Yeah, but all of this is pointless when RAM is as expensive as two CPUs by itself - if it's even in stock. AMD/Intel should focus on that first if they want to save their DIY business at all - which I'm starting to doubt they don't
I am wondering if this so true. What resources and time are needed to increase supply by N times to catch demand.
The same way nobody wanted to invest in mask supply in the US or Western Europe during covid, because producers knew the demand spike wouldn't last and they'll be left with useless equipment to pay after the crisis passed.
If they're wrong and this is actually a permanent spike in demand, then it'll take the industry a while to realize it but eventually they'll collectively figure it out and increase supply. The ones who figure it out soonest and increase supply fastest will profit the most. The ones who figure it out slowest will lose market share.
I am one who think there is a chance it will sustain. AI is useful tool, Opus unlocks N times productivity gain for devs since Opus 4.5, which is available just for 3 months.
This means adaptation is just started, it could expand on all kind of usecases, niches, solving problems, products, etc, which could be N times more demand for compute from what we have now.
If it doesn't decline, than anyone who took that risk and increased their production capacity will benefit greatly, and those who didn't will lose market share.
The incentives here are naturally very well aligned with solving the shortage. If doing nothing is likely to solve the shortage, then they'll do nothing. If increasing supply is likely to solve the shortage, then they'll increase supply. If there's a 50/50 chance of both, then some will increase supply and some will do nothing, and the market will reward whichever group was right and punish the other.
I'm always baffled how seriously some people take that “market and incentives always lead to the greater good” religion despite plenty of empirical evidence against it.
But hey, there are creationists out there too so it's not too surprising I guess.
I never said anything of the sort, I'm just explaining basic economic facts. Feel free to pretend they aren't true if you wish, but if you do then for all our sakes please stay as far away from the levers of economic policy as possible.
In the real world, executives simply aren't being incentivized for all-or-nothing risk taking and the shareholders of public industrial companies don't want the executives to make gambles in an attempt to make a big payout, they want steady yields with limited risks. And the financial actors who would be ready to take this kind of “high risk high reward” bets aren't operating in these kinds of markets.
Just look out of the window, semiconductor and electronics fabs aren't rushing to expand their capacity, they are very familiar with the issue of oversupply and are always very cautious before making capacity investment, because they know it could very well make them go under very quick should the market reverse.
The more time passes, the more convinced I am that nothing did more damage to the broader public understanding of economic dynamics than the average Econ 101 class.
Ford had almost $20 billion in EV car manufacturing investments planned for the mid-late 2020s and the abrupt end of the EV subsidies cost Ford billions of dollars and they have abandoned multiple investments.
If you do nothing, you still are rewarded because you are making pure profit in either scenario. If you invest billions then you are digging out of that for years regardless, and could be in the whole for decades if you bet wrong.
[0] https://www.construction-physics.com/p/how-to-build-a-20-bil...
[1] https://techovedas.com/what-does-it-take-to-build-a-semicond...
Existing fabs are currently reaping the rewards of their previous wise investments in building out this capacity in the first place right before a shortage. Without those past investments, the shortage today would be even worse. Future rewards will be allocated based on who correctly predicts the best use of current resources to meet future demand (whether that includes huge investments to build even more fabs or not is yet to be discovered).
But common among all these replacements is that they're not really planned in detail years before. I may weigh up factors like "I don't really need an upgrade right now" against factors like "the market looks like it'll probably get really shitty later this year", or "I could really really use an upgrade right now" against "but the market is shitty now and we're right before a product launch which will shake things up". But I always react to my current or near-future needs/wants and current or near-future market conditions.
So hearing "the RAM market will be good again in 5 years" is completely irrelevant to me. My decisions are entirely based around how the RAM market is right now and how I believe it will look throughout the year or the next.
Let me describe this in the most simple terms possible: You have speculators speculating about AI products. The speculators are not very smart when it comes to technology, and think RAM is RAM. There is at least three kinds of RAM that are important to this: DDR for system RAM, GDDR for GPUs, and HBM for high density enterprise products, and they are not interchangeable, there is no one-die-fits-all solution.
So, these speculators are like "oh no, more GPUs requires more RAM!", and then just start speculating on all RAM. Which of these RAMs are the ones that they need to worry about? Exclusively HBM, which is a minority in production, DDR and GDDR dominate production.
If you're into inference, and have older machines, you're buying Hxxx or Bxxx cards that use HBM, fit into dual slot x16 configurations, and you're jamming (optimally) 8 of them in. If you're into hardware that is newer, somewhere in the middle of the inference boom, you're using MXM cards. In either situation, the host machine has DDR, but if you're OpenAI, Anthropic, Microsoft, or Google, you're not building (more) inference machines like this.
The first two are buying Nvidia's all in one SBC solution: unified HBM, onboard ARM CPU to babysit the dual GPUs, has its own dual QSFP network controller that can RDMA, etc. No DDR or GDDR involved. Any machines built before this platform are being phased out entirely.
Microsoft is doing the same, but with AMD's products, the MI series that co-locates Epyc-grade Zen 4/5 CCDs with CDNA compute chiplets, running the entire thing off HBM, thus also unified and no DDR/GDDR needed. They, too, are phasing out machines older than this.
Google has a mix: they offer Nvidia all in one SBCs as part of GCP for legacy inference tasks (so your stack that can't run on AMD yet still can run), but also offer the same MI products that Microsoft offers via Azure's inference product, but also has their own TPUs that some of Gemini runs on; the TPUs run on HBM afiact. No DDR or GDDR here.
So, what does AMD or Intel do here? Lets say they waste fab time to make their own dies on the wrong process (TSMC and Intel-Foundry do not have for-RAM optimized processes)... they would be producing DDR and GDDR for a market that almost has its entire demand met. Intel lacks the die stacking technology required to build HBM, and TSMC I think can't do it for that many layers (HBM has 8 to 16 layers in current gen stuff iirc).
Micron, for example, already is bringing two large factories online here in the US to meet the projected growth in demand for the next 20+ years. When these factories finally start producing, it will not change the minds of speculators: they still seem to think AI datacenters need RAM, of any kind, and refuse to understand even the most basics of nuance. Also, when they come online, HBM will be a minority product; the AI inference boom is still just a bump in the road for them.
Nvidia kinda screwed their consumer partners, btw: they no longer bundle the GDDR required for the card with the purchase of the die. There is a slight short term bump in GDDR spot prices as partners are building up warchests to push series 60 GPUs into production, and once that is done, spot prices return to normal (outside of the wild speculation manipulation).
One last thing: what about LPDDR, used by AMD Strix Halo and Apple stuff? Speculation seems to have not actually effected it. I consider it as a sub-category of DDR (and some dies seem to work as either DDR or LPDDR as of DDR5, due to the merger of the specs by JEDEC), but since it isn't something you find in datacenters, it seems to have avoided speculation.
The Ryzen Max CPUs mentioned in the linked article? Uses LPDDR. Doubling down on the Ryzen Max product line might be a brilliant move.
The commenter is also not very smart and does not realize companies making the RAM can trade capacity of one for another and any re-tooling at current price is still profitable.
The commenter also does not realize that is also true for lines currently making SSDs
Flash chips haven't been speculated on nearly as hard, and are suffering from the same sort of weird lack-of-nuance. Samsung, for example, isn't reassigning capacity to meet some sort of phantom datacenter demand that isn't already there, generically, across all datacenters, AI or not.
A lot of SSD price skyrocketing is largely "SSDs have RAM on them for cache", not "SSDs have flash chips, and they're both made at the same fabs"... which oddly effects low end SSDs that don't have external cache.
To make it worse, for the speculators who do understand this, because it isn't some universal homogeneous group, the flash chips that go into enterprise SSDs aren't the same that go into consumer SSDs.
The Big Three still aren't doing some major re-tasking of capacity, as the actual global demand isn't outstripping supply any more than normal. There is no short term problem to fix, speculators are just gonna have to stop hoarding toilet paper like its the start of Covid.
Edit: Oh, and if you want to ask how AMD/TSMC or Intel solve this? They can't, same reason why making their own in-house HBM isn't happening.
Micron killed Crucial to focus on AI.
Micron killed Crucial because Crucial was a weird offering that competed with their own partners. This was always a weird problem, and it just didn't make financial sense to continue with it. One of the analyses I read was Crucial was less than 12% of sales.
Like, don't get me wrong, I've liked many Crucial products over the years, and even recommended some of them, but it was always weird they were trying to out-compete companies like Adata and other major ODMs.
The counterexample of this is Nvidia absolutely trying to kill their partners, and going to first party assembly and sales of products. Nvidia isn't even going to PNY anymore for ODM needs, but going directly to Foxconn.
Micron execs claiming its because of AI is a bit weird and revisionist, because they've been working on exiting the Crucial brand since long before they publicly announced it. The public didn't learn of any such plans until right before the Ballistix brand sunsetting was announced in 2021, but started years before that. Like, I know they're just playing to their shareholders, but its still a bit weird.
Also, with the FEs, their partners are disallowed from making their own FEs, even if they make their own PCB from scratch and not based on any existing Nvidia design. Doesn't matter who makes the FE, it immediately puts partners at a great disadvantage if they can't make one too.
Good luck actually finding them on stock with 128GB+ RAM. I got strix laptop while ago, now price in EU is technically the same, but no stock. Maybe month or three
There is also claw hype. And large gwen3.5 models can run very well on DDR5 CPUs or mac minis...
Micron, Samsung, and Hynix just basically sell you chips that comply with the JEDEC spec, and the DIMM manufacturers further bin them according to purpose. The highest end chips (that are stable at high clocks and acceptable voltages) end up in enthusiast performance products, the ones that don't work well at all but still meet JEDEC spec are sold to Dell/HP/Lenovo/etc for Grandma's Facebook machine, and the ones that are exceptionally stable at thermal design limits are plunked onto ECC DIMMs and sold to servers.
Also, as others have mentioned, its just a fab, and it can make any of the dies they're able to make. Whatever needs to be made to meet demand, they make, they just can't turn on a dime and react to quarterly concerns, and are locked into cycles that may range from 6 months to 18 months.
Side note that is also worth mentioning, sometimes you can order special bins of parts with features that wouldn't normally be available if you're willing to order enough. Recent example being Nvidia buying overclocked GDDR6 chips from Micron with additional features enabled; Micron was more than happy to become Nvidia's exclusive supplier for the custom GDDR chip if Nvidia was willing to buy out the entire run. Stuff like this happens every so often, but isn't the norm.
https://investors.micron.com/news-releases/news-release-deta...
You just need an additional chip to move from "consumer grade" (ie no parity) to "server grade" (ie have parity). ECC support is actually in the memory controller which is in the CPU for the last 15 years. No magik.
There's nothing in that press release that implies that the memory was somehow different (or "consumer-grade"). The _only_ thing they're saying is that they're ending their B2C business and focusing on B2B.
As a small buyer of all of those things supply at nearly any price has gotten very difficult to reliably predict week to week. When a lot of 100 64GB DDR5 sticks shows up available at a vendor, it’s at a take it or leave it price good for a couple hours. If I don’t pull the trigger they have another buyer for it and I might be waiting another month before anything becomes available again. We can no longer JIT for even failure replacement on our edge nodes.
Then you have the NVMe and even SATA SSD shortages. Still a bunch of very useful hardware out there I would love to find a decent deal on 8TB sata so I could repurpose it. Just doesn’t make any sense right now at current pricing and availability. Good luck trying to even find a batch of 12 of these disks at a time.
This goes for both enterprise and even prosumer I was willing to take for some of these uses.
Datacenter customers, for example, have repair parts on hand; boxes of harddrives/ssds waiting to be put in, boxes of consumable parts, DIMMs waiting to replace ones that went faulty, entire machines already racked and waiting to take over for their fallen siblings, etc. Some of these customers added more to the spare parts pile. The big clouds manage their elastic demand of any sort of consumable or repair parts in volumes that are described in terms that fit cargo trucks in a quarterly basis, and they've already compensated.
Now, otoh, you have the truly psychotic people, that fill their basements with toilet paper, just hoarding more than they could ever use in their entire life. We've all seen that story where a guy was going to lose his house because he blew his mortgage money on toilet paper, and was selling it at a loss just to stay afloat. People like this exist in every crisis, and there's gonna be a headline in the near future where someone is gonna lose their house because they had like a hundred trays of DIMMs in their basement.
A few people I know who scrape eBay like its their job for electronics are just waiting for people to start fire-selling DIMMs and SSDs that got hoarded and they couldn't scalp people over; they're expecting half of MSRP or better sometime later this year.
Are you claiming that these speculators are buying DDR5 RAM and warehousing it somewhere? Or what exactly is the mechanism you are proposing here?
To me it seems much simpler - AI companies want HBM, but HBM and DDR5 share the same wafer production process and facilities, but the HBM process is much more fragile and takes three times the wafer production.
There isn't enough DDR5 RAM being produced, so prices go up.
This only effects a very narrow slice of highly budget conscious consumers trying to build high end PCs at razor thin margins.
All 4 of my "top end PCs" have 128GB RAM. Me server (I self host everything is 512GB). Lucky for me all were bought before that insanity.
Did you adjust for inflation ?
There is nothing vague about the question if prices were scaled or not (and in this pretty much unvague coefficient of ~2x between usd in 2000 and 2026), otherwise there is point in comparing these numbers, as there is no point in comparing inches and cm's without declaring beforehand which number is which.
You cannot possibly look at the price of ram now compared to six months ago and be so fixated on including inflation. Obviously inflation occurred and obviously after 20 years it has an impact on price. But we are all on HN and all know what inflation is, so forcing people to drill down on its contribution in order to advance the conversation when it clearly only accounts for a small portion and we all know it’s a factor is absolutely ridiculous. You know this, we know this, and yet here we are still talking about it. I may as well explain what ram is if we want to get this elementary about things.
True or false: ram has become substantially more expensive in the last 6mo in a way that cannot be meaningfully explained by inflation.
There is a very clear, very obvious answer here. Inflation or not.
Have no idea why you are keeping arguing about something which was not my point to begin with.
But that was not my point _whatsoever_. What I said is - every time you bring the explicit numbers (like in GP "$500 for 32GB is about $15/GB which is a high we haven't seen since the mid-2000s") you _absolutely_ have to adjust for inflation to have a meaningful conversation. This is it.
https://web.archive.org/web/20240805053759/https://jcmit.net...
https://thememoryguy.com/dram-prices-hit-historic-low/
Inflation applied manually; https://www.bls.gov/cpi/
https://www.neowin.net/forum/topic/983036-latest-steam-hardw...
Steam hardware survey GPU history: https://www.youtube.com/watch?v=wHTdnIviZTE
Edit: I don’t get your math. If we’re using a very generous definition of “top end,” even neglecting Nvidia and going AMD - which some would argue makes it not top end - you’re talking conservatively: $600 for a GPU, $500 for 32gb of ram, and $500 for a CPU. $1600 before PSU, case, SSD, fan(s), mobo…there’s no world in which you’re coming in under $2k. The SSD and board will put you over immediately.
You’re talking 3/2025 prices, not 3/2026. A compromise, mid-range computer is $1500 to build now.
Bestbuy is selling ram and storage at the same cost as everyone else. I imagine Walmart is not much better. I’m also not sure what you do with all the bundle parts that you don’t need. Do you sell them? Where do you sell them?
What deals did you take advantage of? What did everything cost you in the end/when did you build? If you don’t feel like answering that’s fine but it’s valid to remain skeptical given all the evidence to the contrary. Perhaps you’re just really good at finding deals but you can look around this thread and see that we are all telling the same story. Building a computer has gone up $600+ for common builds over the last 4-5mo on top of the already inflated GPU prices we’ve been experiencing for years. If you put my exact build I did last April into PC part picker it is an additional $500+ to build now, and that’s with an AMD GPU to keep costs down.
It’s strange times when Mac minis are a budget-friendly computer. Building a half decent PC for less than $1500 is a serious challenge now. Things are so volatile valve still hasn’t released or even set a price for the new Steam machine.
I'm not sure what you expecting to hear. What do i do with parts I'm not gonna use? What are you talking about? Don't get the bundle is you're not going to use what the bundle comes with, simple as that. Have you not shopped before?
Currently computer components are not cheap and it does not look like it's getting any better.
I currently have two moderately good laptop that either i sell or keep for back up.
That is to say, if you want a system that keeps up with 32 GB Ram, you'd be already willing to spend alot what with options for noctua fans, water cooling, higher end MOBOs, premium cases, OLEDs etc. If you can't afford that then you won't be buying expensive DDRD5 RAM either.
I built my current PC (9800x3D, 9060, 32gb DDR6) last April for about $1800. It would cost almost $3000 now between storage and ram increases. The economics have completely shifted. Everything is more expensive except basically the PSU and case
AMD has no equivalent to NVidia in the high end, it isn't tax as it is functional monopoly
Ram and storage have ballooned PC costs. That’s the issue. Whether you are buying an AMD GPU or an Nvidia GPU, it is still substantially increasing build costs. Nobody is spending $1500 on an Nvidia GPU and then going “well nothing else matters now.” The ram and storage has gone from $200-$300 to $800-$1000. That’s still a huge portion of the budget. They’ve gone from near-line item status to 1/3rd (or more) of the cost. Affordable builds have become incredibly difficult to achieve
Apples to oranges. Why are you comparing RAM prices to CPU prices? It's different hardware.
$500 for 32 GB is insane. Just 18 months ago, I bought 128 GB of DDR5 for only $480.