Says, not inflation-adjusted. With reason; adjusting those 1960-1980 prices for inflation would make the graph a lot taller.
Pricing "per GB" before 1990 is unrealistic, though; nobody thought in GB or purchased GB quantities, or conceived of GB systems. I remember a moment circa 1973 when I saw an IBM CE about to do an upgrade on a 370 system at Cal Berkeley. He had a box with several carefully-packed, large circuit boards. "So, is that a megabyte?" I asked. "Yup, that's a meg."
The author clearly wasn't implying that these were 1GB chips. They just wanted to show a graph scaled per unit of memory. It could just as well have been per byte, and the graph would've been identical but the values on the left would be changed by a factor of a billion.
You could argue that you'd rather see a "price per typical-sized RAM chip as sold at the time". That would also be a perfectly valid thing to graph (though a bit more subjective), but it doesn't invalidate this one. Since per byte (or GB or whatever you want to say) has continued downward all this time, it makes the recent spike all the more notable.
(I'm not sure it's right to label vacuum tubes and core memory as "DRAM" though.)
The problem with "price per typical-sized RAM chip as sold at the time" is not so much that it's subjective, but you'd get artificial spikes in the data.
For me, the key take home from the graph as stands is that price per GB right now is about the same as 2020. That seems reasonable, it's more expensive than it was, but only outrageous if you forget what it was like only a short while ago.
But back in 2020, 4GB or 8GB sticks were most common, a few years ago it was up to 8GB, 16GB or 32GB, and now 2x8GB seems to be the most common high-end configuration or 2x4GB for low-end again. If you'd jumped from 8GB sticks to 32GB sticks and back again, it would seem like there was a spike up around 2021-2 and that memory was cheaper now than a few years ago.
I think the main driver for the data is that probably consumers or the market decides on a reasonable price for memory, and people buy whatever they can get for that money. When I had a Z80 computer in the mid 80s, 64KB expansion RAM was about £100. For a similar computer but a few years earlier, a 32KB expansion RAM was about the same price. When I had an Amiga in the early 90s, a 512KB expansion RAM was again around the same price. In the 2000s, a couple of MB was around the same price. Maybe 5 years ago, the market was split a bit and a 4GB RAM was around £60 and 8GB around £120, but maybe this reflects "under $100" as the ideal target. A few years ago, it was similar but 8GB for around £80 and 16GB for around £160, now it's "doubled" in price, it's just back to 8GB for £120 again. But whatever the decade, it seems people are prepared to spend about £100 on memory for an average PC.
IMO you'd really need to graph the price of how much RAM is needed to comfortably run contemporary OS + software combinations. That will get you an actual picture of the pain inflicted by the RAM prices.
And yes, RAM demand goes up with the average RAM in computers but it does lag and it's not yet clear if it will go down with increasing ram prices as IT corporations can still afford the more expensive RAM needed for the developers to run the RAM-hungry applications they need to run, which means they won't be dogfooding their software in a normal budget user environment and are less motivated to optimize for a reasonably priced amount of RAM.
Depressing to see how much discussion on HN (!!) has resulted from a objectively terrible graph reading. I mean... in the process of our infiltration by finance bros and VC money, have we genuinely forgotten how exponentials work?
What? Gold underwent a massive revaluation with the end of Bretton Woods in 1971. It was prior to then that government were actively involved in making the price of gold artificially low.
Fun fact: under 31 U.S. Code § 5117 a troy ounce of gold is still valued at 42 and 2/9ths dollars.
Gold became much ‘cheaper’ in the period 1945-70 because there was a series of technical revolutions in South Africa mining (never mind the apartheid …) This is why Breton Woods lasted as long as it did.
The countries like France that conspired in resentment to break it as ‘privilege’ are now effectively in flames.
The right to a civil jury trial in the 7th Amend. to the US Constitution is only available where the value in controversy exceeds twenty dollars (a massive sum in 1791, and a trifle now).
You got it backwards; ending the gold standard was very much a unilateral decision by the United States because Nixon couldn't handle making politically unpopular decisions to cut spending and/or end the Vietnam war. Many countries that had their gold reserves held via US dollars were livid.
Well, there's an arguable point of fiscal policy there, and a conspiracy rathole that no one wants to excavate.
But none of that matters here. As grandparent comment indicates, you're making a pretty fundamental math error. This is a log chart. Government may have devalued currency. It did not do so exponentially.
I wouldn't go so far as to say "nobody". Electric Boat had 2 GB memory in one of its systems at that time, with the hardware capacity to increase to 4 GB. It sounded insane at the time, but it absolutely existed, and thereby seems reasonable to include it in any research of historical pricing.
Yes, you really need "dollars per amount of RAM you need for standard computing tasks." Windows 11 requires a bare minimum of 4 GB of RAM, Window 10 only needed 1 GB.
If what you're interested in is fluctuations in production versus demand then you absolutely do not want a subjective metric. Measures of the form dollars per unit, units per watt, units per flop, etc are what you're interested in.
Discussion of memory in terms of words and their bit length, time to complete a task is more meaningful to intent on use and compaction, see greycode technique. Dollars of slop a unit sacrifice skills in the industrial base for the gain of paper profits at the repeating business meetings.
Have you ever actually tried using Windows 10 with 1GB of RAM? I wouldn't consider it suitable for "standard computing tasks."
And that's the hangup, what do you consider a "standard computing task?" On what OS? Running what software? How well? Plenty of people were still using XP in 2009, so is 256 MB of RAM okay for "standard computing tasks" in 2009?
> Have you ever actually tried using Windows 10 with 1GB of RAM? I wouldn't consider it suitable for "standard computing tasks."
I have [0], and it's actually not quite as bad as you would expect. It certainly wasn't fast, but I had no problem using it for basic web browsing and document editing. The painfully slow hard drive and processor speeds on that computer actually caused more issues than the lack of RAM.
I remember installing Windows 10 on my laptop when I was in high school, it was a decommissioned Thinkpad T410s from my dad's work. 4GB of RAM, Nehalem i5, a very early consumer SSD (OCZ, if I recall). I vividly remember my first thought being "wow this is really slow." Win7 ran like a champ on that machine.
My experience with Win10 on that laptop actually led me to buy a dumb gamer laptop for college. As those all do, it died prematurely, so I ended up back on the T410s for a while. I put KDE Neon on it. It was great!
If you're saying that you can install and use Win10 on a laptop with 1 GB of RAM, well yes I acknowledge that is true. But it's a purely academic exercise, it's not actually a usable computer for the overwhelming majority of people.
Maybe it would have been fine for my grandma. She was using a Pentium II running Windows XP to go on Facebook in the early 2010s.
Well it's complicated. Y2K was a combination of logic issues and the consequences of certain inefficient ways to store dates, like text and BCD. Migrating to binary could fit plenty of dates into the same space or even less.
In particular, 16 bits is enough to store the entire date, year month day, from 1900 into mid 2079. Any date format that couldn't go past 1999 was probably using 24-48 bits.
Abstractions on abstractions on abstractions; background tasks and their abstraction stacks; increased cache and buffer sizes to take advantage of increased typical memory capacity. For an example of the latter, handling TCP on a Commodore 64 is a problem because the memory can only fit about 45 packets with nothing left over, but now you can just allocate a megabyte receive buffer per connection.
Neither do the developers, because until recently, RAM was so cheap it didn’t matter, and we were in a situation where almost no one ever needed to consider “how much RAM will this take?” when writing code.
Like Homer Simpson said about alcohol: "To alcohol! The cause of, and solution to, all of life's problems", we developers can say about AI (assuming AI-assisted coding can save memory in popular programs, OSes, etc)
I once held in my hand the main part of a ferrite core memory module from the early 70s. It was kilobytes at best.
I also recall looking at recommended requirements for Dungeon Keeper 2 - 266MHz CPU, 64MB RAM and thinking "that's absurd - no such device exists!". I was a kid back then, so what did I know?
Later on in college a friend showed us his absolute monster of a laptop with a whopping 8GB of RAM - he could spin up several VMs on one device! Groundbreaking on a (nominally) portable device.
So yeah, safe to say the notion of gigabytes of RAM anywhere close to a regular person belongs firmly to the 21st century.
So total system cost per unit of memory is going up.. 2GB costs in 1985 was $2 million (from the graph), a cray-2 was $16 million (from wikipedia). A GPU server with 8xB200 today can be had for ~$500k (estimate), 1.5TB memory is $25k (from the graph).
The natural unit of measure for integrated circuits is a power of 2 since that's what the systems operate in. It's so natural that early 9 and 36 bit architectures were squeezed into 8 and 32 bits as it just works so much more efficiently.
Long term storage and communications? Those start to introduce things like human division of timings, frequencies, and other analog systems like rotating disks. It still generally makes sense fab actual flash chips in various powers of 2 though. The discrepancy there tends to be various forms of 'overhead' for the translation table / wear level indirection, over-provisioning, and even variations in density caused by different levels of physical cell utilization.
Still, most network stuff ships around packets of 'up to' 1500 bytes ( https://en.wikipedia.org/wiki/Ethernet_frame and lets just exclude jumbo frames ) so arguably it'd be better to talk about all computer measures in binary powers of two, exclude the marketing huckster trying to make things more impressive by shoehorning SI engineering units into a realm that uses binary math.
If my memory serves me correct (no pun intended), when I was a kid I remember bugging my mom to buy me like 2 or 4 1 MB modules, it was at least 50 bucks or 100 bucks each.
Now everyone's going to talk about how cheap everything is by comparison - but someone needs to talk about how oppressively hungry browsers and OSes are compared to in the past. This is no HIMEM.SYS
There’s been a sharp divergence in memory requirements. Talk to developers and they think that 32GB is the bare minimum these days, with 64GB or more preferred. They’ll point to Electron and Chrome tabs and everything else.
Then you sit down with an average computer user on their 8GB RAM MacBook Neo and they’re in love with how fast and smooth it is, even with their chrome tabs and the company Slack up and Spotify in the background.
I still have an older 8GB machine to kick around with on the go when I don’t want to haul the expensive laptop. It’s fine, even for a lot of development.
Some tasks simply require more RAM. Compiling big software, for instance, wants as many CPU cores as it can get, and each compiler instance needs some amount of RAM to run efficiently. It's not unusual for a 32-core build to need 32-64GB of RAM to run at full speed. Work on a smaller program, though, and 16GB is absolutely fine.
Because they use one or two apps at at time, the ones they must spend all their time to perform their job. E.g. Excel and a web app to work on invoices and a stack of paper documents. I see 8 GB on Windows PCs too.
I think it would be better if one has the discipline to just use older machines and play older games and only visit certain websites that can be visited on older version of browsers. A second-hand 16GB laptop can go a long way.
But yeah that probably sucks from time to time, especially for young people.
You can do a lot on old machines but developers also need to optimize a bit. Youtube almost plays on a 20-year-old machine, which means with some effort it'll play just fine. Most the other sites work just fine.
My laptop has 8 GB. I write blog posts, I have a dozen-ish tabs open, I do KiCAD things (including 3D renders!). Works great. I was doing Verilog synthesis on a similar machine in college in 2020.
The truth is that, if you do the same things you were doing with your computer 10 years ago, well then you don't need a new computer!
If all you do is write books, a Pentium III will do the job just as well as a brand new PC.
Of course, the web throws a wrench in this. Word 2003 is still far more capable than Google Docs, yet tons of people opt for the cloud slop because it's convenient and free-as-in-beer. And, Google Docs will continue to become less efficient with time.
Look at it this way: while the upfront cost to scale up production is huge, prices are now high enough to justify it even if demand is expected to drop abruptly later on. So if you can wait 5 years for your next PC, 1TB RAM might go for what 64GB would have cost without the AI demand spike.
Granted, if you need a new system before then, you're SOL.
One thing to look out for is supply capacity curiously going offline in 2030 or whatever. That would hint at market power or collusion.
It’s possible we’ll see a huge price drop on the near term but SSD + Cache + GPU’s seems to have changed the equation where RAM speed is considered more important than size. And from a pure architecture standpoint it makes sense.
Lowest 2012 price listed is 3.7 (2012-10-30) vs highest listed in 2026 is 5.375 (2026-2-1), which overlaps based on the margin for error involved. https://www.usinflationcalculator.com/
And to think that AAA video game was distributed on cartridge with expensive memory in cardboard packaging with a manual while these days that same $60 game is just a digital download from a CDN.
How is it crazy? The marginal cost to produce a copy of a video game is ~$0 and the market is much larger now then it was so fixed costs can be spread around more. The sticker price was always more or less arbitrary chosen - i.e. to maximize profits, which includes how the number is perceived.
> while the upfront cost to scale up production is huge, prices are now high enough to justify it even if demand is expected to drop abruptly later on.
Given the nature of the industry and how critical the product is I think it would make more sense for governments to bankroll fab construction in a way that the public takes on the risk of consumer prices falling below a certain level within some limited timeframe. Mildly subsidized chip production seems like a much better downside than the current sky high prices.
The log scale is nice to compare decades. Wether it's inflation-adjusted or not isn't too important but it's still a factor of 10, which would show in a linear recent graph. The fact that we're comparing GBs instead of the average RAM stick shows how much the price has decreased per GB rather than per unit (much smaller decrease).
But a linear graph that represents only the last decade and where the bottom is 0 (not the min value) would tell a different story, but I guess we already know that story because we're living it.
TIL someone took over the now defunct jcmit dataset[1] (archive[2]). I expected his dataset to die off when his website did, but I guess someone found the data dump on archive.org and revived it. Which raises a question: how will this dataset fare five years from now?
In the first graph, if you hover over the DRAM line you'll notice that the most recent data points are for DDR3. One of the data points from 2025 is a 2 GB stick. This paints a more rosy picture than the situation deserves.
Arguably they already did with the "cloud native" systems. There were plenty of examples personally known to me in the mid and late 2010s of smaller tech companies trying to run production PostgreSQL on 8-16 GB of RAM because they didn't want to pay the cloud RAM tax. Many "cloud native" systems were designed under these (mostly artificial IMO) RAM constraints.
Is that because the amount of available memory is limited for a single process? You can always add more storage and storage access is relatively the same regardless of whether it comes from the SSD inside the server or sits in another rack. Storage is a pretty linear cost when you're a cloud host buying storage in the hundreds of PB numbers. Whereas for memory, if you want the whole thing, you need the whole server even if your process is light on CPU requirements.
One could also blame crypto and AI (they're clearly responsible for some of the volatility in the graph), but I can see the curve flatten in the 2010s, just as Moore's law ended.
1979 to 2009 in the OP graph has a pretty steady drop from 10^7 to 10^1 USD/GB: 6 OOMs in 30 years. Then till before the recent spike it was around 1 OOM in 15 years: 1/3 the rate of progress on a log scale.
When it comes to CPU progress we blame the end of Dennard scaling several years before the knee in this memory curve. I'd guess the story of memory is similar in also hitting technical difficulties, but I don't know.
Heh, unintentional. Another term shorter than spelling out "orders of magnitude" is "decades", but I figure that's less familiar and even more confusing here. "Memory price started out falling two decades per decade..."
Moore's law didn't end in any broad sense and certainly not that far back. It's a tiring piece of misinformation that just won't die.
Progress has consistently become more difficult (ie more expensive) but has generally kept up. The scaling of a couple specific technologies noticably slowed down a few years back but that's not the general case.
The node names aren't representative of the reality.
This graph is the touchstone one should rub all the "RAM and storage are no longer a commodity" bs that micron, sk hynix, samsung, western digital, seagate and others are peddling as of late while the valuation of their companies have gone from "supplier of widely available fungible goods" to "state-of-the-art moat AI backbone tecnology".
turns out things are not that bad! we just rolled back to 2010.
oh, wait, now every app is a browser instance. shit.
EDIT: so, how did I arrive at 2010, you ask? I looked at DDR5 pricing and found the closest pricing per GB in the past. this turned out to be DDR3 memory. I think it's totally fair since it was the latest and greatest thing back then, much like DDR5 is now. although, if we compare DDR3 to DDR3, we still roll back pretty far - a very close to current price was spotted in 2018, '17, 15, '13, and '11.
Will they..? It seems equally (or perhaps more) likely that we'll increasingly see vibe coded browser or Electron based applications as the bar is now lower to build such a thing.
yeah but you also have commercial licensing with Qt specifically :))
or we are going to see an explosion of vibe-coded GPL apps.
anyhow, the likes of Linear and Notion ain't gonna abandon web and go Qt. or!! if we are very lucky, we can see a native app framework that ticks all the boxes of a modern UI framework and is permissively licensed, but we need this crunch to stay there for years.
That doesn't apply so long as you are willing to accept the LGPL. In practice that means you can statically link everything except QT so that the end user is free to drop in a modified QT version if he would like.
If you are going to vibe code you can just pick any language you want. I had a go vibecoding in Rust and it worked perfectly fine. Even better than vibe coding in JS/Python because the type hints give the LLM a faster way to check progress.
The memory manufacturers have made an interesting mistake. The tech giants of the world will be working to replace them from the supply chain as soon as possible. China already makes it's own RAM albeit at 16nm but you can bet they are working to get down to 4nm.
DRAM hit a barrier at 10 nm a few years ago[0], so 16 nm is actually even closer to state-of-the-art. E.g. Micron newest node (1-γ) is their sixth at 10 nm [1] and their first EUV-based node.
The problem is that DRAM is fundamentally based on storing charge in a capacitor and how much charge a capacitor can store is a result of the geometry of the capacitor. So either someone will have to figure out a way to make the same size capacitor take up less space on the RAM chip (this is what broke the previous 20 nm barrier) or someone will have to invent a practical way of making RAM with less than 1 capacitor per bit.
China often exhibits tremendous internal market competition, so it's possible that different Chinese suppliers will race each other to the bottom (Chinese firms are really good at suriving on ultra-thin margins) making prices even lower than 10% below the premium providers.
More competition will drive down the market price, so it'll be 10% below a price that is lower than what we have currently. Obviously it's not gonna go down immediately, but more supply will definitely bring down prices.
Obviously, this is very bad for the existing memory makers, since these boom prices will not last forever, and the Chinese aren't gonna stop selling memory once they are in the market.
Unfortunately, this is unadjusted prices, and this failed to annotate where the cartel years and when the cartel was 'broken up'. Not a bad assignment's work but clearly lacking the domain awareness necessary to report the complete story through graphs.
It's amazing how consistently thr lower memory cost have expanded the set of economic viable applications : cheaper hardware doesn't just improve existing software it also enables software that was not possible before
Tim Cook recently likened memory prices to a "hundred year flood." Looking here, it looks like 1988 was at least as bad, with prices going up 3X+ from $160/MB to $500/MB.
This is probably the first real thing that is affecting me personally with this whole AI business. Having to pay more for device upgrades going forward. I hope the demand settles or new memory production offsets the demand.
The truly absurd part is that datacenters are barely being built and those that are built can't be turned on because they don't have enough power. Satya Nadella admitted recently that they have warehouses full of unused hardware because they a) can't get datacenters built and b) don't have enough power i.e. this whole RAM scandal is a bloody joke. If OpenAI goes bust (their financials are a huge mess and lots of red flags - they might not even survive to IPO!) and then what will happen to all those "inked" deals to buy all that RAM?
If it were possible, it would have been done already. The issue is the capacitors are already tiny, and barely can prevent a single bit decaying before refresh.
do you have a reference to exact / realistic scaling laws for the leakage currents as function of capacitor/dielectric dimensions and access transistor dimensions?
using 4 (or 2^N) voltage levels stores 2 (or N) bits, so we can afford to make the structures larger
why would this approach make sense for NAND flash but not DRAM?
Can you back that up with anything about semi-recent nodes? The voltages are so fragile that I'm not convinced you would actually save space once you adjust the design to handle more levels.
You could also do a computing pr dollar graph - which would be a similar sharp decline over the past decades - however it won’t show anything like the memory price spike of the past few years.
A perfect example of how graphs are often misleading. $/GB is a totally useless unit value because it's an arbitrary size. The unit needs to be tied to the relative usefulness for its time. The y axis should be something like $/average workstation memory or $/requirement for common compute task. It's obvious that ram is expensive right now, but it's not expensive per GB. It's expensive relative to what you need to accomplish a useful task.
But relative usefulness is entirely subjective, making it a meaningless unit. Depending on your use case you may need 256 GB or 0.5 GB.
The audience who would benefit from hypothetical $/usefulness would be people who don’t know what memory is and don’t know what’s inside of their computers, or what it does. This is a fine audience to be in and to serve, but obviously not the audience of that website and not HN.
If you think that audience is under served for memory market statistics, I encourage you to make such a website and serve that audience.
For people on HN, who do you know what memory is, $/GB is a fine metric.
This is assuming that the wide variety of use cases are evenly distributed and that larger use cases are not mostly just a lot of duplicated smaller use cases. If I have a website I will need X amount of ram. If you run a much larger website offering a comparable service you will need some multiple of X, but you don't actually need much more ram per user (assuming you're also accounting for extra infrastructure and not just the web servers). It's the same task just scaled. Relative usefulness is not subjective, you could look at a variety of tasks in different industries. Windows server 2012 had a minimum requirement of 512 MB. Windows server 2025 has a minimum requirement of 2 GB. That's 4x for the same task which totally distorts $/GBs usefulness for being able to tell you anything helpful economically. It's obviously good to collect this data, but you need to pair it with some kind of demand data for it to actually tell you anything.
> you need to pair it with some kind of demand data for it to actually tell you anything.
Again, this is entirely dependant on who is consuming the statistic and for what purpose. For some use cases, yes demand data will be quite crucial. For others it will not. It's quite apparent the site's author doesn't see this as crucial and for the purposes I need to consider memory pricing, I agree.
Gas is priced in $/gal, not dollars per mile or hour of lawn mowing or whatever. The resource and the use are completely different concepts and the resource owner/producer cares not of the buyers purpose for it.
I use a Thinkpad T530 for reasons that are very important to me. It is the only laptop that I have, so it is what I use for every manner of portable computing.
It still does all the things I want it to do, including using modern websites with modern browsers on modern operating systems (including Windows 11).
The 2012 computer running a modern linux install will still work fine. I'm talking more about the specs, specifically memory. I had 8gb of ram in my computer in 2012, the Macbook Neo released this year still has 8gb and is usable for modern day tasks.
We don't _need_ that much ram, we just found new things to do with more.
I guess ‘per GB’ doesn’t really capture it, because the base number of gigabytes available to people (ie- the smallest compatible RAM kit you need to build a computer) and the base number of gigabytes you really need (OS bloat, feeling responsive, etc) have gone up so much.
I don't think that's true - instead software will be exactly as efficient as it needs to be to run on the developers workstations. And companies throwing sky high salaries at software developers aren't as price sensitive when it comes to those workstations as the average user.
This is extremely misleading and not very useful. It makes little sense to use pricing per GB during decades when RAM was at most in MBs. In that case, why not talk about price per TB or PB? Then the line will look pretty much flat and horizontal.
this should really show average install cost. Even phones have 8-12gb because the software is so atrocious. It would be like comparing cars by pricing per horsepower. Nobody is running a 12 horsepower vehicle on the highway, and doing so would be dangerous because of the change in average power & speed.
With respect... I am surprised such a low-quality analysis is published on stanford.edu . What is compared here? What is the purpose of this? What are the conclusions of the analysis? Heck, where is the analysis? By what logic are the prices per GB(!) comparable between 1960(!) and 2026?
I am sorry to being rude, I just don't understand this publishing beyond getting the media exposure.
Says, not inflation-adjusted. With reason; adjusting those 1960-1980 prices for inflation would make the graph a lot taller.
Pricing "per GB" before 1990 is unrealistic, though; nobody thought in GB or purchased GB quantities, or conceived of GB systems. I remember a moment circa 1973 when I saw an IBM CE about to do an upgrade on a 370 system at Cal Berkeley. He had a box with several carefully-packed, large circuit boards. "So, is that a megabyte?" I asked. "Yup, that's a meg."
The author clearly wasn't implying that these were 1GB chips. They just wanted to show a graph scaled per unit of memory. It could just as well have been per byte, and the graph would've been identical but the values on the left would be changed by a factor of a billion.
You could argue that you'd rather see a "price per typical-sized RAM chip as sold at the time". That would also be a perfectly valid thing to graph (though a bit more subjective), but it doesn't invalidate this one. Since per byte (or GB or whatever you want to say) has continued downward all this time, it makes the recent spike all the more notable.
(I'm not sure it's right to label vacuum tubes and core memory as "DRAM" though.)
> I'm not sure it's right to label vacuum tubes and core memory as "DRAM" though.
Core memory does need a refresh after a read, but since it doesn't need refreshing otherwise, I'd mark it as SRAM.
Williams tube memories seem DRAM like enough to me though.
The problem with "price per typical-sized RAM chip as sold at the time" is not so much that it's subjective, but you'd get artificial spikes in the data.
For me, the key take home from the graph as stands is that price per GB right now is about the same as 2020. That seems reasonable, it's more expensive than it was, but only outrageous if you forget what it was like only a short while ago.
But back in 2020, 4GB or 8GB sticks were most common, a few years ago it was up to 8GB, 16GB or 32GB, and now 2x8GB seems to be the most common high-end configuration or 2x4GB for low-end again. If you'd jumped from 8GB sticks to 32GB sticks and back again, it would seem like there was a spike up around 2021-2 and that memory was cheaper now than a few years ago.
I think the main driver for the data is that probably consumers or the market decides on a reasonable price for memory, and people buy whatever they can get for that money. When I had a Z80 computer in the mid 80s, 64KB expansion RAM was about £100. For a similar computer but a few years earlier, a 32KB expansion RAM was about the same price. When I had an Amiga in the early 90s, a 512KB expansion RAM was again around the same price. In the 2000s, a couple of MB was around the same price. Maybe 5 years ago, the market was split a bit and a 4GB RAM was around £60 and 8GB around £120, but maybe this reflects "under $100" as the ideal target. A few years ago, it was similar but 8GB for around £80 and 16GB for around £160, now it's "doubled" in price, it's just back to 8GB for £120 again. But whatever the decade, it seems people are prepared to spend about £100 on memory for an average PC.
IMO you'd really need to graph the price of how much RAM is needed to comfortably run contemporary OS + software combinations. That will get you an actual picture of the pain inflicted by the RAM prices.
And yes, RAM demand goes up with the average RAM in computers but it does lag and it's not yet clear if it will go down with increasing ram prices as IT corporations can still afford the more expensive RAM needed for the developers to run the RAM-hungry applications they need to run, which means they won't be dogfooding their software in a normal budget user environment and are less motivated to optimize for a reasonably priced amount of RAM.
> adjusting those 1960-1980 prices for inflation would make the graph a lot taller.
It won't though. One dollar in 1960 is just about ten dollars today. The graph is already in logarithmic scale so it won't make much difference.
Depressing to see how much discussion on HN (!!) has resulted from a objectively terrible graph reading. I mean... in the process of our infiltration by finance bros and VC money, have we genuinely forgotten how exponentials work?
Exponentials... that's the one with Sylvester Stallone and Jason Statham, right?
Yeah, I heard Statham got remaindered from the new Exponentials though, but don't worry, he's going to be joining The Mod Squad.
[flagged]
What? Gold underwent a massive revaluation with the end of Bretton Woods in 1971. It was prior to then that government were actively involved in making the price of gold artificially low.
Fun fact: under 31 U.S. Code § 5117 a troy ounce of gold is still valued at 42 and 2/9ths dollars.
Gold became much ‘cheaper’ in the period 1945-70 because there was a series of technical revolutions in South Africa mining (never mind the apartheid …) This is why Breton Woods lasted as long as it did.
The countries like France that conspired in resentment to break it as ‘privilege’ are now effectively in flames.
The right to a civil jury trial in the 7th Amend. to the US Constitution is only available where the value in controversy exceeds twenty dollars (a massive sum in 1791, and a trifle now).
You got it backwards; ending the gold standard was very much a unilateral decision by the United States because Nixon couldn't handle making politically unpopular decisions to cut spending and/or end the Vietnam war. Many countries that had their gold reserves held via US dollars were livid.
Well, there's an arguable point of fiscal policy there, and a conspiracy rathole that no one wants to excavate.
But none of that matters here. As grandparent comment indicates, you're making a pretty fundamental math error. This is a log chart. Government may have devalued currency. It did not do so exponentially.
I wouldn't go so far as to say "nobody". Electric Boat had 2 GB memory in one of its systems at that time, with the hardware capacity to increase to 4 GB. It sounded insane at the time, but it absolutely existed, and thereby seems reasonable to include it in any research of historical pricing.
I am curious. which year was it ? also which boat.
Yes, you really need "dollars per amount of RAM you need for standard computing tasks." Windows 11 requires a bare minimum of 4 GB of RAM, Window 10 only needed 1 GB.
If what you're interested in is fluctuations in production versus demand then you absolutely do not want a subjective metric. Measures of the form dollars per unit, units per watt, units per flop, etc are what you're interested in.
Discussion of memory in terms of words and their bit length, time to complete a task is more meaningful to intent on use and compaction, see greycode technique. Dollars of slop a unit sacrifice skills in the industrial base for the gain of paper profits at the repeating business meetings.
Have you ever actually tried using Windows 10 with 1GB of RAM? I wouldn't consider it suitable for "standard computing tasks."
And that's the hangup, what do you consider a "standard computing task?" On what OS? Running what software? How well? Plenty of people were still using XP in 2009, so is 256 MB of RAM okay for "standard computing tasks" in 2009?
> Have you ever actually tried using Windows 10 with 1GB of RAM? I wouldn't consider it suitable for "standard computing tasks."
I have [0], and it's actually not quite as bad as you would expect. It certainly wasn't fast, but I had no problem using it for basic web browsing and document editing. The painfully slow hard drive and processor speeds on that computer actually caused more issues than the lack of RAM.
[0]: https://news.ycombinator.com/item?id=45743066
I remember installing Windows 10 on my laptop when I was in high school, it was a decommissioned Thinkpad T410s from my dad's work. 4GB of RAM, Nehalem i5, a very early consumer SSD (OCZ, if I recall). I vividly remember my first thought being "wow this is really slow." Win7 ran like a champ on that machine.
My experience with Win10 on that laptop actually led me to buy a dumb gamer laptop for college. As those all do, it died prematurely, so I ended up back on the T410s for a while. I put KDE Neon on it. It was great!
If you're saying that you can install and use Win10 on a laptop with 1 GB of RAM, well yes I acknowledge that is true. But it's a purely academic exercise, it's not actually a usable computer for the overwhelming majority of people.
Maybe it would have been fine for my grandma. She was using a Pentium II running Windows XP to go on Facebook in the early 2010s.
That's just as wrong in the opposite direction, y2k was a thing because two bytes were worth the saving in 1980, and we really needed those two bytes.
Well it's complicated. Y2K was a combination of logic issues and the consequences of certain inefficient ways to store dates, like text and BCD. Migrating to binary could fit plenty of dates into the same space or even less.
In particular, 16 bits is enough to store the entire date, year month day, from 1900 into mid 2079. Any date format that couldn't go past 1999 was probably using 24-48 bits.
[dead]
I still don't get where all that memory goes.
Abstractions on abstractions on abstractions; background tasks and their abstraction stacks; increased cache and buffer sizes to take advantage of increased typical memory capacity. For an example of the latter, handling TCP on a Commodore 64 is a problem because the memory can only fit about 45 packets with nothing left over, but now you can just allocate a megabyte receive buffer per connection.
Neither do the developers, because until recently, RAM was so cheap it didn’t matter, and we were in a situation where almost no one ever needed to consider “how much RAM will this take?” when writing code.
Like Homer Simpson said about alcohol: "To alcohol! The cause of, and solution to, all of life's problems", we developers can say about AI (assuming AI-assisted coding can save memory in popular programs, OSes, etc)
More than code they write, the framework and runtime they use.
Mostly used by JavaScript parsers and HTML rendering, the rest is for telemetry.
For windows 11, it seesms to be antivirus scanning. That's what's always blowing up my RAM
[dead]
I once held in my hand the main part of a ferrite core memory module from the early 70s. It was kilobytes at best.
I also recall looking at recommended requirements for Dungeon Keeper 2 - 266MHz CPU, 64MB RAM and thinking "that's absurd - no such device exists!". I was a kid back then, so what did I know?
Later on in college a friend showed us his absolute monster of a laptop with a whopping 8GB of RAM - he could spin up several VMs on one device! Groundbreaking on a (nominally) portable device.
So yeah, safe to say the notion of gigabytes of RAM anywhere close to a regular person belongs firmly to the 21st century.
The graph wouldn't be a lot taller because it's using a logarithmic scale
The Cray-2 had 2GB of RAM in 1985.
So total system cost per unit of memory is going up.. 2GB costs in 1985 was $2 million (from the graph), a cray-2 was $16 million (from wikipedia). A GPU server with 8xB200 today can be had for ~$500k (estimate), 1.5TB memory is $25k (from the graph).
Hmm, so memory is actually still cheap compared to historical highs. At least cheap relative to other computer components.
Back then was it 1000KB or 1024KB?
The natural unit of measure for integrated circuits is a power of 2 since that's what the systems operate in. It's so natural that early 9 and 36 bit architectures were squeezed into 8 and 32 bits as it just works so much more efficiently.
Long term storage and communications? Those start to introduce things like human division of timings, frequencies, and other analog systems like rotating disks. It still generally makes sense fab actual flash chips in various powers of 2 though. The discrepancy there tends to be various forms of 'overhead' for the translation table / wear level indirection, over-provisioning, and even variations in density caused by different levels of physical cell utilization.
Still, most network stuff ships around packets of 'up to' 1500 bytes ( https://en.wikipedia.org/wiki/Ethernet_frame and lets just exclude jumbo frames ) so arguably it'd be better to talk about all computer measures in binary powers of two, exclude the marketing huckster trying to make things more impressive by shoehorning SI engineering units into a realm that uses binary math.
Also maybe you want price per average program footprint size...
IIRC the Cray 2 was offered in a 1GB configuration by the mid 80s.
[dead]
If my memory serves me correct (no pun intended), when I was a kid I remember bugging my mom to buy me like 2 or 4 1 MB modules, it was at least 50 bucks or 100 bucks each.
Now everyone's going to talk about how cheap everything is by comparison - but someone needs to talk about how oppressively hungry browsers and OSes are compared to in the past. This is no HIMEM.SYS
There’s been a sharp divergence in memory requirements. Talk to developers and they think that 32GB is the bare minimum these days, with 64GB or more preferred. They’ll point to Electron and Chrome tabs and everything else.
Then you sit down with an average computer user on their 8GB RAM MacBook Neo and they’re in love with how fast and smooth it is, even with their chrome tabs and the company Slack up and Spotify in the background.
I still have an older 8GB machine to kick around with on the go when I don’t want to haul the expensive laptop. It’s fine, even for a lot of development.
How do you explain this discrepancy? Is it because the OS is agressively fencing in and pruning these wasteful software?
Some tasks simply require more RAM. Compiling big software, for instance, wants as many CPU cores as it can get, and each compiler instance needs some amount of RAM to run efficiently. It's not unusual for a 32-core build to need 32-64GB of RAM to run at full speed. Work on a smaller program, though, and 16GB is absolutely fine.
Because they use one or two apps at at time, the ones they must spend all their time to perform their job. E.g. Excel and a web app to work on invoices and a stack of paper documents. I see 8 GB on Windows PCs too.
Containers is often the reason. You start a container and you are immediately pulling in a quarter to half a gig or more (often the latter).
[dead]
I think it would be better if one has the discipline to just use older machines and play older games and only visit certain websites that can be visited on older version of browsers. A second-hand 16GB laptop can go a long way.
But yeah that probably sucks from time to time, especially for young people.
Second-hand? lol my main driver has 16gb and just peachy. What do you folks do that needs so much ram to browse the web.
You can do a lot on old machines but developers also need to optimize a bit. Youtube almost plays on a 20-year-old machine, which means with some effort it'll play just fine. Most the other sites work just fine.
My laptop has 8 GB. I write blog posts, I have a dozen-ish tabs open, I do KiCAD things (including 3D renders!). Works great. I was doing Verilog synthesis on a similar machine in college in 2020.
The truth is that, if you do the same things you were doing with your computer 10 years ago, well then you don't need a new computer!
If all you do is write books, a Pentium III will do the job just as well as a brand new PC.
Of course, the web throws a wrench in this. Word 2003 is still far more capable than Google Docs, yet tons of people opt for the cloud slop because it's convenient and free-as-in-beer. And, Google Docs will continue to become less efficient with time.
> only visit certain websites that can be visited on older version of browsers. A second-hand 16GB laptop can go a long way.
My desktop has 8 and I have no problem keeping multiple tabs open using up-to-date Firefox.
Look at it this way: while the upfront cost to scale up production is huge, prices are now high enough to justify it even if demand is expected to drop abruptly later on. So if you can wait 5 years for your next PC, 1TB RAM might go for what 64GB would have cost without the AI demand spike.
Granted, if you need a new system before then, you're SOL.
One thing to look out for is supply capacity curiously going offline in 2030 or whatever. That would hint at market power or collusion.
Memory prices per GB were cheaper in 2012.
It’s possible we’ll see a huge price drop on the near term but SSD + Cache + GPU’s seems to have changed the equation where RAM speed is considered more important than size. And from a pure architecture standpoint it makes sense.
They weren't though when you adjust for inflation. If you took inflation into account, ram is cheaper now by $0.89/GB for DRAM compared to 2012.
Lowest 2012 price listed is 3.7 (2012-10-30) vs highest listed in 2026 is 5.375 (2026-2-1), which overlaps based on the margin for error involved. https://www.usinflationcalculator.com/
Even being vaguely in the same ballpark is a wild regression when you consider the difference in density.
So you are trying to compare the lowest with the highest and not the current price. RUn that number with the current price.
The most recent price listed is from a month ago and the prior prices are increasing. So really you can roughly compare years but not Today.
There’s plenty of data to say prices in 2012 and 2016 overlap, which is wild.
It's crazy that a AAA video game cost $60 when I was a child in the 90s and costs $60 today, sticker price! Not adjusted for inflation!
And to think that AAA video game was distributed on cartridge with expensive memory in cardboard packaging with a manual while these days that same $60 game is just a digital download from a CDN.
How is it crazy? The marginal cost to produce a copy of a video game is ~$0 and the market is much larger now then it was so fixed costs can be spread around more. The sticker price was always more or less arbitrary chosen - i.e. to maximize profits, which includes how the number is perceived.
> while the upfront cost to scale up production is huge, prices are now high enough to justify it even if demand is expected to drop abruptly later on.
Given the nature of the industry and how critical the product is I think it would make more sense for governments to bankroll fab construction in a way that the public takes on the risk of consumer prices falling below a certain level within some limited timeframe. Mildly subsidized chip production seems like a much better downside than the current sky high prices.
The log scale is nice to compare decades. Wether it's inflation-adjusted or not isn't too important but it's still a factor of 10, which would show in a linear recent graph. The fact that we're comparing GBs instead of the average RAM stick shows how much the price has decreased per GB rather than per unit (much smaller decrease).
But a linear graph that represents only the last decade and where the bottom is 0 (not the min value) would tell a different story, but I guess we already know that story because we're living it.
TIL someone took over the now defunct jcmit dataset[1] (archive[2]). I expected his dataset to die off when his website did, but I guess someone found the data dump on archive.org and revived it. Which raises a question: how will this dataset fare five years from now?
[1]: https://www.jcmit.com/mem2010.htm [2]: https://web.archive.org/web/20250716092935/https://jcmit.net...
The dataset about memory prices now has a memory preservation problem. Very meta.
In the first graph, if you hover over the DRAM line you'll notice that the most recent data points are for DDR3. One of the data points from 2025 is a 2 GB stick. This paints a more rosy picture than the situation deserves.
So a price per GB today is about the same as it was in 2010. 16 year regression, wow!
Drawing a line backward from today's high water mark only goes back to 2018.
2010 prices were significantly higher.
The chart is also not inflation adjusted, which would bring the equivalent date forward even further.
Nowhere near a 16 year regression.
Nominal. The inflation-adjusted price today is 2/3 of what it was then.
Two wrongs do not make a right?
sure but you also need more gb these days for various tasks so it's not 1:1
I wonder if developers will start trying to do more with less in certain areas
Arguably they already did with the "cloud native" systems. There were plenty of examples personally known to me in the mid and late 2010s of smaller tech companies trying to run production PostgreSQL on 8-16 GB of RAM because they didn't want to pay the cloud RAM tax. Many "cloud native" systems were designed under these (mostly artificial IMO) RAM constraints.
Is that because the amount of available memory is limited for a single process? You can always add more storage and storage access is relatively the same regardless of whether it comes from the SSD inside the server or sits in another rack. Storage is a pretty linear cost when you're a cloud host buying storage in the hundreds of PB numbers. Whereas for memory, if you want the whole thing, you need the whole server even if your process is light on CPU requirements.
It's not 1:1 when you consider inflation either. Ram is still cheaper when inflation is a factor.
I'm going to tell my kids about the terrible 20s where DRAM prices stayed flat (or increased!) throughout the decade
One could also blame crypto and AI (they're clearly responsible for some of the volatility in the graph), but I can see the curve flatten in the 2010s, just as Moore's law ended.
Can you blame Moore's Law ending? The graph at https://en.wikipedia.org/wiki/Moore's_law looks steady up to the 2020s.
1979 to 2009 in the OP graph has a pretty steady drop from 10^7 to 10^1 USD/GB: 6 OOMs in 30 years. Then till before the recent spike it was around 1 OOM in 15 years: 1/3 the rate of progress on a log scale.
When it comes to CPU progress we blame the end of Dennard scaling several years before the knee in this memory curve. I'd guess the story of memory is similar in also hitting technical difficulties, but I don't know.
Moore's law is about transistors doubling every interval¹ *on the most economical package*.
Wikipedia is misquoting it, and extraordinary expensive chips being more capable doesn't change the economical situation.
I am tickled that OOM can mean "out of memory" in another context. You clearly meant "orders of magnitude".
Heh, unintentional. Another term shorter than spelling out "orders of magnitude" is "decades", but I figure that's less familiar and even more confusing here. "Memory price started out falling two decades per decade..."
Well if you underestimate your memory requirements by orders of magnitude you'll be out of memory for sure.
Moore's law didn't end in any broad sense and certainly not that far back. It's a tiring piece of misinformation that just won't die.
Progress has consistently become more difficult (ie more expensive) but has generally kept up. The scaling of a couple specific technologies noticably slowed down a few years back but that's not the general case.
The node names aren't representative of the reality.
Why has there been such an obvious repeating price cycle in the last 20 years?
Is that due to node sizes or generations or fabs coming online or what?
Memory semis are a classical example of a cyclical industry: simulaneous capacity investments -> overproduction -> price crash -> ...
This cycle is the first one than truly breaks the trend. It seems that the industry NEVER needed thus nuch memory for this long.
Also, given the history, producers are afraid to overivest, and newer players from china are lagging behind for now.
Cyclical industries are very common.
This graph is the touchstone one should rub all the "RAM and storage are no longer a commodity" bs that micron, sk hynix, samsung, western digital, seagate and others are peddling as of late while the valuation of their companies have gone from "supplier of widely available fungible goods" to "state-of-the-art moat AI backbone tecnology".
turns out things are not that bad! we just rolled back to 2010.
oh, wait, now every app is a browser instance. shit.
EDIT: so, how did I arrive at 2010, you ask? I looked at DDR5 pricing and found the closest pricing per GB in the past. this turned out to be DDR3 memory. I think it's totally fair since it was the latest and greatest thing back then, much like DDR5 is now. although, if we compare DDR3 to DDR3, we still roll back pretty far - a very close to current price was spotted in 2018, '17, 15, '13, and '11.
Yeah but now apps will have to start shaving off memory and maybe going native again. So it'll end up okay.
Will they..? It seems equally (or perhaps more) likely that we'll increasingly see vibe coded browser or Electron based applications as the bar is now lower to build such a thing.
Vibe coding also lowers the barrier of maintaining multiple native pathways. Also of adopting QT instead of electron.
yeah but you also have commercial licensing with Qt specifically :))
or we are going to see an explosion of vibe-coded GPL apps.
anyhow, the likes of Linear and Notion ain't gonna abandon web and go Qt. or!! if we are very lucky, we can see a native app framework that ticks all the boxes of a modern UI framework and is permissively licensed, but we need this crunch to stay there for years.
> you also have commercial licensing with Qt
That doesn't apply so long as you are willing to accept the LGPL. In practice that means you can statically link everything except QT so that the end user is free to drop in a modified QT version if he would like.
You can even statically link QT if you really need to as long as you dump the unlinked objects on some server somewhere.
If you are going to vibe code you can just pick any language you want. I had a go vibecoding in Rust and it worked perfectly fine. Even better than vibe coding in JS/Python because the type hints give the LLM a faster way to check progress.
Except you didn't when you consider the prices aren't adjusted for inflation.
The memory manufacturers have made an interesting mistake. The tech giants of the world will be working to replace them from the supply chain as soon as possible. China already makes it's own RAM albeit at 16nm but you can bet they are working to get down to 4nm.
DRAM hit a barrier at 10 nm a few years ago[0], so 16 nm is actually even closer to state-of-the-art. E.g. Micron newest node (1-γ) is their sixth at 10 nm [1] and their first EUV-based node.
The problem is that DRAM is fundamentally based on storing charge in a capacitor and how much charge a capacitor can store is a result of the geometry of the capacitor. So either someone will have to figure out a way to make the same size capacitor take up less space on the RAM chip (this is what broke the previous 20 nm barrier) or someone will have to invent a practical way of making RAM with less than 1 capacitor per bit.
0: https://semiengineering.com/dram-scaling-challenges-grow/
1: https://www.techpowerup.com/333111/micron-announces-shipment...
what makes you think China RAM makers will sell their chips at the old memory prices and not just 10% below the current market price
China often exhibits tremendous internal market competition, so it's possible that different Chinese suppliers will race each other to the bottom (Chinese firms are really good at suriving on ultra-thin margins) making prices even lower than 10% below the premium providers.
More competition will drive down the market price, so it'll be 10% below a price that is lower than what we have currently. Obviously it's not gonna go down immediately, but more supply will definitely bring down prices.
Obviously, this is very bad for the existing memory makers, since these boom prices will not last forever, and the Chinese aren't gonna stop selling memory once they are in the market.
Unfortunately, this is unadjusted prices, and this failed to annotate where the cartel years and when the cartel was 'broken up'. Not a bad assignment's work but clearly lacking the domain awareness necessary to report the complete story through graphs.
It's amazing how consistently thr lower memory cost have expanded the set of economic viable applications : cheaper hardware doesn't just improve existing software it also enables software that was not possible before
>it also enables software that was not possible before
Which is not always a good thing.
There is something wrong with these graphs: they indicate nand price to be back to 2020 level but in 2020 I got nand for half the current price.
Do we have a chart of memory production per year? (Are any large expansions, by incumbents or new entrants, planned for the near term?)
Tim Cook recently likened memory prices to a "hundred year flood." Looking here, it looks like 1988 was at least as bad, with prices going up 3X+ from $160/MB to $500/MB.
This is probably the first real thing that is affecting me personally with this whole AI business. Having to pay more for device upgrades going forward. I hope the demand settles or new memory production offsets the demand.
The truly absurd part is that datacenters are barely being built and those that are built can't be turned on because they don't have enough power. Satya Nadella admitted recently that they have warehouses full of unused hardware because they a) can't get datacenters built and b) don't have enough power i.e. this whole RAM scandal is a bloody joke. If OpenAI goes bust (their financials are a huge mess and lots of red flags - they might not even survive to IPO!) and then what will happen to all those "inked" deals to buy all that RAM?
is multi-level DRAM worth considering? storing multiple voltage levels per DRAM capacitor?
If it were possible, it would have been done already. The issue is the capacitors are already tiny, and barely can prevent a single bit decaying before refresh.
do you have a reference to exact / realistic scaling laws for the leakage currents as function of capacitor/dielectric dimensions and access transistor dimensions?
using 4 (or 2^N) voltage levels stores 2 (or N) bits, so we can afford to make the structures larger
why would this approach make sense for NAND flash but not DRAM?
[flagged]
If you care about only capacity and cost yes, but not if you care about performance.
Can you back that up with anything about semi-recent nodes? The voltages are so fragile that I'm not convinced you would actually save space once you adjust the design to handle more levels.
You could also do a computing pr dollar graph - which would be a similar sharp decline over the past decades - however it won’t show anything like the memory price spike of the past few years.
aaah, the 90s price crash. Good times.
Which manufacturers quickly thereafter put to an end:
https://en.wikipedia.org/wiki/DRAM_price_fixing_scandal
Would be nice to have hard drives on the same chart
Usefulness per gigabyte should be taken into account too. We could do way more with 32MB in the 90s than 4GB today
A perfect example of how graphs are often misleading. $/GB is a totally useless unit value because it's an arbitrary size. The unit needs to be tied to the relative usefulness for its time. The y axis should be something like $/average workstation memory or $/requirement for common compute task. It's obvious that ram is expensive right now, but it's not expensive per GB. It's expensive relative to what you need to accomplish a useful task.
But relative usefulness is entirely subjective, making it a meaningless unit. Depending on your use case you may need 256 GB or 0.5 GB.
The audience who would benefit from hypothetical $/usefulness would be people who don’t know what memory is and don’t know what’s inside of their computers, or what it does. This is a fine audience to be in and to serve, but obviously not the audience of that website and not HN.
If you think that audience is under served for memory market statistics, I encourage you to make such a website and serve that audience.
For people on HN, who do you know what memory is, $/GB is a fine metric.
This is assuming that the wide variety of use cases are evenly distributed and that larger use cases are not mostly just a lot of duplicated smaller use cases. If I have a website I will need X amount of ram. If you run a much larger website offering a comparable service you will need some multiple of X, but you don't actually need much more ram per user (assuming you're also accounting for extra infrastructure and not just the web servers). It's the same task just scaled. Relative usefulness is not subjective, you could look at a variety of tasks in different industries. Windows server 2012 had a minimum requirement of 512 MB. Windows server 2025 has a minimum requirement of 2 GB. That's 4x for the same task which totally distorts $/GBs usefulness for being able to tell you anything helpful economically. It's obviously good to collect this data, but you need to pair it with some kind of demand data for it to actually tell you anything.
> you need to pair it with some kind of demand data for it to actually tell you anything.
Again, this is entirely dependant on who is consuming the statistic and for what purpose. For some use cases, yes demand data will be quite crucial. For others it will not. It's quite apparent the site's author doesn't see this as crucial and for the purposes I need to consider memory pricing, I agree.
> The unit needs to be tied to the relative usefulness for its time.
That requires baking in assumptions, and makes the data less general.
You can go from $/gb to $/usefulness fairly trivially by adding assumptions, but you can't go the other way.
Gas is priced in $/gal, not dollars per mile or hour of lawn mowing or whatever. The resource and the use are completely different concepts and the resource owner/producer cares not of the buyers purpose for it.
A useful task isn't a fixed thing though. Everything the 2012 computer did you can still do today with the same amount of ram we had back then.
No, not if it involves a web browser. Most web sites today will not work on a 2012 web browser.
The PC stopped existing in isolation, for most useful tasks now, it needs an Internet connection.
I use a Thinkpad T530 for reasons that are very important to me. It is the only laptop that I have, so it is what I use for every manner of portable computing.
It still does all the things I want it to do, including using modern websites with modern browsers on modern operating systems (including Windows 11).
The T530 was released in June of 2012.
The 2012 computer running a modern linux install will still work fine. I'm talking more about the specs, specifically memory. I had 8gb of ram in my computer in 2012, the Macbook Neo released this year still has 8gb and is usable for modern day tasks.
We don't _need_ that much ram, we just found new things to do with more.
[dead]
Were we really paying more for RAM per GB in the late 2010s than we are now?
Just really doesn't feel like it. Interesting.
We're paying more now than we paid in the late 2010s, but less than we paid in the early 2010s.
I guess ‘per GB’ doesn’t really capture it, because the base number of gigabytes available to people (ie- the smallest compatible RAM kit you need to build a computer) and the base number of gigabytes you really need (OS bloat, feeling responsive, etc) have gone up so much.
And at the data center level too (AI). Not that we don't need some, but I don't see how this won't be another boom bust cycle eventually.
Only if you're buying DDR3 now, DDR4 and DDR5 are both more expensive than the 2017 peak.
At this rate somebody's going to get killed over three megabytes of hot RAM any day now.
This post had me wracking my brain asking, "Did any of Gibson's books have someone get mugged for some hot RAM?"
Edit: Actually Johnny Mnemonic had 325MB in his head and he had the whole Yakuza out to kill him.
I knew I paid more per GB in 2018 for DDR4 than even today's inflated prices of DDR5...
“All that is human must retrograde if it does not advance.” -Edward Gibbon
My fellow humans, we have retrograded.
It certainly doesn't look as bad as it really is when presented on a log scale chart.
Going up is worse though because software has gradually got less and less memory efficient.
It'll get more efficient now people aren't upgrading. Software will be exactly as efficient as it needs to be to run on most peoples computers.
I don't think that's true - instead software will be exactly as efficient as it needs to be to run on the developers workstations. And companies throwing sky high salaries at software developers aren't as price sensitive when it comes to those workstations as the average user.
This is extremely misleading and not very useful. It makes little sense to use pricing per GB during decades when RAM was at most in MBs. In that case, why not talk about price per TB or PB? Then the line will look pretty much flat and horizontal.
The line would look the exact same if you switched it to price per PB. Only the labels on the y-axis would change.
Nice to know we still have a ways to go till 1960s pricing!
It’s weird to see supply and demand battle moores law.
Note that the chart scales by powers of 10
So, it’s on a logarithmic scale
Well, that was depressing.
TIL about "HBM".
Except no one was buying 1 Trillion $ for a GB of RAM in 1960. Even Professor Frink would agree:
https://www.youtube.com/watch?v=ykxMqtuM6Ko
the problem with this analysis is that it doesn't account for memory speed, which doubles with each generation of ddr
this should really show average install cost. Even phones have 8-12gb because the software is so atrocious. It would be like comparing cars by pricing per horsepower. Nobody is running a 12 horsepower vehicle on the highway, and doing so would be dangerous because of the change in average power & speed.
this is interesting. but i’d be more interested to see a graph starting at the point when developers got their own computer.
then the price of ram over time for whatever the daily functional workstation a developer would have needed then.
i mean this is a graph of the price of GIGS of ram from a time period when the space shuttle needed like 1 MB.
[flagged]
[flagged]
[dead]
[dead]
"on a galactic timescale, the prices haven't risen at all..."
DRAM and chill
With respect... I am surprised such a low-quality analysis is published on stanford.edu . What is compared here? What is the purpose of this? What are the conclusions of the analysis? Heck, where is the analysis? By what logic are the prices per GB(!) comparable between 1960(!) and 2026?
I am sorry to being rude, I just don't understand this publishing beyond getting the media exposure.
Likely your expectations are mismatched because this is just data and not an analysis.
Let f(t) be the price of RAM per GB from 1960 to 2026. Let F(t) be the price of RAM per byte in the same period.
At every point in time t, f(t) is the price per 1 GB of RAM which is 1GB/1B times the price per byte of RAM.
Because 1GB/1B is non-zero, it follows that f(t)=1GB/1B F(t).
It also follows that ratios are preserved, ie
f(t1)/f(t2)= 1GB/1B F(t1)/F(t2)
As long as f(t2) is non-zero, which in this context is never the case.
Visually, the two graphs are the same except scale is different.