Comment by bgnn
2 months ago
I agree, forgetting happens naturally. For example, it would be pretty difficult to produce vacuum tubes now. But I doubt this is applicable for CMOS technologies. Most of the steps down to finfets (TSMC 16nm) are rather well known. Yes we don't know the exact recipe of TSMC, Samsung or Intel, but it's not like alien technology. I read technical papers from all these fabs regularly and it's more open than what would people expect. Of course they keep their secrets too, for the cutting edge. There's so much knowhow out there that it's quite probable we can get there again in a short time if TSMC vanished fron earth all of a sudden.
It's not just about secrets... it's about how many techniques and processes simply aren't documented. There's no need (someone knows how, and is in the business of training new hires), no capacity (they're not exactly idle), and no perception that any of this is important (things have kept working so far).
Could they eventually replicate a CMOS technology? No one doubts this, but the latest lith process took how many years to develop, and only one company makes those machines anywhere in the world? Nearly microscopic molten tin droplets being flattened mid-air so that it can radiate a particular wavelength of UV?
That's not something they'll have up and running again in 6 months, and if it were lost, regression to other technologies would be difficult or impossible too. We might have to start from scratch, so to speak.
> For example, it would be pretty difficult to produce vacuum tubes now.
Vacuum tubes are still made. They’re used extensively in instrument amplification.
But I think this bolsters your point!
There's a guy who makes them in his garage. They're not really conceptually hard to make as such, they're just fiddly, delicate, labour-intensive and mostly replaced by astoundingly cheaper and often better (outside of a few niches) solid state options.
If there were some kind of interdiction on silicon (an evil genie or some kind of Butlerian Jihad perhaps?), the market would remember and/or rediscover the thermoelectric effect and throw money/postapocalyptic bartered goods at glassblowers pretty sharpish.
If that status continued, I'm sure we'd see developments in that space in terms of miniaturisation, robustness, efficiency, performance, etc., that would seem as improbable to us as a modern CPU would seem to someone in the no-silicon timeline. You may never get to "most of a teraflops in your pocket, runs all day on 4000mAh and costs three figures" but you could still do a meaningfully large amount of computation with valves.
>There's a guy who makes them in his garage.
Savant-tier, obsessive, dedicates his life to it "guy" does it in his garage over a period of how many years, and has succeeded to what point yet? Has he managed even a single little 8-bit or even 4-bit cpu? I'm cheering that guy on, you know, but he's hardly cranking out the next-gen GPUs.
>the market would remember
Markets don't remember squat. The market might try to re-discover, but this shit's path dependent. Re-discovery isn't guaranteed, and it's even less likely when a civilization that is desperate to have previously-manufacturable technology can't afford to dump trillions of dollars of research into it because it's also a poor civilization due to its inability to manufacture these things.
5 replies →