see the opposite of what I wrote. But given the current industry, I fully trust them so somehow fill their hardware.
Market history shows us than when the cost of something goes down, we do more with the same amount, not the same thing with less. But I deeply hope to be wrong here and the memory market will relax.
I hate to mention Jevons paradox as it has become cliche by now, but this is a textbook such scenario
Then, mostly by chance, I saw that my local Microcenter had some pre-builts for sale, and I ended up picking one up for <$5k that had "best in slot" components across the board, including a 5090 and even a high-end power supply.
The last time I built a gaming PC was upwards of a decade ago, and at that time the prevailing wisdom was to never buy a pre-built unless you had a massive amount of disposable income and couldn't spare even just one weekend to dedicate to a hobby project that could benefit you for years. Now, it was absolutely a no-brainer.
That's still the case, and always will be — with a pre-built you're at the very least paying for someone to assemble it for you, so it's always going to be more expensive as a baseline.
Beyond that, the chance they've chosen good components and haven't tried to screw you over on less flashy ones like the motherboard and power supply is low.
That's not to say it's literally impossible to ever find a good deal. You very well might have. Doesn't change anything though.
Except isn't it possible that pre-built companies actually get better deals on hardware bought in bulk, and therefore could offset the labor costs with cheaper materials?
[0] https://techwireasia.com/2026/04/chinese-memory-chips-ymtc-c...
>CXMT still trails Samsung, SK Hynix, and Micron by approximately three years in advanced DRAM node development, and yield rates on new production lines remain the variable that determines whether capacity targets translate into reliable supply. Liu notes that lines launched in the second half of 2026 are unlikely to change the global supply-demand balance until 2027.
The Verge article talks about demand exceeding supply in 2028. Your article suggests it'll take until 2029 before Chinese production catches up to current technology.
It'll help drive prices down in five yearss, but the Chinese memory production won't be ready and efficient enough to prevent the shortages from continuing to grow.
Basically, the optimizing that can happen is that I ditch heavy tools in favour of lighter ones, and hopefully enough other people do the same to help lighter tools with finances/dev resources.
But software optimisation helps all hardware and that doesnt drive sales.
Linux however, they dont have to worry about that. Maybe it is finally the era of Haiku OS as the ghost of BeOS rises!
Assuming China takes TSMC in one piece (unlikely without internal sabotage in the best case scenario), it would still probably take years before it produces another high end GPU or CPU.
We would probably be stuck with the existing inventory of equipment for a long time…
The risk with China taking over Taiwan is that they mostly expedite their own production research by a couple of years.
Have you seen how many states and countries look enviously at Silicon Valley’s tech companies, China’s manufacturing dominance, or London’s financial sector and try to replicate them?
Turns out it’s way harder than you’d expect.
Hell, Intel can’t match TSMC despite decades of expertise, much greater fame, and regulators happy to change the law and hand out tens of billions in subsidies.
Anyone trying to spin up a competitor to TSMC would have to first overcome a significant financial hurdle: the capital investment to build all the industrial equipment needed for fabrication.
Then they'd have to convince institutions to choose them over TSMC when they're unproven, and likely objectively worse than TSMC, given that they would not have its decades of experience and process optimization.
This would be mitigated somewhat if our institutions had common-sense rules in place requiring multiple vendors for every part of their supply chain—note, not just "multiple bids, leading to picking a single vendor" but "multiple vendors actively supplying them at all times". But our system prioritizes efficiency over resiliency.
A wealthy nation-state with a sufficiently motivated voter base could certainly build up a meaningful competitor to TSMC over the course of, say, a decade or two (or three...). But it would require sustained investment at all levels—and not just investment in the simple financial sense; it requires people investing their time in education and research. Dedicating their lives to making the best chips in the world. And the only reason that would work is that it defies our system, and chooses to invest in plants that won't be finished for years, and then pay for chips that they know are inferior in quality, because they're our chips, and paying for them when they're lower quality is the only way to get them to be the best chips in the world.
They have the other system.
> A wealthy nation-state with a sufficiently motivated voter base could certainly build up a meaningful competitor to TSMC over the course of, say, a decade or two (or three...).
Demand increased, everyone built new fabs, then prices dropped and they couldn't pay off their investments. Many went out of business. It happened in the 80s, it happened in the 90s, it happened in the 2000s.
Now there's only three manufacturers left, and they know very well that demand for their product tends to be cyclical.
https://imgur.com/a/cDLoeZm
I've been in the industry for 30 years and I've worked at companies with fabs were demand was high and customers would only get 30% of what they ordered. Then just 2 years later our fab was only running at 50% capacity and losing money. It takes about $20 billion and 3-4 years to make a modern new fab. If you think that AI is a bubble then do you want to be left with a shiny new factory and no products to sell because demand has collapsed?
The lawsuits in the past prove that statement to not be basically but actually.
From now on, RAM will always be super costly for consumers, because they can't make massive deals like Apple/OpenAI/etc. We are the bagholders.
Have they really ever been cheap? Also Tesla 3 is cheaper now, Yaris is still cheap as well.
even if gaming is and will remain very popular for years, it and the desire to upgrade gaming rigs is still a discretionary activity with more price elasticity of demand than corporate uses for RAM in the dawn of the AI age. gamers live on the margin of this market, where low prices will stimulate upgrades and high prices will lead to holding out. The complaints about price are real, but that segment of the market is some combination of less large and less important.
letting the market set prices ensures that the chips go to the critical markets and uses. less critical uses will not allocate funds for purchases.
Everybody’s getting pinched, not just the gamers.
At the moment, nothing is certain. Could this last? Sure. Could it not last? Yup.
All computers in my household are 8+ years old.
Think I will scrap my PC and sell its parts.
I wonder if there are any niche companies building decent rigs with DDR3 and 5/6th generation Intel CPUs out there, it is cheap and might be a business opportunity?
There's a future where RAM makers tool up for this massively increased demand, then the AI companies go broke as the bubble bursts, so RAM is cheap as. So laptop manufacturers get on that and start making laptops with 1TB+ memory so we can run decent LLMs on the local machine. Everyone happy :)
I don't want to pay more because of AI companies driving the price up. That is milking.
Another thing I've been thinking about is what happens when the next generation of NVidia chips comes out? I suspect NVidia is going to delay this to milk the current demand but at some point you'll be able to buy something that's better than the H100 or B200 or whatever the current state-of-the-art for half the price. And what's that going to do to the trillions in AI DC investment?
I'm interested when the next bump in DRAM chip density is coming. That's going to change things although it seems like much of production has moved from consumer DRAM chips to HBM chips. So maybe that won't help at all.
I do think that companies will start seeing little ot no return from billions spent on AI and that's going to be aproblem. I also think that the hudnreds of billions of capital expenditure of OpenAI is going to come crashing down as there just isn't any even theoretical future revenue that can pay for all that.
They'll just spend whatever they were planning to spend and get more performance.
We have RAM shortage now, we will have very cheap RAM tomorrow. It’s not like production is bottlenecked by raw materials. Chip companies just need to assess if the demand by AI companies will last so it’s better to scale up, or perhaps they should wait it out instead of oversupplying and cutting into their profits.
There are two RAM suppliers...
I cannot stand how you and people like you try to justify everything by supply and demand. Also you act like it's some natural law of nature. It's not a law of nature- if you took an economics class you would realize it's try to maximize PROFIT. It's not for the good of the people.
All of these things are a CHOICE that people are making to now completely screw the average person for, again, the needs of big corporations and the top 0.01%.