5 ms·

The geopolitical bit of all this is the real wild card. No one really knows what's going to happen there.

by davidw 2y ago

UncleOxidant 2y ago
Yeah, you've got Apple, Nvidia and AMD heavily exposed to that geopolitical risk, but seemingly not willing to help lessen it by investing in US fabs. Whenever I note this people here and elsewhere say "they're smart to be fabless" but somebody's gotta run fabs and more fabs need to be run in the US and quickly. I realize industrial policy isn't in vogue these days but it might be a good idea for the future of US semiconductor production to arrange some marriages between some of these companies that have a lot of capital and have a lot of need to reduce their geopolitical risk. Our brand of capitalism isn't great at looking ahead more than a quarter or two.
- jjtheblunt 2y ago
  Apple is heavily invested in the new TSMC fab in the US east of the north end of Phoenix, last i read; not sure how arranged that marriage really is, but perhaps tax incentives for TSMC facilitated Apple plans to heavily utilize it. (trying to find linkable articles...like this https://www.theverge.com/2022/12/6/23497417/apple-tsmc-phoenix-fab-plans-biden-amd-nvidia https://www.theverge.com/2022/12/6/23497417/apple-tsmc-phoen... ) ; p.s. i wonder if only i compulsively balance lisp parens
- sroussey 2y ago
  Apple could invest in an Intel Foundry spinoff, but there is no reason to invest in the Intel of today.
- Panzer04 2y ago
  I thought historically companies always wanted "second sources" for things like this. Surely all the fabless semis can't be blithely running off a cliff with TSMC as their only parts source? That would be a catastrophic risk to take down the track if there's only a single vendor for advanced chips and they've decided to hike prices and collect all of the benefits alone from their advanced processes and share none of them with their customers.
  - lotsofpulp 2y ago
    Buyers will obviously always want more sellers. That is not a sufficient condition for sellers to exist, however. Selling what TSMC sells might not be easily reproducible.
    - Panzer04 2y ago
      As it currently stands, it is reproducible. In 5 years, it might not be, if Samsung/Intel throw in the towel. I don't really see that happening yet, but at some point if TSMC maintains its lead and the alternative options don't get the sales to continue investing, TSMC could well become completely unassailable at the leading edge.
      - lotsofpulp 2y ago
        Intel seems like it’s at least a few years (but I would bet many years) behind TSMC on obtaining/training the necessary people with necessary expertise with the necessary management changes. It took TSMC a long time and many expensive, risky bets to beat Intel. I imagine it will take Intel just as many resources and time to catch up to TSMC. Hence the divergence in their market caps.
        Panzer04 2y ago
        I would frame TSMCs lead to be as much down to a failure st intel rather than a success at TSMC. Intel invested a ton of money into making a non-EUV 7nm process work and basically got it wrong. Now they have EUV, it seems largely a matter of doing basically the same things as TSMC. Not to denigrate TSMC at all, they have excellent execution and are being well rewarded for that. Intel supposedly will be on par with TSMC in the next year or two with their newest processes. Even when they were stuck on 10nm, they were fabbing competitive parts with it for ages. I don't see the same "years behind" narrative you see, personally. Interestingly, even TSMC revenue is only 70b usd. Intel made 55b revenue last year (down markedly from earlier years as well). Yes, that combines semiconductor design and fab income, but Intel still has the financials to make things work over the next few years. Funnily enough, if TSMC starts hiking prices (which presumably they will at some point), that would also aid Intel, since there are only really three players, and Samsung is hardly in a better position than Intel is.
moffkalast 2y ago
The geopolitical bit is that TSMC is building local US and EU fabs in case of a Taiwan invasion. Once those are up and Intel becomes redundant the ad hoc subsidies might just dry up.
- davidw 2y ago
  Sure, but "is building" could be very far from "is able to smoothly switch production if the main plants and some/many of the associated people are war casualties"
- hakfoo 2y ago
  I have to wonder if a TSMC-owned fab in Phoenix is still a different geopolitical risk factor than an Intel-owned one. It reduces the deterrence factor of the TSMC facilities on Taiwan: if they get damaged or sabotaged during an invasion, the entire industry doesn't get blown back to 2010. But now I'm imagining they'd have a board of directors having to meet overseas and claim their legitimacy to trading partners, almost like the "governments in exile" during wars.
keepamovin 2y ago
Less so that thought. TSMC is Chinese via it being Taiwanese. No disparagement of regional uniqueness or actual ownership, just its success is a success of the Chinese people (into which Taiwanese are included). China will not substantially harm TSMC when it is so successful at destroying American dominance in a key industry. Taiwan and China are very much aligned when it comes to TSMC. Maintaining a US native foundry might actually be a way to prevent war because one of the key triggers might be the moment where all major chip players totally depend on TSMC, which gives Chinese/Taiwanese massive leverage. Economically, might be a loss leader. But strategically, important, probably.

The geopolitical bit of all this is the real wild card. No one really knows what's going to happen there.

by davidw 2y ago

The geopolitical bit of all this is the real wild card. No one really knows what's going to happen there.

Yeah, you've got Apple, Nvidia and AMD heavily exposed to that geopolitical risk, but seemingly not willing to help lessen it by investing in US fabs. Whenever I note this people here and elsewhere say "they're smart to be fabless" but somebody's gotta run fabs and more fabs need to be run in the US and quickly. I realize industrial policy isn't in vogue these days but it might be a good idea for the future of US semiconductor production to arrange some marriages between some of these companies that have a lot of capital and have a lot of need to reduce their geopolitical risk. Our brand of capitalism isn't great at looking ahead more than a quarter or two.

jjtheblunt 2y ago
Apple is heavily invested in the new TSMC fab in the US east of the north end of Phoenix, last i read; not sure how arranged that marriage really is, but perhaps tax incentives for TSMC facilitated Apple plans to heavily utilize it. (trying to find linkable articles...like this https://www.theverge.com/2022/12/6/23497417/apple-tsmc-phoenix-fab-plans-biden-amd-nvidia https://www.theverge.com/2022/12/6/23497417/apple-tsmc-phoen... ) ; p.s. i wonder if only i compulsively balance lisp parens
sroussey 2y ago
Apple could invest in an Intel Foundry spinoff, but there is no reason to invest in the Intel of today.
Panzer04 2y ago
I thought historically companies always wanted "second sources" for things like this. Surely all the fabless semis can't be blithely running off a cliff with TSMC as their only parts source? That would be a catastrophic risk to take down the track if there's only a single vendor for advanced chips and they've decided to hike prices and collect all of the benefits alone from their advanced processes and share none of them with their customers.
- lotsofpulp 2y ago
  Buyers will obviously always want more sellers. That is not a sufficient condition for sellers to exist, however. Selling what TSMC sells might not be easily reproducible.
  - Panzer04 2y ago
    As it currently stands, it is reproducible. In 5 years, it might not be, if Samsung/Intel throw in the towel. I don't really see that happening yet, but at some point if TSMC maintains its lead and the alternative options don't get the sales to continue investing, TSMC could well become completely unassailable at the leading edge.
    - lotsofpulp 2y ago
      Intel seems like it’s at least a few years (but I would bet many years) behind TSMC on obtaining/training the necessary people with necessary expertise with the necessary management changes. It took TSMC a long time and many expensive, risky bets to beat Intel. I imagine it will take Intel just as many resources and time to catch up to TSMC. Hence the divergence in their market caps.
      - Panzer04 2y ago
        I would frame TSMCs lead to be as much down to a failure st intel rather than a success at TSMC. Intel invested a ton of money into making a non-EUV 7nm process work and basically got it wrong. Now they have EUV, it seems largely a matter of doing basically the same things as TSMC. Not to denigrate TSMC at all, they have excellent execution and are being well rewarded for that. Intel supposedly will be on par with TSMC in the next year or two with their newest processes. Even when they were stuck on 10nm, they were fabbing competitive parts with it for ages. I don't see the same "years behind" narrative you see, personally. Interestingly, even TSMC revenue is only 70b usd. Intel made 55b revenue last year (down markedly from earlier years as well). Yes, that combines semiconductor design and fab income, but Intel still has the financials to make things work over the next few years. Funnily enough, if TSMC starts hiking prices (which presumably they will at some point), that would also aid Intel, since there are only really three players, and Samsung is hardly in a better position than Intel is.

moffkalast 2y ago

The geopolitical bit is that TSMC is building local US and EU fabs in case of a Taiwan invasion. Once those are up and Intel becomes redundant the ad hoc subsidies might just dry up.

davidw 2y ago
Sure, but "is building" could be very far from "is able to smoothly switch production if the main plants and some/many of the associated people are war casualties"
hakfoo 2y ago
I have to wonder if a TSMC-owned fab in Phoenix is still a different geopolitical risk factor than an Intel-owned one. It reduces the deterrence factor of the TSMC facilities on Taiwan: if they get damaged or sabotaged during an invasion, the entire industry doesn't get blown back to 2010. But now I'm imagining they'd have a board of directors having to meet overseas and claim their legitimacy to trading partners, almost like the "governments in exile" during wars.

keepamovin 2y ago

Less so that thought. TSMC is Chinese via it being Taiwanese. No disparagement of regional uniqueness or actual ownership, just its success is a success of the Chinese people (into which Taiwanese are included). China will not substantially harm TSMC when it is so successful at destroying American dominance in a key industry. Taiwan and China are very much aligned when it comes to TSMC. Maintaining a US native foundry might actually be a way to prevent war because one of the key triggers might be the moment where all major chip players totally depend on TSMC, which gives Chinese/Taiwanese massive leverage. Economically, might be a loss leader. But strategically, important, probably.