7 ms·

It seems we will run out of hardware by March? "Hard drives already sold out for this year" - https://www.theregister.com/2026/02/20/ai_blamed_again_as_hard_dr

It seems we will run out of hardware by March? "Hard drives already sold out for this year" - https://www.theregister.com/2026/02/20/ai_blamed_again_as_hard_drives_sell_out/ https://www.theregister.com/2026/02/20/ai_blamed_again_as_ha... Time for an AI tax on the hyperscalers.

jsheard 8mo ago
> It seems we will run out of hardware by March? What happens when an unstoppable force (building everything in Electron because hardware is cheap) meets an immovable object (oh no hardware is expensive now)?
- ieie3366 8mo ago
  [dead]
- Betelbuddy 8mo ago
  2026 will be the year of Rust...
  - NoiseBert69 8mo ago
    Due to lack of memory leaks which will stop increasing RAM prices?
    - Betelbuddy 8mo ago
      The efficiency... https://users.rust-lang.org/t/energy-consumption-in-programming-languages/95625 https://users.rust-lang.org/t/energy-consumption-in-programm...
      - xnorswap 8mo ago
        I see that's from almost 10 years ago, it would be interesting to see how that's changed with improvements to V8, python and C# since. Also, Typescript 5 times worse than Javascript? That doesn't really make sense, since they share the same runtime.
        embedding-shape 8mo ago
        Why is that so unbelievable? TypeScript isn't JavaScript, and while they have the same runtime, compiled TypeScript often don't look like how you'd solve the same problem in vanilla JS, where you'd leverage the dynamic typing rather than trying to work around it. See this example as one demonstration: https://www.typescriptlang.org/play/?q=8#example/enums https://www.typescriptlang.org/play/?q=8#example/enums The TS code looks very different from the JS code (which obviously is the point), but given that, not hard to imagine they have different runtime characteristics, especially for people who don't understand the inside and outs of JavaScript itself, and have only learned TypeScript.
        xnorswap 8mo ago
        Enums are one of only a few places where there is significant deviation, I don't believe that makes it 400% less efficient.
        embedding-shape 8mo ago
        Maybe read the paper and see if you can figure out their reasoning/motivation :) https://dl.acm.org/doi/10.1145/3136014.3136031 https://dl.acm.org/doi/10.1145/3136014.3136031 One thing to consider, is that with JavaScript you put it in a .js file, point a HTML page at it, and that's it. TypeScript uses a ton more than that, which would impact the amount of energy usage too, not to mention everything running the package registries and what not. Not sure if this is why the difference is bigger, as I haven't read the paper myself :) But if you do, please do share what you find out about their methodology.
      - Zababa 8mo ago
        This image comes from running the different versions of the benchmark games programs. Some of the difference between languages may actually be just algorithmic differences, and also those programs are in general not representative of most of the software that runs.
    - gck1 8mo ago
      That, and also because rust compiler is a very good guardrail & feedback mechanism for AI. I made 3 little tools that I use for myself without knowing how to write a single rust line myself.
      - Imustaskforhelp 8mo ago
        I can see that a reality but I am more comfortable using Golang as the language compared to rust given its fast compile times and I have found it to be much more easier to create no-dependices/fewer-dependencies project plus even though I wouldn't consider myself a master in golang, maybe mediocre, I feel much easier playing with golang than rust. The resource consumption b/w rust and golang would be pretty minimal to figure out actually for most use cases imho.
    - nicoburns 8mo ago
      Because it's more memory efficient than most other languages. So you can achieve the same result with lower RAM requirements.
- throwaw12 8mo ago
  consumer RAM is not what's creating shortage. Data centers doesn't run electron to train the model or for inference
  - malfist 8mo ago
    Sure, consumer ram isn't causing a shortage, but it's affected by the shortage.
  - bayindirh 8mo ago
    Every RAM producer is stopping their consumer grade RAM production to provide ECC-RAM and VRAM now. Micron discontinued and closed down Crucial brand as a whole. So, getting systems with higher RAM capacity is getting harder (from laptops to smartphones). So, for a couple of years, we need to stop using Electron so much and use what we have efficiently. Data centers, esp. AI hyperscalers do not care about efficiency for now, because they can suffocate consumer-grade part of the hardware marketplace and get anything and everything they want. When their bubble pops, or the whole capacity ends, they need to learn to be efficient, too. For reference, a well-optimized cluster runs at ~90% efficiency even though they have thousands of users. AI hyperscalers are not there. Maybe 60% efficient, at most. They waste a lot of resources to keep their momentum.
    - spockz 8mo ago
      I have a silent hope that because of this change we all will get ECC ram and that consumer CPUs will get proper support for them.
      - bayindirh 8mo ago
        AMD's RYZEN already supports it. ASUStor's latest generation of NAS devices come with AMD x86_64 processors and ECC RAM as a standard, but ECC RAM in SODIMM format was not cheap, even when the RAM was cheap, either.
        spockz 8mo ago
- fullstop 8mo ago
  I guess we have to get creative again.
  - interleave 8mo ago
    I actually think you're right here. Resource constraints have often helped me come up with stuff that I'm actually proud of.
    - fullstop 8mo ago
      Absolutely. It's why I find working on a microcontroller with 1KB of memory so much more rewarding than, say, a Raspberry Pi.
- pjmlp 8mo ago
  We go back to the demoscene days, being creative with what we have instead of shipping Electron junk.
  - nozzlegear 8mo ago
    Inshallah
    - spiderfarmer 8mo ago
      If the FSM wills
- UltraSane 8mo ago
  Stop using Electron to save massive amounts of RAM.
  - riku_iki 8mo ago
    its not easy choice if you want mature crossplatform UI framework.
    - izacus 8mo ago
      There's plenty of those.
      - riku_iki 8mo ago
        like what?
        izacus 8mo ago
        Ask your favorite LLM? :D
- andix 8mo ago
  Maybe we need to let go of our auto-scaled 100 pod service mesh for a todo list app, and just deploy it bare metal on 2 servers.
  - fireant 8mo ago
    You joke, but I remember seeing a talk by Wunderlist CTO who has pretty much that. Also polyglot company and microservices in random languages. Can't find the talk now, but https://www.infoq.com/news/2014/11/gotober-wunderlist-microservices/ https://www.infoq.com/news/2014/11/gotober-wunderlist-micros... mentions 60 services at least.
    - andix 8mo ago
      I need to get more ideas for my side project. A todo list app with micro services, but everything in bash scripts. So far it's just 6 services. https://github.com/andi0b/vibe-todo https://github.com/andi0b/vibe-todo
- lima 8mo ago
  Everything gets more expensive?
infecto 8mo ago
Why tax something that the market will figure out? This is normal and things will sort themself out.
- dakolli 8mo ago
  Because this perfect version of capitalism you think exists, doesn't. We live in a world with markets dominated by cartels of tech companies who don't play by the rules. Every other industry that impacts society in a negative way typically pays some sort of specialized tax to offset that, I don't know why these tech oligarchs shouldn't have too. It's wild how people just want to let them do whatever they want. Everyone says we need to deregulate tech, and certain industries to get ahead of China.. Isn't it funny how their largely government controlled economy (to a degree) is annihilating the west on all fronts economically. We need far more regulation. China will defeat the West solely because it regulates its billionaires, not the other way around like we have it in the West. And I hope so, the world is rooting for you China.
  - infecto 8mo ago
    Way to put words in people mouths. Markets are imperfect but I do believe on average they are one of the better tools to solve supply and demand issues. I don’t know who will come out winners but I do agree that China did well taking the playbook from Singapore and navigating their country through incredible amounts of growth. They are still facing depressing housing prices and deflation in other parts of the economy. There are absolutely areas where markets breakdown, thinking problems where impacts are on longer horizons but for simple supply and demand like what we are seeing today, things will sort out in a couple years.
- gck1 8mo ago
  We see that market is very irrational now and it can stay irrational for long enough to destroy everything we know in tech. By the time market figures things out, you may no longer have services, and hardware that you use daily. When such amounts of stupid money are pumped into a single industry, even if all AI companies went out of business tomorrow, it's going to take years for things to go back to normal. FWIW, I'm not advocating taxes, as I think that won't really do anything. I don't know what the solution is either.
usrusr 8mo ago
Reads a bit like the Paperclip Maximizer appearing way ahead of schedule? Implemented not as AI, but as emergent behavior in the ways of the financial class (that happens to be about AI, singularity and all that).
Cthulhu_ 8mo ago
> Time for an AI tax on the hyperscalers. Like the purchase price + increased cost? The thing is that these parties are sitting on billions and billions of investor money, they don't care that hardware is 400x as expensive. Which companies like nvidia have capitalized on a few years ago, they were already able to price their hardware at a 400% markup compared to pre-crypto times, and shift their focus from consumer graphics chips to datacenter compute chips, causing their revenue to go up 6x (if my interpretation of [0] is correct) [0] https://www.macrotrends.net/stocks/charts/NVDA/nvidia/revenue https://www.macrotrends.net/stocks/charts/NVDA/nvidia/revenu...

jsheard 8mo ago

> It seems we will run out of hardware by March? What happens when an unstoppable force (building everything in Electron because hardware is cheap) meets an immovable object (oh no hardware is expensive now)?

ieie3366 8mo ago
[dead]
Betelbuddy 8mo ago
2026 will be the year of Rust...
- NoiseBert69 8mo ago
  Due to lack of memory leaks which will stop increasing RAM prices?
  - Betelbuddy 8mo ago
    The efficiency... https://users.rust-lang.org/t/energy-consumption-in-programming-languages/95625 https://users.rust-lang.org/t/energy-consumption-in-programm...
    - xnorswap 8mo ago
      I see that's from almost 10 years ago, it would be interesting to see how that's changed with improvements to V8, python and C# since. Also, Typescript 5 times worse than Javascript? That doesn't really make sense, since they share the same runtime.
      - embedding-shape 8mo ago
        Why is that so unbelievable? TypeScript isn't JavaScript, and while they have the same runtime, compiled TypeScript often don't look like how you'd solve the same problem in vanilla JS, where you'd leverage the dynamic typing rather than trying to work around it. See this example as one demonstration: https://www.typescriptlang.org/play/?q=8#example/enums https://www.typescriptlang.org/play/?q=8#example/enums The TS code looks very different from the JS code (which obviously is the point), but given that, not hard to imagine they have different runtime characteristics, especially for people who don't understand the inside and outs of JavaScript itself, and have only learned TypeScript.
        xnorswap 8mo ago
        Enums are one of only a few places where there is significant deviation, I don't believe that makes it 400% less efficient.
        embedding-shape 8mo ago
        Maybe read the paper and see if you can figure out their reasoning/motivation :) https://dl.acm.org/doi/10.1145/3136014.3136031 https://dl.acm.org/doi/10.1145/3136014.3136031 One thing to consider, is that with JavaScript you put it in a .js file, point a HTML page at it, and that's it. TypeScript uses a ton more than that, which would impact the amount of energy usage too, not to mention everything running the package registries and what not. Not sure if this is why the difference is bigger, as I haven't read the paper myself :) But if you do, please do share what you find out about their methodology.
    - Zababa 8mo ago
      This image comes from running the different versions of the benchmark games programs. Some of the difference between languages may actually be just algorithmic differences, and also those programs are in general not representative of most of the software that runs.
  - gck1 8mo ago
    That, and also because rust compiler is a very good guardrail & feedback mechanism for AI. I made 3 little tools that I use for myself without knowing how to write a single rust line myself.
    - Imustaskforhelp 8mo ago
      I can see that a reality but I am more comfortable using Golang as the language compared to rust given its fast compile times and I have found it to be much more easier to create no-dependices/fewer-dependencies project plus even though I wouldn't consider myself a master in golang, maybe mediocre, I feel much easier playing with golang than rust. The resource consumption b/w rust and golang would be pretty minimal to figure out actually for most use cases imho.
  - nicoburns 8mo ago
    Because it's more memory efficient than most other languages. So you can achieve the same result with lower RAM requirements.
throwaw12 8mo ago
consumer RAM is not what's creating shortage. Data centers doesn't run electron to train the model or for inference
- malfist 8mo ago
  Sure, consumer ram isn't causing a shortage, but it's affected by the shortage.
- bayindirh 8mo ago
  Every RAM producer is stopping their consumer grade RAM production to provide ECC-RAM and VRAM now. Micron discontinued and closed down Crucial brand as a whole. So, getting systems with higher RAM capacity is getting harder (from laptops to smartphones). So, for a couple of years, we need to stop using Electron so much and use what we have efficiently. Data centers, esp. AI hyperscalers do not care about efficiency for now, because they can suffocate consumer-grade part of the hardware marketplace and get anything and everything they want. When their bubble pops, or the whole capacity ends, they need to learn to be efficient, too. For reference, a well-optimized cluster runs at ~90% efficiency even though they have thousands of users. AI hyperscalers are not there. Maybe 60% efficient, at most. They waste a lot of resources to keep their momentum.
  - spockz 8mo ago
    I have a silent hope that because of this change we all will get ECC ram and that consumer CPUs will get proper support for them.
    - bayindirh 8mo ago
      AMD's RYZEN already supports it. ASUStor's latest generation of NAS devices come with AMD x86_64 processors and ECC RAM as a standard, but ECC RAM in SODIMM format was not cheap, even when the RAM was cheap, either.
      - spockz 8mo ago
fullstop 8mo ago
I guess we have to get creative again.
- interleave 8mo ago
  I actually think you're right here. Resource constraints have often helped me come up with stuff that I'm actually proud of.
  - fullstop 8mo ago
    Absolutely. It's why I find working on a microcontroller with 1KB of memory so much more rewarding than, say, a Raspberry Pi.
pjmlp 8mo ago
We go back to the demoscene days, being creative with what we have instead of shipping Electron junk.
- nozzlegear 8mo ago
  Inshallah
  - spiderfarmer 8mo ago
    If the FSM wills
UltraSane 8mo ago
Stop using Electron to save massive amounts of RAM.
- riku_iki 8mo ago
  its not easy choice if you want mature crossplatform UI framework.
  - izacus 8mo ago
    There's plenty of those.
    - riku_iki 8mo ago
      like what?
      - izacus 8mo ago
        Ask your favorite LLM? :D
andix 8mo ago
Maybe we need to let go of our auto-scaled 100 pod service mesh for a todo list app, and just deploy it bare metal on 2 servers.
- fireant 8mo ago
  You joke, but I remember seeing a talk by Wunderlist CTO who has pretty much that. Also polyglot company and microservices in random languages. Can't find the talk now, but https://www.infoq.com/news/2014/11/gotober-wunderlist-microservices/ https://www.infoq.com/news/2014/11/gotober-wunderlist-micros... mentions 60 services at least.
  - andix 8mo ago
    I need to get more ideas for my side project. A todo list app with micro services, but everything in bash scripts. So far it's just 6 services. https://github.com/andi0b/vibe-todo https://github.com/andi0b/vibe-todo
lima 8mo ago
Everything gets more expensive?

infecto 8mo ago

Why tax something that the market will figure out? This is normal and things will sort themself out.

dakolli 8mo ago
Because this perfect version of capitalism you think exists, doesn't. We live in a world with markets dominated by cartels of tech companies who don't play by the rules. Every other industry that impacts society in a negative way typically pays some sort of specialized tax to offset that, I don't know why these tech oligarchs shouldn't have too. It's wild how people just want to let them do whatever they want. Everyone says we need to deregulate tech, and certain industries to get ahead of China.. Isn't it funny how their largely government controlled economy (to a degree) is annihilating the west on all fronts economically. We need far more regulation. China will defeat the West solely because it regulates its billionaires, not the other way around like we have it in the West. And I hope so, the world is rooting for you China.
- infecto 8mo ago
  Way to put words in people mouths. Markets are imperfect but I do believe on average they are one of the better tools to solve supply and demand issues. I don’t know who will come out winners but I do agree that China did well taking the playbook from Singapore and navigating their country through incredible amounts of growth. They are still facing depressing housing prices and deflation in other parts of the economy. There are absolutely areas where markets breakdown, thinking problems where impacts are on longer horizons but for simple supply and demand like what we are seeing today, things will sort out in a couple years.
gck1 8mo ago
We see that market is very irrational now and it can stay irrational for long enough to destroy everything we know in tech. By the time market figures things out, you may no longer have services, and hardware that you use daily. When such amounts of stupid money are pumped into a single industry, even if all AI companies went out of business tomorrow, it's going to take years for things to go back to normal. FWIW, I'm not advocating taxes, as I think that won't really do anything. I don't know what the solution is either.

usrusr 8mo ago

Reads a bit like the Paperclip Maximizer appearing way ahead of schedule? Implemented not as AI, but as emergent behavior in the ways of the financial class (that happens to be about AI, singularity and all that).

Cthulhu_ 8mo ago

> Time for an AI tax on the hyperscalers. Like the purchase price + increased cost? The thing is that these parties are sitting on billions and billions of investor money, they don't care that hardware is 400x as expensive. Which companies like nvidia have capitalized on a few years ago, they were already able to price their hardware at a 400% markup compared to pre-crypto times, and shift their focus from consumer graphics chips to datacenter compute chips, causing their revenue to go up 6x (if my interpretation of [0] is correct) [0] https://www.macrotrends.net/stocks/charts/NVDA/nvidia/revenue https://www.macrotrends.net/stocks/charts/NVDA/nvidia/revenu...