The memory shortage is causing a repricing of consumer electronics

https://davidoks.blog/p/ai-is-killing-the-cheap-smartphone

by d0ks • 2 months ago

544 648 2 months ago

simonw• 2 months ago

The headline here under-serves the article in my opinion: this is a fascinating, deep explanation of how the memory market works and why increased demand for HBM (used by big GPU racks) hurts the availability of wafers for DDR and LPDDR (used by laptops and phones).

cm2012• 2 months ago

This author has outstanding articles regularly

gblargg• 2 months ago

This also put a lot into perspective:

these memory makers have learned a very particular lesson from the unforgiving history [deep drops in demand] of their industry: always leave demand unmet

I can't blame them for keeping some reserve demand ready so they keep having customers over the years.

Ilaurens• 2 months ago

I'm no expert analyst on this topic, but I'm worried this time might be different for them. China finally has the technological capability to challenge this time around and they will gobble up the unmet demand, allowing them to get more experience and more money to catch up.

Aerroon4• 2 months ago

I wonder if this is actually true in the long-term though. If they were to flood the market with lots of high capacity memory, then I think our programs would start using more memory too. As a result we might end up needing more memory faster compared to if they keep demand unmet.

Just consider that a chat application today takes more memory than a full 3D game with thousands of users (including chat) + the operating system used to ~20 years ago. If 128 GB of RAM were the norm then there's a chance we might expect to buy 128 GB of RAM too.

But I suppose it's really a question of how many dollars we expect to spend on memory rather than the specific amounts of memory.

nashashmi• 2 months ago

always leave demand unmet

And then allow small low cost manufacturers to get the rest of the market... like China has been doing this whole time.

pseudohadamard2• 2 months ago

It's more than that though. Memory is a series of boom-and-bust cycles. Also, it costs around $15 billion and takes 2-3 years to build a fab. So if you were to start now, you'd have a pretty good chance of hitting the next bust right after spending $15 billion. It's not necessarily some sinister conspiracy, it's because execs are reluctant to be the ones to take the blame for setting fire to $15 billion. Looks kinda bad on your CV.

illiac7862• 2 months ago

How does that work in practice though? There’s a central “RAM Oversight Council” that checks that no one sells too much?

nullpoint420• 2 months ago

Agentic workflows and computer use. You still need regular servers for the agents to use.

sys_64738• 2 months ago

I can't say I've noticed specifically. I have two tracfone accounts and a Cricket account so used to use the Android phone with Cricket after two months - a free phone. But tracfone was bought by Verizon and they being them immediately changed the unlock period from two months to a year. So that to me kills my use case of free smartphones as I don't want to spend any money on phones.

mrandish11• 2 months ago

The deep dive on memory market dynamics and the LLM bubble distortion is great. But another cause of declining smart phone sales is simply that the devices have matured and aren't improving at nearly the same rate.

From 2008 to around 2015, upgrading every two years could make a meaningful difference. From ~2015 to ~2020 upgrading every three years might be worth considering. I just upgraded my top of the line flagship after nearly six years. And I actually looked for compelling reasons to buy a new phone every year since 2023. There just weren't any.

Frankly, this latest flagship phone is pretty underwhelming. It's slightly faster at a few things. The battery lasts a little longer. The screen can get a little brighter. The camera is supposed to a little better. But those are just the claimed improvements. I haven't actually noticed any of them in daily use because they weren't issues with my 2020 flagship phone either. Otherwise, the new phone is almost exactly the same size, same weight, same resolution, same look and same capabilities. I only upgraded because I was long out of contract and it was a only a couple hundred bucks for a $1400 MSRP phone with a new contract and a trade-in of the old phone.

trvz• 2 months ago

First, the iPhone 17 Pro is a huge improvement.

Second, the article doesn’t focus on phones we buy. There won’t be a shortage of those.

georgebcrawford• 2 months ago

Improvement in what way and over what previous phone? The parent mentioned a number of metrics.

taneq• 2 months ago

What does it improve, in practice? For myself, I need my phone to make phone calls, take photos/videos, occasionally run apps, and to be a wifi hotspot. My iPhone 6S did all of these well enough that I only upgraded it recently because I dropped it and bent the power button. My new phone has a slightly nicer camera and better battery life, that’s about it.

Brybry• 2 months ago

Near the end of the article it says:

We’re already at the point where marginal buyers in the poor world are getting priced out of the smartphone market. We’re rapidly approaching the point where buyers in the rich world feel the same thing.

So it predicts that phones we buy are next.

keithnz• 2 months ago

Ever since I started working from home I barely use my phone, I just buy a mid range phone, and it seems just as capable as high range phone in most every practical way. The only real thing that seems a bit better is the camera, but I don't actually use it that often. The only thing I've really been a bit tempted by is Lidar on Apple, but more for dev fun that normal practical purposes.

krackers• 2 months ago

The only thing I've really been a bit tempted by is Lidar on Apple

Maybe in that vein, one thing I wish phones would do to differentiate themselves is just add more sensors. I want my phone to be the tricorder from star trek. iPhones should have first party support for generating point clouds and measuring distances using lidar. Their microphones are probably already calibrated, why not expose that as a decibel meter. Same for light sensors. Phones used to have IR emitters, why not add those back in?

Also the iPhone still only has 240fps slow motion, I've found samsung's 960fps really useful in capturing transient phenomenon or even measuring mundane things like LED flicker.

Conversely the Pixel seems to be the only one shipping with an IR thermometer, and they'll probably remove it given most people don't seem to care. That's something I would've found useful in ad-hoc situations where I've had to make do with the back of my hand.

Air quality detection (especially pm2.5, CO2, and CO levels) would be great but I don't know if those sensors can be miniaturized enough to fit.

stephen_g• 2 months ago

Yeah I'm still using an iPhone 12 from late 2020 and it's honestly still fine in terms of its processing power, cameras, features etc. for anything I use it for. It does need a new battery but even that has degraded way slower than earlier phones I've had.

I was using an iPad Pro from late 2018 (mostly just for casual web browsing, reading documents and watching video on, I still do all my real work on laptop/desktops) as well until this year, and would have kept using it if I hadn't accidentally dropped it in water. I don't really notice much difference at all between my old one (when it worked) and the new iPad Air I replaced it with, except for the battery being a little better and having a bit more ram being nice (websites in background tabs are less likely to be purged from memory when I come back to them).

swiftcoder3• 2 months ago

Yeah, I mostly upgraded to get the magnetic charger on the back of my iPhone. Weren't a lot of other compelling reasons - the already pretty great cameras improved a bit, the screen is a bit nicer, but otherwise its much the same slab of glass.

aurareturn4• 2 months ago

  The MacBook Pro on which I’m writing this piece needs memory that can keep up with a powerful processor running many programs at once: so it uses a standard called DDR, “double data rate,” which runs at a reasonably high voltage and offers high bandwidth. The processor on my iPhone is less powerful, so it needs less data at any given moment; but voltage matters enormously, since every milliwatt allocated to memory is drained from the battery. So smartphones use LPDDR, “low-power double data rate,” a variant of DDR engineered to operate at lower voltages.

The last MacBook Pro to use DDR was in 2019. All Apple Silicon Macs use LPDDR.

cherioo• 2 months ago

Maybe author is writing this from an Intel Mac!

aurareturn4• 2 months ago

Quite possible. He did say his “powerful” MacBook Pro CPU is faster than his iPhone.

I’m pretty sure even an iPhone 11 chip is more powerful than a 2019 MacBook Pro CPU in ST. An iPhone 15 is more powerful than the fastest 2019 MacBook Pro Intel CPU in MT.

I suppose he can be using a 2019 MacBook Pro or older and an iPhone 14 or older and compares only MT speeds.

windowsrookie• 2 months ago

Apple has been using LPDDR in the MacBooks since at least 2015. I remember it was one of the complaint of the 2016-2017 MacBook Pros. They were still using LPDDR3 because LPDDR4 wasn't ready for production yet (despite regular DDR4 being available). The 2018 MacBook Pro's finally switched to LPDDR4.

fredoralive• 2 months ago

The early 2005 PowerBook G4 was the last pro notebook with DDR surely, as the MacBook Pro your referring to seems to use DDR4.

jimbokun• 2 months ago

This article expanded my understanding of the memory industry dramatically.

For anyone who doesn’t follow the market closely, this is about a good a primer as you could hope for.

blehn• 2 months ago

But aren't there plenty of used expensive phones from the last 5-7 years that are more or less equivalent replacements for new cheap phones? Apple alone sells 250 million phones a year.

_--__--__• 2 months ago

The article takes a while to get there, but it is focusing on a set of companies I hadn't heard of ("Transsion, Oppo, Vivo, and Lava") that buy components from last gen smartphones to make cheap devices to sell in the African and South(/east) Asian markets.

Presumably the supply of 5+ year old used phones that fully work is not enough to meet that demand, which is why these frankenstein Android companies exist.

frm882• 2 months ago

You are underestimating these phones. Transsion is in the top 5 smartphone shippers worldwide together with Samsung, Apple, Oppo and Xiaomi. There is nothing Frankenstein about them: https://techeconomy.ng/transsion-ranks-in-top-5-worldwide-sm...

LeifCarrotson• 2 months ago

What was most surprising about all this to me was this line:

So modern DRAM manufacturing is an extraordinarily complex and expensive process. Building a single state-of-the-art DRAM fabrication facility, a “fab,” will cost you about $15 to $20 billion; acquiring all the necessary equipment, like lithography tools and etching machines, will cost you another few billion; and then it’ll take you a few years of producing substandard and defective memory chips before your yields start to look competitive.

Extraordinarily complex and expensive! And yet I look at all the money being shuffled around between Nvidia and Google and Microsoft and Amazon and Apple and can't help but think that this is a tiny amount in comparison to what they're moving around on the stock market buying shares in each other.

Apple in particular has $20B in its couch cushions and is very vertically integrated and hardware-focused. Apple silicon is currently made by TSMC, but it seems they'd be a prime candidate to spin up their own memory fab.

I suppose the biggest problem to current executives at each company is the "few years" until that investment yields results, in the short term it's better to pay through the nose and buy GPUs with HBM at any price.

swatcoder• 2 months ago

And what happens if the market settles back down or the leading memory tech pivots away from what you invested all this capital and time chasing?

You'd need a very strong, very particular forecast to make such a costly bet. And conversely, it may say something about their internal forecasts that they're not making the bet.

shimman6• 2 months ago

Why should society care about people making profits? Society would greatly benefit from cheap abundant ram than FAANG shares going up. I'm kinda sick about only caring that some billionaire makes more money and would rather you know... actually improve conditions to better society.

This is why China is eating the West. Quite easy to start an electronics company when you have such an abundance of suppliers, compare this to America where there is maybe one or two players in the entire nation.

Quite pathetic, but we live in a pathetic world so it tracks.

lesuorac• 2 months ago

And conversely, it may say something about their internal forecasts that they're not making the bet.

Idk if you can read into it that way.

All these companies have cafeterias but you don't see them investing into farmland so they can get their bananas a few cents cheaper.

But also why bother spending 20B on a fab when you can invest 20B into TSMC and let them build the fab?

dyauspitr6• 2 months ago

Meta spent more on the Metaverse, it’s all the most certainly something they can afford to take a hit on and they won’t because memory and CPU usage is only going to go up from this point on.

tw042• 2 months ago

And conversely, it may say something about their internal forecasts that they're not making the bet.

It says they are no longer worried about being punished for monopolistic behavior and have bet a “ballroom donation” will exempt them from another round of punishment.

I feel like folks around here have already forgotten about the last time the memory suppliers quietly agreed to keep raising prices and stop competing with each other.

https://en.wikipedia.org/wiki/DRAM_price_fixing_scandal

Eric_WVGG• 2 months ago

I’m of the (possibly unpopular) opinion that ~80% of these new data centers are never going to see the light of day.

None of the economics or supply needs of these things make any sense. The water is generally not there, electric transformers are next to impossible to acquire. I just read about one in Utah that’s supposed to be 2.5x the size of Manhattan?? https://www.theverge.com/ai-artificial-intelligence/933687/u...

I could be wrong, but I think a lot of this was planned by software guys who are used to brute-forcing their way through “impossible” problems. But those were software problems, where the limits are mostly theoretical. In the world of atoms, limitations are real.

richforrester2• 2 months ago

Literally the most overwhelming thought, reading this, is "man, capitalism is a mistake"

The amounts of money circulating whilst some of us struggle to make rent ...

Nothing fair, or just, about this world we live in

mhb• 2 months ago

Courtesy of TFA and capitalism:

"In 1985, if you were a reasonably affluent American, the best computer that you could afford was the IBM PC AT. The PC AT would cost you about $6,000—$19,400 in 2026 dollars—and thus represented about a quarter of the median American’s annual income; and it ran on an Intel 80286 processor, capable of something like 900,000 instructions per second. Today, if you find yourself in a market stall in Nairobi or Lagos, you’ll be able to find a cheap smartphone—like the Tecno Spark Go, manufactured by China’s Transsion—for somewhere between $30 and $120. That phone will run on a processor capable of billions of calculations per second."

KnuthIsGod• 2 months ago

China going capitalist ( while remaining authoritarian ) has helped lift 800 million out of desperate poverty.

I hope that India too can emulate this in my lifetime. I was born in Kerala and would love to see Indians live in a country that is as wealthy per capita ( PPP adjusted ) as Singapore or failing that even as wealthy as the USA.

Capitalism has it's problems. But you struggling with rent is entirely your self-inflicted problem...

WalterBright36• 2 months ago

Nothing fair, or just, about this world we live in

We all make different choices, and so have different consequences.

For example, I squandered the proceeds of a business deal on a new car. If I'd bought MSFT instead, I could have bought 100 new cars.

Linosaurus• 2 months ago

, is "man, capitalism is a mistake"

Man, we need a better term for this criticism.

The US right now is more harsh to live in than the US pre-Reagan, which was more harsh than many countries with a strong welfare state, but all of them are capitalistic.

sophrosyne427• 2 months ago

The amounts of money circulating whilst some of us struggle to make rent ...

It may be expressed in terms of money, but these are assets and capital. You wouldn't find a fab a very comfortable or satisfying place to live.

The reason the numbers get so big is the gargantuan accumulated savings that are prepared to fund the similarly gargantuan ventures. The large numbers are supported by maintaining and sustaining the capitalized value by profitable investment and depreciation.

When you talk about rent, while it may be expressed in the same unit (money), the economic function played is very different. Here, it is consumption rather than saving/investment. If the dollar equivalent of all those assets went towards paying rent, while it would secure housing for some time, it would eventually become completely exhausted, and so too would housing security. Large pools of assets are simply not a way to get stable housing, it would only consume the large savings society has accumulated to be able to undergo those big ventures. It would be short-sighted and harm us in the future.

The cause and solution to rental instability is to be found in housing, building, and land policy that restricts the production of housing.

tensor• 2 months ago

No, capitalism isn't a mistake, unregulated capitalism is a mistake. The solution is simple, don't let companies merge/acquire after a certain size. Capitalism works when there is a healthy competitive market. It doesn't work when there are 1-3 big companies all fixing prices.

xbmcuser5• 2 months ago

Part of this is a lie high prices for equipment has a lot to do with monopolies in the whole supply chain and one of the major reason I am rooting for China to get parity on node as that would mean it would have been able to break all the monopolies and we get competition on the whole production from uv machines, wafer, on equipment for storing etc. Renewables are rapidly taking the world to very cheap or free energy but it would be bad if the way to best use the energy for production is controlled by a few corporations

s081486922• 2 months ago

China becoming self sufficient in chip fab could be the trigger point for WW3 so lets not root too hard

m463• 2 months ago

Apple in particular has $20B in its couch cushions

I wonder how long it REALLY took them to move from intel to apple silicon, which they don't even make.

It might be easy, like a consumer deciding to generate their own electricity (pv on the roof)

or it might be slightly harder, like a consumer deciding to generate their own electricity by drilling for oil, refining it, etc...

prewett2• 2 months ago

Well, they acquired PA Semi in 2008, they used their first chip (the A4) in the iPad in 2010, and they progressively iterated using the chip in phones until they announced the switch from Intel in 2020 (concurrently, more or less, with the A14). So it took them 12 years and 10 different chips.

mschuster91• 2 months ago

I wonder how long it REALLY took them to move from intel to apple silicon, which they don't even make.

Apple has much, much more control over their silicon than they ever had with Intel.

With Intel, they were stuck with whatever Intel could supply - and that, particularly regarding power management, wasn't much.

In contrast, for ARM, Apple has been one of their longest-ever partners [1], with Apple holding one of the very rare and probably very expensive "Architecture Licenses" that allow Apple to not just build CPUs out of pre-made IP blocks, but also design their own cores and, and that is the crucial thing, extend the ARM instruction set on their CPUs.

The latter is what made the ARM move feasible, there would have been no way of getting high-performance emulation of x86 without a few specific extensions.

[1] https://www.macrumors.com/2023/09/06/apple-inks-new-deal-arm...

mrandish11• 2 months ago

it seems they'd be a prime candidate to spin up their own memory fab.

While Apple et al certainly have the money to tilt up their own fab, they're savvy enough to understand the memory market's long history of constant boom/bust cycles. I still remember the huge DRAM shortage in late 80s forcing my startup at the time to delay launching our new product for a year.

People assume Apple cares about vertically integrating cost but they're actually focused on integrating margin. Apple has billions in cash on hand and when they think about what to do with it, a key metric is Return on Capital, especially the margin that capital will generate. Since a core metric public companies are judged on is blended margin, they are looking for ways their bags o' cash can be put to work generating revenue at margins that will pull their current average margin up vs down.

Averaged over time, mainstream memory devices are historically one of the worst margin areas of the semi market. It's super expensive to tilt up a fab on a new node but once you do, turning the crank faster to make a lot more chips isn't too hard because mainstream DRAM tends to be quite uniform. So when a fab on a new node and/or RAM generation first opens, the margins tend to be pretty great. But as the node matures and/or the RAM generation goes from 'new' to 'commodity', competition heats up as everyone gets better at making more faster. Then they're tempted to maximize revenue by cutting prices until their mature fab is at 101% utilization. And that eventually drives margins down until someone's selling near cost to sustain their low-price-enabling volume - with occasional dips below cost when they get stuck holding excess inventory. That's why cash-rich companies with high margins like Apple are delighted to buy DRAM built with Other People's Money. As long as the DRAM market is under competitive pressure, Apple gets to shop their huge orders around to get the absolute lowest price on RAM that was built with other investor's low margin dollars.

baxtr• 2 months ago

To reframe your great comment:

Is it fair to conclude that DRAM is basically a commodity that can be specified well enough by a set of parameters?

If so it won’t allow you to get any competitive advantage in your products and thus wouldn’t be a business you want to be in as Apple.

kopirgan2• 2 months ago

This has become like opec's best years. Only there's no cartel at least not openly.

But there's another key bottleneck. Even with all the money in the world, getting those machines that etch the RAM could be a multi year ration shop queue. And they're not making those companies every day!

Tuna-Fish• 2 months ago

I still remember the huge DRAM shortage in late 80s forcing my startup at the time to delay launching our new product for a year.

That one was caused by manipulation by politicians, not market forces. Micron started a price war with Japanese memory manufacturers, the Japanese cut prices to compete, Micron sued them for "dumping". The saga ended with the 1986 U.S.–Japan Semiconductor Agreement, which, among other things, created production controls that limited the total dram supply. The level was set based on then current demand, and due to the rapid growth of demand at the time it almost instantly caused a massive global supply deficit.

The agreement also caused the rise of the South Korean memory industry, because the Japanese companies offloaded their now surplus equipment for cheap.

spockz• 2 months ago

The only reason I can see Apple do this is if it enables them to sell entry level devices with vastly more ram than the competition can afford. Say entry level MacBook Pro with 256GiB ram to facilitate running frontier level local models. If that is an edge they want to have.

stasomatic3• 2 months ago

And yet, Apple had to drop base configs from their lineup. They weren't selling $599 Minis at cost. They could take someone over and inflict damage on competition.

stouset• 2 months ago

While I think Apple is a prime candidate to do something like this financially, I’ve never had the impression that they ever want to get involved in something “just” for vertical efficiency. There’s always a long-term vision where they can leverage it to gain a competitive edge that their competitors can’t match (the PA Semi acquisition being a perfect example of this).

RAM doesn’t seem like something where simply owning the manufacturing could lead to a disproportionate competitive edge. It would just be a vertical efficiency gamble that may or may not pay off. Of course that could simply be a failure of my imagination.

dyauspitr6• 2 months ago

Apple, Google Microsoft Amazon should all set up memory and CPU fabs. I don’t understand why they don’t do it. $400 million ASML machine is chump change and you can almost certainly find/train locals or provide incentives to immigrants to come here as workers to fill the roles there.

eecc3• 2 months ago

Well, isn’t Musk doing something similar with his new solar panels fab?

Frankly it’s ridiculous how we (the West) dropped renewables like a hot potato because it became synonymous with subsidizing China’s dominance in the field.

dboreham• 2 months ago

30 years ago that's how the industry was structured. IBM was the largest memory manufacturer. MBAs spent decades undoing the arrangement you're now suggesting.

andriy_koval2• 2 months ago

fabs probably was low margin business compared to selling Ads before current AI boom.

MagicMoonlight2• 2 months ago

It’s hilarious that they’ll spend a trillion renting GPUs instead of just making their own.

Even if you were nowhere near state of the art, being able to produce millions of your own cards every year at cost would save you a lot of money.

Curosinono• 2 months ago

They are not just shuffling money around though.

Nvidia sells all the stock, prices for GPUs moved up not down.

Companies like Apple don't want to be in every industry. Its risk, its cost, its expertise you need.

And sure you can throw money at it but you need people. Alone hiring all the experts is probalby not that easy. Who knows how to setup a DRam Factory?

It takes years to build this up

dspillett• 2 months ago

And yet I look at all the money being shuffled around between Nvidia and Google and Microsoft and Amazon and Apple

A lot of that is “weird money” created by the act of passing it around between the entities or “Holywood accounting” style money that exists when convenient and will vanish the instant it might be taxed or need to be extracted from the cycle to pay for something tangible. Trying to use a large amount of that money for tangible long-term building projects risks piercing the vail.

s081486922• 2 months ago

Tesla is building their own fab

Huge gamble - If they pull it off I wouldn't be surprised if other companies follow

gmerc• 2 months ago

And then everyone will scream unfair when the Chinese roll up the market from the bottom.

ksec2• 2 months ago

$20B a single Fab. And you don't just have one Fab. You need multiple. And again I have to state this every time, the most expensive Fab is empty Fab. You need to fill it, and you need to fill constant and consistently for years.

but it seems they'd be a prime candidate to spin up their own memory fab.

They will be the prime candidate to work with memory makers for additional capacity. Not to start their own one. Here is a $20B cheque and 5 years guarantee of orders. Go and make me some memory for his price.

I mean this is the same story for iPod and NAND. It seems we are repeating what we should have learned.

tw042• 2 months ago

You needed to fill it constantly in the “old times” when memory was ubiquitous and cheap to justify the finance payments on the debt you incurred to build it.

When you’re Apple and can build it with cash, and ram is currently so insane it saves you a decade of memory price gouging in a single year, the math changes quickly.

GeekyBear• 2 months ago

For a commodity part, Apple has a history of buying a manufacturing partner a dedicated manufacturing line in exchange for locking down a guaranteed supply at a lower overall price.

It's one of the early moves Tim Cook made after joining Apple to lock down enough flash memory supply to move iPods from hard drives to flash.

WarOnPrivacy• 2 months ago

I'm a newborn shill for Ulefone. They come unlocked and the manufacturer supports rooting. The devices are rugged, heavy on features and are (still) reasonably priced.

Pics:https://duckduckgo.com/?ia=images&origin=funnel_home_website...

Two other underappreciated handset brands are Doogee and Blackview. Gorgeous devices and solidly built. From what I recall they're friendly to root.

ls6122• 2 months ago

I’m really wishing I had overbuilt my NAS last year. As it stands I feel lucky to have even built it at all given I bought all the parts in the last week of September.

My 4090 and 12900k are gonna have to last till 2029 at this rate won’t they…

rationalist9• 2 months ago

Won't processors come down in price because people are buying less of them because people are buying less computers because of higher RAM prices?

I've heard that is happening for motherboards.

ls6122• 2 months ago

I’m not gonna build new until I can get a GPU and RAM as well lol. That’s gonna be a few years for sure.

phil21• 2 months ago

Processors are currently becoming the next bottleneck on the server component side of things. If consumer SKUs start seeing even less volume than they are today, manufacturers will simply produce more for the datacenter.

Certain popular AMD SKUS are already 120 days lead time and growing. If a vendor will talk to you at all.

I expect motherboards to get cheap until they get to be really expensive niche products should the current situation last a few more years. At least as we know them today as PC enthusiasts and/or prosumers. I could very much see that market evaporating entirely and moving towards large integrators instead. Gamers and such were already a rather small niche to begin with.

Right now the saving grace is a server board doesn’t look a whole lot different than a consumer board. But that may change significantly faster than anyone predicts should current trends with SoCs and the like continue. You may be in a place soon where you get to choose your CPU/RAM/storage at purchase time like you do with a MacBook today.

Viacol• 2 months ago

At Xiaomi's latest smartphone launch event, Xiaomi founder Lei Jun said that memory prices are likely to keep rising over the next two years, which could drive smartphone prices up as well. His conclusion was pretty direct: everyone should just replace their phones now. Kind of a depressing story.

cryptoegorophy• 2 months ago

Wanted to get Steam Frame VR headset and probably it will be delayed due to this. Sad

nyc_data_geek1• 2 months ago

He could be right, or he could just be a phone company CEO trying to get you to buy a new phone, or he could be both.

amazingamazing• 2 months ago

Maybe if we're lucky we get more memory efficient software. ehh who am I kidding.

hnthrowaway6323• 2 months ago

Might get some ingenious creations from poorer countries, like how in China they mod RTX GPUs to have more RAM because of export controls. Maybe some souped up Xiaomi devices with insane amounts of memory from old donor parts? Custom ROMs/postmarketOS for ultra low end devices that wouldn't normally get them but needed because they're going to need to last a few more years?

RachelF• 2 months ago

Yeah, modern software towers of libraries literally eat memory.

MS Teams uses around 1000MB of RAM to do exactly the same things that Microsoft Messenger could do in 8MB.

nottorp2• 2 months ago

Also 12 processes right now on my mac for some reason.

It's haaard to do state machines.

BuyMyBitcoins• 2 months ago

Last week I spent some extra time on a ticket in order to write a slight refactor that saves some RAM. If I had asked my boss for an extra day to implement this I would have been chastised for not delivering faster.

Thankfully my colleagues saw the value in what I was doing. I smuggled the optimization into my PR with their approval. Anecdotal, but there are still people who care about efficiency out there.

That being said, unless your manager is John Carmack, or you work in embedded systems, time spent on reducing memory footprints is seen as wasteful by the business.

pbasista• 2 months ago

time spent on reducing memory footprints is seen as wasteful by the business

I think that there is a way to change that.

If an application runs significantly better on lower end hardware while delivering the same results, the customers should prefer it. It is just a matter of promoting it that way.

stego-tech3• 2 months ago

This is the biggest memory repricing cycle I've ever seen in my life; some degree of high price/limited availability and "free RAM with purchase of Doritos" cycle is always expected, but this has been the worst one yet.

As other commenters have pointed out but I might have missed in the article, compute maturation is amplifying memory constraints right now and making it worse. Device upgrade cycles are getting longer because most compute-based products have matured, with CPUs not seeing substantial gains and memory usage really only expanding at the absolute top end of workloads pre-LLMs (3D and HPC in particular). An iPhone 14 still has almost all the features of the iPhone 17, because the compute capabilities are remarkably similar; Geekbench shows a performance delta of ~25-30% between the 14 and 17 Pro Max models, which is pretty paltry considering the devices are separated by four years of manufacturing improvements. This extends into desktops, laptops, tablets, STBs, and more, with only VR devices and larger ARM/RISC-based kit seeing more substantial uplifts as general designs improve.

So with compute stagnating and memory constrained, my money is on vendors taking this as an opportunity to gradually shift away from a yearly release cadence and slow down to a biennial cycle that alternates between budget and flagship launches every other year. Even if LLMs fail spectacularly and all that memory capacity becomes available, HBM memory likely isn't to find its way into many consumer devices (just ask AMD how it worked out for them on consumer GCN GPUs).

The name of the game, especially for consumers, is efficiency - "potato builds", as I've been calling them. Software and services optimized for lower power, smaller-specced devices of increasing age instead of pandering to flagship devices with poorly optimized code or engines for the sake of new shinies (like Raytracing). Between the memory shortage, shifting geopolitics, rising costs, and stagnant wages, consumer purchasing power is going to be squeezed like a vice for the foreseeable future, and businesses will need to adapt around that reality.

aurareturn4• 2 months ago

  So with compute stagnating and memory constrained, my money is on vendors taking this as an opportunity to gradually shift away from a yearly release cadence and slow down to a biennial cycle that alternates between budget and flagship launches every other year.

My bet is that vendors will simply discontinue their low margin phones, which are usually the budget phones.

For example, Apple might make fewer iPhone 18 and let it sell out frequently. They’ll use their RAM supplies mostly for the Pro phones.

I don’t think Apple will stop releasing new iPhone Pros every year. The business is too big.

stego-tech3• 2 months ago

I mean, that's the correct short-term read, but if LLMs in hyperscalers remain commercially viable to the point of tying up memory for several years, and if that necessitates an expansion of memory fabrication to satiate unmet demand, and if that demand ends up getting hoovered up by AI companies again due to their unmet or delayed demand from technological adoption, then Apple et al may not have much of a choice but to adopt such a profound strategy change.

There's a lot of 'ifs' there to be sure, but they'd be fools not to at least discuss the possibility internally and understand their options.

swiftcoder3• 2 months ago

For example, Apple might make fewer iPhone 18 and let it sell out frequently. They’ll use their RAM supplies mostly for the Pro phones.

Apple has a second option that may not be open to most other vendors - as they've just demonstrated with the MacBook Neo, they could cut the RAM in half on the budget models. One good cycle of optimising the hell out of their (almost entirely native) software stack, and iOS would once again sing on a 4GB SKU.

skiing_crawling• 2 months ago

Even if LLMs fail spectacularly

Haven't they already proven to be extremely useful? In some areas they are definitely here to stay, coding/software and search (retrieve and summarize information). There's a bunch of places where they are surely shoehorned in, overhyped, and don't belong, but there's also equally many places where they might still be transformative but aren't used yet.

But overall I think the technology is well proven.

stego-tech3• 2 months ago

I always leave room open for failure, and that approach has generally served me well personally and professionally. I have never been punished for having an exit strategy.

Besides, the marketplace is still in its infancy for LLMs, with a lot of unanswered questions. A lot of those questions surround the commercial viability of frontier models on bespoke hyperscaler data centers with limited usage outside of LLMs specifically should those economics be non-viable. Since that's where the memory is being tied up into, that means it's a critical question to answer in order to determine long-term investment needs into further memory fabrication.

bigstrat2003• 2 months ago

Haven't they already proven to be extremely useful?

Most certainly not. The accuracy issues mean that they can't really be used effectively for coding or search, the two things you mentioned.

yxhuvud• 2 months ago

They work well functionally, but financially anything can still happen.

TheOtherHobbes• 2 months ago

It's impressive that somehow, as if by coincidence, we're seeing the biggest inflationary drivers for decades, perhaps for centuries, all happening simultaneously.

The Iran war is spiking the price of oil and will likely cause shortages of pretty much everything if it isn't ended.

The Ukraine war is helping with that by destroying Russian refining capacity.

The memory shortage is set to do the same to consumer electronics, which are absolutely essential to the modern economy.

Meanwhile the AI fad is seeing huge layoffs. At the same time as the AI Big Cos are beginning to show signs of ending the subsidised free lunch phase and moving to a utility model, which will raise prices for every company that is hooked on AI.

Also tariffs. Although I'm not sure if anyone knows what's happening there.

And farms are failing. Climate change will accelerate that, so there will be food shortages within a few years.

If it's not cynical and deliberate, it's an astounding confluence of (literally) catastrophic mismanagement.

7thpower• 2 months ago

Farms are failing where? More than usual?

bdangubic• 2 months ago

they are not failing, our Socialist President provides billions in handouts for Farmers so they gonna be just fine

crooked-v• 2 months ago

New Jersey just declared a state of emergency after an unprecedented cold snap wiped out crops across the state, up to 90+% for some farms.

tootie2• 2 months ago

US tariffs are mostly dead. Ruled illegal and being refunded. Second attempt at tariffs has also been stopped. And inflation over the past 5 years is still milder than the late 70s and definitely milder than what has been seen in Turkey or Argentina or Zimbabwe.

gpt52• 2 months ago

Not to mention that some of the effects that OP cited are either deflationary (like layoffs and automation), or clearly cyclic (like the memory boom bust stages, experienced just a few years ago post COVID).

threecheese2• 2 months ago

Ah, the ole’ “were doing better than Germany ca 1937” argument. /s

lurking_swe• 2 months ago

refunded to whom? the businesses? great! where’s my refund as a consumer who paid those higher costs (passed along to me)?

mr_toad2• 2 months ago

Meanwhile the AI fad is seeing huge layoffs.

This is really the only point I disagree with. Layoffs are being blamed on AI, but they are really a hangover of the covid hiring boom, and subsequent bust.

All the tech money is going into building data centres (and the gas turbines that power data centres), and it turns out that programmers don’t have the relevant skills to build gas turbines.

ghaff2• 2 months ago

There are a bunch of factors that are intertwining. And unfortunately, some specific demographics are mostly those being most caught in the crosshairs.

ghaff2• 2 months ago

And building gas turbines (or even designing them) are probably not jobs that pay salaries developers got accustomed to, especially in Sillicon Valley, over the last 10-20 years.

Jobs at GE in Schenectady may not be what those developers are looking for even if they were qualified.

threecheese2• 2 months ago

A few months ago I’d have been right there with you; however recently there have been a number of high profile multi-thousand-head layoffs attributed to budget pressure of deploying AI - and so technically it’s more of an indirect rather than “a robot stole my job” replacement, but still causal. Those tokens will be “doing” labor, rather than the human.

wartywhoa232• 2 months ago

Layoffs are being blamed on AI, but they are really a hangover of the covid hiring boom.

The world's most convenient virus.

array_key_first3• 2 months ago

Most companies are showing high revenue and profit and still laying off huge segments of their workforce. Largely due to the stock market - a lot of non-AI players are facing lowering stock price.

mgh2• 2 months ago

Don't forget the threat of Ebola, hantavirus, Taiwan war

jurgenburgen4• 2 months ago

The Taiwan war that the world is sleepwalking into is going to wipe out so much chip capacity that the current inflation will look like a blip.

fakedang2• 2 months ago

And farms are failing. Climate change will accelerate that, so there will be food shortages within a few years.

This is the key one to watch out for. Prior food-based conflicts have sparked revolutions and civil wars. People can tolerate not having electricity, people can tolerate not having internet, people can tolerate not having gas.

But a lack of food, due to a wheat shortage which in turn was sparked by Siberian wildfires destroying a whole year's wheat crop in Russia, precipitated the Arab Spring and the subsequent civil wars, as well as the ascendancy of ISIS.

Ray202• 2 months ago

Climate change will accelerate that

The climate is changing in a positive direction for farming. Farming is easier now than ever before. There's literally no chance of food shortages. Unless, of course, there's another attempt at building socialism.

Schlagbohrer9• 2 months ago

Coffee crops have already been failing the last few years due to global warming, and now with the fertilizer crisis that is hitting major coffee producers hard this year (Kenya, Ethiopia, Indonesia, all famous for coffee and all dependent on Gulf fertilizer).

mannanj2• 2 months ago

So where are all these rich bunkers and storage caches again? Isn't there like a map of these things, to help them preserve their thousands of lbs of food for their future generations of course. No need to have backup feed for the starving masses.

pfannkuchen• 2 months ago

It would be a great scapegoat for further monetary expansion related inflation.

Step 1) Instigate price increases for temporary reasons, leading to noticeable price increases.

Step 2) As temporary reasons go away, increase money supply. Prices stay the same, you get to blame the causes from step (1) while not mentioning that the prices could have gone down when the problem went away, or blaming greedy companies for not lowering the prices.

ponector2• 2 months ago

Destruction of russian refineries should increase proposition of crude oil on the market and drive price down a bit.

Schlagbohrer9• 2 months ago

William Gibson coined the term "The Jackpot" to describe an apocalypse made up of many different disasters all happening near in time, overwhelming civilization. Pandemic, war, mismanagement, corruption, resource scarcity, climate breakdown, ...

ecshafer• 2 months ago

The book "1177 B.C.: The Year Civilization Collapsed" essentially argues that as being the root cause of the Bronze Age Collapse. The eastern Mediterranean and middle east were sophisticated, highly integrated cultures with a large amount of trade. There was a series of droughts, earth quakes, fires, etc. that re-occurred over a period of time, causing agricultural output to drop, leading to rebellions, leading to wars, leading the trade routes to collapse, and everything collapsing roughly at the same time.

shimman6• 2 months ago

The more academic term is polycrisis.

gib4442• 2 months ago

And AI causing electricity infrastructure price rises for domestic consumers - aka forced indirect investment in AI companies.

Oops, sorry, wrongthink, I mean "due to the war in the Middle East", even though a large majority of gas in the UK comes from elsewhere (gas and electricity prices in the UK are supposedly very tightly coupled ?)

mjh2539• 2 months ago

If Malaysia can't obtain 30% of its fuel because of the Hormuz closure, they don't just reduce their consumption of fuel by 30%, but make up for most of the deficit by buying it for a higher price somewhere else. But the fuel they buy to make up for the deficit simply means there's less fuel for other folks buying in that market to purchase, which drives up prices. This process repeats until prices stabilize worldwide, which ultimately results in higher prices for fuel in the UK.

WillAdams• 2 months ago

Presumably food shortages will first be felt by livestock, so expect meat to get much more affordable (as animals are slaughtered to avoid the need to feed them) for a while, then expensive (when there are fewer head). Animal feed represents over 50% of acreage.

DoctorOetker2• 2 months ago

we can generate full protein mixtures with hydrogenotrophs, fed green hydrogen from solar, the photosynthesis -> protein is 10x more efficient than plants.

90% of acreage could be slashed this way, either for rewilding or for development.

Soylent Green, but veritable, not old people

deadbabe• 2 months ago

Keep going…

xinayder• 2 months ago

Maybe it's time to go back to the time where computers and entertainment were niche and develop efficient resource-constrained devices....

I hate it that we had decades of progress to have computers become a very expensive hobby because some dudes high on fentanyl think some text prediction model that destroys the planet is worth a trillion dollars.

DaveZale• 2 months ago

I've been buying a "new" used iphone for $100-150 now, every few years, for over a decade. My "new" used Dell laptop I bought a few months ago for $40 which became a linux system in just an hour. All good here.

doodlebugging• 2 months ago

I appreciated the detailed breakdown of the memory crunch and how it will affect parts of the industry and consumers. Very good article.

I'm not one of those people who chases all the new great things. I wait until things wear out or become completely obsolete before upgrading. I just get comfortable doing things the same way every day and see no reason to waste money on SaaS shit or anything else wastes my time or money.

I think the memory shortage will present opportunities for those willing to take advantage of the situation. A lot of DRAM is going into GPUs for data centers in AI work. Those units have a limited lifetime online and they will be rotated out and replaced with new units as performance degrades. I think this will be a lot like Li-ion batteries in that many of these GPUs will be perfectly fine for home pcs or small business workstations or for other less intensive use cases and the RAM will be performant enough that a viable recycling industry should arise from this AI buildout.

Funny enough, one day the local AI noise-making, power-wasting, water-wasting data centers will be the best places to score high-tech components and many of us will have one right down the road. That should set a lot of people up as recyclers redistributing reconditioned components to those who build their own systems.

p0w3n3d• 2 months ago

I heard that one company buying all the product or a majority of the product is called a monopoly practice.

  If a dominant buyer locks up most of the supply chain through exclusive contracts, it prevents rival companies from getting the materials they need to survive, which violates laws like Section 2 of the Sherman Antitrust Act.

supriyo-biswas• 2 months ago

It's called a monopsony[1], regardless, antitrust regulation is famously dead under the current US administration.

[1] https://en.wikipedia.org/wiki/Monopsony

gbriel• 2 months ago

Regulation is dead under this current US administration (unless you pay a bribe). It's so weird this is where we are right now and most of the tech leadership has to play along even though it's technically illegal?

nathanaldensr• 2 months ago

Antitrust has been dead for decades. It spans administrations and parties.

_kidlike• 2 months ago

very cool word, but I think the etymology stated in Wikipedia is wrong.

the second part of the word just means purchase (in a weird ancient Greek tense). there's no relation to fish whatsoever.

(I'm Greek)

Animats2• 2 months ago

This is another place where modern capitalism struggles in an area that requires large capital investment producing payoff after years of startup costs, and the buy side is volatile. We've seen this recently with rare earths, copper, and some other minerals. DRAM is almost in the same category.

China has an advantage here, once domestic DRAM production finally gets going. DRAM policy can be set strategically. China's economic planners may choose to provide DRAM to domestic manufacturers rather than export parts, even if exporting parts would be more profitable in the near term. That's already being done in raw materials. Conversely, if external suppliers have lower prices, there may be a policy decision to buy domestically to keep the domestic manufacturers going. Done with the goal of leveling production, this can work. Done stupidly, it becomes a money drain, of course.

Probably China controls the DRAM market around 2030 or so.

ahepp• 2 months ago

what share of the DRAM market does China control today?

Animats2• 2 months ago

About 7.6%.[1] That's Changxin Memory Technologies. That makes them #4 in DRAM, after Samsung, Hynix, and Micron. They're still only a third the size of Micron, the last US manufacturer. They have a lot of scaling up to do, but they can make current-generation DRAM.

Changxin just became very profitable, due to high DRAM prices. They're planning an IPO. And, of course, building more fabs. Yangtze Memory Technologies isn't as far along, but they're building a new fab and planning an IPO.

Both of these companies were created in 2016 as part of a national effort to become a force in the DRAM industry. It took a while to get the technology working. Now it works, and they can scale up.

[1] https://chinabizinsider.com/how-the-economics-of-the-global-...

agumonkey• 2 months ago

Who else thinks that this constraint will force new creative solutions. We got compression algorithms due to limitations.

my 2 cents

bit1993• 2 months ago

If electronics become too expensive for consumers, wouldn't the demand for AI also go down because consumers cant access it?

larsnystrom• 2 months ago

I don’t think AI demand is driven by consumers.

s17tnet• 2 months ago

Users will have even have more dump devices and will be forced to use remote and cloud services.

einpoklum3• 2 months ago

I work in healthcare imaging. The DRAM crunch is translating into significant increases in costs - because reconstructing images from raw scans is a compute-itensive process; and that gradually translates into product price increases. And then, healthcare providers may well opt for cheaper, lower-tier medical scanners, since they can't afford the better ones; and then finally you get less accurate scans if you're suspected to have cancer or whatever.

So, that's another way we are financing the LLM machine and the trillion-dollar valuations of those corporate behemoths.

einpoklum3• 2 months ago

So, if you were wondering how come LLM services are cheap or gratis - now you know another group of people bearing the costs: The poorer half of humanity, who will now not be able to afford a smartphone, i.e. not having access to a computer. Sure, that doesn't kill you, but it is quite significant personally and socially.

azan_3• 2 months ago

That's the worst critique of AI I've ever heard, and I've heard a lot of it.

Hackbraten• 2 months ago

Care to share your constructive rebuttal with us?

nuancebydefault• 2 months ago

So the availability of cheap phones is going down because of the cost of RAM.

What about the RAM consumption trend of the last 10 years? I think it is very feasible to produce phones with the same amount of RAM as was the norm 10 years ago. The only compromise would be using older algorithms and features that consume less of it, and to take a bit of effort on keeping an eye on memory consumption in the development phase. There's a lot of opportunity. We can even leverage AI these days to optimize existing software for RAM usage.

account425• 2 months ago

RAM usage doesn't just depend on the software that comes with the phone but also (or even more so) on the apps the user installs and on the content - so this is something that would need an ecosystem-wide adaption not just a single vendor.

ponector2• 2 months ago

Unless you can use apps created 10 years ago the is no sense to produce phones with old specs.

Think of desktop, browser and electron applications. No way one can comfortable run modern software on 2015 machine. Maybe Linux could help, but anyway

dijit• 2 months ago

I think the reasoning is incomplete.

RAM isn't the only reason to upgrade. CPU's are more efficient, screens are significantly better and support variable-refresh-rates (and high refresh rates), battery technology has not stood still: even if they didn't degrade over 10 years.

There's plenty of reasons to upgrade to a new system even if it has the same memory capacity on paper. The memory will probably be faster and lower voltage too.

bitmasher9• 2 months ago

My initial reaction was that old compute levels are probably good enough for modern applications, but after thinking through all of the common use cases you’re probably right.

I think it’s AV that makes the difference. The modern smartphone does a ton of processing on the image sensor, and the modern laptop is expected to output multiple 4k video signals (zoom camera + desktop sharing) while accepting multiple video signals, without dropping below 120hz.

ryukoposting• 2 months ago

No way one can comfortable run modern software on 2015 machine

My personal laptop is a ULV 6th gen i5 with 8GB of RAM. Gets the job done, slowly but usably. That's about 9 years old. It runs Ubuntu. A dozen or so Firefox tabs, Kicad, VS Code, Slack. Computer things.

I'm of the opinion that application developers should be given middle-tier consumer systems as work machines. The whole notion of "dogfooding" is "dog-something else" if your devs experience the product using different hardware from the customer.

anthk• 2 months ago

Use ZRAM, seriously.

kopirgan2• 2 months ago

Very well written sub.

Any prediction on when it'll end? Can Chinese companies scale up to scare the big 3 into increasing capacity or lose price control?

automatic61312• 2 months ago

It will end slowly, and then very suddenly.

The demand for AI simply doesn't exist at the real prices. It barely exists at the current subsidized rates - Microsoft, Google, Meta, Anthropic, OpenAI are spending hundreds of billions to make mere billions.

And then these data centers will be worthless, future ones won't get built, memory demand will evaporate on the spot.

realusername3• 2 months ago

I have the same opinion, these AI companies can't even work financially with the cheap ram prices, they certainly won't with high prices.

And ram producers are betting on it, they will just milk the AI companies until they collapse.

andybak2• 2 months ago

I'm yet to see any convincing argument that inference is subsidised in any substantial way. Training and speculative expansion are where the spend is from what I can see.

fittingopposite3• 2 months ago

Very interesting and thoughtful comment. I agree. When demand from hyperscalars drops rapidly after the tide goes out and everyone realizes that they can't continue building out the data centers, then fabs will be left with overcapacity that will flood the market. Wondering what this will mean for local LLMs when good RAM is available for cheap?

zozbot2346• 2 months ago

The Chinese memory makers can scale up but the amount of capacity they currently have is negligible. They're in the same boat as everyone else, it's just that nobody expects them to curtail production on purpose to raise prices even higher, whereas that's expected behavior (according to the article) from the established memory makers.

cbdevidal2• 2 months ago

The optimist in me hopes this means developers will pay more attention to memory usage in their apps.

Zopieux• 2 months ago

Well, not if you believe that all future software will be slop.

cbdevidal2• 2 months ago

I don’t. Also from my optimist side :-)

Havoc2• 2 months ago

Amused by all the comments about optimisation. If anything the vibe coding era will lead to the opposite

LatencyKills2• 2 months ago

I have a friend who is VP at a major telecom company. He has no technical experience but has been using Claude to create data analysis apps. He was complaining that it took three hours to process certain datasets, so I took a look.

He had Claude essentially create a 300MB json file and was doing all of the data processing on that data directly.

It never occurred to him, or Claude, that there were other ways to operate on that data. It took me less than 10 minutes to get that processing time down to under a minute.

These are the type of issues that worry me about vibe coding.

sokoloff11• 2 months ago

What would have happened five years ago? I suspect this VP would have gone entirely without analysis tools for these datasets or would have a department of people slowly grinding away at his long list of tickets, tickets that were poorly specified and hard to get clarity on because by the time they were worked on, he’d lost some of the context of his original request.

Now, he makes small apps that scratch his own itches, while everything is fresh in his mind and he can clarify or learn “hmm, that’s not actually what I want” and the cost is some tokens and the occasional job that runs in 3 hours instead of not existing at all.

WithinReason• 2 months ago

It took me less than 10 minutes to get that processing time down to under a minute.

You could have done it with Claude in 1 minute :) It probably never occurred to the VP to ask Claude for performance optimizations

esikich• 2 months ago

People have done this long before vibe coding was a thing. I've seen this everywhere. Not every business employs devs or can afford to. So people hack things together with powershell or python. It has always been this way.

Schlagbohrer9• 2 months ago

Dude I will just include a skills.md for optimisation, and include a line in my prompt saying "also make the code really well optimized". Problem Solved sunglasses emoji

ReptileMan2• 2 months ago

LLMs are quite good at optimization. Here is a simple prompt - rewrite this crap from electron to rust + whatever time tested ui framework. And now we have 80% less ram usage.

Havoc2• 2 months ago

To be fair electron to rust would probably work even with LLM ram overhead

21asdffdsa12• 2 months ago

I find this deeply ironic, the revolutionary product is eating the devices the consumer who where supposed to use it on. Saturn is hungry!

ngcazz• 2 months ago

I feel like this has just accelerated what was the ideal scenario all along: high hardware prices push more people to rent

Schlagbohrer9• 2 months ago

It will result in more people having to "rent" their computer from a corporate cloud. An extension of what already exists for poor gamers who stream their own games from a cloud renderer.

deltoidmaximus• 2 months ago

From an outside view OpenAI buying all that raw RAM certainly looked like they were buying it mostly to prevent anyone else from using it more than that they had an immediate need. I recall a Microsoft exec complaining that many of the GPUs they purchased were just sitting in a warehouse as they didn't have the power supply to actually do anything with them.

HPsquared• 2 months ago

I wonder if aggressive swapping could help a bit. Higher performing SSDs with endurance. Unless flash memory is also being squeezed...

zozbot2346• 2 months ago

Yes, it would be viable. Intel Optane/XPoint is wearout-resistant and has DDR3-like performance for 10% of current DRAM prices (and that's on the secondary market, since it's not being manufactured new). But it's far more power-intensive per bit read/write than even SSDs, let alone DRAM.

wartywhoa232• 2 months ago

TIL: price rise (eng.) = repricing (newspeak).

softwaredoug• 2 months ago

Add to this if China keeps adding pressure to Taiwan, escalating in a soft blockade, we might see something worse than what’s happening in Hormuz.

emsign• 2 months ago

AI is choking the tech industry and the economy at large. This already has negative effects on the supply chain and profits.

heisenbit• 2 months ago

We always knew the limits to Moore‘s Law are first and foremost economic. Given an industry used over decades to predictable lowering of price per compute function and thus swallowed any advance for new user functions and overhead when the limits are reached there is going to be a squeeze. AI scaled up at the time the production capacity became more inelastic.

Maybe it is time not just shrink transistors but also software bundles. I can see decades of possible progress hiding in plain sight behind a browser screen.

onlyrealcuzzo3• 2 months ago

AI data centers are eating like 80% of memory.

Making user space applications more memory efficient is not even going to be a rounding error on memory demand.

I am with you that it needs to happen, but it's not going to solve a memory shortage.

H8crilA2• 2 months ago

It would make memory-poor phones more viable. Like why can't we have a 512MB, or even 256MB RAM phone. Although I doubt that the software effort would be cheaper than just buying the extra RAM. It's definitely much more uncertain.

zozbot2346• 2 months ago

That's newly fabbed memory vs. an existing stock. The stock is quite massive, so optimizing existing use and enabling it to be repurposed can be meaningful.

ReptileMan2• 2 months ago

God of war 2 ran on 32MB of ram. There is enough ram in the devices for any reasonable workload. Just the developers on all levels of the stack weren't bothered to give rat's ass for optimization.

The current crunch and constrain in compute is great time to show once again some ingenuity. The lowest level of smartphones from couple of years ago have more computing power than XBOX 360. That should be enough to run Whatsapp smoothly.

fguerraz• 2 months ago

This is great news, maybe people are going to start caring about their electronic gadgets more and not treat them as disposable? Maybe longevity is going to become a criteria again?

Schlagbohrer9• 2 months ago

The blokes in Agbogbloshie will hopefully make more money from recycled e-waste.

It has been a disturbing trend on TikTok to see all the livestreamers in the united states showing how to recover gold and metals from e-waste... it used to be an exceptionally dirty and low margin business only done by the world's poorest. Now it has become valuable enough, and the economic situation in the usa has deteriorated enough, that americans are doing it in their homes and backyards.

clearstack• 2 months ago

HBM is what makes this bifurcated. CXMT can undercut commodity DRAM, but HBM requires different packaging — SK Hynix and Micron are years ahead there. ASP gap is roughly 5x

stasomatic3• 2 months ago

"and then it’ll take you a few years of producing substandard and defective memory chips before your yields start to look competitive."

As a complete know nothing about the fab industry, I am always puzzled by this. Do the fabs need to be seasoned like an iron cast skillet or something?

JrProgrammer• 2 months ago

I can imagine that the process of fabrication relies on tweaking a tremendous amount of parameters. In a sense seasoning a skillet and perfecting it to the max

ACCount372• 2 months ago

Kind of.

There's a mysterious fab entity known as "the Recipe" - the product of long iterative dialing in of the fab's operating parameters. Of which there are a great many. A modern fab performs hundreds of manufacturing steps, with thousands of tweakable parameters, and they may interact in non-obvious ways to affect the outcomes. This is what's discovered and adjusted as the fab runs.

The difference between having the Recipe and not having the Recipe is the difference between 96% yield and 12% yield.

Changing the process (i.e. 4nm to 2nm) is the most sure way to lose the Recipe. The fab knowledge you spent months and years of engineering work discovering will no longer apply. But you can also lose the Recipe by replacing fab hardware, by changing the suppliers, by an act of god, and more.

stasomatic3• 2 months ago

Can a recipe be acquired and reproduced to perform correctly from the start? Say one line is performing well, can its parameters be copied over to a new one and start producing?

My head annoyingly refuses to accept the non-deterministic outcomes when applied to electronics. In other disciplines, let's say bow or musical instrument making, you need to harvest the wood, age it, dry it, pray to a deity, work with its natural im/perfections etc, but this is "just" bits.

fittingopposite3• 2 months ago

Like Taleb says: "I've never seen a shortage not followed by a glut." Just wondering when this time is and got how long end consumers should wait for prices to fall again strongly.

Ylpertnodi• 2 months ago

I'll wait forever, if necessary.

andriy_koval2• 2 months ago

Did he ever check housing market in certain developed areas?

shdh2• 2 months ago

Then China needs to ramp up DRAM production, because the big three won't in a significant enough fashion to bring the glut

fittingopposite3• 2 months ago

No it's not about interesting supply. The logic is that the demand will just plummet when the general economic situation turns because we have over provisioned the datacenters

DeathArrow2• 2 months ago

One of the most remarkable things about the last few decades is how cheap computers have gotten.
In 1985, if you were a reasonably affluent American, the best computer that you could afford was the IBM PC AT. The PC AT would cost you about $6,000—$19,400 in 2026 dollars—and thus represented about a quarter of the median American’s annual income;

In early 1980s a PC AT was SOTA. I guess that if you buy a SOTA computer today it might cost you close to $19,400.

DeathArrow2• 2 months ago

All of which is to say: things are going to get a lot worse before they get better.

I have a 14900KS, a CPU from 2 years ago and is still amongst the strongest in benchmarks like Cinebench. And 128 GB DDR4 RAM. I feel like I can wait a few more years to upgrade.

shdh2• 2 months ago

Same, I have 128GB of DDR, a 3090, and 5950x

I was about to upgrade earlier 2025 but then held off, guess I shoulda just pulled the trigger

pjdesno2• 2 months ago

Jatin Malek on Twitter had perhaps the best explanation of the DRAM crunch:

"The reason why RAM has become four times more expensive is that a huge amount of RAM that has not yet been produced was purchased with non-existent money to be installed in GPUs that also have not yet been produced, in order to place them in data centers that have not yet been built, powered by infrastructure that may never appear, to satisfy demand that does not actually exist and to obtain profit that is mathematically impossible."

crazygringo3• 2 months ago

That's just a snarky way to describe all business investment that requires purchasing things made in the future. It's entirely normal.

You could literally rewrite the quote to be about iron and about building railroads for trains and passengers that don't exist yet. See how silly that would be?

Except the "profit that is mathematically impossible" part. That's just made up and false. It's entirely possible that we are actually underestimating demand, and there is going to be tons of profit. Nobody knows for sure, but profit is very, very, very possible.

d3rockk• 2 months ago

Except the "profit that is mathematically impossible" part. That's just made up and false. It's entirely possible that we are actually underestimating demand, and there is going to be tons of profit.

JP Morgan says $650 billion in annual revenue required to deliver mere 10% return on AI buildout is equivalent to $35 payment from every iPhone user, or $180 from every Netflix subscriber 'in perpetuity.'

Very, very, very unlikely it makes profit, which why AI keeps getting overhyped by CEOs.

Shog9• 2 months ago

Uh... The history of railroads is littered with folks losing their investments for exactly this reason.

Was your point that owning the steel companies is the path to riches even when most of the railroads fail?

croes2• 2 months ago

Where does the money for that profit come from?

Who buys anything made by AI if he could do it by AI themselves?

Who can afford AI if their customers do with AI what they do?

Who creates the next man made code needed as training data to prevent model collapse?

Kim_Bruning• 2 months ago

You could literally rewrite the quote to be about iron and about building railroads for trains and passengers that don't exist yet. See how silly that would be?

Couldn't possibly happen with railroads!

https://libertystreeteconomics.newyorkfed.org/2016/02/crisis...

array_key_first3• 2 months ago

Didn't that actually happen with railroads? Like, there was a railroad bubble and lots of people lost a lot of money?

It happened with the Internet, too, and the Internet is pretty useful I think.

bob_theslob646• 2 months ago

Can you link to this?

ilaksh• 2 months ago

Surely some significant fraction of the billions and billions are being spent on intensive R&D to for example increase scale of MRAM modules or radical new paradigms like nitrides-based ferroelectric compute-in-memory?

I think there's a reason that University of Michigan researcher jumped after that ferroelectric breakthrough.

I am hopeful that some of this massive pressure is going to push us into the next paradigm, which is likely some kind of "true" compute-in-memory system. That may increase performance and efficiency for AI by up to 100 X.

Levitating• 2 months ago

Luckily innovation in smartphones has also stagnated. I just bought a 6 year old phone (Poco X3 NFC) secondhand for 60 euro. That's barely a downgrade from my Poco X7 from last year.

Poco X3 has excellent support for Ubuntu Touch, PostmarmetOS and LineageOS etc.

nashsimulator• 2 months ago

This is such a detailed article and lesson in manufacturing complex parts and their fixed costs vs variable costs accounting lesson, and a lesson in how a product we use daily goes through the manufacturing chain and its' price, availability etc.. is influenced by a complex chain of factors.. I agree with the top poster, the title is an undersell of the full article..

wg02• 2 months ago

That's why it is very important for these AI companies to go bust for the collective good.

gbgarbeb• 2 months ago

It'll definitely happen. After all, the collective hate for the oil barons 100 years ago put them out of business in a jiffy. /s

wg02• 2 months ago

Oil companies were profitable.

Oil always burned provided there's oxygen.

Price of oil was crystal clear how much you'll get for what you're paying.

And nothing else was affected with supply of oil.

No harm in keeping comparisons inline with the subject matter at hand.

tim333• 2 months ago

So it will be hard to avoid a great repricing of consumer electronics in the coming years

I would have thought in the coming years they'll build more chip plants or the AI bubble will crash or some such. If there's an effect it should be greatest now.

fsckboy• 2 months ago

since any given day here there will be a majority unconcerned or actually defending code bloat, well now, this uniinteresting topic is just chickens-coming-home-to-roost. All the benefits of Moore's law are not gone, today's prices reflect a very large N of an exponential decline of cost per bit; these price increases are linear, not even a 2nd order polynomial, so why all the angst? go to your room and clean up your code.

ecommerceguy• 2 months ago

So does this mean all of the ram I've got piled up in a box is not worthless anymore?

arikrahman• 2 months ago

It's bad when even Samsung is not even able to negotiate an agreement with Samsung.

try-working2• 2 months ago

The Big 3 memory producers are really super charging their Chinese competitors CXMT and YMTC. They will IPO this year and they will be able to get massive funding for their build outs.

If the Big 3 had been more prudent, they could have kept profits lower and avoided this, but now they've probably dug their own grave and will see massive decline in profits starting from 2027/2028 when Chinese capacity really comes online.

lmz4• 2 months ago

The article has examples from history of overexpansion and crashes. Would rapidly expanding (or keeping profits lower(??)) be beneficial long term for the Big 3? I bet they think it's better to bank the surplus now. The Chinese will have to get their investment back somehow too after the expansion spend.

try-working2• 2 months ago

Well, they're about to lose massive market share to CXMT and YMTC.

georgeburdell• 2 months ago

Profit is not a concern for the Chinese

rldjbpin• 2 months ago

while it looks like outlandish fantasy today, looking forward to their competitors dumping consumer memory in a couple year's time.

regardless, this industry goes through such cycles (which have killed weaker companies and made the situation worse). high time for the memory "OPEC" to disband.

anigbrowl• 2 months ago

I feel confident in predicting there will be a glut in 5 years.

KronisLV• 2 months ago

Two predictions.

The people responsible for throwing the consumer segment under the bus will see no repercussions for this - nobody is going to jail, consumers get screwed.

The memory prices will never return to the pre-shortage prices. They’ll use this as an excuse to subtly increase the base prices and the orgs that sell the devices will also do a similar increase, also past what the BOM changed by.

DoctorOetker2• 2 months ago

did I understand the article correctly?

"Because of all the peripheral circuits and all the vertical channels, a single gigabyte of HBM consumes more than three times the wafer capacity that a gigabyte of DDR or LPDDR does. Every gigabyte of HBM produced is, in effect, three gigabytes of commodity memory not produced."

Is this true or conflating wafer area with process steps?

If the bottleneck isn't process steps but wafer area, you'd expect the semiconductor boule growers to rake in all the profit, not the memory makers themselves (although they would see more revenue). I'm not hearing about Umicore or other semiconductor corps raking in big under a wafer shortage.

Thats why I ask the question if this is actually correct.

If correct, it really makes one wonder what would be more productive for machine learning community X PB of HBM vs 3X PB of conventional (LP)DDR RAM, or somewhere in between. By giving a couple of big players carte blanche financially, has allowed the decision to divide in 3 global memory capacity (in units of terrabytes manufactured per day say)? The singularity that slows itself down?

The memory shortage is causing a repricing of consumer electronics

Related Stories