Great news for the US. Great news for Apple, AMD, and Nvidia.
Not so good news for Taiwan. The fact that the whole tech world is dependent on chips produced in Hsinchu is a huge advantage for the safety of Taiwan. Moving the fabs and talent further away from the island will not benefit the people of Taiwan at all. TSMC with its global influence has been a huge factor for guaranteeing safety of the island and peace in the Taiwan strait. On the other hand, TSMC is a corporation like every other and does what's best for their business. Most likely it's a huge win for everyone holding their stock. Are we going to see fabs in Central/Eastern Europe next? I'd hope so.
By the time the fab comes online in 2024 (which seems optimistic), 4nm should be a half-step generation or so behind.
TSMC also has a dozen or so fabs in Asia, so it's not clear how the single plant in Arizona is going to meet capacity requirements or how often that plant will be retrofitted with newer equipment.
This plant and the federal subsidies backing it seem more like a way for defense contractors to domestically source relatively recent fab processes over the next couple decades rather than something intended solely for consumer products.
The fact that consumer facing companies are interested in using the fab when it comes online doesn't necessarily mean they'll still be using it 10 years from now unless TSMC keeps it up to date.
This isn't my field, but I'm not sure what's the excitement about an Arizona plant that will optimistically start production a year later with an older process.
There is still a lot of demand for the not-latest-generation, "long tail" fab processes. Much of the long tail manufacturing was (is?) what the chip shortage was about, rather than the current generation.
A good example is the automotive industry. It doesn't typically move quickly or frequently onto newer generation processes as that requires R&D time and other expenses. However, these have been some of the worst shortages, with auto manufacturers still impacted by lead times (although, this is starting to clearing up fwiu).
Targeting the current-best process, rather than the next generation, alleviates some compounding of risk that would be incurred by logistics concerns of turning up a new site and putting that new site on a process that they don't yet have full confidence in because it's not seen production yet.
Does "one generation before current" count as "long tail", though?
The chip shortage in products like cars has been for much simpler chips, AFAIK.
The article says Apple, AMD, and Nvidia are looking to source from the new plant. Aren't they canonical examples of companies that are always looking for the latest and greatest? Why would they be interested in an prev-gen chip?
Automotive industry is still using 90nm chips. It's not even the generation before the current one. They have long production runs and require stability, and they also prefer to use standardized parts across many production runs.
Defense also uses older chips. Don't ask what kind of chips are in Amraam missiles or the F-35 -- although the F-35 is getting a technology refresh now (after a 14 year production run).
Only in bleeding edge consumer devices does it make sense to keep changing chips. In other systems, production runs can last 20 years and the life of the asset can be 40 years or more, and you want the same spare parts available throughout the entire expected life of all assets produced. And then when you look at fixed assets, such as thermal power stations, then you are looking at even longer time horizons.
That's what I thought. If consumer products tend to use the bleeding edge, which will be the state-of-the-art chips produced in Taiwan, then I still don't get:
1. What's so exciting about a prev-gen fab possibly being completed in 2024?
2. Who will buy the chips produced by the 4nm Arizona fab, and why?
The only plausible answer offered for 1 in this thread has been that it's too difficult to jump straight into the bleeding edge, and this is the most the US can do to lay the foundation for an eventual catch-up with the state of the art in chip fabrication.
If mission critical applications (military, automotive, aerospace) are using older chips, and smartphones/computers are using the latest, then everything else is using chips between those two extremes... computer peripherals, office machines, toys, audio/video systems, communication equipment, lights, HVAC, anything rechargeable. The list is massive, you name an industry and they will probably be buying 4nm over the next couple decades.
Because overall the us chip production capacity and technical reach has been reducing over time. We need to establish and pull over a much larger amount of production so our strategic needs can be met inside the us. This helps us avoid problems like our key sources being cut off or destroyed with our allies like Taiwan and South Korea, or China does more and more production and technical leadership and decides to squeeze us.
Intel has sinking in at least technical leadership for a long time. It's obviously a priority a need for the us to on-shore more chip prod.
Lower end consumer electronics are fine with chips that aren't the absolute latest. The point isn't that they won't find buyers, but rather that they will earn less money building a fab in a location with much higher labor costs and producing chips that are a bit behind.
My guess is that the decision to build fabs in the U.S. was the result of geopolitical strong-arming by the U.S., but TSMC will still find buyers and that the fab will at least pay for itself.
Starting in 2022, the high-volume iPhone has been switched to the N-1 generation SoC, while the bleeding edge SoC is reserved for the lower volume iPhone Pro/Max.
Many companies (including NVidia and AMD) are adopting chiplet designs for which different parts of the chip can be manufactured on different process nodes. The most performance critical aspects of chips will be manufactured on 3 nm nodes, but most of the chip will be manufactured on 4-10 nm nodes. It's too expensive (in cost/wafer) to use 3 nm for everything.
> The article says Apple, AMD, and Nvidia are looking to source from the new plant. Aren't they canonical examples of companies that are always looking for the latest and greatest?
CPUs and GPUs get all the press, but there are dozens of other chips used in even the latest models that use older technology. For all we know Apple could be buying those chips for USB cables, or pencils.
You don't start at the cutting edge immediately. You start where it is better understood. Then you can still pay the enormous investments needed for getting to cutting edge and furthering it, if you want. Also, from an economic perspective, there is still plenty of demand in the market for non-cutting edge node sizes, i.e. in embedded the priority is more around the fact that the chip is not changed from a specific design that was made 4 years ago than having the latest and greatest.
From a strategic perspective, it is absolutely important that you have the capability to build chips in your country, of a reasonably modern node size. You need this for weapons manufacturing, for a working government apparatus (governments use computers now), for sending messages to your population. And if China bombs Taiwan, then this fab will become the cutting edge, instead of you having zero chip manufacturing capabilities.
> TSMC with its global influence has been a huge factor for guaranteeing safety of the island and peace in the Taiwan strait.
This narrative is echoed around the internet, but if you study the actual history of the Taiwan Strait Crises (which started before the semiconductor was invented), it never comes up in official discussion or analysis. See Kissinger, for example.
Cynically, this narrative was possibly promoted by the USG/Taiwan-lobby to put lipstick on what has always been a naked display of power politics. Washington wants its First Island Chain to contain China [1], and Beijing doesn't want Washington to have it.
> Cynically, this narrative was possibly promoted by the USG/Taiwan-lobby to put lipstick on what has always been a naked display of power politics.
I also suspect this narrative is at the very least supported if not conceived by Taiwan as propaganda. But we don't need to be cynical. What this narrative does is establish in the minds of Americans a shared interest in Taiwan and, arguably, even a shared identity. That's not intrinsically bad, malicious, nor even disingenuous on the part of the Taiwanese. Especially for a country as large and resource rich (in every meaning of the term) as the United States, all overseas interests and identities are built principally on fictions; some built very deliberately, but many which grew organically. IMO, the high technology dependency narrative fits comfortably between pure ideology (democratic solidarity!) and pure real politick, e.g. oil. Unlike the case with oil interests, the narrative speaks to a coevolution of our industrial and economic bases on equal terms in tandem with political ideology, and thus posits a shared future. And given how quickly and easily the narrative has spread, it has a strong organic character to it--even if Taiwanese political strategists planted the seeds, the soil was more than accommodating.
Such a crafted narrative to me seems more like an invitation than manipulation. (Either way, admittedly such characterizations are dependent on one's chosen perspective.) Nonetheless, it should be recognized for what it is; taken literally it leads to erroneous conclusions. If the U.S. aggressively defends Taiwan in an invasion attempt, it won't be because of TSMC and fears of a chip shortage; it'll be because the American public has become invested in the idea of a free and democratic Taiwan, and willing to believe and accept that the US's long-term self-interests are furthered by putting itself and its citizens in the way of considerable, even existential harm. The story of TSMC would just be one of many--albeit an important one--along the road which brought the nation to that state of mind.
> If the U.S. aggressively defends Taiwan in an invasion attempt, it won't be because of TSMC and fears of a chip shortage; it'll be because the American public has become invested in the idea of a free and democratic Taiwan,
That, and the concern about China’s long-term ambitions, and what sort of precedent letting them take Taiwan unchecked would set.
However for this, japan and s Korea would be in play. They have to submit to the power that control the sea or 1/2 of Pacific Ocean down to Australia (with a base in South Pacific say).
I was very surprised to find in the recent New Yorker long article about Chinese-Taiwanese Relations [1] that TSMC was only mentioned at the very end. I was expecting this to be central to the story. I recommend reading the whole thing to anyone who has an hour to spare.
"One of the most important deterrents to war is Taiwan’s role in producing semiconductors. Seventy per cent of the world’s most advanced chips are manufactured there, many of them at the Taiwanese Semiconductor Manufacturing Company. [...]China is similarly reliant on the highest-end chips produced in Taiwan; it doesn’t have the equipment or the expertise to manufacture them. If China seized control of Taiwan’s semiconductor factories, it could conceivably force local workers to run them. But the factories depend on a constant flow of Western material, software, expertise, and engineers, without which production would cease in a matter of weeks. Pottinger told me, “If the Chinese took the factories, there’s no way the West would help run them.”"
The real reason we care so much... just look into how power is created by international laws allow landmasses to project power X number of miles out into the ocean. That control plays right into the most influential lanes of commerce in the world.
Sometimes, real life feels like a game of Command and Conquer or Risk, when you zoom back out and boil things down to their most simple form.
If anything the semiconductor industry raises the stakes for the US. The vast majority of chips are manufactured in Taiwan, China, and South Korea. A China that can capture Taiwan can also blockade South Korea, and therefore put the entire world economy at its mercy.
I don't see anything changing yet in terms of Global Influence. TSMC will still have their leading edge, state of art, and majority of capacity in Taiwan for the foreseeable future.
The 4nm US Fab, based on N5, will be two years behind in 2024, where TSMC will be producing N3 class in volume and N2 in 2025. Given the N3 Class are long node before moving to something more exotic and expensive N2 with GAFFET. I would expect the US Fab to be upgraded and start producing N3 in 2026 on US Soil.
>Great news for the US. Great news for Apple, AMD, and Nvidia. Not so good news for Taiwan. The fact that the whole tech world is dependent on chips produced in Hsinchu is a huge advantage for the safety of Taiwan. Moving the fabs and talent further away from the island will not benefit the people of Taiwan at all.
Naive question, can we relocate the people of Taiwan?
Think about that for a second: why would they want to be relocated? And when is the last time you can remember someone deciding "let's relocate these other people that I'm not part of" and that being okay? I mean, one of these relocations is literally called "march of tears".
35 million or so Latin Americans have relocated themselves to the US, even piled into dark crowded trucks and given up their life savings to dangerous coyotes, to do so. I would imagine the the specter of Communist China knocking on your door and breathing down your neck would be a motivator for many Taiwanese. And there would be a great opportunity for the US to welcome them with open arms as allies
They aren't currently living that life so not that many people would want to leave, and leading an exodus of a couple percentage of their population would only hurt their position even more. Most people just rarely ever want to leave their homes until after a disaster has been realized. There is risks anywhere you live, they know their fight, and they are currently making a decent life. Rolling the dice somewhere else just wouldn't be that attractive of a position when it could easily be an immediately worse situation.
You're assuming the population of Taiwan views itself in the same position vis a vis CCP as the government of Taiwan does, which is not a historically probable claim, for all that the situation is quite complex.
Taiwan is a legitimate and robust representative democracy and can quite simply vote itself right into joining with the mainland. The fact that it’s not doing just that tells us that there isn’t a basis to claim that the wishes of populace are not taken into an account by the government.
(And they have in fact just voted in a more pro-China government, so the comment may have a short shelf life).
And not speaking to you directly, but contributing to the general discussion - US opposes China taking over Taiwan using force specifically. If Taiwan were to willingly join with China, one can imagine some conspiracy scenarios, but setting conspiracies aside, there isn’t anything anyone can and will do to prevent it.
A better example might be the wave of Vietnamese refugees who fled the country as the communists conquered the south. Still, being an ally isn’t about accepting a country’s refugees after they’ve been overrun by the enemy. It’s about helping them not be overrun in the first place.
Humans have strong ties to their land. Try getting a sick American rancher to go to the big city for treatment. The people of Taiwan will fight to the death or at least to the point that subjugation is inevitable. And then there will be a resistance.
They're not closing shop in Taiwan, rather manufacturers are shifting away from China and the US government is probably encouraging/subsidizing new plants there. So this is overall a good healthy development that should have happened many years ago.
Doubtful. This is obviously in exchange for continued military support from America. Protection from the world's greatest Navy is well worth a single chip plant.
My initial thoughts on this came the from the opposite side in terms of Taiwans security.
One of the major bounties for invading Taiwan would be in the acquisition and control of their chip manufacturing. Which - as you mention, the entire tech world is dependant on.
By moving some of that capability offshore, that incentive is gone.
> By moving some of that capability offshore, that incentive is gone.
China's main motivation isn't about controlling TSMC, that's just a useful side-objective if it comes to pass. Their main motivations are:
1) break the first island chain barrier and gain a naval base with unhindered access to the Pacific, and
2) shut down a high-functioning Chinese democracy that is a constant reminder to the people of mainland China that democracy works for Chinese people and that they don't actually need the CCP.
These are also the reasons the US and Taiwan's other allies like Japan will continue to defend the country even if it moves some chip production to safer locales.
Perhaps it isn't about controlling TSMC, but control of TSMC would be a very strong leverage over the west.
Not that it is the only or even the most powerful leverage China has over the west, but it is still a massive one, and it has quite a lot of second-order effects.
> but control of TSMC would be a very strong leverage over the west.
They'll never control TSMC by invading Taiwan. TSMC will sabotage their equipment if it looks like a CCP invasion is about to succeed. CCP knows that, it's not their main reason for wanting Taiwan.
China wants to focus more of their economy on internal demand. Something like TSMC, which would allow them to fab CPUs and own more of the value chain in things like laptops and cellphones would help them reach that goal.
Everyone is aware that this would be scorched earth, right? If the fabs somehow survive the initial wave of strategic bombing both the US and Taiwan have an interest in preventing them from falling into enemy hands. In addition, the US will then place China under a trade embargo. And that's assuming that the US doesn't actively engage PRC forces.
To say this would make the world worse off is a drastic understatement.
China absolutely wants to be reunited with Taiwan. Taiwan is a constant reminder of the century of humiliation (Translated term that is basically China's term for Opium wars to WWII). Taiwan is an integral part of the Chinese nation and for nationalistic reasons, it would be an issue even if Taiwan had no other benefits.
> Taiwan is an integral part of the Chinese nation
No it isn't. China under the Communist Party has never ruled or controlled Taiwan (since 1949). Taiwan democratized, developed, and got wealthy first, completely independently of China. Taiwan has never in modern history been an integral part of China.
There's a difference between a nation and a state, although the two are nearly always synonymous in the modern world.
Taiwan and China being one nation is the policy of the governments of both China and Taiwan.
Of course Taiwanese nationalism is it's own thing now, but the Taiwanese people seeing themselves as not Chinese is a relatively recent phenomenon - it's been functionally independent for less than 100 years, and before that it was a Japanese colony like several others that have since been reabsorbed by China.
I'm not supporting Chineses irredentism, and I don't think the parent comment was either. Taiwan should remain independent. What the parent was explaining is why China would want Taiwan no matter what - it's a historical part of China that is relatively recently separated, so they want it for purely nationalistic reasons. It's no different from Serbia and Kosovo, or Russia and swaths of Ukraine. The economics don't matter if all you care about is your wounded national pride.
> Taiwan and China being one nation is the policy of the governments of both China and Taiwan.
No one in Taiwan and not even the current government consider China/Taiwan together. When KMT fled and created the constitution they claim China. Taiwan is now stuck in limbo because the people just want to live their lives in peace and already consider themselves Taiwanese and independent. But if they change the constitution then China will consider it a formal act of independence and use it as an excuse.
You're speaking to the "state" part of the above comment. Yes, clearly the PRC and ROC are not the same state. "Nation" is not necessarily the same thing.
I’m saying despite the constitution. People in Taiwan don’t claim China to belong to them. This is a relic of the past that is stuck in writing that people suffer from.
The pro-independence party just suffered a major loss in local elections. The prime-minister who invited Nancy Pelosi for a provocative visit had to resign. Looks like Taiwanese people are not very enthusiastic about becoming a new Ukraine.
Greens never do well in local elections. Local governments have no say at national level. It seems the only people who say they suffered a major blow are pro CCP bots.
For clarification: It was the President of Taiwan Tsai Ing-wen(as far as I can tell) who invited Nancy Pelosi for a visit. She (the President of Taiwan) did resign as head of her party, but not as president.
China has been a country for longer than the history of European civilization. They do not subscribe to the view that nothing that happened before 1945 matters the way Americans seem to.
China also makes the claim of 1000s of years of history but Taiwans inclusion in that history is about 200 years. And of that less than 10 was a province that China still didn’t govern or control. It was more or less just something they said to deter Japan. So historically China has never ruled over Taiwan. Japan has more claim to Taiwan than China as it actually ruled and controlled Taiwan.
No, China has not been a country for longer than European civilization. That CCP propaganda claim is the equivalent of saying that the European Union has been around for 8000 years because Plovdiv, Bulgaria was founded in the 6th Millennium BC.
Without the staff from TSMC, the Chinese Government probably can't run those facilities. I suspect there isn't much probability that china could mobilize and take those facilities as-is with no sabotage.
ATM PRC assaulting Taiwan would hobble western defense industry. By moving more manufacturing into NA & EU that alleviates it. It helps Taiwan as well who depends on western defense industry.
The Chinese government doesn't run those facilities now, and I agree that they won't be able to do so in the event of a Taiwan invasion either.
That's not the point though. Who runs those facilities now? TSMC aka Taiwan aka a western ally. Who stands to lose the most from TSMC facilities being burned to the ground? Taiwan (obviously) and the west.
As a layman with no background in international relations, to me it seems like TSMC is simply an extra bargaining chip for the Chinese government. Which is why I am all about the idea of building more TSMC facilities in places that are less susceptible to being invaded. And yes, the Arizona plant is just a drop in the bucket compared to their facilities in Taiwan, but you gotta start somewhere, and something is better than nothing in this case imo.
The Chinese government also wouldn't be able to run those fabs without ongoing support from ASML and other key foreign vendors. The production machinery is extremely complex with many specialized parts and a significant software component. Reverse engineering and duplicating everything would take years.
It seems like there are a few key parts that could be removed that would make those ASML machines utterly useless even if Chinese engineers spent time reverse engineering the rest. Surely there is no need to remove or destroy the whole thing as is suggested in other comments.
The equivalent of popping out the Intel/AMD/ARM CPU to disable a computer, and leaving the motherboard, RAM etc.
I've read that TSMC fabs are filled with explosives to set off if China invades so they wouldn't get their hands on tech and equipment. I'm pretty certain that's true, given how important TSMC tech is.
The difference is both sides want that gas to keep flowing (for the time being at least).
Both sides don't want China to take over the technology to produce state of the art chips and control the distribution of those chips. And in case of a war, no way is china continuing to sell those chips to the US defense dept. The profit is negligible compared to their strategic value. Different goals will produce a different result.
That meme makes zero sense considering TW don't want to be a third world economy dependant on exporting fruit even if PRC successfully invades. There's reason TW media was telling US think tanks to leave TSMC alone when US Army War College analysis suggested US should consider bombing TSMC... or exfiltrate TSMC engineers (before children no less) in event of war. As long as fabs and downstream supply chain supplies said fabs are intact, the island will have leverage to remain viable modern economy to support relatively affluent lifestyle. Note the point on downstream supply chain, there's sufficiently exclusive niche semi industries sustaining TSMC on TW that makes it as critical as ASML. Don't expect any Arizona TSMC fabs to operate smoothly without them. If anything, expect TSMC and TW + PRC to collude to threaten TSMC US fabs if try to sanction TSMC TW from making chips in event of successful PRC takeover. The people whose making bank off TSMC will want to so regardless of who rules the island. Ultimately short/medium term also in US interest to keep fabs going because not enough fab capacity will be reshored off island for long time, and the interest groups hurt most is US high tech industry who extracts disproportionate value add from TW fabs. Imaging every company that depends on leading edge chips turning into Huawei/ZTE because 95% of production goes kaput. Currently, cratered TSMC fabs actually works in sanctioned PRC's favour because it dramatically closes relative gap of who has access to high end semi. PRC vastly better off in balance where everyone is mostly stuck on 28nm+ instead of one where US has unfettered access to leading edge.
US military doesn't care. But the TWnese care, which US thinktankers/media, and I'm guessing US based commenters like you seem to forget. Hence the disconnect on why people seriously contemplate these TW will blow up TSMC memes. And why TW media reminding US, that if they're going to evacuate anyone off the island first, it's not going to be their semi engineers, it's going to be women and children. Or that more generally, they're not interested in blowing up their lively hood to stick it to the PRC. Like how in UKR war, it's RU whose blowing up UKR infra and industry when they decided it was better to scorch earth long term.
Yes, Taiwanese obviously care more about their children. But that doesn't change that the US military won't allow the Chinese to obtain TSMC or the knowledge of their engineers.
Sure, except original comment also highlights that the incentives of destroying TSMC is backwards. It's PRC who benefits most from denying US access to TW semi supply chains or engineers not vice versa. Denying TW to US closes relative semi gap for PRC, leveraging PRC control of TSMC in case of successfuly invasion to compel US to lift sanctions also closes relative semi gap for PRC. US has leverage via sanctions during peacetime, but PRC has leverage via threatening access or destruction of east asian semi supply chain during war. Utlimately it's in both US and TW interests for TSMC + co to survive because they extract most value / benefit, but not necessarily for PRC. And for TW, ensuring TSMC+co survival =/= paperclipping them to US. All the interest calculates points towards PRC/TW denying US access, and US wanting continued access since US fabs will still be dependant on TW inputs as much as current TW fabs or future PRC controlled TW fabs will be dependant on US/EU/JP inputs.
But this is missing the obvious. Yes the US would prefer being able to maintain its tech advantage over China by continuing production and receiving of state of the art chips. And this is clearly better for them than choosing that gap by destroying TSMC fab. But if China takes over TW, that's not an option. The options are either flatten the gap or flip the gap in China's favor as it now controls those chips while the US falls back to tech multiple generations old. There is no other strategic option for the US flatten the gap is the only choice, flipping the gap is the worst case scenario for the us.
This deal is in US and TW best interest. TW is still valuable as cutting edge is still made on island. So production and economics so benefit them. US still provides military protection / defense. They are still long term partners will aligned goals. And if China does invade TW, US can continue a long fight and win by falling back only a generation or two to is smaller but important domestic production to continue supporting its defense technology needs.
I can't speak for them, but I have to assume RW also sees its best interests if China invades to lose TSMC plants but US win the war and they maintain democracy, rather than keep TSMC plants but controlled by China and US lose the war and go under full control of mainland rule.
Plants can be rebuilt, just like Marshall plan or what will happen in Ukraine. TW wants long term freedom from China - strengthening US defense's tech position helps this the most.
The obvious option in event of PRC control is everyone deferring to mutual leverage, where US controls fab inputs, TW controls outputs, PRC negotiating access to % of high end production by controlling island. Meanwhile buys both US and PRC time to secure independant semi chain, TW gets to still profit off semi, much like pre techwar arrangment. That's the strategically rare win-win-win scenario where everyone gains / avoids loss vs having accessible leading edge setback 20 years. Of course PRC wins most in this arrangment, and US/TW relatively loses, but if PRC wins in TW scenario vs US, that's going to be least of worries / concessions. The least obvious option is US deciding to scorch earth tech that supports their most competitive high tech industries, responsible for huge % of competitive advantage, especially if that leaves PRC the biggest producer of mature nodes (where things are trending). Even less sensible considering long term timeline after which experts believes PRC can reach semblance of semi parity.
This is short/medium term balance / interest calculation irrespective of the deal - which I didn't comment on - substantial capacity won't shift off TW in the timelines US forecast PRC will make move. Leading edge isn't particularly relelvant to defense hardware ends which use mature nodes, nor does anyone project long fight with how irreplaceable modern platforms are. It's why PRC (and chip partners US is trying to coerce into export controls) views Oct chip curbs as attack on their commercial sector, not military since PRC also have small batch 7nm production capability.
> TW also sees its best interests
It's also obvious "best" TW interest is to hedge having continued leverage over being leading ledge supplier regardless of who wins. There is no scenario where giving away their silicon dominance makes sense, unless coerced to, which is why US had to unilaterally announce export controls - because CHIPS4 partners weren't biting and actively pushing back. Least of all because no one is sure TW can win local war against PRC. Wealthy TW industrialist/elites/chip talent would rather be wealthy under PRC control after war then be fruit farmers because their industry got glassed. Read between line of Morris Chang stating US efforts to reshore chips manufacturing doomed to fail - it's not commentary on US talent - but TW isn't ceding industry to be Americanized. IMO don't expect these plants to launch without hitch. Prolonged war is probably least in TW interests because island simply can't sustain past a few months on domestic resources. There's reason TW defense posture is not prepping for prolonged war like US wants, deep down leadership and industry knows best interest for TW is to survive short/sharp war as developed economy and work with whoever wins.
> strengthening US defense's tech position
Except, again, TW offshoring leading edge fabs doesn't substantively strengthen US defense indy that depend on mature nodes with respect to PRC war. As stated, US fabs will STILL depend on TW inputs for forseeable future - it's not just ASML that's bottle neck - niche suppliers and expertise will be stuck on TW long term. Also consider "partners" pushing back on unilateral US curbs or SKR, JP building out their own indigenous defense industry - very few are actually interested in giving/strengthining US further leverage especially at their cost. And let's be real, fabs aren't going to be rebuilt on TW if they get leveled when they can be reconstructed in more secure locations.
> By moving some of that capability offshore, that incentive is gone.
I'm not sure if that incentive was significant enough. In my eyes, PRC's interest in Taiwan is at best orthogonal with TSMC's manufacturing capabilities.
I thought the same thing. I believe it is more nuanced than that.
PRC wants Taiwan regardless of its production value (geopolitical & PRC narrative) though I'm sure they would like the control of chip manufacturing. However if the US did move all of its strategic production off island there would be less value accrued to defending outside that said keeping a close presence on China expansion is important to the US. So it would still have value maybe a bit less so.
Please excuse the human aspect of the population as we are separating that part of the discussion.
More bleakly, if TSMC is destroyed as a result of invasion, the rest of the world is denied access. This might be to their liking. With some of the capability in the US, they can no longer deny the rest of the world.
They don't really have a choice. Either they build fabs in other countries or those countries will simply invest/subsidize their own fabs (including Intel) to avoid a single point of failure for chip supply.
>Great news for the US. Great news for Apple, AMD, and Nvidia. Not so good news for Taiwan. The fact that the whole tech world is dependent on chips produced in Hsinchu is a huge advantage for the safety of Taiwan. Moving the fabs and talent further away from the island will not benefit the people of Taiwan at all.
Is it possible part of why China wants control of China is to control these factories, so if they move production to America paired with demanding existing diplomatic agreements be honored, China will be less obsessed with seizing, for lack of better phrasing, a unique resource?
The CCP’s intensifying rhetoric about unification paired with Taiwan’s low level of military readiness is somewhat troubling and I’m not sure how much TSMC is going to help matters on its own. A lot of Taiwanese elites have vested interests on the mainland, and CCP subversion and espionage are reportedly rampant. At some point Taiwan is going to have to be willing to actually defend their sovereignty.
Does China still want Taiwan if it's no longer technologically relevant? My understanding is that China is making similar strides in its own chip industry to the point where TSMC won't be as nearly as useful to them in 10 years as it is now.
Taiwan also serves the role as an unsinkable battleship right next door to Chinas mainland. Its military value extends far beyond its chip production capabilities in order for the US to check Chinas influence in SEA region.
The U.S. military doesn't stage any assets in Taiwan, not since shortly after rapprochement with China and establishment of the current strategic ambiguity. The U.S. military is very careful about both what and who it officially permits to land on Taiwan. Occasionally (on the order of years) there are borderline cases, such as a research vessel with U.S. Navy ties docking in Taiwan, and it becomes a huge thing in the Chinese media. AFAIU, high-level officers are rarely if ever given permission to enter Taiwan; direct military liaisons in Taiwan are limited to lower level staff officers. When this protocol is broken it's a tit-for-tat situation designed to send a message, and doesn't change anything of substance in how the relationship operates.
Anyhow, the U.S. has no need nor desire to establish a presence on Taiwan. Okinawa and Korea are more than close enough for its purposes, peaceful or otherwise. And if the U.S. were to try to establish a presence, you can be sure China's invasion fleet would reach Taiwan before any significant U.S. materiel could make it ashore.
Taiwan does far more than just TSMC. That is just one single company, albeit a very essential one. For instance, I don't like to buy hand tools (impact sockets, torque wrenches, etc) made in China, but Taiwanese tools are a higher quality and I have no hesitation purchasing them.
I suspect that the hand tools made in China aren't low quality because China doesn't have the tech to make them better, but instead because making cheap low quality tools is more profitable.
In fact, when taking apart China-made, China-designed products, I am frequently very impressed at cost-cutting measures that are taken with minimal impact on the functionality of the product.
Many computer/electrics manufacturers and OEMs, like ASUS, Acer, Foxconn, Wistron, Pegatron, MediaTek, Synology, etc are HQ'd in Taiwan. Though PRC company may be able to supersede them, unlike TSMC.
> Essential for you. But overall tsmc eclipses everything else Taiwan makes.
TSMC’s revenue is less than a tenth of Taiwan’s total export revenue. They might be the most profitable of Taiwan’s companies, but Taiwan has a huge economy.
And it terms of impact on the world economy, Hon Hai precision, better known in the west as Foxconn, would like to say hi.
China wants Taiwan orthgonally to any economic value it would provide. Reunification is pretty core to their mythos at this point.
In fact, I wouldn't be surprised if making Taiwan globally less economically relevant is core to their strategy for reunification. While the US is dependent on Taiwan for chips, Taiwan has a very important defense partner that China can't really compete with. But China has a way better chance of securing Taiwan if the rest of the global community only cares for ideological reasons.
Technically, since Taiwan isn’t recognised as a country they can’t be a signatory of the nuclear non-proliferation treaty. They only thing stopping them developing nuclear weapons is political pressure.
the issue is "is it healthy for the country" which being taken over by china is not, so you aren't really addressing either the question I was responding to, or my question.
Well that's debatable. It depends entirely on how you define "healthy". We tend to look at things from a Western perspective, but that's rather presumptive isn't it?
What's presumptive is to call something debateable without making a persuasive or even barely logical argument, throwing in a random jab at perspective without offering any opinion to talk about, and then calling your opponent presumptive.
Basically you have no argument and no position except to be rude.
> WASHINGTON, March 11 (Reuters) - Ukraine's two leading suppliers of neon, which produce about half the world's supply of the key ingredient for making chips, have halted their operations as Moscow has sharpened its attack on the country, threatening to raise prices and aggravate the semiconductor shortage.
Every Russian soldier, vehicle, or round of ammunition expended in Ukraine for increasingly diminishing returns is a Russian soldier, vehicle, or round of ammunition that will never harm a NATO member.
> Every Russian soldier, vehicle, or round of ammunition expended in Ukraine for increasingly diminishing returns is a Russian soldier, vehicle, or round of ammunition that will never harm a NATO member.
Sure, and it's also 10 dead Ukrainian soldiers that will never return to their families, and a few thousand people living an extra day in war, and a few hundred more Ukrainian refugees that must be subsidized in Europe, a few dead Ukrainian civilians that die of Russian shelling, and a few dead Ukrainian civilians in Donbas that die of Ukrainian shelling, and the further demilitarization of both NATO and the US as they run out of munitions and equipment to send (and lack the industrial base to make up for the shortfall). It is also a few more percent lopped off of European GDP as the continent plunges into a steep recession due to the energy crisis.
It is only by weighing of the lives of the people in the conflict at zero that the U.S. justifies prolonging this proxy war and continuing to arm and subsidize Ukraine.
It is truly willing to fight Russia until the last Ukrainian, and until Europe is completely de-industrialized.
> It is only by weighing of the lives of the people in the conflict at zero that the U.S. justifies prolonging this proxy war and continuing to arm and subsidize Ukraine.
Ukraine is literally fighting for their sovereignty, America is helping them push the Russians, who are raping and torturing citizens out of the occupied territories.
Every weapon to Ukraine is time less under Russian occupation, which is a good thing.
You seriously believe that the US or EU care about the sovereignty of other nations? America has invaded or attacked over 120 nations since 1991. We currently have troops in Syria extracting oil and are still occupying Iraq. We just helped organize another invasion of Haiti, after murdering their President and replacing them with a US puppet. We just tried to overthrew the government of Ethopia by backing the Tigray rebels. We overthrew the government of Ukraine twice, in 2004 and 2013. We overthrew the government of Iraq and are financing violent rebels to overthrow the government of Myanmar. Only strong, large nations that can stand up to the US and other rival great powers have any hope of achieving sovereignty. Ukraine has never had it.
So please stop with this new found interest in the "sovereignty" of US client states. We have tape recordings of US state department officials deciding who will be in the Ukrainian cabinet and who wont, of what their policies can and can't be. One phone call from Washington and there is a third coup in Ukraine with Zelensky out the door and another US puppet installed. So Ukraine has zero sovereignty, and that's true regardless of what happens in this war. The only question is whether it will continue to be ruled from Washington or whether it will be ruled from Moscow. That is why this is a US-Russia proxy war, with Ukraine stuck in the middle.
> We have tape recordings of US state department officials deciding who will be in the Ukrainian cabinet and who wont. of what their policies can and can't be.
If you could provide a source for this, that’d great.
> One phone call from Washington and there is a third coup in Ukraine with Zelensky out the door and another US puppet installed.
Zelensky was elected in an election, I know you likely don’t know this so I just thought you should know.
> So Ukraine has zero sovereignty, and that's true regardless of what happens in this war.
It’s actually not true and patently false.
> The only question is whether it will continue to be ruled from Washington or whether it will be ruled from Moscow.
Even if these were the options (and they aren’t cause Americas not looking to rule Ukraine). The choice is clear, being ruled by Moscow which leads to the rape and torture of your civilians(including children), or being ruled by America which leads to not that.
> That is why this is a US-Russia proxy war, with Ukraine stuck in the middle.
Russia keeps changing the reasons they invaded the current reasons have nothing to with America but I’m sure they will change next week.
> Zelensky was elected in an election, I know you likely don’t know this so I just thought you should know.
In an environment in which rival political parties were banned, torture centers were established, death squads were roaming the streets arresting dissidents, etc, and there was strict control of media.
But in some sense elections in client states don't matter. Zelensky was elected on a peace platform and immediately was forced to abandon his stance on implementing Minsk 2 when he got into power - 70% of the population wanted it implemented. Minsk 2 was the agreement Ukraine signed with Russia to avoid war. Minsk 2, which was brokered by the EU after the violent anti-Russian Maidan coup. It called on Ukraine to grant autonomy to Donbas and to respect the rights of the Russians living there. It also called on Ukraine to recognize Russian control over Crimea. In return, the Donbas provinces would return to Ukraine and Russia would recognize the coup government, and there would be no war. Poroshenko stated he had no intention of ever implementing Minsk 2, but signed it to buy time for Ukraine to arm itself to fight a war with Russia. US Senators McCain and Lindsey Graham visited Ukraine and urged them to defeat Russia in the war. This was 2015. The US was clear that Minsk 2 would never be implemented by Ukraine and that it was just a lie to buy time as they geared up for war.
So in a place like Ukraine, it doesn't matter who is elected as long as the US controls the country. Zelensky was elected on a peace platform and pro-Minsk 2 platform, and immediately changed his position after the election. This is something you should also research if you want to be informed.
> The choice is clear, being ruled by Moscow which leads to the rape and torture of your civilians(including children), or being ruled by America which leads to not that.
No, it is the Maidan regime that set up SBU torture centers, burned people alive, buried people alive, funded various Neo-Nazi groups like Aidar, Right Sektor, Azov, banned the use of Russian language, disappeared pro-Russian citizens, banned pro-Russian parties, banned pro-Hungarian parties, etc, and they were committing and continue to commit mass atrocities. There were many articles by Amnesty International and other human rights groups about this.
> Russia keeps changing the reasons they invaded the current reasons have nothing to with America but I’m sure they will change next week.
This is just made up nonsense. Putin gave a speech explaining the reasons for the invasion - the liberation of pro-Russian Donbas region, preventing Ukraine from hosting NATO troops/bases, and the denazification of the country. These were the three reasons. You can try listening to this speech rather than making up your own reasons, and then listening to his other speeches across time (he gives many speeches about Ukraine), and seeing if there is a change. He has been fairly consistent -- much to the anger of the Russian population which wants a far more hardline position on Ukraine, particularly in light of all the ethnic cleansing and atrocities committed by this US client regime. I would say, however, that as the war drags on, there is a real chance that Ukraine will end up partitioned into nothing, which is an object lesson to other regimes that wish to be U.S. clients. This is probably why Taiwan kicked out the pro-US party in the recent elections -- they don't want to be a battering ram the US uses against China and then discards when their powerful neighbor attacks them. Again, in all of this, it is Ukrainian people that suffer, whether they be of Russian, Hungarian, Polish, or Ukrainian ethnicity. The US doesn't treat its client states very humanely.
But I take it that now we agree on the larger point that this proxy war has nothing to do with "sovereignty".
This does not claim at all what you think it does.
> In an environment in which rival political parties were banned, torture centers were established, death squads were roaming the streets arresting dissidents, etc, and there was strict control of media.
What?, Zelensky was elected in 2019, 5 years after the Maiden revolution you know that right?. Do you have _any_ proof at all, that isn't thinly veiled Russian propaganda?.
> So in a place like Ukraine, it doesn't matter who is elected as long as the US controls the country. Zelensky was elected on a peace platform and pro-Minsk 2 platform, and immediately changed his position after the election. This is something you should also research if you want to be informed.
Theres literally zero proof the US controls the country, the Ukrainian government didn't even believe the Russians were going to invade. Probably because the Russians themselves continued to say they wouldn't and the Ukrainians put misplaced trust or maybe misguided hope in the words of the Russians.
> No, it is the Maidan regime that set up SBU torture centers, burned people alive, buried people alive, funded various Neo-Nazi groups like Aidar, Right Sektor, Azov, banned the use of Russian language, disappeared pro-Russian citizens, banned pro-Russian parties, banned pro-Hungarian parties, etc, and they were committing and continue to commit mass atrocities. There were many articles by Amnesty International and other human rights groups about this.
We literally have videos of the Russians committing war crimes against Ukrainian civilians, including execution and castration and we have many many many independent reports of torture and sex crimes including against children.
It would be nice if you source any of what you say, your stream of conscientious is not really a source.
> preventing Ukraine from hosting NATO troops/bases
aka, preventing Ukraine from protecting itself from future Russian attacks.
> much to the anger of the Russian population which wants a far more hardline position on Ukraine, particularly in light of all the ethnic cleansing and atrocities committed by this US client regime.
This is literally fake news, there is no proof at all that isn't just straight Russian propaganda of any 'genocide' in Ukraine outside of the one that Russia is committing.
> I would say, however, that as the war drags on, there is a real chance that Ukraine will end up partitioned into nothing, which is an object lesson to other regimes that wish to be U.S. clients.
I dunno, Russia is currently sending out under equipped mobilised citizens with T-62s and keeps recruiting from prisons whilst Ukraine is still fielding T-64s and better equipped and trained soldiers. Russia keeps getting more and more desperate every week, their army has been largely reduced to a pathetic bunch of barely trained cannon fodder lately.
I see the Russian army collapsing before Ukraine gets 'partitioned' by some mythical boogey man that doesn't exist.
> But I take it that now we agree on the larger point that this proxy war has nothing to do with "sovereignty".
This war has everything to do with Ukrainian natural resources and sovereignty, the Russians don't see the Ukrainians as a seperate people or as a real country and feel their position as the worlds gas station is threatened by the natural resources in Ukraine.
on Minsk-2 why would anyone believe what Russia says in any international agreement they already promised to not invade Ukraine or even threaten there sovereignty under the Budapest Memorandum?.
> America has invaded or attacked over 120 nations since 1991.
[citation needed]
> We currently have troops in Syria extracting oil
We have a few troops in Syria, but while the Syrian regime and its allies continue to claim the thing about stealing oil, no evidence of any kind has been offered to support that that has been happening since Biden ended the Trump policy around oil extraction.
> We just tried to overthrew the government of Ethopia by backing the Tigray rebels.
No, we didn’t. In fact, the US government has condemned abuses by all sides, including the TPLF, and sanctioned actors on all sides, including in the TPLF, for abuses in the conflict.
> We overthrew the government of Ukraine twice, in 2004 and 2013.
We didn’t overthrow the government of Ukraine at all. The main organized, non-grassroots actor in the overthrow of the regime that stole the 2004 election appears to have been the Ukrainian security services.
> We overthrew the government of Iraq
The first not-lie in your entire recitation.
> and are financing violent rebels to overthrow the government of Myanmar.
Myanmar has…quite a variety of different internal conflicts and rebellions, but while a number of actors have urged the US to back forces working against the regime that recently seized power in a coup, it is not doing so.
> Only strong, large nations that can stand up to the US and other rival great powers have any hope of achieving sovereignty. Ukraine has never had it.
I dunno, Ukraine’s doing a pretty good job of standing up to a supposed great power right now (and its only getting substantial Western aid in doing that now because it initially did much better than anyone expected without it.)
> The only question is whether it will continue to be ruled from Washington or whether it will be ruled from Moscow.
No, the only question is whether the Putin regime will give up its imperial ambitions before it collapses, or because it collapses.
Fair point, but the rest of the world will still be customers for chips made in Taiwan, and US defense promises to Taiwan are likely worth more if the economic flows are not just unidirectional.
No the reason they want to invade is to project power into the wider pacific, nationalism, imperialism and the people and human capital not just limited to semi conductors. The factories will be destroyed in any invasion, the US will guarantee it, I promise.
I doubt the parent has access to American war plans, but it's reasonable to guess that the US would prefer for China not to have intact TSMC plants because it provides enormous leverage. It's the same as blowing up Nordstream II. This is standard war stuff, and if the US is at war with China, heavy sanctions, etc, we wouldn't be able to buy the chips anyway. Why not drop a cruise missile on it?
Personally I hope such a thing doesn't happen. If I had to guess, Taiwan will eventually come under mainland China's control, but I hope this is done very slowly and in a bloodless way following a referendum by the Taiwanese themselves (e.g., only after certain guarantees of autonomy are made and the Taiwanese opt for the 'easy route'). I doubt mainland China will accept this thorn in their side indefinitely.
It may end up being a question of choosing to be Hong Kong or choosing to be Ukraine.
China doesn't want this to be like the Ukraine conflict either. I don't know if and how guarantees of some autonomy can be made that carry some significant degree of trust, but it's the least bad solution for all parties to make that happen.
Mutually Assured (Economic) Destruction. If destroying the TSMC factories is off the table, then China is incentivized to invade (to capture leading edge chip production). If the fabs are destroyed, it would take years to rebuild, which will cripple Chinese production of consumer goods using those chips.
Personally, I would be shocked if the US military doesn't also have a plan to "relocate" strategic personnel to the US in the event of invasion by China.
I’m not sure it’s proven China can build better products under their system than countries with freedom. Maybe only DJI are doing this in the consumer space today. The West/Japan/SK/Taiwan are still massively ahead in building the more complex components that China then assembles.
It appears the West is slowly untangling their economies from Chinese dependency and about 1/3 of stuff is made there today. With zero covid making them an unreliable manufacturing partner I doubt this will increase in the short term. I am pretty sure high end fabs will be returning to the west very soon and America could not have a better competitor than China to push them forward to new heights. I would bet on free countries producing most on the innovation over the next 20 years.
The reason China want Taiwan is because it poses an existential threat to the CCP, same story with Hong Kong, Taiwan has the same cultural roots as mainland China, yet they pursued democracy and free markets, despite not having access to the abundant natural resources of the mainland, its GDP per capita is roughly double that of the mainland, and its citizens enjoy a more free society.
What greater example is there of the abject failure of Chinas communist rule?
It is categorically NOT a huge win for the stock. This is all happening because of US subsidies.
The US has more expensive workers who are overall worse at the job and don't have the expertise to be compete with the people in Taiwan.
The subsidies only make this a viable option for political reasons. The decision to create a fab in Arizona is strictly speaking unprofitable and essentially a economically irrational move for the company.
It is ONLY being done because of US demand and political tension from China. For a shareholder this move is not good when looking at it in terms of profit.
I know this is a hard pill to swallow but it's true.
Don't joke about Europe. The best place for tsmc to expand is actually china. But this won't happen for various reasons that we all know about.
> The US has more expensive workers who are overall worse at the job and don't have the expertise to be compete with the people in Taiwan.
Putting the rest of your statement aside, this is a very silly thing to say. The United States invented the IC and started silicon age. Integrated circuits made in Silicon Valley were literally on the moon at the same time that Taiwan was still an incredibly poor country living under martial law.
Maybe today there aren't the exact people in the US to compete with Taiwan on this chipmaking process, but that doesn't mean the US lacks the ability to compete. It's not about general country-wide work ethic. If the right person to get the job done is a one-in-a-million person... well the US has 330 of them vs 23 in Taiwan.
Countries change over time. If the US has the ability to compete with Taiwan in semiconductor manufacturing, we sure haven’t shown it in recent decades.
IBM and Intel are competitive with TSMC and Samsung when it comes to ability to cram transistors onto wafers. This idea that only Taiwan/TSMC knows how to fab is light years from reality.
Well, that is certainly the Chinese government's take on things.
Your posts suggest that the US should just give up and let Asia -- more specifically China -- dominate semiconductors forever because US workers are lazy, fat, and stupid. Am I characterizing your position correctly?
I am simply stating the truth. It is from the perspective of a tsmc shareholder not a patriotic American who wants to beat china for no other reason then being the best.
As a tsmc shareholder one part of your post is correct. US workers are unfortunately lazier and slower and more expensive. Not necessarily stupider. You characterized this part of my position partially correctly.
As for what the US should or should not do, I never commented on that. Your patriotism and defensiveness specifically injected rivalry into your response. I literally have no opinion on what Taiwan or China or the US should do. I am neutral on that front.
> US workers are unfortunately lazier and slower and more expensive
I guess we just kinda stumbled into being one of the wealthiest and most developed countries in the world while having the lazier/slowest/most expensive workforce.
> The US has more expensive workers who are overall worse at the job and don't have the expertise to be compete with the people in Taiwan.
GMAFB. Intel has several of their fabs in the Phoenix metro area, they’re another 3-5 major players in the area, and there’s a talent pipeline from the local University into these companies.
Intel’s problems aren’t an inability to fab, it’s an inability to translate their design language into the new, smaller process.
It's universally well known that Asian workers work harder and can be paid less.
It's not just about intels capabilities. It's about economic wage standards. The cost is just too high in the US.
That being said tsmc workers in Taiwan are by far more capable then Intel this is proven by the 3nm process of which Intel is completely incapable of achieving.
You have posted many times in the thread saying the same thing, but slightly moderated because you got flagged.
You said elsewhere "things change" regarding labor force quality. They do. Asian labor across the board, but especially in China, has been rapidly increasing in cost while for example USA labor is stagnant in overall cost. Apple is medium-term going to be priced out of China just by labor costs. It is actually smart in a real-politik sense (and a business sense you deny) for labor sourcing to start looking a lot more broadly at different countries on a cost basis. USA is rich in some measures, but in terms of purchasing power and compensation of much of working class, it no longer is.
As for the quality of USA workers you've commented on a lot, I'll give you there is a serious decline in education. Saying they are slow or lazy shows you don't know anything about USA. The vast majority of the country is working itself to death and the life expectancy is cratering. As sad and reprehensible as it is, from the kind of logic you're using, a desperate and broken workforce is a GREAT business opportunity.
There is something beyond "American Exceptionalism" and "Asian Exceptionalism" and I think you really need to find it.
I always feel there's some implicit racism or belief in cultural superiority or something at play in these discussions. Anyone who has gone to grad school can see pretty plainly that the top schools are stuffed with Chinese and Indian nationals. They're capable and they work hard. Some of them, not a small number, go back home. Further, western industry set up shop in Asia for their own reasons and brought their expertise over.
The west had a lead, but 'we' trained Asians at our top institutions and worked closely with Asian manufacturers so that they can make our most sophisticated products more cheaply. There are a lot more people in Asia, and high relative poverty and cultural practices encourage a higher degree of scholastic achievement. Of course they're beating us now.
Outside of a very explicit and intense effort to develop domestic talent and retain foreign talent (or bloody wars), the west probably won't ever really lead ever again. This was the obvious outcome decades ago, but these things take time. The gap will grow and will extend up the value chain--western nations will do protectionism to try to slow this (e.g., Huawei, current chip restrictions), but cat's out of the bag.
I don't think it's a good or a bad thing from a global perspective. It just is. The great power competition that may result, wars, etc, is a very bad thing. The US in particular should compete as best it can, but it's best for everyone if we learn to live in a multipolar world.
> Anyone who has gone to grad school can see pretty plainly that the top schools are stuffed with Chinese and Indian nationals. They're capable and they work hard.
People in grad school tend to be hard working and smart, that’s how they got there. Hypothesis: foreign nationals have to be even smarter and harder-working to secure places in US universities, hence the stereotype.
It wasn't my experience that they were harder working or more capable than the domestic talent. Because of the immigration benefits, some had additional motivations native-born Americans didn't have, but there are easier ways to immigrate. There were some differences culturally in how certain things were approached, but I thought capability wise there was not an obvious and significant disparity in terms of capability of students by national origin.
What was noticeable is that there were a lot of these students and to the extent that the US does not retain them, we are training the workforce of our competitors. To the extent we do retain them, we are poaching the talent of our competitors. My guess is this is tough to balance and where we're at now is making total global innovation higher but lowering the proportion of the pie over which the US has dominion.
What percentage of the chip cost is the wage cost? This might be relevant for a 90uM process, but I think at 4nm, the wages are a minuscule part of the production costs.
Should be a huge portion of the cost. The material is just silicon. The expertise and know how that goes into this is where most of the money goes. For Taiwan, this expertise is better, faster and cheaper.
Those billions are also mostly R&D. It's also from a country external to Taiwan. ASML, the Ductch.
Yes, Obviously there's a lot that goes on in this process, more then "just" silicon. But if you want a summary, then cost expertise and knowhow, eclipses actual process. In a sense it is just "silicon" when measured relative to the ability and knowledge required to transform that material.
Chip-making isn't the same as sewing together cheap trinkets, and the Chinese economy has changed to support a growing middle-class so the reality of Chinese labor costs has drifted from the stereotype in recent years (especially in the domain of skilled labor).
Not to mention the lack of seismic activity and humidity that AZ offers.
The main reason is the looming threat of invasion of Taiwan from China. Such an action is catastrophic for Taiwan and the US. I thought this was obvious. Guess not.
The US is an easier espionage point for the US to steal technology from Taiwan. Most likely it will happen. But simple espionage isn't enough for this technology to fully transfer. The expertise and knowhow is just too challenging.
Do not let your patriotism blind you from the moral grayness that operates within the US as well.
Does TSMC "own" the technology? Because their fabs are wholly dependent on ASML, a Dutch company. And they're using that to produce tech that's designed in the west as well.
Right, only culturally European people know how to do anything. Those Taiwanese are just factory workers, no expertise at all, that's why Intel is in such good shape these days. Chipmaking is obviously so easy the only reason we let them do it is to keep the price down. /s
They obviously have expertise, the point is that the parent was implying that all the tech involved is exclusively TSMC's, which is also wrong. The parent literally said the US is trying to "steal" TSMC's technology...
So maybe read my comment in context instead of implying racism right off the bat SMH.
> The best place for tsmc to expand is actually china.
The chinese are vastly worse at fab than americans. Why do you point out that american workers are worse than TW, but neglect to mention that Chinese are even worse?
Not trying to make a judgment about people from USA, CN, TW as a whole - mostly just a function of experience.
Because they're not. PRC has worst fabs due to equipment sanctions, and now expertise sanctions (first by TW, then US). But said sanctions were placed at all because the cost and quality (east asian work ethic etc) of PRC semi workers weren't in question. Versus US workers being described as "group of giant babies" by TSMC employees ranting on PPT about doing more work for less pay than Americans in Arizona. Experience doesn't matter if you don't work hard enough. It's precisely because PRC workers, who can keep up with TSMC "military" pace, have demonstrated potential to run great fabs that led to companies like TSMC expanding in PRC in the first place, and why US had to do export controls, and do massive CHIPs subsidy to get fabs on US soil.
Good thing a stock price is about the future growth of a company. Hell, many public companies are valued at billions but still take a yearly loss. They are "subsidizing" their own growth in a sense.
After flooding took out most of the world's magnetic hard drive manufacturing capacity, it seemed clear that absolute efficiency in the immediate sense was the enemy of a robust long term manufacturer. I'm not sure why human v human threats are being heeded when nature v human ones were not, but diversifying your locations absolutely makes sense. There are some geographic constraints to where locating fabs make sense. And cost of living has to be balanced with the need for knowledge workers.
If you ran TSMC and you wanted to be sure no single disaster or war destroyed your whole manufacturing capability, where would you put the fabs? China's coast seems too close disaster wise and the same war that would dust your TW masks might destroy those too. Middle east maybe for climate and shipping logistics? Then one in the EU, perhaps spain? North america is probably third choice, and between climate and cartels that probably means the US. If third pick is paying you to go there, maybe it does make good sense.
I think you are certainly right, but why focus on Asia exclusively as a second fab location? One earthquake could flood all of them. As I said, oil giants in the middle east trying to diversify seems pretty ideal. I only suggested EU for the knowledge workers and it faces a different ocean.
From a geopolitical standpoint, US investment in Vietnam’s manufacturing capability would put pressure on China. China is trying to move up the value chain with Vietnam already eating their lunch on the low end.
Long sigted. The investment starts from scratch. Why not invest in some places cheaper? Why not invest in a place that has better expertise?
Overall the long term vision is to wrestle some control away from China. That is the long term bet the US and tsmc are ultimately making and what's driving the decision. But economically this is a bad bet.
My first job out of school was final test for a semiconductor in Phoenix that had a small fab in Gilbert. I remember we took a rabbit suited tour there as part of the new hire orientation, it was pretty neat. The one thing though that stuck with me is how much water a fab needs on a daily basis. I can’t remember the exact number, I just remember thinking how stupid it was to build fabs in Phoenix.
But then again, the amount of lawns and greenways in Phoenix compared to a place like Tucson, it’s pretty clear most folks in Phoenix don’t much care about water conservation.
I'm from Arizona and have thus wondered this and researched the perceived water dilemma myself. Fabs do need a considerable injection of water to start, but their systems are so advanced and their logistics so efficient that they end up reusing so much of it over time. Couple this with the fact that Arizona also has a great infrastructure already in place for water reuse and conservation. When Arizona had only 700,000 people in the 1950s, they used more water than they do now for 7,100,000 people. And it's still in the top 3 fastest growing states in the US (both by new relative to existing population, but also by total incoming population volume). Models also indicate that the once-in-a-century drought is coming to an end in the next few years, with huge rainfall amounts the last two years in the state.
As an Arizona native, occasionally my paranoia about living in a desert and simultaneously living through world wide climate change begins to really worry me.
Inevitably I come to a conclusion that is very similar to yours. Arizona is pretty low on priority for water from the Colorado River and does a great job with water reclamation.
That being said, I worry if I'm just believing what I want to hear
Though you need a lot more AC during the summer. Something solar could provide but until it's more ubiquitous most power would still come from fossil fuels.
It's like a swimming pool. It takes a lot of water to initially fill it, but then it's cleaned and mostly recirculated/reused. In 2020, Intel used less water per chip than they did in 2010.
Does the CHIPS act have anything to do with making this happen?
It's funny to think that the supply chain gurus at Apple & NVIDIA may be doing the work of geopolitics in the service of just defending their bottom line from disruption.
There is also the inflation reduction act that is very enticing to European industry giants that are facing hardship or downright insolvency due to their own governments stupid energy sanctions game. In fact NOW suddenly Macron and Scholz find it unfair that the US is snatching these company's. I can't say I have much sympathy for Europe. Well played US.
I wonder to what extent you can get subsidies from both acts. I assume it shouldn't be a problem.
> due to their own governments stupid energy sanctions game
Having morals and accepting hardship for standing up for them is not stupid. It's a principled stance most Europeans agree with. Governments need to step in and help whenever needed (with price caps for consumers, advantageous loans to businesses) to smooth it out, but it was that or appeasement, and we all know very well it doesn't work.
I don't think Europe has much of a choice. It's either sanctioning Russia and suffering ourselves, or allowing Ukraine to be taken with the horrors that come with that. If it would end there, maybe but still no, but it won't end there. Who is next on Russia's list?
The choice was made decades ago on the alter of faux environmentalism to export the environmental cost of energy production to another nation so the EU could claim moral superiority in the climate change battle. Completely decimating domestic energy sources.
Things were already heading that direction before the CHIPS act. The geopolitical issues with China have been a major concern for a while now. Not to mention the general business continuity concerns with everything coming out of a handful of countries/factories.
Apple spent something like a half-trillion on share buybacks. They could have averted this crisis a decade ago by thinking strategically. They didn't. Why give them credit where little or no credit is due?
>Apple spent something like a half-trillion on share buybacks. They could have averted this crisis a decade ago by thinking strategically.
I don't understand the fixation on this. Share buybacks strengthen the stock price of the company, which the company can leverage in the future to raise more capital if needed (by re-releasing new shares to the market). From that perspective, they are better than dividends.
Besides, right now, Apple isn't strapped for cash as they have around ~50 billion on-hand and could raise more if they wanted to. So Apple can still invest more in fab processes if that's what they want.
"Share buybacks strengthen the stock price of the company, which the company can leverage in the future to raise more capital if needed (by re-releasing new shares to the market). From that perspective, they are better than dividends."
You are confused. A 100 dollar company with 100 shares outstanding, each worth 1 dollar, is not better able to raise capital than a 100 dollar company with 50 shares outstanding, each worth 2 dollars.
And even if it were a reverse stock split can convert the former to the latter.
In the case where there are 50 shares outstanding and each are worth 2 dollars, that would be true, but if there's a constant appetite for the public to invest in a company and a diminished number of shares, that would drive up the price.
That might be how you imagine the stock market works based on first principles but Investopedia tells us
"The most liquid stocks tend to be those with a great deal of interest from various market actors and a lot of daily transaction volume. Such stocks will also attract a larger number of market makers who maintain a tighter two-sided market."
In other words the more appetite for the stock, the more liquid the stock, and therefore the easier it is for a buyer to acquire the stock. This is presumably because stock market indexes will refuse to include an illiquid stock, meaning there will be less demand for the stock, and the stock price will drop.
Apple did a 4 for 1 stock split in 2020. Do you really think they screwed over their investors because you know something they don't about how stocks are valued?
> From that perspective, they are better than dividends
This is kinda circular when you think about it. 'Strengthening the stock price' is just artificial scarcity, and incentivizes further financialization (by incentivizing borrowing with the stock as collateral) rather than simply distributing the profit or reinvesting it.
The data says their approach worked for them, but not necessarily for their customers, who regularly are made to shoulder indignities other companies do not dare to impose.
This is part of the marketing delusion, multiple people have done A/B testing to clearly show Apple Build Quality is either on par with the rest of the industry or in many specific examples WORSE, people that claim this often have ZERO experience with non-apple products or are comparing different classes like a budget consumer level Lenovo to the pro line of apple, of course of $500 unit will be of less quality than a $2500 unit
For an example of poor build quality / engineering Check out Louis Rossmans' "Think Different" video where he outlines some of the various engineering problems Apple has had
>>- it mostly just works.
I see the often repeated, I have had android phone for decades and never had any issue with them, they just work as well. I am not sure what is "not working" on an Android, I can also say i have never had an Android phone that I had to hold "correctly" in order for the antenna to work...
Again I think this stems from comparing different classes of phone, Apple only makes Mid to High eng phones, (which is the market I am in for Android devices I tend to buy the Flagship from the manufacturer) they do not make budget and entry phone, these low end phones will have more problem. Would it be better to just price out entire segments of the population? That does not seem to be socially moral
I will assume I’m not going to change your mind with my comment, but anecdotally I was a pixel user for years (Nexus 5 (boot loop break) Pixel 2, Pixel 3, tried to order a Pixel 6 but the site didn’t work on launch day) before switching to the iPhone 13. I build applications for both Android and iOS. My wife has been an iPhone user for years. Both iPhones she has owned are still in use by someone in our extended family (iPhone 7, X)
My pixel phones did not take high quality videos without crashing and over heating. This iPhone can do 4k 60fps for an hour without a hiccup. The build quality of this phone feels like it came from an alternate reality to me. Pixels in the same price point do not hold up against it on any metric that matters to me (stability, battery life, software reliability, hardware longevity)
Again, this is just anecdotally and I try not to be an Apple fanboy, but being the family tech point person, Apple has made my life marginally easier.
Your build quality comparison is off, a cheap $500 laptop should be compared to prior years Macbook Air M1 which is $300 more, not $2000 more. A $800 macbook has far better build quality than any $500 Windows laptop, probably better than most $800 laptops. Its Windows that's the issue for me. Having too many bad experiences and strong bias against Microsoft, I choose Apple because its not MS and it does work vs. tinkering with stuff on Linux.
Apple phones are usually behind in whiz-bang features that Samsung and some others have but my opinion is killer features for smart phones are in the rear view mirror by a few years. Any of them are just fine and its more about the ecosystem and purchase history at this point (i.e. switching and you'd have to buy things again).
>Having too many bad experiences and strong bias against Microsoft
Well at least you admit your bias, as someone that works in enterprise and manages 1000's of windows computers i can say i would never want to attempt that with Apple who does not have the enterprise tooling that MS does
before I was in enterprise I felt the same about windows and primarily used Linux, I would never use a Apple which shows my bais... If apple was the last computer on earth I would crush it...
>>Any of them are just fine and its more about the ecosystem and purchase history at this point
I agree with that, and that should change, but I value my freedom too much ever to surrender it for Apple's walled garden
Ironic you talk of a walled garden and work in Enterprise IT creating a ... walled garden for your users (enterprise tooling like pre-installed images, not being able to install apps as admin, etc.) because its easier to support.
Oh I don't for a second think Apple stuff would ever compete with 1000's of devices being managed from an Enterprise IT department. I lived in that environment for a while and would dual boot to linux, run VirtualBox or get a 2nd PC from IT and wipe it and install linux.
>A $800 macbook has far better build quality than any $500 Windows laptop
Took a peak at the apple store, there is no $800 Macbook for sale. So we're comparing, an older (used?) higher-end Macbook to a budget Windows laptop? Why not compare an older higher-end Windows laptop to an older Macbook? Why not compare a new $500 budget Windows laptop to a $500 used Macbook?
>Its Windows that's the issue for me. Having too many bad experiences and strong bias against Microsoft
You have a have a subjective preference - fine - we all do. So why pretend there is some objective measure here of Apple's superiority? Apple sells very expensive higher-end devices. The reality is that the quality of those devices is about on par with other devices in that class.
Comparing a refurb Macbook to a budget Windows laptop doesn't really show that Apple devices are of "higher quality" - which was the argument OP was trying to make.
That's true... I guess the argument that as a student you can get an M1 macbook air for 900 or that Mac minis start at 700 (non student) also wouldn't work (because minis aren't macbooks and not everyone is a student).
> I can also say i have never had an Android phone that I had to hold "correctly" in order for the antenna to work...
It feels like you're trying to revive decade-old flamewars here. FWIW, on my Pixel 5 (and 2 before that), if I reverse my phone (hold normally but with camera at bottom) I get a drop in both cellular and wifi signal strength - more so if I'm in a location with marginal signal to begin with.
What a passionate comment about Apple. I guess some people are just this invested in Apple enough to write book-long comments on every single thread even tangentially related to Apple.
When I worked as an Apple laptop repair tech, that was almost nonexistent. 2/3 logic boards shipped in to be used for repairs on customer computers were faulty.
"- it mostly just works"
Imagine a laptop perpetually-stuck on OSX 10.2.4 that can't ever upgrade or it borks the system - that was my experience dealing with Apple laptops years ago as a repair tech. School system iBooks and PowerBooks were the worst offenders, and Apple had no discerning way to let you know which image belonged on which hardware, so we were always having to reimage laptops in hopes we picked a proper image for them.
Build quality, that's a riot. Here's a riddle for you: my $300 Thinkpad and $1,500 Macbook Pro each hit the concrete from waist height. Which laptop can I open up and keep using like nothing happened?
Build quality != resistance to drops. You can have a rugged, impervious laptop that will survive a drop from an airplane onto concrete, but if the keyboard flexes and mushes when you use it, that's not exactly 'good build quality'. Manufacturers optimize for different parts of the UX and it's valid for customers to pick which device they like based on how well the product does in whichever area they care about.
It's also impossible to bring back horses as main vehicle of transportation until you have a civilisation collapse and can't viably produce machinery and fuel in large scale.
I find it unfortunate that we are going back into a partitioned world but let's hope it brings competing ideas at least. I'm even a bit excited about it, as long as it stays a cold war and doesn't turn into WW3 and stays as a competition in everything like during the cold war.
> It's also impossible to bring back horses as main vehicle of transportation until you have a civilisation collapse
That’s a bad read of the situation. It’s not like no one makes chips.
The reasons people cite for why manufacturers don’t come to America are largely political. The reality is that manufacturing is alive and well but those with industry knowledge aren’t American, and America has most left low margin and high labor manufacturing not all manufacturing. In this case, America is literally paying the Taiwanese to bring their knowledge to America.
Is the world collapsing? No one knows but this can be more easily seen as “vertical integration” of a national economy. No one said Samsung was doomed when apple made their own mobile chips, and there’s no reason that the global economy is doomed just because the richest and most powerful economy in the world wanted a strategically important, high margin business.
I'm speaking about manufacturing in general, I don't think that anybody said that high margin and high tech manufacturing can't be done in the west - stuff like chip manufacturing never left the USA, they simply fell behind.
I mean, that's true. If the 'we' is American companies. They had to bring one in from another country to manufacture here. We stopped being a manufacturing country decades ago.
It's possible. With subsidies. The US is literally paying tsmc for this to even be viable.
Tsmc is not coming here because it's an economically wise move. They are coming here knowing it's an economically irrational move.
Politics is the reason for this move. It is not a clear cut win.
Your post has the aroma of patriotism. Patriotism blinds us to harsh truths. Why do you love how people were wrong when they said we can't make chips in the US anymore? You shouldn't feel love or hate for any of these statements.
> Tsmc is not coming here because it's an economically wise move.
No, it's an economically wise move because of the politics. This plant allows TSMC to gets the CHIPS subsidy, but maybe more importantly it allows them to become part of the military industrial complex which requires that products are made domestically. This fab will print money.
The subsidies make the plant break even. Defense spending is basically another subsidy. Most of the weapons the US makes is useless in a civilization that has mostly been at peace for a long time. Hard to say how far defense will go though in terms of spending.
You are definitive about defense spending in a recession. This is wrong. Aspects of the government will begin pulling back. We do not know how this will effect tsmc.
This move from a profitablity standpoint is break even as far as we know. It makes no economic sense.
Btw. This isn't something I'm making up. The CEO of tsmc mark liu stated this to Nancy pelosi during her visit. Also her husband is dumping tsmc stocks for unknown reasons.
> Most of the weapons the US makes is useless in a civilization that has mostly been at peace for a long time
It's not useless, it's all about power projection. When you go to rob a bank and take a gun with you that you will never use then you don't call that gun useless because thanks to projecting your deadly power with it you were able to successfully rob the bank.
Inflation is a problem right now. A huge problem. The recession is actually the result of the government pulling back. It is a necessary action.
Interest rates will increase spending will be pulled back to stop inflation. A recession is the tool being used to stop inflation.
The CEO did not ask for money. The money was already given. The CEO was simply stating the status quo. Helping her come to terms realistically with what is truly going on with the Arizona plant. Look it up.
Your post has an enormous aroma of patriotism. Where were you when Taiwan was subsidizing TSMC? They were only able to outcompete Western fabs due to the enormous amount of government help they got in the early days. Those Taiwanese leaders sure were stupid to have invested in chip manufacturing.
Patriotism for America at best. I'm American. Just not biased. Patriotism is a form of bias. When you have none you see the truth more clearly.
Taiwan subsidized tsmc for tech. Once the tech was established the subsidy ended.
Right now the US is subsidizing tsmc for simply switching locations. Once the location is switched the economic output of this plant will be negative so the subsidies have to remain.
The US chip market is not capable of making competitive chips. Not without economic assistance. This is categoric fact.
So cool. When I drive to San Diego to see family and friends I drive right by these massive new facilities on the new Route 303 bypass outside of Phoenix.
I am so happy to see more super high tech manufacturing happening back to my country. While I am in general a fan of globalization, I also believe that for resiliency every country should maintain their own unique cultures as well as have some independence so that with a reduction of life style they are still be able to survive independently from the rest of the world.
Where in Arizona specifically? Have they even decided? Did the government decide to pay for the factory and enroll their for-profit prisoners to work for $0.10/per hour, and divert the remaining water supply to the factory? Arizona seems like a terrible choice for a chip factory, unless they've selected an area that gets natural water... and there aren't many.
Hmm, my understanding was that fabs needed an initial supply of water, but then could cycle that same water for a long time.
A quick Google[0] suggests that they've only started doing this heavily pretty recently, but I'm guessing new fabs in Arizona will implement state of the art water-recycling.
That said, the numbers in this article suggest that 98% recycling would drop usage to the equivalent of ~6000 homes, which still feels significant.
Low seismic activity is the benefit of Arizona that I've seen cited in the past, which I'd guess outweighs the water sourcing issues.
That article is somewhat wrong. 10 million gallons of water is not enough for 300k households. It's maybe enough for 300k people. 10 million divided by 300k is 33 gallons. The average use per person even in urban areas is around 40 gallons per day - so that's generally ignoring outdoor irrigation. It's more like enough water for 200k-250k people.
So 2% of that is maybe around enough water for 5k people. Arizona has 7 million people. And residential doesn't even use most of Arizona's water.
Arizona is actually a good choice. It is geologically stable, doesn't suffer from natural disasters, is somewhat close to Silicon Valley, and has a large enough talent pool to staff TSMC's facilities.
Surely somewhere in the midwest would be more geologically stable and have plenty of access to water. Not sure the proximity to Silicon Valley makes any difference. It's not like factory workers are going to commute from California to Arizona.
Phoenix has long been a hub for semiconductor manufacturing, so it makes sense from the perspective of there is an existing skilled worker pool and supplier base. Motorola had large sites there, and Intel has a bunch of fabs and is building more in the area.
I believe the primary reason was tax-related, but the secondary reason is that there is already substantial personnel in the area who are experts in chip design and fabrication (Intel), with a strong pipeline for new talent from local universities for those skills.
Surprisingly, the area is also one of the largest Taiwanese communities in the U.S., which is a bonus for the engineers and their families who will be relocated from the "mothership".
Also, it's actually quite common for high-level Intel staff to fly back and forth from Portland and Phoenix. I am not sure if the same would be said about Taiwan and Arizona, but if they have staff in SV it wouldn't be too far-fetched.
Access to talent is a huge factor. One of the biggest obstacles of creating a US fab was the lack of access to cheap talent (in contrast to Taiwanese engineers that work long hours for relatively cheap money and are widely available - of course comparing to the US, in Taiwan they are among the highest earners).
Intel has tons of fabs in phoenix so TSMC can steal some of their employees. I think intel chose arizona because of the lack of natural disasters and tax incentives. The water is mostly reused so not really that big of a deal.
Yes, actually more like 2 generations once chips are commercially available, the Arizona factory has never been advertised as a cutting edge node.
But there are 100s of media headlines basically implying that by listing high profile customers and not mentioning the date of opening. Those customers will for the most part be using this factory for auxiliary chips.
I don't know much about semi conductors. maybe i'm just an old dinosaur. But, why is the whole world so hellbent on getting the very latest chips? Couldn't we just make do with 10 year old chip designs? My Iphone6 can do just about everything that my wife's latest iphone11 can do (at least that a casual observer can see). Cars from 10 years ago were'nt that different from now (other than buttons being replaced with screens), etc. The nintendo from the 90s is just as entertaining as the latest nintendo, etc.
Generally smaller nanometre designs are a lot more power efficient. Your old iPhone will do just fine yes - but imagine if it had twice the battery life?
Now why they just keep massively increasing power as efficiency goes up is beyond me......I think a lot of people would love a new iPhone with a processor a bit faster than your old iPhone 6 but with insane battery life.
It's not necessarily about having the latest chip tech from a consumer perspective but more about TSMC's business demand. Businesses like Apple are only signing contracts for the best chip designs, and they are a massive customer for TSMC. Moving production to the US is great because it means apple has a more reliable supply source in the wake of increasing China/Taiwan tensions. It makes sense for TSMC as well because Apple and others were considering alternative manufactureres (Intel) in the wake of those political problems.
But as for regular consumers, I think you're right. Most of us don't care whether we have a 10nm or 4nm chip in our devices, we just need good battery life.
We are headed to a regime where more redundancy has to be baked into the fabbing business so that Taiwan can go offline (if/when China invades) without making phones and computers and datacenters completely un-upgradeable.
If the company is Taiwanese is there not still a concern with dependency on China and leaking of information. I realise it's important to have local manufacturing, but is this still essentially a "Chinese" company considering the disputed territory thing? Excuse my ignorance here about how such things work.
Taiwan is "Chinese" ethnically, being populated[1] by descendants of many waves of Han colonization over the last few centuries.
Concern over "China" has nothing to do with ethnicity, it's a geopolitical fight with the government of the People's Republic of China, which does not[2] rule Taiwan.
[1] To be clear: there are also descendants of indigenous "Taiwanese" living there, who are austronesian and not Han. Ethnicity is complicated and everywhere is a melting pot.
[2] In practice. Obviously "legally" both Taiwan and the PRC consider themselves the true government of the other's territory.
I know I said I wouldn't engage, but just to point out: nothing in the linked page substantiates "doesn't actually have anything to do with". I'm not interested in getting into an ethnicity argument, but please don't misrepresent source material.
People in Taiwan don't walk down the street saying "im han im han im han"... They don't call themselves 'han chinese', they don't call themselves 'chinese'.
It's a Taiwanese company with no dependency on China. TSMC has the knowledge and expertise to make 4nm chips anywhere - you could view this as their information 'leaking' to the US.
TSMC does not depend on China..
China only claims Taiwan's sovereignty but Taiwan have de facto independence.
A bit like a stalker going around telling people the person they're stalking is their girlfriend...
I was thinking about the PRC government having influence on Taiwan, I've no idea what their relationship is. Of course I have nothing against Chinese/Asian individuals, I just don't know what level of control China might seek to exert on Taiwan I guess.
What is the secret of TSMC? I believe a Dutch company makes the machinery that TSMC uses. Why can't other companies use it like TSMC does? Does TSMC hold some crucial patents?
This is great news, especially given the geo-political risks related to Taiwan. I still hope for the best, but with majority of chips produced in Taiwan we need to be ready for alternatives.
Given that the chipmaking process is quite water intensive and Arizona is a literal desert in the midst of a major drought maybe this wasn't the best possible US location for the fab?
Phoenix is a massive hub for semi-conductor manufacturing and data centers partly because there are no natural disasters.
A lot of water is required to start but a huge majority is recycled. Heat can be a problem in the summer but it's extremely predictable and commonly dealt with.
chip noob here, i thought we'd already reached the theoretical limit of chip fab sizes? lots of people are talking about 3nm and smaller in this thread? can someone explain?
It's been a few years but I've known a number people who worked on the production lines at Intel's fabs in Hillsboro, Oregon. Many, if not all of them worked 3 twelve hour shifts each week. As I understand it the schedule is pretty common in production facilities of this type.
Note: I'm not commenting on the suitability of the practice, just on the fact that it's not unique to certain countries in Asia.
This move doesn't make sense under a capitalistic point of view. This is geopolitics. The US protecting its interests, corporations playing along - they don't have a choice anyway.
Hopefully we'll do something like that here in the EU too. We're experiencing the hard way how's like depending on the energy and resources of warmongering criminal dictators; it would be wise to start moving away from technological dependence on China asap.
Actually, Europe is doing pretty good due to having a monopoly on making chip making machines via ASML in the Netherlands which relies heavily on Carl Zeiss in Germany.
Fun fact by the way, at some point Japan and Europe were very close in state-of-the-art semiconductor manufacturing while the US didn’t have the capabilities anymore because it was outcompeted by Japan. At that point, the US donated funds and 20 years of research to ASML since it was better than letting Japan win the race, according to Chip War by Chris Miller.
This is largely because Europe became infatuated with austerity and its industrial policy has suffered as a consequence. You want big, cutting-edge fabs? You’re going to need to spend public money getting them off the ground. TSMC’s success is in part due to Taiwan itself designating semiconductors as a key strategic economic interest years ago and making investments/tax breaks accordingly
Interestingly, in the 1970s, before any fabs on the island, Taiwan arranged for semiconductor engineers from the iconic but slowly dying American company RCA to transfer their technology to a visiting Taiwanese team, establishing what would later become TSMC. RCA pioneered so much: radio, TV, color TV, NBC... And just as it was starting to decline and die, its semiconductor knowledge was transferred to Taiwan!
> This is largely because Europe became infatuated with austerity and its industrial policy has suffered as a consequence.
Was austerity broad-based or specifically targeted toward industrial areas? I'd always assumed austerity meant cutting back on public benefits/pensions/etc. but not strategic areas like this which is why I'm curious.
That's an interesting thesis - but, for example, Greece with its "let's piss money away without control for anything and everything" somehow hasn't become an industrial powerhouse either... so it might be that austerity (or lack thereof), in general, isn't really such an important factor?
Greece has other, deeper problems to fix before it makes any attempts in creating any form of industry. The "pissing money away" happened because of internal problems, which can be attributed to corruption and cultural aversion to any form of entrepreneurship that goes beyond the scale of mom 'n' pop stores.
I am not trying to absolve Greece from its liabilities, just pointing out that Greece is a bad example for austerity not playing a significant role in slowing industrial development.
_leading-edge chip manufacturing_ must be seen like "defence": making money out of it is optional, but it has to stay _really_ leading-edge and should be ready to produce at scale for other failing "friendly" part of the world. Since South-Korea and Taiwan did just that, and the others not, they are now alone on the global market.
To believe the "supply/demand" rule of the economy can magically make the money flow decently and properly is _REALLY_ dangerous, it cannot apply to everything.
Why do we even want cutting edge fabs in Europe? We have no companies that design cutting edge logic chips here. Literally none. Why invest 20 billion dollars (or what is the price of a cutting edge fab these days?) to create supply without demand?
Or do we seriously expect that US companies will generate significant demand even though TSMC and Samsung are already building heavily subsidized fabs in the US?
> or what is the price of a cutting edge fab these days?
Intel says they're going for two factories of 20 bln indeed [1], Samsung for 17 bln in Taylor, Texas [2], and Micron claims to go for 100 bln (over time) in Clay, New York [3].
> Why invest 20 billion dollars [...] to create supply without demand?
Why do we have to pay about 5 dollars per month for a VPS with only 1 vCPU, 1 GB RAM, and 10 GB storage? I'm certainly hoping these specs to all increase 10-fold over the next 10 years for the same price.
Re your last point: cloud specs are already far beyond that. Cloud companies just pocket the difference. Cloud lets hosting companies benefit from Moore’s law, not you.
Look at the machine you can build for one months’ typical AWS cost for a medium size SaaS company.
Cloud also charges insanely high rates for bandwidth.
I doubt that. I was not talking about AWS. The price I mentioned is from Hetzner (by heart though, so I might be a bit off) which is pretty cheap. I have also tested multiple budget VPS providers and they all don’t dare to go below aforementioned price even though there is a lot of competition in the VPS market. Sometimes the more budget providers provide more vCPUs but in my tests those usually turn out to be extremely slow.
It's quite low, but the real costs are on datacenter space and connectivity anyway, I have no idea what their cost structure looks like.
I would expect any real user to switch into renting servers as soon as small VPSs aren't enough. (But yes, the fact that there is a market of large VPSs tells people don't to that. I don't think I will ever understand this, as I don't understand most people usage of AWS.)
Most buyers are simply not savvy enough to pick the most effective hosting. I worked for a small e-commerce operation years ago, they had essentially no technical expertise in-house, but they knew their products and their market. Odds are very low that their VPS arrangement was optimal but how would they know? As long as the site stayed up and the orders came in.
> Or do we seriously expect that US companies will generate significant demand
Yes I think so, presuming that there is not an over-supply of capacity.
US companies would much rather rely on an EU country than one that is being threatened with invasion over a small gap of sea. US local supply will never be enough.
> US companies would much rather rely on an EU country
Agreed
> US local supply will never be enough.
But you won't just have US supply. You'll have US and Taiwanese supply, and I don't believe that Taiwan will happily let TSMC (the only cutting edge foundry left in the world, when you take Samsungs abysmal yields into account) build foundry redundancy in the western world.
But we'll see, you could definitely end up being right. I just hope we'll invest at least an equal amount of money into chip design.
ARM came out of europe. With the right industrial policy, chip manufacturing could be onsourced. The dutch already make most of the equipment that makes the fabs/chips.
It's not necessarily about supply and demand. These past few years have shown what shortages of chips can do to the supply chain. It's a strategic vulnerability if Europe does not at least think about this.
I guess it depends what fabs you build but we have seen very large numbers of manufacturers desperate for components. A fab anywhere in Europe could easily supply any factories in Europe so there should be demand.
On the other hand, if people are trying to build the cutting-edge, there might not be as much local demand since it is probably only needed for the latest IT equipment, most of which is built in the Far East.
Do you have any sources for your claims or are you making stuff up?
Many, many Europeans do work in factories. Just ask the Germans.
What's wrong with working in chip factories anyway? They produce some of the highest margin products in the world and since they are highly automated, working in a chip factory requires certain knowledge and education on physics, quality assurance, automation, material science, and certainly give you experience that makes you a valuable worker with future perspects rather than a replaceable cog in a dead end job as is the case for the Europeans working in most other factories that are a few steps away from being off-shored to lower cost areas.
I have personal experience having been in a low end research fab.
The bunny suits and the protocols are elaborate.
It was only for few hours and it was quite uncomfortable, hard to see or get a sense of things around you.
Regular users would generally stay for several hours to make it worth it. No break, water, toilet or food, probably come in with an empty bladder and empty stomach and stay the whole day.
From what I have read, it requires specialised training and intermediate if not advanced level skills and relatively high level of education. You work on the same machines for years and they pay is not necessarily that high compared to the trouble that you put in.
In fact, even in Taiwan the challenge is that people often switch to chip design or software instead. This definitely gets harder as people age.
Yes it’s extremely toxic there’s a reason why manufacturing moved from the US (and probably EU/UK but I’m not certain). It’s horribly toxic and you can read this about Samsung[1].
Asia has what some would call almost slave labor and a complete lack of care for workers. Many countries don’t care about pollution either.
US and European countries will gladly clean up manufacturing at home while shifting to countries who could care less about employees or environmental impacts.
So I’ve seen conditions in some poorer nations in Asia be described as similar to slave labour, but we’re talking about Taiwan and South Korea aren’t we? These are high-income countries, so I’d be really surprised if they had such conditions.
I could believe EU has some stricter environmental regulations than both, though
Tons of people go work in factories making cars, chemicals, and food. If they could provide better shifts and working conditions they can potentially attract talent. Oil fields attract people who wouldn’t have gone into the industry if they hadn’t been offered better pay and family benefits.
Do you have hard numbers how much it would impact the bottom line to offer fab employees good wages and work-live balance? It's not like fabs are employing fast armies of low skilled labourers in sweatshop (even if you do sweat under PPA).
Oh man if you don't consider it a donation if it has strings then don't do nonprofit work. Juggling buckets of money from people who donated for specific things with specific conditions is just part of the job.
For this industry, you want as many verticals as you can to ensure supply. Have the early parts of the chain is great, but producing the end product is necessary too.
You also see this problem the other way around, when a US company produces a chip design, but then actually gets it fabricated, packaged, integrated into a product, boxed, all in another country
Already being planned by Intel. They're building a new leading edge fab in Magdeburg, Germany [1]. This is backed by the new European Chips Act [2]. Wheels have started turning very quickly with regards to home-grown semiconductors in 2022.
It will be whatever the latest node is at completion, so most likely Intel 3 or Intel 20A [1]. It was Pat Gelsinger's pledge to get Intel's node roadmap back on track when he joined as CEO just under 2 years ago. From what they've been submitting papers wise to industry forums, IEEE Synopsium and the like, they're on track to deliver 20A by 2025. I recommend following the progress of Intel's new "Intel Foundry Services" business segment.
Intel's "10nm+" (non-EUV) processors compete rather fiercely with AMD's "5nm" (EUV) processors. The latter are weirdly not 2x better, not even close, on any metric. Hm.
Anecdotal evidence, but I do know that my 12th gen Intel can burn through 10% of battery in 10 minutes. That is to say that Intel is doing really poorly in power consumption.
It really depends on the model. I see, for example, Kaby Lake Y models with a TDP of 4.5W . Though, in general, yes, many report that Intel is not taking power consumption as a priority (which does not mean that they do not offer niche "ultra/extremely-low power" products).
I can't explain the downvotes, but for what it's worth both Apple and AMD offer far more power for the same TDP when talking about low-wattage parts, do they not? Thats the whole benefit of TSMC's process advantage (and also why Nvidia also moved to them from Samsung this generation). In practice of course, to get the performance uplift consumers expect, AMD and Intel and Nvidia are all pushing TDPs to the limit where they can, but for products that want performance-per-watt, the TSMC-based ones are king, from what I've seen.
That said, it's all a moving target, and anyone who thinks Intel isn't producing good parts today is wrong -- they are.
there is intersuccession of Intel and AMD. AMD seems to offer more models, but a proper analysis of their stand-outs and similarities is required to get a proper picture.
TDP is not a good metric to normalize against because it's a thermal design target only tangentially related to power limits. Power limits are generally much higher. For example the AMD stock limits result in a power target that is generally ~40 % higher than the specified TDP. Lots of motherboards will disable that limit automatically though and let the CPU draw as much power as it likes. I don't know how current AMD mobile platforms work, Intel Alder Lake has four or five different power limits. There's an instantaneous power limit, that's like 100 W. Then there's PL2, the short-term power limit (for a few seconds), which is like 30-40 W. PL1 is the sustained limit and depends on the laptop and is set to match the capabilities of the cooling system (or lack thereof) in an Intel-specified range (like 6-25 W). That's typically marketed as the TDP for laptops. There's an additional power limit for the iGPU and something somewhere is doing the divvying up of the currently available package power into iGPU power and CPU power (which doesn't seem to work that well under Linux).
In the current desktop generation AMD and Intel are much closer than in Intel 12th vs AMD Zen 3. That's mostly because 13th gen has pretty much the same limits as before, but is more efficient, while Zen 4 has massively increased the power targets throughout the stack - by 60 to 90 % - to keep up with Intel (plus the marketing advantage of getting through the 5 GHz wall)
E.g. a 5600X had a PPT of 76 W, a 7600X has a PPT of 142 W. A 5950X was 142 W, a 7950X is 230 W.
While they may be powerful, and they definitely are, they’re pushing out quite a bit more heat and requiring a decent amount more power to get there.
Though it’s impressive how much efficiency they’ve squeezed out of that node regardless.
But to say that TSMCs process doesn’t confer large advantages is silly. Intel even uses them for their new discrete graphics cards — they wouldn’t have been able to compete otherwise (among other less interesting business reasons)
Intel is making the right capex moves to position itself for a future situation where Taiwanese production could get interrupted. Intel is right alongside TSMC and Samsung at the leading edge and it would be an exaggeration to suggest otherwise. Marginal differences in road map and quality over the short term don't tell us the full picture in a business where course-corrections take years and years to bear fruit.
> If Intel is not good enough for US, why would it be good for EU?
Who said they're not good enough? They're slightly worse than TSMC, but still put out competitive chips for desktops, laptops and servers. Slightly more expensive, slightly higher power consumption but still, perfectly acceptable and still the market leader in some areas.
> It's clear that EU did what US wanted instead of what would be in its interest.
No, it's not clear. Why would the EU do that? They did the best they could given the very limited choice present.
Intel is still holds the market on laptops and desktops at about 3/4. They simply push more volume. TSMC makes desktop chips for Apple, but Apple isn’t the majority. The rest of TSMC’s production is almost exclusively GPUs and phones, and this could change. If Intel executes well on their next nodes, and should TSMC slip at all, Intel could pull ahead. Never count Intel out. People thought Intel would die after the z80 ate their lunch, after the Athlon 64 ate their lunch, and now they say it again… I think it’s all just noise.
Intel has lots of leading edge fabs in the US and continues to build them. They made some business mistakes around not opening up their chip design the way ARM enabled ... but the tech capacity is all there. Now they are doing contract production for chips that they did not design so I expect them to catch up in a few years.
Meanwhile Chris Miller in Chip War argues that Intel will fail again since nobody wants to give their secrets to Intel. If you’re Apple, for example, then handing over chip designs to Intel isn’t ideal in terms of competition.
It's not that odd that international companies would build local facilities for major markets they're selling into. Toyota sold cars to the US, and building plants here made sense relative to that for a variety of economic and political reasons. Everyone makes more money together so everyone's happy with the arrangement.
I find it more odd to presume that corporations have some nationalistic competitive interest.
Cars are big and heavy, and so building them near the buyer has a logistics benefit. Chips have higher value per weight and size by orders of magnitude.
The most common reason is typically cheaper labour. That’s why Nike produced in Japan and then moved to even cheaper countries. Related might be that some countries have a lot of knowledge workers in their 40-50s who can then manage factories in other countries. Japanse companies producing cars in Europe is an example of that.
For the current trend to move to the US. That has to be government funding. Luckily, fabs require large investments at the start which might break-even more expensive labor. Also, maybe some more automation in the fabs could also help in making the fabs cost competitive.
While a factor, labor costs alone are the primary reason. In reality is more about the wider regulatory burden on a company, including environmental controls
That said it does not explain why a Japan Company can make cars profitably in the US but a US Manufacturer can not.
If it was only labor why would Toyota not have a factory right next to GM's in Mexico and import cars from Mexico to the US?
I thought Japanese car factories in the US were all non-union (as opposed to the US brands)? Toyota does have a Mexico factory at least according to Wikipedia.
Correct, the Honda/Toyota plants in the USA are non-union.
Interestingly enough the “Clean Energy for America” bill pushes for additional tax incentives on EV's built by companies using union labor:
More specifically, the proposal says that electric vehicles assembled in the United States would qualify for a $10,000 tax credit while EVs that are built at facilities whose production workers are members of, or represented by, a labor union would be eligible for the full $12,500 credit.
Aren't clauses like this just a clear cut sign of corruption?
It doesn't even make a lot of sense as an incentive to have workers unionize, since why would workers care what price what they're selling cars for. The law would only exist to benefit existing unions.
I guess its that the Japanese Company can make cars profitably anywhere but making cars for the US market in the US means there's no/less import taxes, also because the factories are huge they often get local tax breaks, and the cars may also eligible for EV/hybrid tax credits (I think Toyota have sold over 200K so there cars can no longer get this).
I think making cars in Mexico for the US became less worthwhile when Trump pulled the US out of NAFTA.
>US became less worthwhile when Trump pulled the US out of NAFTA.
lol, trump did not "pull the US Out of NAFTA" not only is that not with in the power of the president to do, it was never on the table, NATA was replaced with USMCA, and only made a few small changes, for Automotive that means 75% of the vehicle components must be made in MX, US, or Canada, up from 64% under NAFTA
Nothing in the law would impact making at car in the US vs MX, and was aimed at preventing increased parts from China or other non-north American nations, and all of the changes were passed by congress with wide bipartisan support
OK - but he signed a deal with Mexico that replaced it and didn't the new deal mean that more parts had to be from the US and half the car factory workers needed to pay $16 an hour - which would make Mexico less attractive?
It makes sense. ASML are in the Netherlands, as are NXP. ST are in France. Germany has quite a bit of semiconductor production for the automotive industry.
It'll be slow and expensive to get there though. There's not many places in the EU where you could simply drop a Shenzen in, demolishing historic buildings and natural environment along the way.
"STMicroelectronics N.V. commonly referred as ST or STMicro is a Dutch multinational corporation and technology company of French-Italian origin headquartered in Plan-les-Ouates near Geneva, Switzerland"
The lengths French and Italian corporations will go to to avoid taxes and red tape in their home countries, it's almost poetic. Even Airbus is now headquartered in Leiden, Netherlands instead of it's original place of Blagnac, France.
I get that many individuals and companies see taxes as wasted money or theft, but taxes pay to fund the education of the workforce that will work for said corporations and fund the infrastructure used by said corporations to build and transport their goods, including stuff like courts which companies can use do protect their IP or military force projection to protect their assets and investments abroad.
Without the skilled workforce or infrastructure the company will for sure not be competitive.
What most big companies are now doing to get insanely wealthy is have most governments foot the bill for the infrastructure, protection, and education of their workforce, while all the profits from the fruit of their labor go directly to shareholders and the governments don't see a dime in taxes so stuff like education, healthcare and infrastructure is falling apart while corporate profits have never been higher.
You are mixing chicken and eggs here. We had education and skilled people before huge governments stealing massive amounts of taxes from people existed. The disposable income from the population gave rise to bureaucrats who love to tax everything under and pretend that they world cannot run without them. That's pure twisted logic.
Yes, those pesky medieval castles all over the place. And narrow cobbled streets, only two riders abreast. Those are the major problems of Europe, not bureaucracy and indecision at the highest level.
How did we outsource it? We still have semiconductor manufacturing in EU. It's just way behind the cutting edge, and not because it's cleaner than the one in US or Taiwan.
Semiconductor manufacturing is dirty business. I live in the EU next to one such fab and whenever local environmental concerns are raised about the fab, since nasty chemicals are sometimes found in the river the plant uses, the official response is always "the exact process is confidential, we cannot allow external inspectors inside, but we can pinky swear we conform to all regulations via self audits" and the government rolls with that as the unofficial response is "stop bothering us about the environment or we relocate production to Asia and you're left with a bunch of unemployed engineers and tax hole in your city coffers".
They're not going to build a fab in downtown Heidelberg :-)
There are plenty of industrial parks available near transportation, water, and electricity in Europe to build this out. This also aligns with how the modern chip ecosystem works where the fab components are coming from Europe, the advanced chips are coming from Taiwan, other chips are coming from across SEA, and final assembly is happening in China. What I suspect will happen is that Europe will start printing advanced chips and most of them will be shipped to China for integration. If a conflict/dispute breaks out, switching to European assemblers will be expensive but totally doable with a short-term investment (2-3 years??)
> We're experiencing the hard way how's like depending on the energy and resources of warmongering criminal dictators
I'm not sure we're learning, though.
At the moment, it seems that we are not only still buying the energy and resources from said warmongering criminal dictators (thus funding their war of aggression), but also setting up institutions to patrol gas pipelines, so that said dictator cannot blow them up as easily and so that we can buy even more resources from said dictator. You couldn't make this up.
In light of this, I have little hope for rational policy making regarding China and chip production, unfortunately.
I see this repeated as a matra, in a way of Carthago delenda est.
But it is not true. Said warmongering dictators do not need dollars or euros to wage any war. The entire chain of the war machine is in local currency, backed by local resources. They do not need anything there, that has to be bought for dollars or euros. They are not some third would countries that have to buy their weapons abroad.
By repeating this mantra, we are only lying to ourselves.
> The entire chain of the war machine is in local currency, backed by local resources.
Not really. There are already signs that the Russian military industry is in hot water now because there are no Russian semiconductor fabs that can supply the type of chips needed for anything beyond dumb ballistic missiles [1]. And it's not just chips, but also other basic electronic components or modules whose manufacturing has long since gone to China and other countries, some of which are under the scope of international sanctions against Russia.
> They are not some third would countries that have to buy their weapons abroad.
They are buying a ton of drones from Iran, for example, or Soviet-era stocks of artillery munition from North Korea [2].
> Said warmongering dictators do not need dollars or euros to wage any war.
Oh yes they do. No country on this planet is self-sufficient, not even the US. And it's not just about military equipment, it's about basic necessities of life, especially medicine and food. Russia needs money to buy these abroad, and for that they need foreign currency.
Western reports have to be taken with grain of salt; it was demonstrated that western journalists know exactly zero about the war and what you read is more wishful thinking than news. For example, there were reports of Russia running out of gas in march, and out of ammo in april, and our (I'm EU citizen, that's why "our") Ursula was talking about cannibalizing chips from wash machines. And yet, here we are, this all turned out to be nonsense.
Same goes for "iranian" drones. Iran would have no idea how to start integrating them with russian C4ISR (note how the "mopeds" are observed by lancet drones); non-integrated drones would be OK for tactical, but not operational level of war-waging.
Russia is closest thing to autarchy that you can find on this planet (US isn't even a player here, US deindustrialized itself in the name of cost cutting). The 2014 sanctions only helped in building such economy. Medicine is easy to clone, if you are not bothered with intellectual property (see also India) and they are net food exporter. Sure, some french cheese or wines could be missing, but they are not necessary for the war.
But back to our topic: none of this means that buying energy is financing the war. It is there for conditioning western population to get used to more expensive energy (basic economic theory: the supply was cut, demand was preserved, the equilibrium moves). The oligarchs are going be laughing all the way to the bank. It is just surprising that someone with intelligence to be discussing on HN would be taking part of such conditioning, without realizing it.
> Not really. There are already signs that the Russian military industry is in hot water now because there are no Russian semiconductor fabs that can supply the type of chips needed for anything beyond dumb ballistic missiles […]
I don't know what your definition of «the type of chips needed for anything beyond dumb […]» is, but – if I were to infer – all chips manufactured for military needs are the dump chips, be it in China, or in Russia, or in South Korea, or in the US etc.
Military does not chase cutting-edge, smallest nanometer manufacturing facilities nor do they look for fancy 3D stacked L1 CPU caches and alike, the ones we encounter in consumer tailored microchips (MC's). MC's produced for the military sector 1) are always several generations behind the consumer counterparts; 2) are slower; 3) get subjected to extreme and very rigorous testing, e.g. getting baked in specially designed ovens; 4) they come with hardened shells to later get subjected to irradiation; 5) likely something else. Surviving specimens make it into missiles and elsewhere, for all is required for a missile is a chip that will be guaranteed to not have failed a mission.
Freescale (ex-Motorola) and Texas Instruments are two of the largest MC manufacturing contractors in the US. They have separate lines set up for consumer and military needs, with the military contracts taking a priority. I can't be bothered to check whether Intel or AMD have clandestine US DoD contracts but it is safe to assume so. When Motorola used to manufacture their own MC's, they also had two separate lines for the highly sought after DSP's, 56k and 96k series. There were two versions, a hardened one (prohibited for export out of the US), and the consumer version (with somewhat more relaxed export controls for the 56k series but not for the 96k series). Tolerance specs of the hardened version were classified at the time.
Back onto Russia. To cut it short, it is a case of hypocrisy on both sides as propaganda has been busy working on both sides. Soviet Union (and later, Russia) has been self-sustained, self-contained and has been manufacturing their own chips since late 1960s - early 1970s with always prioritising military needs over the consumer needs. They have never chased the latest designs or developments, but their stuff has been reliable where required.
It is not easy to assess their current situation due to the information disclosure suppresion and also due to prior reports of embezzlement on a unfathomable scale having taken place in Russia specifically when it comes to military contracts. What it is known with a fairly high degree of certainty is that Russia had commenced a 90 nm manufacturing facility as early as 2014[•]. There have also been sketchy reports that they have since moved on to a 28 nm process. Finding a reliable source is not easy, though.
Either way, chips that have been produced to a 90 (or 28 nm) process and have been subjected to hardcore testing requirements are good enough to drive missiles (likely, other military equipment too). Provided chip manufacturing facilities are still operating in Zelenograd, their output will be prioritised in the current political environment, and it will receive a priority funding in the local currency. One ought not to underestimate the adversary and ought to be wary of the creativity they may come up with once having been cornered.
Emperor Poo of All Russia has been demonstrating the world that he is willing to drive his never be, imaginary empire into the ground at any cost – in order to fulfil his ultimate wet dream of crowning himself as the first Galactic Emperor Poo, and, since Russia has been making their own silicon wafers, the MC manufacturing situation is not all that black or white – until more is known for sure.
> They do not need anything there, that has to be bought for dollars or euros. They are not some third would countries that have to buy their weapons abroad.
they had their local currency forever. yet they developed very fast exactly at the same time as we started buying their goods. its not a coincidence. and yes China is a third world country by all metrics, its not because you have very modern centers like Shanghai, Beijing and more big cities like that that there is not utter misery in the countryside that would make you blush
Its not like those who are paid in roubles would have any use for any other currency. Everything they need can be paid for in roubles, turtles all the way down.
Sure, you won't get iphone, porsche or gucci wares for that, but those are not necessary to wage the war.
> Sure, you won't get iphone, porsche or gucci wares for that
This is where your argument breaks down. Russia imports vast amounts of manufactured goods that it doesn't have the capacity to make itself, including basic military and basic consumer goods, and for which it needs something that their trade partners might conceivably want. Which ain't roubles, for the most part.
This just doesn't reflect reality at all. There are vast drops in the domestic production of things like cars (-85%), motors (-70%), and white-goods (-50%) due to a collapsed import chain. (Russian government source: https://rosstat.gov.ru/storage/mediabank/87_01-06-2022.html) Key economic indicators are flashing red - e.g. non-tax/oil tax revenues are down 20% YoY (Russian Finance Ministry figures from last week). If these are the official Russian government stats, it's likely the real numbers are much worse.
Cars and motors are dual use goods (so the collapse of Russian domestic manufacturing of them is militarily relevant), but even setting those aside and looking purely at single-use military goods, there's persuasive evidence that the Russian military is increasingly reliant on Iranian drones and equipment (https://www.al-monitor.com/originals/2022/10/russias-use-ira...). Vast numbers of Russian equipment have been confirmed destroyed or captured by observers such as Oryx, and those are just the visually confirmed losses you can check the evidence for yourself (https://www.oryxspioenkop.com/2022/02/attack-on-europe-docum...). No country would be able to replace those losses without switching to a full war economy, which Russia has not done (and which would only be a necessary but not sufficient condition for replenishing these kinds of losses).
It's a bad idea to depend on the outside for energy and resources in general. Not just on warmongering criminal dictators. Buying US gas for a fortune is far from ideal, for example. Not blaming anyone, obviously any country will look after themselves first. If resource scarcity starts hitting hard, we won't be able to expect external countries to just send us resources as if nothing happened, even if they are free countries and good allies.
And that's without even mentioning that who knows which countries will be free and which will be dictatorships in 20-30 years.
You now that back in 1300BC tin to make bronze in Egypt or Greece was imported from far away Afghanistan? Do you know that in 10000BC people traded seashells in Siberia, thousands of km away from the ocean? It's a good idea to be as independent as possible, but no country was ever entirely self-sufficient. Not even North Korea.
We are already on the way, as the EU is securing investments of 43 billion euro for the semiconductor industry[0]. We also have ASML, which TSMC are heavily relying on, since no one makes the machines they do.
> Hopefully we'll do something like that here in the EU too.
Sure! France is betting everything with investing for an Intel factory here, the non-fashionable manufacturer that is getting excluded of the market because their chip designs are outdated. If my tax revenue can be used to maintain old actors afloat and create jobs to teach the French how to do things that don’t perform, I’m more than happy.
And it’s not even a technological dependence strictly speaking, as pretty much all of the technology actually comes from the west. It’s just manufacturing.
They're also building a fab in Japan and that one seems to be progressing much faster.
About the Invasion of Taiwan, I kinda doubt it. They just had local elections and the president Tsai said it would be a referendum on her stance against China[1]. IF that is really the case then it seems like the Taiwanese people have voted against being turned into another Ukraine.[2]
I am not currently in Taiwan, but a couple of weeks ago I could see 3 military cargo planes land in Taipei every day. I'd be curious how it looks today.
One thing to remember is that 17(or more?) billion of those arms that are destined to Ukraine were originally intended for Taiwan[3]. With all those (western) reports of the US running out of ammunition because Ukraine uses in 3 days what the US produces in 1 month[4], and given how isolated Taiwan is on a map I wonder how wise the whole endeavour really was.
I get that there are a lot of Tech people in the valley from Taiwan that have a more hawkish view on the relationship between Taiwan and China, but can we acknowledge for a moment that a) the valley is not representative for the world or any countries population and b) we also have to be a bit realistic about the facts on the ground.
> With all those (western) reports of the US running out of ammunition because Ukraine uses in 3 days what the US produces in 1 month[4], and given how isolated Taiwan is on a map I wonder how wise the whole endeavour really was.
Maybe that’s one of the reasons why china was so supportive of the invasion. It’s a test of Americas ability to supply a war (in proxy).
The US has less "skin in the game" now that the TSMC core tech can be found on US soil - that's what the parent and me are referring to. Taiwan is not able to defend itself without US aid.
ps. I'm not in the valley and it's not about "hawkish" views, it is pure power-play.
> Taiwan is not able to defend itself without US aid.
If you look at a topographic map of Taiwan, it's not so obvious. Rough terrain coupled with the fact that an amphibious assault is needed to even get to Taiwan, and then the troops there need to be resupplied by sea and air (both of which require infrastructure which can be sabotaged), make Taiwan a very good defensive position. Of course it couldn't last forever without external help, but even on it's own it's plenty to cause a massive embarrassing bloodbath.
Island a series of plains with no depth fragmented by rivers from rain + high mountains. Essentially a series of sequential turkey shooting galleries from air. PRC will be the ones blowing up bridges and infra to cut island into piece meal bastions to further restrict operation space of TW. The mountains themselves are incredibly tall, which is a nightmare for defenders limited to light arms against attackers who'll be droning them with relative impunity. The foliage helps, but SAR / sensors tech filling that gap fast. Before/if PLA even bother with landing, they're going to shape conditions to be as uncontested as possible. Which likely means embarassing one sided bloodbath as PLA drone operators in air conditioned mainland offices glassing 100,000s of relatively soft ROCA defenders in plains with no rear except rough mountains rougher than what Vietcon / Taliban operated from. Meanwhile, rest of island - the home front - will have critical infra distrupted, after calories and clean water runs out, they'll be inviting PLA to resupply island vias sea and air. Capabilities of attacker determine whether geography is blessing or curse to defenders. For TW, it's increasingly curse.
Taiwan is a small island, it lacks lot of basic natural resources, like oil, iron ore, etc. One simple thing that China could do is to send their navy to cut off all the supply lines: it won't take long before Taiwan is degraded to third world country. Without the US navy to counter Chinese they do not stand a chance.
You're talking about it being hard to resupply Chinese troops over the air, how about resupplying the whole Taiwanese population?
Oh absolutely, I can't say I disagree with you. But I'm just saying that it also means that if the US drops Taiwan they do have a lot of politicians that are sympathetic to mainland China. So it doesn't necessarily have to be an invasion at that point.
> b) we also have to be a bit realistic about the facts on the ground.
Realistic facts on the ground for you:
The prime majority of rural Taiwanese who don't speak English are even bigger sinophobes, and have even less relation with the mainland.
Most Sinophilic area in TW is Taipei, where the highest concentration of migrants from China live, and from where the lion share of immigrants to US comes from.
And the mainland is full of Russians pretending to be Chinese like you used to say? You yourself have a massive personal chip on your shoulder against both Taiwan and the mainland. Your anecdote means very little.
That would certainly be a difficult vote, but I think the second alternative would be more like being treated like Hong Kong. Certainly not nice and probably a bit worse than Hong Kong, but a very long shot from the Uighur situation.
What? Is the USA going to sell their chips to China? Because when Taiwan gets invaded the Fabs will be destroyed and/or made inoperable with no way to get them functional again.
> Because when Taiwan gets invaded the Fabs will be destroyed and/or made inoperable with no way to get them functional again.
You can't know that. Taiwan might plan to do this, which China surely anticipates, thus in any invasion plans would be made to stop it - paratroopers dropping on top to seize control quickly during the night, covert operatives swooping in to take out critical personnel in charge of sabotage, etc. etc.
> Is the USA going to sell their chips to China?
China is already working on improving their own industry and reducing reliance on imports.
China is decades behind and will stay so because the industry is not standing still either.
The chip machines need maintenance which the Chinese cannot do themselves. They don't have the knowledge. They'd need ASML (European company) for that.
They are lightyears behind. 20 years if not more. These chip machines is work of many decates of iterative work. You cannot simply step in and produce similar tech.
You don’t need to actually replace them with equal alternatives, just enough so the digital economy doesn’t implode. Even 2015 is probably a target year in terms of performance that doesn’t destroy the entire electronics industry.
If they are "lightyears behind", and everyone on HN thinks that China will invade Taiwan in order to get chips, then why the ** is the US trying to cripple China's domestic semiconductor industry?!? Isn't this a self-fulfulling prophecy? If HNers are so concerned about Taiwan's peace and independence then why aren't HNers protesting more against the US' effort to cripple China's semiconductor industry so that China has no incentive to invade Taiwan for chips?
These questions are only half rhetorical. I really want to hear what people have to say about this.
> and everyone on HN thinks that China will invade Taiwan in order to get chips
Most everyone things China will invade Taiwan because they want to own the land. It was once part of "china" but now is a (politically unrecognized) separate nation.
Everyone thinks that the only reason not to invade is that warfare will surely destroy all semiconductor manufacturing on the island.
The US wants to cripple the Chinese semiconductor industry to ensure that Taiwan can't be invaded without severely hurting the Chinese economy.
I may be naive, but I think R&D in those cutting-edge sectors is not something you can just throw money at and then get results, you need to create the foundation for it first.
Is there evidence that links more open societies and liberal economies to technological progress?
TSMC isn’t the only thing they need to replace. ASML isn’t either. There’s multiple levels of sole suppliers for chipmaking that nobody can replace. China can’t replicate an mRNA vaccine either.
Good enough for what? The free world isn't standing still so China will always be several generations behind. The market wants the latest and greatest.
It's only if the free world can't compete will China catch up, in which case it will get what it deserves.
Good enough for 70% of dometic market demand, which analists have found is satisified by 14nm.
The thing is, "latest and greatest" is actually a niche demand, even if we don't feel like that's the case because of phones and laptops. The market for non-phone, non-laptop, boring unsexy applications that don't require more than 14nm is apparently much bigger.
who says there is only one way to invade Taiwan? a maritime blockade would be super effective to suffocate Taiwan in a matter of months and destroy their economy.
The US does not the balls anymore to do anything against China, unfortunately. (Nor the financial interests to do so, with most of the US debt being owned by China)
Keep in mind that much of the West is in a recession that looks quite bad. People will be careful about losing their job maybe even accepting much lower wages as a result.
If you would have said " Invasion of Taiwan likely" that might have been productive. But this is not reddit, just making up "confirmed" is neither mature nor appropriate for a technical audience.
Give away to a foreign country? You realize that TSMC is a publicly traded company and they have a shareholder responsibility in managing their operational risk?
David Ricardo never said that England should depend on Portugal for wine, and Portugal on England for clothes.
Edit: apologies, I read the original point as referred to the USA as a decisor, not to Taiwan (as the poster intended). The former point remains, about the clear opportunity for strengthening ties.
With respect, this seems like a case in which credentials are important.
The geopolitical context involving the relevant actors brings a number of difficulties the knowledge of which must be assumed, and that even Mainland recognizes.
I find it comical that the US educates more than 90% of top engineers yet we don’t control 90% of crucial chip-making assets.
We need to implement some type of conditions for anyone seeking education here in the US especially in institutions that are backed by US tax dollars. We need to stop handing out education to the very people who are dead set on competing against our national interests. In other words, stop training the enemy.
The US actively kicks them out of the country: once your student visa expires, GTFO. Seems crazy to me- in my dictatorship, we should do the opposite: confiscate the passport of anyone in a PhD program until 5 years after the completion of their degree.
For that matter, it is similarly "«comical»" that if one wanted lean modular furniture (and not even "scientifically" covering every reasonable need) one has to go to the Swedes
> training the enemy
That is much more complex than those terms. For one, knowledge is transversal ("zero" is not e.g. a "cultural appropriation", etc).
> we need to stop handing out education …stop training the enemy
Seems a little harsh and nationalistic? Not an American so maybe I’m way off base here, but in a country where people already pay exorbitant prices for higher education, what would you propose?
Unless I misunderstood you, I don’t think banning foreigners will solve your problem.
Poor Taiwanese people... their security insurance has just been cancelled.
I wonder how many days after the first successful batches of chips coming out of the Arizona fab will China invade Taiwan.
I do hope not, but realistically: with high-end chips being made on US soil, the US will have very little interest in protecting Taiwan, apart from maybe blocking China from also acquiring the tech.
One plant does not change the game that much. You still need TSMC to keep working on whatever the next tech will be as well as all their current production in Taiwan which is fully booked.
I agree that it's not a simple plant opens => China invades, but this does feel like we're starting to see the dominos line up.
If it's about talent, it's a lot easier to quickly import people than to import a massive fab plant, and I assume we'll be building talent (either domestic or imported) as we build out the related infrastructure and industry.
It would certainly be disruptive, but I assume part of the US's drive is to reduce dependency on Taiwan, and consequently exposure to the threat of China.
If the US stops caring about Taiwan, it's both safer for China to invade Taiwan (less pushback from the US), and less geopolitically valuable (less damage to the US), but China's interests aren't focused entirely on the geopolitical when it comes to Taiwan.
My cousin is doing a contract with TSMC. Basically they fly over 100 Americans per year to Taiwan and train them for 1-1.5years. Then they fly them back to the usa to work in the US factory.
The problem is TW compensation and work conditions are terrible. Many of them quit before completely their agreements, so they aren’t actually training that many Americans.
First, your phrasing could read as concern trolling style gloating, which you probably didn't intend.
China may invade Taiwan within my lifetime, but it won't be triggered by anything to do with TSMC.
China and the US are both involved in the Taiwan conflict due to history, ideology, and current economic relationships. Nothing material about that changes with TSMC building facilities in the US. If anything US ties grow stronger.
TSMC is not something China can acquire with military power. It's not a building in a RTS game you can just take over and operate yourself. It's a huge number of engineers and a globe spanning high tech supply chain. All that grinds to a halt the moment missiles fly into Taiwan.
People really blow out of proportion the importance of TSMC. Taiwan is valuable to the west because of its strategic location. TSMC could disappear tomorrow and the US will still have to defend it. The day Taiwan fall is the day the US loses its dominance.
Not in this lifetime. Unless you are talking about occupying China / storming their territory, that would not go well. But defeating China at sea and in the air and in other countries? They'd have no chance.
China has 5 times as many people as the US. Their technology is behind, but they are modernizing and having 5 times as many resources and no public backlash to wasting soldiers lives goes a far way.
