If you want the personal thanks from Altman without having to actually give up your RAM, you can download some here first https://downloadmoreram.com and then upload that to Altman.
This is silly, and I know Sam Altman said that consumers having RAM is not an efficient usage, but the real villains here are the memory manufacturers that have time and again have been proven to collude on price fixing and whatnot, and are now not scaling up production according to demand.
I don’t think it’s very productive to go on a witch hunt on someone who said something stupid and is being taken out of context (albeit he should have known not to say that).
It was taken out of context for sure but still find this site amusing. Perhaps even prophetic, can we build enough ram and vram for future requirements when theres only a few holding the keys.
Obligatory: "we're curing cancer r-right?" No Nat we're making slop videos with it.
To date, five manufacturers have pleaded guilty to their involvement in an international price-fixing conspiracy between July 1, 1998, and June 15, 2002, including Hynix, Infineon, Micron Technology, Samsung, and Elpida.[3]
Hynix is charged with carrying out the conspiracy by:
* Participating in meetings, conversations, and communications in the United States and elsewhere to discuss the prices of DRAM to be sold to certain customers;
* Agreeing, during those meetings, conversations, and communications, to charge prices of DRAM at certain levels to be sold to certain customers;
* Issuing price quotations in accordance with the agreements reached; and
* Exchanging information on sales of DRAM to certain customers, for the purpose of monitoring and enforcing adherence to the agreed-upon prices.
* Under the plea agreement which must be approved by the court, Hynix has agreed to cooperate with the government in its ongoing investigation of other DRAM producers.
KOREAN COMPANY--HYNIX--AGREES TO PLEAD GUILTY TO PRICE FIXING
AND AGREES TO PAY $185 MILLION FINE FOR ROLE IN DRAM CONSPIRACY
This is remarkably different than someone claiming donuts are spaceships when the company itself was charged and pleaded guilty, and it's a matter of public record.
If I said water was wet you are not obliged to Google it either. If someone continues to assert that water cannot make you wet even after being informed that water makes things wet because they hadn't been provided proof, and they didn't bother to check, that might be on them.
Treating the 4th time the same conspiracy would be happening—after 3 previous times proven to be true—as a nutty conspiracy theory is itself nutty. Detached from reality.
You're free to doubt it, it's not absolutely certain, but pretending the chance is near zero is just not even remotely as rational as you seem to think.
All are responsible - the AI greedy companies and
the suppliers.
If the USA were a real democracy they would do something about
it. After all this even violates capitalism, e. g. free market
leading to reduced prices. But the USA is run by a criminal
gang, pocketing away money - the TechBros and their cronies. They
just make any legislation they want to, with a bit sugar coating
on top to not appear too overly evil or greedy. But they are.
Yep. Not only that but the quote (even if taken out of context) is factual. >99% of people load up their computers with comical amounts of ram when they really, really don't need it. This excess then only fuels the creation of memory inefficient apps. People are in shock when I tell them they could take the whole OS state when running any AAA game, reduce memory usage by an order of magnitude (if optimized correctly) and they couldn't spot a single difference.
I get that I'm falling for low quality bait here, but what about him saying that it's terrible for people to have RAM is "fake propaganda"? The entire exact quote is:
> From a global allocation perspective, actually, it’s quite terrible to have everybody have a bunch of the world’s limited supply of DRAM sitting in their own computer and being used 5% of the time. You want maximum utilization in the cloud if you’re trying to deal with this very supply-constrained environment.
That's a ghoulish point of view that puts real living humans on a lower hung than the poor poor corporations who just want RAM to be so expensive that everyone needs to buy compute from them. Convenient for the corporations!
That quote is accurate, but becomes fake when the context is missed out. He was talking about people using local AI models versus cloud. Nothing to do with "you shouldn't have your own PC".
He's an easy target because of who he is, so I understand why the entire Internet ran with the fake story (even LTT!). But on Hacker News you'll definitely get pushback if you post misleading stuff.
Except that's nonsense, any company that could make more ram absolutely would because the damand is insane. There is no universe or economy where it would be profitable to artificially lower your production to keep prices high when you can just make more ram.
It depends on how much inelastic demand there is. For example:
Let's say there are 10 companies that need 1 GB of RAM each in a life-or-death scenario. They would be willing to pay $100,000,000 per GB to get 1 GB of RAM.
At the same time, there are 90 more companies that need 1 GB of RAM each to make certain quarterly targets. They are willing to pay $10,000 per GB to get 1 GB of RAM.
If you make 9 GB of RAM, you can sell it for $900,000,000. If you make 99 GB of RAM, you can only sell it for $990,000, because the 10 life-or-death companies only need to bid high enough to push out a quarterly-target company.
It's a bit of a contrived example, but it shows you how restricting supply can boost total revenue if there is a segment of the market that has inelastic demand. (this is why medicine pricing has to be regulated - in the medical industry, there are lots of inelastic demand scenarios that can lead to predatory pricing)
Until making more means huge capex building extra factories. Semiconductors is famously a boom and bust industry, building too much in the boom will kill your company.
It was the same with cloud and software as a service. We underuse the power of our local machines and pay high cost for modest resources that are bandwidth constrained for most users.
One consequence has been lower power requirement for local machines. This is partly why a 2015 MacBook is still perfectly serviceable for basic use in this day and age.
The AI boom first reversed the trend, to now try to get back to that cycle.
I never left the PC era, but the RAM prices kind of hurt me. I don't
know when I would want to buy a new PC - it seems to be 3x the cost
for barely any real benefits over the status quo.
most games which require that big PC are poorly optimised and often times they arent even fun. these companies focus on visual fidelity and give all their engineers insane machines to build the games on. that has beenn destroying PC gaming way before AI/LLMs came into the picture.
I am using AI now to make a game engine that will run and make 'nice enough' graphics on simpler machines. things like laptops with iGPU and 4gb of ram.
U can still get quite ok graphics, and make very complex and rich gameplay systems using such a machine. It just needs to be the target machine while developing so the right tradeoffs are made.
You know what that era has in common with fossil fuel? It takes a very specialized infrastructure and supply chain that you can limit with lobbying efforts and gatekeeping so that all the money only flows to you and your cronies.
There has been a relatively strong local model community since the days of llama-3. IMO, the number of non-technical people interested in qlora, adapters, parameter fine tuning etc is remarkable.
> Will we go back to PCs? As in: are we going to get cheap local hackable frontier AIs at some point?
Forget about AI, the entire PC market is getting killed over this.
As an avid gamer I'm unsure if in 5 years I can even buy a gaming PC anymore. Not even because of cost, but it's unclear if anyone will still make the components anymore.
It's interesting trying to be an AI "moderate," which as the owner of a 16gb card I definitely am. Still working on the balance, but it really does feel like I'm going to land on something like "use external AI services to build a few tools, but use my local rig to actually handle everything personal once those tools are built."
Its interesting because I was at Costco the other day and they weren't even selling PCs unless they were gaming ones. It seems most consumer electronics have switched over to laptops.
Also, this reminds me so much of the download more RAM ads back in the day.
Most of it has switched to mobile phones and tablets. It doesn't help that almost everywhere digital classroom means tablet instead of laptop. Sure, the control is convenient for the teacher, but it also means we are raising our children in walled gardens. So much so, that whole generations don't even leave that garden in their spare time.
I'm still waiting for the day a mac air style shell exists where I plug my phone in and just get a real desktop experience. The chips in phones these days are plenty powerful. I always thought it would be where we were going, but alas, no major manufacturer seems to want to make it feel like a seamless, premium experience. Samsung got sorta close, but the peripherals involved never felt great or didn't make sense from a cost perspective.
I wonder what it would look like if Apple offered a slimmed down shell from the Neo for ~$300 that required a connected iPhone. Maybe as iOS and MacOS converge we'll get there.
not true. it wont be worth it anymore to implement Path Tracing or Ray tracing which is absolutely not needed for a prime gaming experience. its deterimental even because ull only get into games with a certain type of person who can afford a big machine.
Many games that come out still, and are quite successful, do not require such technologies.
It also won't be worth it to make games for next gen consoles if the prices end up being what people are saying. So in the end they'll just keep doing what they're doing right now, and not increase the hardware requirements of their games.
1. Indy developer releases a game or trailer of a game.
2. Within an extremely short time (days to weeks at most) there are clones on the market.
3. These games often outperform the original.
It's absolutely not something that's gone on for ages.
It might be a novel gameplay idea / twist on a genre, and mere days after launch instead of being able to stand on its own and gather attention it's drowning in a pool of lookalikes. It's no longer distinct.
The speed of being able to slop together something like this is less than 1% of what it used to be.
Clones are not going to be the problem. The slopalanch will be the problem. Take a look at https://www.reddit.com/r/playmygame/. There's a new game every few minutes and pretty much zero of them are getting any attention. And this is only the beginning.
So it's easier than ever to make a game, but it's more difficult than ever to get anyone to try what you've made. I'd say that's just as bad, and in some ways worse, than having your game cloned.
For someone looking to play a new game it will be difficult to sort through all the bad games to find the good ones. And for anyone pouring their soul into a game they will more likely than not have an audience of one.
Our eventual destination is probably going to be a) pick up a device, b) ask your bot to make up a game on the spot, c) play. It sounds so lonely and depressing.
If the AI bubble bursts and prices normalize because demand for AI hardware is reduced, then these companies might pay more attention to other markets like gaming again. But who knows the future anyway.
intel makes gpus :). Maybe games will be finally made again in a way that a PC can run them without needing a 3k GPU in there to get toward half the refreshrate of your monitor.
gaming does not equate to only things like the last of us or cyberpunk which need extreme machines. a lot of visual fidelity can be reached still without having such a monster. Just need to not use certain features which are currently pretty much exclusive to Nvidia...
Have you seen the current prices on Steam Deck? Almost $800 for the base model.
And I assume they will only continue to go up. When it came out, purchasing one seemed like an easy decision. But I paid half as much for one with the same CPU/RAM (and upgraded my SSD for cheap).
Regarding graphics capabilities, haven't we reached "good enough" like a decade ago? Does anyone actually want to see ultra realistic gore when attacking an enemy? I'd also submit that photorealistic art style is not the most popular one either nowadays except for a handful of special sims like race car driving.
I was floored to discover that my i3-1215U with integrated graphics can easily run PS2 games at 60ish FPS. These are triple A games, just a little older.
If I started today and played 10 games I can emulate every single day for the rest of my life restricting myself to purely emulation, I wouldn't run out of new games for...five and a half years. (assuming my rough calculation of ~20k available games).
Drop it to a game a day and I will likely never run out of games given my life expectancy.
Consider that I will find some games I like to play for more than a day and there are easily well more than enough games in emulation alone to entertain me for the rest of my life on existing entry level hardware.
Will I ever play MODERN SHOOTER {CURRENT YEAR} on this device? Certainly not, but I can't see a reason to bemoan lack of access to such a small portion of the available industry.
Like, suppose you're a reader. You might be right that you might enjoy your favorites in their original languages, but you'd be silly to be worried about your future ability to enjoy books just because you'll likely never get to experience reading 三体 as a chinese citizen fluent in the language.
Not just the Console and PC's market, the Phone market to. On my country before, IDR 2.000.000, you can have a 8GB phone with G100 Ultimate. How about Now? On that price, you'll just got the 4/128 with Unisoc T7250.
Today I opened newegg for the hell of it and figured I'd look at the cost of a 2 TB NVMe ... Fuck man. This is seriously depressing.
BUT! I feel that lost hope could/will turn into a new forms of "min-maxing" or whatever where we will see people focusing on small, efficient software to make the most of what we have to weather the drought. We got spoiled with cheap hardware. Time to make the most of all that cheap hardware we have.
Though of course, replacing this hardware as it breaks down will be expensive. Perhaps we could go back to 1/2/4 GB RAM sticks and 256/512 GB Disks. Then run minimal Linux installs, minimal GUIs, minimal services, minimal software.
I have an external USB-C 4 TB portable SSD called the Samsung T7 Shield 4TB Black.
I bought it at the middle/end of 2024, and I paid 3790 NOK (~357 USD at the time).
The lowest price online in my country for the same product new online today is 10233 NOK (~1085 USD)!
270% of what it was before, it’s absolutely bonkers.
I was even considering putting it up for sale a few weeks ago. But of course what I would be able to get for it would be no where near $1000 when selling second hand. Instead I might if I am lucky get slightly more than I originally paid which would be nice and all except that if I then regret selling it it’s still expensive to replace, or I have to also buy it second hand from someone else that may have worn theirs out considerably more compared to what I have done to the one I currently have. Mine is very well taken care of since the start, and although it’s nearing capacity now I haven’t written and deleted data over any over on it.
"... we will see people focusing on small, efficient software to make the most of what we have..."
Been doing that for myself for the past 15 years
Advances in hardware are only impressive and useful for me when I avoid the so-called "tech" company "engineer" BS
The reason I stop using old computers is never due to software. It's always outdated proprietary cables, failing power adapters, screens, hinges, and so on
Sometimes "minimalism" is simply for _speed, resource usage and reliability_ (that's what I get from small software, anyway) in contrast with prevailing marketed garbage that is bloated and sluggish by commparison
Other times, claims of "minimalism" might be trying to capitalise on some silly trend
Personally I like what the computers I own can do for me and learning to make them do more. I have a brain, I don't do much "guessing" and I don't really need "inference" on a daily basis, certainly not for such a high cost to the environment
I have to think, it can’t take much time before the Chinese DUV fabs come online and non-frontier chip prices crater. Sure, you might not be able to get 5090 equivalents, but the 30xx series was DUV and so that should be replicable in a few years.
Whatever quality bar / process node China gets to you know they will scale it out super hard.
(Some quick investigation suggests that the Yuliangsheng DUV machines currently under construction are 28nm, which would map to 2014’s GTX 900. Perhaps I’m being overly bullish but I assume they will recapitulate/clone the tech faster than the rate it was originally discovered.)
Yes but then you'd have to give closed-ecosystem-ass Apple money, which is worse than giving it to Nvidia who at least support linux properly. And you still get prompt processing speeds of a leisurely sloth.
They absolutely can't do what, or more of what the biggest models can do. There's nothing frontier about small local Qwen or Gemma models.
2899$ gets you a machine that is just good enough to run some small models, slowly, with terrible context, requires a ton of tinkering and fine tuning, managing and constantly downloading a bunch of huge file blobs, etc.
Partaking those $2899 in most of the large cloud models would bring you much more productivity, much less boilerplate and time spent in setup, etc.
So.. just weird accusing people here of "not partaking" in AI when you're just essentially toying around if you're actually comparing local Qwen and Gemma's with frontier models.
Yes, we will. Eventually, specialized AI hardware (ASICs) will emerge that will be good for AI, but not much anything else. PCs are general purpose computers that are good at many things, but excel at few. There are quite a few startups working on it already.
Yupppp. My current conspiracy theory, with all of the increasing hardware costs and how easy decompilation projects are, is that there is going to be a major push to eradicate consumer hardware and keep all software in the cloud.
Everybody will have a mini PC for the sole purpose of streaming content, with no storage to keep anything on it.
I would have laughed at the thought a few years ago, but now I am slowly starting to adopt some of the viewpoints of American second amendment hardliners.
It's just unfortunate that producing a half decent computing platform is significantly more difficult than a rifle (and that in the case of said hardliners, the capital is backing them, here it's working against us).
Why would component vendors allow shortages to continue indefinitely for the benefit of hyperscalers and AI companies, especially given what may be years of pent-up demand in other markets?
And why would the game industry, considered as a whole, want to incur higher costs to exclusively deliver its product in a way that provides less value to the majority of its customers, many who won’t be willing to pay the higher prices required to maintain similar profits?
It's worse than centralization. These AI models have gobbled up all the world knowledge. If for some reason the Internet dies we will have no option but to be dependent on these proprietary, for profit, AI models for any information we need.
You might say that this is impossible but look at current trends. A lot of information today is getting shared in the walled gardens of Internet. You have to login to social media to view posts. The Twitter, Reddit threads that were accessible freely now require you to log in. Similar thing is happening across other networks. People are no longer asking questions on Stackoverflow. All those queries are gonig to AI. If someone chooses to not use AI, their options will be limited in future.
The free sharing of information on Internet is I think coming to an end.
Not only if the internet dies. If the country or countries managing the AI decide they don't like your country and ban you from using their (major) AI providers, in a world where most knowledge and work tanks will require access, you're screwed.
Not even talking dictatorship scenarios here. The USA might simply decide it doesn't like your country. Or maybe heavily cripple the models you can use, or censor them, for whatever reason.
I guess this is an argument for local AI. Probably not feasible for the less technically savvy.
I have high hopes for AMD Halo / NVidia Spark or sth. similar - relatively open computers (unlike Apple) with unified memory. Now if those prices just came down a bit... :-/
Don't think of it as a step back, it's just a point in a familiar cycle:
New application of computing requires a lot of compute -> compute centralizes -> compute improves, smaller, faster, cheaper -> compute decentralizes -> new application of compute is found -> (return to beginning of cycle)
Can you think of examples of this cycle? I know this is roughly what happened with mainframes, but I struggle to think of other times it went like that.
cloud? I feel like that's an example of things going the other way. It used to be every company would have either on on-prem or a colocated server at the nearest network hub - a fairly de-centralized system, and then everything went centralized with cloud platforms.
CGI rendering. To do Pixar quality animations you used to need racks and racks of servers and it still took months. Now anyone can do it, and we saw tons of slick animations on YouTube from indie studios as a result (before the AI slop cloning started I mean)
- Large centralized compute is more powerful than portable compute
- Wired is better than wireless
- The public wants new things
- The tech industry grows to use/“use” the available processing power
As long as you’re not continuously comparing to AI in large centralized compute, what you’re really looking for is “when does portable compute become powerful enough to run AI models that are capable of meaningful work” and I’d say we’re already there and it’s getting better.
This began with cloud. When software began running as a front-end in a browser, people began complaining about centralization.
But then it became normal.
The monopolization of knowledge and expertise is the next step. I don't mean as if it were intended, but it is the logical progress given the technological evolution. The cause is inventive structures: the tech is neutral (in my opinion), but the way in which it is used and deployed is subject to and disciplined by capital.
From a quick skim of the page source (including a dynamically created iframe) it just does a little animation and locally allocates some useless arrays in JavaScript to waste your "uploaded" RAM.
Sam Altman's argument was that people tend to not run their consumer electronics at 100% 24/7, and so views the time that's not happening as wasted potential for RAM that could be sitting in a data center. I sincerely reject that notion, but from that... utilitarian perspective it's hard to argue that spreading out the world's limited RAM supply across thousands of personal machines would fix any issues
Just give us back our RAM prices so that everyone can afford 256GB DDR5 and run their models locally!
And Sam's (correct) point is that that doesn't work. If everyone tried to run the same amount of inference locally with the same size models that they're currently doing, the shortage would be even worse.
It sort of is, no? To agree with it, you need to share the goal of maximum utilization. Sam Altman seems to take that as a given. But clearly people disagree and not just with RAM. Most kitchens and bathrooms are unused most of the time, but that doesn't mean every home having one is bad.
No, he was not saying that was his goal! That's what everyone seems to misunderstand about this quote.
It's like a highrise developer saying "people think our buildings take up a lot of land, but actually houses are much worse from a land use perspective". Which is objectively true, and yet does not mean the developer thinks the entire world should be turned into houses, or even that they dislike houses!
Except he was the developer who signed a deal with land owners to buy 40% of the land with the intention of building houses to rent and caused condo prices to 5x.
The quote assumes that all of humanity has the same goals as a corporation does: that every individual person desires maximum utilization of components for the sake of company shareholders. Individual people want a computer and just can't buy them because the entire world's memory chip supply, flash storage supply, and increasingly hard drive supply is being swallowed into the maws of corporations who would love nothing more than to sell you tiny bits of that compute through a monthly subscription in perpetuity.
Not everyone subscribes to the "cloud compute is more important than everything else" outlook, but everyone is affected by it nonetheless.
In that statement, Altman also insinuated that most businesses switching from SAAS provided by OpenAI, Anthropic, or the VC-funded companies that build on top of them to locally-hosted software won't kill both OpenAI and Anthropic.
Given that he was in a fluffy interview at the time, he's not going to say that OpenAI and Anthropic would die, but we all know that -unless the "Pacing The Frontier" garbage succeeds in its goal of creating law and regulations that massively obstruct LLM manufacturers who aren't the four or five currently-largest ones in the US- that's how it'd turn out. Once OpenAI and Anthropic are dead, the VC money pouring into the LLM manufacturing and services industries evaporates, and all that demand for high-speed VRAM vanishes with it.
So, no, switching to locally-hosted software will reduce RAM prices... just not immediately.
The quote wasn't weird and made total sense in Altman's POV. It was just egoistic and myopic regarding how people feel about ownership and independence.
It has nothing to do with his POV or even ego. It was an objectively true fact. GPUs in data centers do get more utilization than those used locally. That's all he was saying.
Especially when you take into account that people are largely unconcerned about true facts, and will gladly misquote or misinterpret you just in order to be outraged.
He said/implied that we would have shortage either way (without data center buildout, but everyone doing private/local inference), which doesn't sound far fetched.
Also, a lot of things are true but what truths you choose to broadcast can tell us what you care about. E.g it's also true that we'd all be richer if sam altman shared is wealth with us, lets take a few seconds to think why he's not saying this truth, after all it's an objectively true fact.
The whole context of Sam Altman and other tech billionaires is weird. The modern tech advancements are deeply rooted in exploitation, racism and eugenics, concentrating military, political and economic power, and stealing agency and power of resistance from the common people.
>so consumers are left to feel the guilt of having unused RAM in their computers
Funny WEB Site!
Not me, I have 2 8G DDR4 sticks sitting in a draw that I purchased a few of years ago when they were cheap. I was going to use them to upgrade my W541 to 32G, but I have one of the first models that the max is 16G. So saving them in case I have RAM issues.
If Sam wants them, he can pay me 5000 USD each for them, otherwise I will proudly hang on to them. If I knew his email I would send him a pic of them saying "hahaha" :)
It's the proof that people have too much money and should instead give it to Sam Altman that will make a more efficient use of it for our collective wellbeing®.
$15 is cheaper than a movie ticket and the creator probably got more enjoyment out of it. Also, sometimes getting a few giggles out of people is return enough out of an investment.
The real solution is even simpler than having everyone upload their RAM to him: Sam can ask Claude (or ChatGPT, if OpenAI takes diversity hires) to build him a DRAM fab, using ultrathinking, making no mistakes.
200 comments
[ 2.4 ms ] story [ 56.9 ms ] threadI don’t think it’s very productive to go on a witch hunt on someone who said something stupid and is being taken out of context (albeit he should have known not to say that).
Obligatory: "we're curing cancer r-right?" No Nat we're making slop videos with it.
where is the proof?
>and are now not scaling up production according to demand.
fabs are being built by maaany companies, but it takes years
https://en.wikipedia.org/wiki/DRAM_industry_price_fixing
To date, five manufacturers have pleaded guilty to their involvement in an international price-fixing conspiracy between July 1, 1998, and June 15, 2002, including Hynix, Infineon, Micron Technology, Samsung, and Elpida.[3]
Simple as that.
If I said all donuts are spaceships, readers are not obligated to go and google it.
US DOJ link: https://www.justice.gov/archive/opa/pr/2005/April/05_at_207....
This is remarkably different than someone claiming donuts are spaceships when the company itself was charged and pleaded guilty, and it's a matter of public record.
Theres never any "proof."
Theres no "proof" on the internet.
All we ever have is "evidence."
You're free to doubt it, it's not absolutely certain, but pretending the chance is near zero is just not even remotely as rational as you seem to think.
just asked about some kind of proof when world is full of conspiracy theories, nothing wrong with that, so stop being aggressive and snarky
and i knew about stuff from 2000, but thought there was something recent
If the USA were a real democracy they would do something about it. After all this even violates capitalism, e. g. free market leading to reduced prices. But the USA is run by a criminal gang, pocketing away money - the TechBros and their cronies. They just make any legislation they want to, with a bit sugar coating on top to not appear too overly evil or greedy. But they are.
> From a global allocation perspective, actually, it’s quite terrible to have everybody have a bunch of the world’s limited supply of DRAM sitting in their own computer and being used 5% of the time. You want maximum utilization in the cloud if you’re trying to deal with this very supply-constrained environment.
That's a ghoulish point of view that puts real living humans on a lower hung than the poor poor corporations who just want RAM to be so expensive that everyone needs to buy compute from them. Convenient for the corporations!
He's an easy target because of who he is, so I understand why the entire Internet ran with the fake story (even LTT!). But on Hacker News you'll definitely get pushback if you post misleading stuff.
Let's say there are 10 companies that need 1 GB of RAM each in a life-or-death scenario. They would be willing to pay $100,000,000 per GB to get 1 GB of RAM.
At the same time, there are 90 more companies that need 1 GB of RAM each to make certain quarterly targets. They are willing to pay $10,000 per GB to get 1 GB of RAM.
If you make 9 GB of RAM, you can sell it for $900,000,000. If you make 99 GB of RAM, you can only sell it for $990,000, because the 10 life-or-death companies only need to bid high enough to push out a quarterly-target company.
It's a bit of a contrived example, but it shows you how restricting supply can boost total revenue if there is a segment of the market that has inelastic demand. (this is why medicine pricing has to be regulated - in the medical industry, there are lots of inelastic demand scenarios that can lead to predatory pricing)
that could've been crummy for someone; you should warn that your cute little complaint actually eats ram; half is clearly fiction but the OOM isn't.
Do you also click on the links of the Nigerian princes?
Will we go back to PCs? As in: are we going to get cheap local hackable frontier AIs at some point? I hope so.
:insert Will Smith presenting meme:
The Cloud
Lot of people using their office-notebooks at home, even if its not allowed. Most people out of tech do not have big hardware at home anymore.
The AI boom first reversed the trend, to now try to get back to that cycle.
I am using AI now to make a game engine that will run and make 'nice enough' graphics on simpler machines. things like laptops with iGPU and 4gb of ram.
U can still get quite ok graphics, and make very complex and rich gameplay systems using such a machine. It just needs to be the target machine while developing so the right tradeoffs are made.
But yes, it has been a test of just how few tokens you can get by on.
Which would not be all too bad. It would also mean the end of stupid SaaS subscription services, and mass surveillance, adtech. And shitty propaganda.
And your local AI could emulate the good parts of the internet.
Forget about AI, the entire PC market is getting killed over this.
As an avid gamer I'm unsure if in 5 years I can even buy a gaming PC anymore. Not even because of cost, but it's unclear if anyone will still make the components anymore.
NVIDIA isn't interested, nor is AMD.
5 years ago it would have been science fiction
Also, this reminds me so much of the download more RAM ads back in the day.
I wonder what it would look like if Apple offered a slimmed down shell from the Neo for ~$300 that required a connected iPhone. Maybe as iOS and MacOS converge we'll get there.
Many games that come out still, and are quite successful, do not require such technologies.
1. Indy developer releases a game or trailer of a game. 2. Within an extremely short time (days to weeks at most) there are clones on the market. 3. These games often outperform the original.
It's absolutely not something that's gone on for ages.
It might be a novel gameplay idea / twist on a genre, and mere days after launch instead of being able to stand on its own and gather attention it's drowning in a pool of lookalikes. It's no longer distinct.
The speed of being able to slop together something like this is less than 1% of what it used to be.
So yes maybe it happens more often now, but it’s still not new.
Yes, that's literally the point of the people you're responding to. Has HN lost reading comprehension, or are you arguing in bad faith?
It used to happen occasionally. Now it happens instantly. That delta is the problem.
So it's easier than ever to make a game, but it's more difficult than ever to get anyone to try what you've made. I'd say that's just as bad, and in some ways worse, than having your game cloned.
For someone looking to play a new game it will be difficult to sort through all the bad games to find the good ones. And for anyone pouring their soul into a game they will more likely than not have an audience of one.
Our eventual destination is probably going to be a) pick up a device, b) ask your bot to make up a game on the spot, c) play. It sounds so lonely and depressing.
gaming does not equate to only things like the last of us or cyberpunk which need extreme machines. a lot of visual fidelity can be reached still without having such a monster. Just need to not use certain features which are currently pretty much exclusive to Nvidia...
And I assume they will only continue to go up. When it came out, purchasing one seemed like an easy decision. But I paid half as much for one with the same CPU/RAM (and upgraded my SSD for cheap).
If I started today and played 10 games I can emulate every single day for the rest of my life restricting myself to purely emulation, I wouldn't run out of new games for...five and a half years. (assuming my rough calculation of ~20k available games).
Drop it to a game a day and I will likely never run out of games given my life expectancy.
Consider that I will find some games I like to play for more than a day and there are easily well more than enough games in emulation alone to entertain me for the rest of my life on existing entry level hardware.
Will I ever play MODERN SHOOTER {CURRENT YEAR} on this device? Certainly not, but I can't see a reason to bemoan lack of access to such a small portion of the available industry.
Like, suppose you're a reader. You might be right that you might enjoy your favorites in their original languages, but you'd be silly to be worried about your future ability to enjoy books just because you'll likely never get to experience reading 三体 as a chinese citizen fluent in the language.
There should be a big enough back catalog of great games for you to enjoy without having to buy a new gaming rig.
BUT! I feel that lost hope could/will turn into a new forms of "min-maxing" or whatever where we will see people focusing on small, efficient software to make the most of what we have to weather the drought. We got spoiled with cheap hardware. Time to make the most of all that cheap hardware we have.
Though of course, replacing this hardware as it breaks down will be expensive. Perhaps we could go back to 1/2/4 GB RAM sticks and 256/512 GB Disks. Then run minimal Linux installs, minimal GUIs, minimal services, minimal software.
I bought it at the middle/end of 2024, and I paid 3790 NOK (~357 USD at the time).
The lowest price online in my country for the same product new online today is 10233 NOK (~1085 USD)!
270% of what it was before, it’s absolutely bonkers.
I was even considering putting it up for sale a few weeks ago. But of course what I would be able to get for it would be no where near $1000 when selling second hand. Instead I might if I am lucky get slightly more than I originally paid which would be nice and all except that if I then regret selling it it’s still expensive to replace, or I have to also buy it second hand from someone else that may have worn theirs out considerably more compared to what I have done to the one I currently have. Mine is very well taken care of since the start, and although it’s nearing capacity now I haven’t written and deleted data over any over on it.
Been doing that for myself for the past 15 years
Advances in hardware are only impressive and useful for me when I avoid the so-called "tech" company "engineer" BS
The reason I stop using old computers is never due to software. It's always outdated proprietary cables, failing power adapters, screens, hinges, and so on
Sometimes "minimalism" is simply for _speed, resource usage and reliability_ (that's what I get from small software, anyway) in contrast with prevailing marketed garbage that is bloated and sluggish by commparison
Other times, claims of "minimalism" might be trying to capitalise on some silly trend
Personally I like what the computers I own can do for me and learning to make them do more. I have a brain, I don't do much "guessing" and I don't really need "inference" on a daily basis, certainly not for such a high cost to the environment
I have all the "convenience" I can handle
I have to think, it can’t take much time before the Chinese DUV fabs come online and non-frontier chip prices crater. Sure, you might not be able to get 5090 equivalents, but the 30xx series was DUV and so that should be replicable in a few years.
Whatever quality bar / process node China gets to you know they will scale it out super hard.
(Some quick investigation suggests that the Yuliangsheng DUV machines currently under construction are 28nm, which would map to 2014’s GTX 900. Perhaps I’m being overly bullish but I assume they will recapitulate/clone the tech faster than the rate it was originally discovered.)
Have an agentic enough setup and they'll do anything that the biggest closed models can. They'll also do more, making them the true frontier.
A sensible point of entry is a 2899$ Mac mini.
If you're not partaking that's only on you.
> If you're not partaking that's only on you.
There's honestly nothing sensible about that for a lot of developer budgets across the globe. Don't assume most folks have $3k to burn.
They absolutely can't do what, or more of what the biggest models can do. There's nothing frontier about small local Qwen or Gemma models.
2899$ gets you a machine that is just good enough to run some small models, slowly, with terrible context, requires a ton of tinkering and fine tuning, managing and constantly downloading a bunch of huge file blobs, etc.
Partaking those $2899 in most of the large cloud models would bring you much more productivity, much less boilerplate and time spent in setup, etc.
So.. just weird accusing people here of "not partaking" in AI when you're just essentially toying around if you're actually comparing local Qwen and Gemma's with frontier models.
Everybody will have a mini PC for the sole purpose of streaming content, with no storage to keep anything on it.
It's just unfortunate that producing a half decent computing platform is significantly more difficult than a rifle (and that in the case of said hardliners, the capital is backing them, here it's working against us).
And why would the game industry, considered as a whole, want to incur higher costs to exclusively deliver its product in a way that provides less value to the majority of its customers, many who won’t be willing to pay the higher prices required to maintain similar profits?
You might say that this is impossible but look at current trends. A lot of information today is getting shared in the walled gardens of Internet. You have to login to social media to view posts. The Twitter, Reddit threads that were accessible freely now require you to log in. Similar thing is happening across other networks. People are no longer asking questions on Stackoverflow. All those queries are gonig to AI. If someone chooses to not use AI, their options will be limited in future.
The free sharing of information on Internet is I think coming to an end.
Not even talking dictatorship scenarios here. The USA might simply decide it doesn't like your country. Or maybe heavily cripple the models you can use, or censor them, for whatever reason.
I guess this is an argument for local AI. Probably not feasible for the less technically savvy.
New application of computing requires a lot of compute -> compute centralizes -> compute improves, smaller, faster, cheaper -> compute decentralizes -> new application of compute is found -> (return to beginning of cycle)
Maybe arcade games?
- Large centralized compute is more powerful than portable compute
- Wired is better than wireless
- The public wants new things
- The tech industry grows to use/“use” the available processing power
As long as you’re not continuously comparing to AI in large centralized compute, what you’re really looking for is “when does portable compute become powerful enough to run AI models that are capable of meaningful work” and I’d say we’re already there and it’s getting better.
- local latency is lower than network latency
- latency is important for action games
But then it became normal.
The monopolization of knowledge and expertise is the next step. I don't mean as if it were intended, but it is the logical progress given the technological evolution. The cause is inventive structures: the tech is neutral (in my opinion), but the way in which it is used and deployed is subject to and disciplined by capital.
But, the “good enough for most lay people” intelligence bar will for sure miniaturize, especially as inference ASICs get more specialized.
But yes, we both agree that what was the frontier many years ago will fit on your desktop.
On a less sarcastic note, I wonder if a community based compute donation could actually help offset knowledge and compute centralisation.
Incentive is difficult but anti-AI sentiment is still as strong as ever. Maybe an A-v-A (in the spirit of H-v-H) campaign might pull in some.
also, spare LLM credits from donations for use by community groups
I would be shocked if it did anything at all.
"Iniital commit"
Lol, little cringe "Look guys, how human I am!!"
Just give us back our RAM prices so that everyone can afford 256GB DDR5 and run their models locally!
This solves the datacenter problem and it's impacts, RAMageddon, SSDgeddon, etc..
and they can just sell us the improved models without the hassle of doing the compute themselves..
What's the problem? With the proper ToC's they would still get our data.. ¯\_(ツ)_/¯
And Sam's (correct) point is that that doesn't work. If everyone tried to run the same amount of inference locally with the same size models that they're currently doing, the shortage would be even worse.
It's like a highrise developer saying "people think our buildings take up a lot of land, but actually houses are much worse from a land use perspective". Which is objectively true, and yet does not mean the developer thinks the entire world should be turned into houses, or even that they dislike houses!
isn't communal living the most efficient thing resource-wise though?
Not everyone subscribes to the "cloud compute is more important than everything else" outlook, but everyone is affected by it nonetheless.
Given that he was in a fluffy interview at the time, he's not going to say that OpenAI and Anthropic would die, but we all know that -unless the "Pacing The Frontier" garbage succeeds in its goal of creating law and regulations that massively obstruct LLM manufacturers who aren't the four or five currently-largest ones in the US- that's how it'd turn out. Once OpenAI and Anthropic are dead, the VC money pouring into the LLM manufacturing and services industries evaporates, and all that demand for high-speed VRAM vanishes with it.
So, no, switching to locally-hosted software will reduce RAM prices... just not immediately.
But saying people shouldn’t want something that was abundant and you made scarce is an asshole thing to say.
Also, a lot of things are true but what truths you choose to broadcast can tell us what you care about. E.g it's also true that we'd all be richer if sam altman shared is wealth with us, lets take a few seconds to think why he's not saying this truth, after all it's an objectively true fact.
Funny WEB Site!
Not me, I have 2 8G DDR4 sticks sitting in a draw that I purchased a few of years ago when they were cheap. I was going to use them to upgrade my W541 to 32G, but I have one of the first models that the max is 16G. So saving them in case I have RAM issues.
If Sam wants them, he can pay me 5000 USD each for them, otherwise I will proudly hang on to them. If I knew his email I would send him a pic of them saying "hahaha" :)
Ok.
Plus it's far more efficient to buy tons of gpu's for the shelves rather than actually use them.