8 ms·

A Texas semiconductor boom on the horizon?

delfinom 3y ago
It's debatable if it'll be much of a boom anywhere in the US. Most uses of chips are in cheap consumer junk overwhelmingly made overseas for bottom dollar. Second after that is actual serious uses for cars, not $1 electronics and more. And once you get into that category you have a finite market that is pretty much mature in terms of consumer reach. For all intensive purposes, the "EV demand surge" is over and existing semiconductors have caught up. There's also a real fear that US sanctions have caused China and other countries to intensify investment in their own semiconductor manufacturing (the EU definitely is), which will ultimately hamper the demand for US fabs.
- edgyquant 3y ago
  It may be debatable but the idea the chip surge is “basically over” is laughable. By your logic no nation should ever try to do anything economically because the economy is already done forming, the end of history is here.
  - baz00 3y ago
    Yup the chip surge isn't anywhere over due to geopolitics and demand, both of which are serious driving forces now.
  - EricE 3y ago
    Hey - no poking holes in this line of reasoning allowed. It’s a major tent pole of the global boiling religion - pesky human innovation blows too many holes on the “we are doomed - sacrifice your rights to fix it” narratives.
    - diputsmonro 3y ago
      I'm not really sure how these are related at all. Technical innovations for electric vehicles are totally different than the technical innovations required for fighting climate change. We can be poised to make progress in one sector and skeptical about progress in another. Blindly hoping that some breakthrough will happen to fix all our problems and just barreling along carelessly is worse than what you're describing.
  - kevin_thibedeau 3y ago
    The primary motive is to have redundant fabs for defense critical infrastructure that is vulnerable to Chinese attack. That's why Micron is being gifted a new fab for DRAM production on the other side of the country in the midst of ongoing layoffs. Boosting production isn't the goal and CHIPS act will be a net money loser.
  - User23 3y ago
    That was the ruling ethos of the United States Government from the fall of the Berlin wall on. Look how that worked out for us.
  - corethree 3y ago
    History has plenty of examples of nations and economies completely ending. In fact it's so consistent that it's basically a guarantee that the current economy you're living in will be "done" eventually. Therefore it's not a laughable conclusion but a valid possibility. We've seen two major economies descend into this "done" zone with no hope for recovery: Russia and Japan. Additionally no one is saying to "not to do anything". Try all you want he is simply predicting the outcome of trying.
- justrealist 3y ago
  The point is economic industrial independence for the US once Taiwan lights up. You're kind of missing the point.
  - bsaul 3y ago
    not sure why you got downvoted, that seemed like a pretty reasonable thing to consider.
    - User23 3y ago
      I can assure you that the Chinese strategists are. Wang Huning is way to smart not to have and he's just one that's somehow famous enough to be known in the West. I'm sure there are many more strategists of if not equal at least similar caliber working on furthering China's (hopefully) peaceful rise.
      - corethree 3y ago
        It's too chaotic. There are many who want peace but there are many who want war and there are many who will flip their opinion and this is on both sides of the line, the US and China. Anything can occur.
        justrealist 3y ago
        What we saw in Ukraine is that arguments about geopolitics, economy, military readiness etc can be completely negated by the person in charge acting on a strategic or emotional whim. Like all the experts in the world can concur that China is better off with peace in the Pacific, but if Xi wants a war, he's getting a war.
        User23 3y ago
        Xi already believes he has a war, it’s just in the early stages. See this[1]. [1] https://en.m.wikipedia.org/wiki/Unrestricted_Warfare https://en.m.wikipedia.org/wiki/Unrestricted_Warfare
- adamc 3y ago
  Just an idiom note: It's "intents and purposes" not "intensive purposes". Easy to mis-hear, but something you want to remember before it makes you look ill-educated in a work memo.
  - anonuser123456 3y ago
    This really jives with me.
  - retSava 3y ago
    It's also exactly the kind of humor I love. Intentional misuse of idioms, homonyms, intentional dangling modifiers, and the like. It's ok if people think I'm stupid, I can always prove them right!
    - robmsmt 3y ago
      reminds me when a friend said "escape goat" rather than "scapegoat"
- nwiswell 3y ago
  > Most uses of chips are in cheap consumer junk overwhelmingly made overseas for bottom dollar I don't understand your point. While true, those chips are made at historical process nodes using antiquated equipment. The plan isn't to buy antiquated equipment and try to compete. Friendly nations with low cost of labor and low geopolitical risk (ie., other than Taiwan) are perfectly capable of meeting our demands for old chips, given sufficient ramp time. The lion's share of the value in semiconductors is at the leading edge. These are the chips that are used in phones, laptops, PCs, servers, GPUs, et cetera. The United States has never been interested in competing on volume. We already are the world leader in producing (including manufacturing!) the cutting edge equipment that makes leading edge chips -- that is what we have denied China access to. The open question is whether we can "onshore" the actual manufacturing of leading edge chips again. Strictly speaking, Intel has always kept their most cutting edge manufacturing onshore (Oregon), but I understand that has more to do with IP and cycle time reasons than cost. An additional note: it is not necessary that a leading edge fab built today becomes a trailing edge fab in 5 or 10 years. There is a lively secondary market for used equipment specifically because of demand for the commodity chips you mentioned, so given that US labor costs are high, the old equipment would likely be sold and shipped to a low cost nation, and the US fabs repopulated with new equipment.
- corethree 3y ago
  You realize the top of the line chips are coming from Asia right now? The US right now does not have the ability to fab top of the line chips. Only Asia does. I mention this because it's basically one of the biggest hurdles for a boom other than cost which you didn't mention. The US has to exceed Asia in terms of cost AND raw technological prowess if it wants a boom because it is currently behind by a huge margin.
hooverd 3y ago
Faster calculators incoming?
- MengerSponge 3y ago
  I have a student who uses a TI-84. She's a high schooler taking university courses. Hers has a backlight and a USB-rechargeable battery, but it's basically the same machine otherwise. The downstream costs of how we do standardized tests are bonkers.
  - radicaldreamer 3y ago
    With LLMs being able to run locally on consumer hardware, I think we’re headed back to hand written tests pretty soon.
    - chrisco255 3y ago
      Yeah, even before LLMs, the more advanced TI calcs like the TI-nspire can have Calc and Physics apps loaded on them that make tests trivial to cheat on.
      - andomggazzz662 3y ago
        That's why they were banned back when I took the tests. Have they loosened the restrictions on the n-spire? The standard back in the early 2010s was still definitely TI-84 being about the most powerful calculator you could use for most tests that mattered. Those calculators are intentionally kept slow by TI because of these requirements. My impression of the n-spire was that it had always been marketed much more towards professionals. The first one I saw in person was at an electrical engineering internship where my boss had just bought one.
        chrisco255 3y ago
        No, they're still banned, but I got one early on when they were first released before they were consistently banned, and I realized shortly after that you could load some apps on there that basically allowed you to plug questions in and get the answer, since many problems are formulaic.
bob1029 3y ago
> Underscoring this momentum, Samsung and UT announced a plan to formalize a years-long partnership between the two organizations to support a semiconductor manufacturing talent development pipeline. I think this is a fantastic plan. Having an actual employer who will more-or-less take you by default is a much better value proposition for college than some other paths... The proximity of the university campus to the existing A2/S2 lines should allow for quality field trip opportunities. I know how Samsung's management feels about strangers up on the catwalk, but that was the #1 thing that inspired me - Seeing the actual starship with my own eyes.
- kurthr 3y ago
  Reliable water and electricity, as well as geology, are really important too. It's worth checking into those first two, but it's probably fine for the next 10-20 years.
  - foota 3y ago
    In Texas? The state currently dealing with a power crisis and looking to restrict the ability to build renewable energy?
    - shrubble 3y ago
      They just had a big power crisis (summer AC, maximum GWh usage) and survived.
      - dehugger 3y ago
        Is that supposed to be reassuring? Why not choose a state with better energy management, so "recently survived an energy crisis" isn't even part of the conversation?
        0xBDB 3y ago
        Because there aren't any of those. New York utilities have multiple similar events per summer, for example, as does California. https://www.caiso.com/Documents/intensifying-heat-leads-to-another-conservation-call.pdf https://www.caiso.com/Documents/intensifying-heat-leads-to-a...
mcpackieh 3y ago
"Boom on the horizon" is oddly pessimistic imagery for the message they're trying to convey.
- User23 3y ago
  Not really, hysteresis is definitely a thing for hugely expertise and capital intensive projects like ramping up bleeding edge semiconductor fabrication.
  - mcpackieh 3y ago
    I get "[economic] boom on the horizon [near future]". But the phrase "boom on the horizon" invokes imagery of mushroom clouds.
    - MiguelX413 3y ago
      Not really
cratermoon 3y ago
Wasn't this the same claim made back when officials were hyping up SEMATECH, formed in 1987? SEMATECH just kind of fizzled, moved to NY and became part of SUNY Polytechnic in 2015. I wouldn't say that was much of a 'boom'.
sacnoradhq 3y ago
Disclaimer: Austinite here unrelated to UT. UT needs a wider incubation, financing, and entrepreneurial (undergrad/grad-student/community) ecosystem like Stanford. Critical mass of players brings with it more interested parties, and with it, that elusive "innovation". It should be okay to make money, okay to turn a PhD thesis into a startup, okay to launch side projects, and cool to fail and learn lessons. Not all paths must lead to a "corporate career".
- mistymountains 3y ago
  Is this your diagnosis? That things are more corporate in terms of outcome? Genuinely curious, as a Bay Area native, why UT/Texas can’t begin to compete given how many issues face SF currently.
  - KRAKRISMOTT 3y ago
    Austin's native industry is oil and gas, not internet. The software tech sector is relatively new.
    - TexanFeller 3y ago
      Oil and gas? I haven't met a single person working in that in this town. You have us confused with Houston.
  - 15457345234 3y ago
    [dead]
oneTbrain23 3y ago
If you visit Texas universities 5 years ago, there are many China students there. Their PhD candidates are mostly non white and non Americans. Fast forward today, these have significantly drop. Foxconn had to shutdown their silicon plant because of severe constraint skilled labor. Also, China has just mass produce all the lucrative 14+ nm at a fractions pricing which have started to decimate American revenue income in that market. Already many tech workers in Cali unable to find jobs after layoff 6mths ago. Good luck if Texans able to defy all these ecobomics fundamentals and achieve boom. I know solar and Solyndra failed badly. Even Tesla energy divisions now start to fizzle out.

It's debatable if it'll be much of a boom anywhere in the US. Most uses of chips are in cheap consumer junk overwhelmingly made overseas for bottom dollar. Second after that is actual serious uses for cars, not $1 electronics and more. And once you get into that category you have a finite market that is pretty much mature in terms of consumer reach. For all intensive purposes, the "EV demand surge" is over and existing semiconductors have caught up. There's also a real fear that US sanctions have caused China and other countries to intensify investment in their own semiconductor manufacturing (the EU definitely is), which will ultimately hamper the demand for US fabs.

edgyquant 3y ago
It may be debatable but the idea the chip surge is “basically over” is laughable. By your logic no nation should ever try to do anything economically because the economy is already done forming, the end of history is here.
- baz00 3y ago
  Yup the chip surge isn't anywhere over due to geopolitics and demand, both of which are serious driving forces now.
- EricE 3y ago
  Hey - no poking holes in this line of reasoning allowed. It’s a major tent pole of the global boiling religion - pesky human innovation blows too many holes on the “we are doomed - sacrifice your rights to fix it” narratives.
  - diputsmonro 3y ago
    I'm not really sure how these are related at all. Technical innovations for electric vehicles are totally different than the technical innovations required for fighting climate change. We can be poised to make progress in one sector and skeptical about progress in another. Blindly hoping that some breakthrough will happen to fix all our problems and just barreling along carelessly is worse than what you're describing.
- kevin_thibedeau 3y ago
  The primary motive is to have redundant fabs for defense critical infrastructure that is vulnerable to Chinese attack. That's why Micron is being gifted a new fab for DRAM production on the other side of the country in the midst of ongoing layoffs. Boosting production isn't the goal and CHIPS act will be a net money loser.
- User23 3y ago
  That was the ruling ethos of the United States Government from the fall of the Berlin wall on. Look how that worked out for us.
- corethree 3y ago
  History has plenty of examples of nations and economies completely ending. In fact it's so consistent that it's basically a guarantee that the current economy you're living in will be "done" eventually. Therefore it's not a laughable conclusion but a valid possibility. We've seen two major economies descend into this "done" zone with no hope for recovery: Russia and Japan. Additionally no one is saying to "not to do anything". Try all you want he is simply predicting the outcome of trying.
justrealist 3y ago
The point is economic industrial independence for the US once Taiwan lights up. You're kind of missing the point.
- bsaul 3y ago
  not sure why you got downvoted, that seemed like a pretty reasonable thing to consider.
  - User23 3y ago
    I can assure you that the Chinese strategists are. Wang Huning is way to smart not to have and he's just one that's somehow famous enough to be known in the West. I'm sure there are many more strategists of if not equal at least similar caliber working on furthering China's (hopefully) peaceful rise.
    - corethree 3y ago
      It's too chaotic. There are many who want peace but there are many who want war and there are many who will flip their opinion and this is on both sides of the line, the US and China. Anything can occur.
      - justrealist 3y ago
        What we saw in Ukraine is that arguments about geopolitics, economy, military readiness etc can be completely negated by the person in charge acting on a strategic or emotional whim. Like all the experts in the world can concur that China is better off with peace in the Pacific, but if Xi wants a war, he's getting a war.
        User23 3y ago
        Xi already believes he has a war, it’s just in the early stages. See this[1]. [1] https://en.m.wikipedia.org/wiki/Unrestricted_Warfare https://en.m.wikipedia.org/wiki/Unrestricted_Warfare
adamc 3y ago
Just an idiom note: It's "intents and purposes" not "intensive purposes". Easy to mis-hear, but something you want to remember before it makes you look ill-educated in a work memo.
- anonuser123456 3y ago
  This really jives with me.
- retSava 3y ago
  It's also exactly the kind of humor I love. Intentional misuse of idioms, homonyms, intentional dangling modifiers, and the like. It's ok if people think I'm stupid, I can always prove them right!
  - robmsmt 3y ago
    reminds me when a friend said "escape goat" rather than "scapegoat"
nwiswell 3y ago
> Most uses of chips are in cheap consumer junk overwhelmingly made overseas for bottom dollar I don't understand your point. While true, those chips are made at historical process nodes using antiquated equipment. The plan isn't to buy antiquated equipment and try to compete. Friendly nations with low cost of labor and low geopolitical risk (ie., other than Taiwan) are perfectly capable of meeting our demands for old chips, given sufficient ramp time. The lion's share of the value in semiconductors is at the leading edge. These are the chips that are used in phones, laptops, PCs, servers, GPUs, et cetera. The United States has never been interested in competing on volume. We already are the world leader in producing (including manufacturing!) the cutting edge equipment that makes leading edge chips -- that is what we have denied China access to. The open question is whether we can "onshore" the actual manufacturing of leading edge chips again. Strictly speaking, Intel has always kept their most cutting edge manufacturing onshore (Oregon), but I understand that has more to do with IP and cycle time reasons than cost. An additional note: it is not necessary that a leading edge fab built today becomes a trailing edge fab in 5 or 10 years. There is a lively secondary market for used equipment specifically because of demand for the commodity chips you mentioned, so given that US labor costs are high, the old equipment would likely be sold and shipped to a low cost nation, and the US fabs repopulated with new equipment.
corethree 3y ago
You realize the top of the line chips are coming from Asia right now? The US right now does not have the ability to fab top of the line chips. Only Asia does. I mention this because it's basically one of the biggest hurdles for a boom other than cost which you didn't mention. The US has to exceed Asia in terms of cost AND raw technological prowess if it wants a boom because it is currently behind by a huge margin.

hooverd 3y ago

Faster calculators incoming?

MengerSponge 3y ago
I have a student who uses a TI-84. She's a high schooler taking university courses. Hers has a backlight and a USB-rechargeable battery, but it's basically the same machine otherwise. The downstream costs of how we do standardized tests are bonkers.
- radicaldreamer 3y ago
  With LLMs being able to run locally on consumer hardware, I think we’re headed back to hand written tests pretty soon.
  - chrisco255 3y ago
    Yeah, even before LLMs, the more advanced TI calcs like the TI-nspire can have Calc and Physics apps loaded on them that make tests trivial to cheat on.
    - andomggazzz662 3y ago
      That's why they were banned back when I took the tests. Have they loosened the restrictions on the n-spire? The standard back in the early 2010s was still definitely TI-84 being about the most powerful calculator you could use for most tests that mattered. Those calculators are intentionally kept slow by TI because of these requirements. My impression of the n-spire was that it had always been marketed much more towards professionals. The first one I saw in person was at an electrical engineering internship where my boss had just bought one.
      - chrisco255 3y ago
        No, they're still banned, but I got one early on when they were first released before they were consistently banned, and I realized shortly after that you could load some apps on there that basically allowed you to plug questions in and get the answer, since many problems are formulaic.

bob1029 3y ago

> Underscoring this momentum, Samsung and UT announced a plan to formalize a years-long partnership between the two organizations to support a semiconductor manufacturing talent development pipeline. I think this is a fantastic plan. Having an actual employer who will more-or-less take you by default is a much better value proposition for college than some other paths... The proximity of the university campus to the existing A2/S2 lines should allow for quality field trip opportunities. I know how Samsung's management feels about strangers up on the catwalk, but that was the #1 thing that inspired me - Seeing the actual starship with my own eyes.

kurthr 3y ago
Reliable water and electricity, as well as geology, are really important too. It's worth checking into those first two, but it's probably fine for the next 10-20 years.
- foota 3y ago
  In Texas? The state currently dealing with a power crisis and looking to restrict the ability to build renewable energy?
  - shrubble 3y ago
    They just had a big power crisis (summer AC, maximum GWh usage) and survived.
    - dehugger 3y ago
      Is that supposed to be reassuring? Why not choose a state with better energy management, so "recently survived an energy crisis" isn't even part of the conversation?
      - 0xBDB 3y ago
        Because there aren't any of those. New York utilities have multiple similar events per summer, for example, as does California. https://www.caiso.com/Documents/intensifying-heat-leads-to-another-conservation-call.pdf https://www.caiso.com/Documents/intensifying-heat-leads-to-a...

mcpackieh 3y ago

"Boom on the horizon" is oddly pessimistic imagery for the message they're trying to convey.

User23 3y ago
Not really, hysteresis is definitely a thing for hugely expertise and capital intensive projects like ramping up bleeding edge semiconductor fabrication.
- mcpackieh 3y ago
  I get "[economic] boom on the horizon [near future]". But the phrase "boom on the horizon" invokes imagery of mushroom clouds.
  - MiguelX413 3y ago
    Not really

cratermoon 3y ago

Wasn't this the same claim made back when officials were hyping up SEMATECH, formed in 1987? SEMATECH just kind of fizzled, moved to NY and became part of SUNY Polytechnic in 2015. I wouldn't say that was much of a 'boom'.

sacnoradhq 3y ago

Disclaimer: Austinite here unrelated to UT. UT needs a wider incubation, financing, and entrepreneurial (undergrad/grad-student/community) ecosystem like Stanford. Critical mass of players brings with it more interested parties, and with it, that elusive "innovation". It should be okay to make money, okay to turn a PhD thesis into a startup, okay to launch side projects, and cool to fail and learn lessons. Not all paths must lead to a "corporate career".

mistymountains 3y ago
Is this your diagnosis? That things are more corporate in terms of outcome? Genuinely curious, as a Bay Area native, why UT/Texas can’t begin to compete given how many issues face SF currently.
- KRAKRISMOTT 3y ago
  Austin's native industry is oil and gas, not internet. The software tech sector is relatively new.
  - TexanFeller 3y ago
    Oil and gas? I haven't met a single person working in that in this town. You have us confused with Houston.
- 15457345234 3y ago
  [dead]

oneTbrain23 3y ago

If you visit Texas universities 5 years ago, there are many China students there. Their PhD candidates are mostly non white and non Americans. Fast forward today, these have significantly drop. Foxconn had to shutdown their silicon plant because of severe constraint skilled labor. Also, China has just mass produce all the lucrative 14+ nm at a fractions pricing which have started to decimate American revenue income in that market. Already many tech workers in Cali unable to find jobs after layoff 6mths ago. Good luck if Texans able to defy all these ecobomics fundamentals and achieve boom. I know solar and Solyndra failed badly. Even Tesla energy divisions now start to fizzle out.