9 ms·
How to build a $20B semiconductor fab
- gjkood 2y agoI have been listening to Kai Ryssdal's Marketplace on NPR/KQED the last few days on my commute home from work. The topic for the last several days was on the CHIPS And Science Act and the new Semiconductor Fabs being built by TSMC and Intel in Phoenx, AZ. It will be several years before the plants already being constructed will go into production but there is a whole ecosystem of current construction, education of the future workforce that will need to be hired in the future. Not to mention all the ancilliary companies that are needed to support these gigantic plants in the area. The dollars from CHIPS Act are not only bringing in the manufacturing plants but will be essential to bring this lost capability back to the US in the scale needed both from an economic and national security perspective. It was great listening to the show and the impact the CHIPS Act on people's lives already happening now and in the future. For anyone interested the links to the specific shows are available as podcasts here. https://www.npr.org/podcasts/381444600/marketplace https://www.npr.org/podcasts/381444600/marketplace
- wuj 2y agoTSMC has moved many engineers from Taiwan to Phoenix. Entire towns were built with accomodation, schools, ethnic grocery stores from scratch. It would be interesting to see this initiative's cultural and economic impact on Phoenix in the years to come.
- psaux 2y agoI did not know this, I have family in Tempe close by. Do you have any references I can pass along to them?
- wuj 2y agoHere is an article from Phoenix Business Journal: https://archive.ph/dYCAF https://archive.ph/dYCAF > With Taiwanese transplants moving to north Phoenix in droves, Arizona officials — and a local baker — are working behind the scenes to make them feel welcome
- gjkood 2y agoI believe the new term for Phoenix, AZ and other major cities in AZ is "Silicon Desert". You can see a map of the many companies in the high tech space in AZ at the following site. https://siliconmaps.com/silicon-desert/ https://siliconmaps.com/silicon-desert/
- davidgay 2y agoMajor semiconductor manufacturing in Arizona is not exactly new: Motorola had fabs there in the 70s (or earlier?), Intel's presence dates to 1980, etc. The article [1] below from 2001 says that Arizona was "3rd in chipmaking"... [1] https://go.gale.com/ps/i.do?id=GALE%7CA79561583&sid=googleScholar&v=2.1&it=r&linkaccess=abs&issn=0038111X&p=AONE&sw=w&userGroupName=anon%7Eadfe2077&aty=open-web-entry https://go.gale.com/ps/i.do?id=GALE%7CA79561583&sid=googleSc...
- zdw 2y agoBeing an AZ resident, this map is hilarious as it skips out the ~100 mile distance between Chandler and Tucson. Tucson is more well known for optics, but more focused on space science than on use in silicon fabs.
- jonhohle 2y agoI’m not sure where “towns” (plural) are being built. The first plant is across the highway from a master planned community that was built 20 years ago, long before any fabs were being discussed in that area. There certainly is a lot of development there, but it’s not like a factory town or anything. There’s an outdoor recreational shooting facility across the street. I can only assume that is a huge culture shock for anyone coming over from Taiwan.
- navigate8310 2y agoHere's another fascinating article about the cultural struggles faced by TSMC in order to expand in Phoenix: https://restofworld.org/2024/tsmc-arizona-expansion/ https://restofworld.org/2024/tsmc-arizona-expansion/
- onthecanposting 2y agoHonestly, this sounds like normal life being a junior employee at a big company. An ossified clique of old guard managers pick winners and losers with little regard for merit? Unbelievable bureaucratic overhead? Your boss expects you to flatter and fawn all over him because he allows you to work 50-60hrs a week at 80% the median salary without overtime pay? Sorry to say, this is just what they're all like. Do good work, make friends, and try to get a job at a small company where people treat each other decently and incompetents can't hide in the crowd.
- wslh 2y agoCould we say this is the "Manhattan or Apollo 11 Project for Chips"? For its articulation. BTW, about this topic, I always recall "A View To Kill" James Bond's (1985) movie [1]. The top hit in the soundtrack from Duran Duran [2] is also recommended and playing in radios even today. Seems like Intel passed the torch long time ago but don't forget to read the mantra book: "Only the Paranoid Survive" [3]. [1] https://en.wikipedia.org/wiki/A_View_to_a_Kill https://en.wikipedia.org/wiki/A_View_to_a_Kill [2] https://open.spotify.com/track/6I4snLrVOrJsLdd43isc27?si=7ba0c55e669d4747 https://open.spotify.com/track/6I4snLrVOrJsLdd43isc27?si=7ba... [3] https://en.wikipedia.org/wiki/Andrew_Grove#Books https://en.wikipedia.org/wiki/Andrew_Grove#Books
- jacobsimon 2y agoI’m hugely supportive of the new factories and investments, but I’m curious why they decided on Phoenix which seems to be affected dramatically by climate change and incidentally has a very different climate and culture from Taiwan.
- 11thEarlOfMar 2y agoThere are already a number of fabs there. Intel has a big presence. Government support, infranstructure and an experienced (although insufficient at this point) workforce are already there. The reason there is a chip fab activity in Phoenix goes back to 1949 [0] when Motorola built a lab there. In 1952, they started making semiconductors and eventually chips. [0] https://go.gale.com/ps/i.do?id=GALE%7CA79561583&sid=googleScholar&v=2.1&it=r&linkaccess=abs&issn=0038111X&p=AONE&sw=w&userGroupName=anon%7Ef79c90d7&aty=open-web-entry https://go.gale.com/ps/i.do?id=GALE%7CA79561583&sid=googleSc...
- icey 2y agoA few possible reasons: Cheap & reliable electricity (https://www.axios.com/local/phoenix/2024/04/26/arizona-power-outages-electic-grid https://www.axios.com/local/phoenix/2024/04/26/arizona-power...), low seismic activity (https://www.spglobal.com/mobility/en/research-analysis/briefcase-taiwan-earthquake-puts-spotlight-back-on-chip-supply.html https://www.spglobal.com/mobility/en/research-analysis/brief...), existing semiconductor supporting industries (as mentioned elsewhere in this thread), and a politically favorable business environment.
- brailsafe 2y agoSnowbirds from Winnipeg will quickly get outcompeted for foreign vacation rentals/properties lmao
- trwm 2y agoYes this is called a positive externality. It is why outsourcing produced much more devastation than was promised and why onshoring will create much more work than expected. The end result currently will likely be stagflation since like always politiciand do the wrong thing even when doing the right thing.
- VHRanger 2y agoThat's not why outsourcing created devastation. The issue without outsourcing is that the benefits are widespread (lower prices!) but the drawbacks are concentrated (factory town is now a hellhole). And our political system is incapable of redistributing correctly even though the net effect is highly positive. The seminal study on the topic is the "China shock" paper from Autor et Al.: https://www.nber.org/papers/w21906 https://www.nber.org/papers/w21906
- trwm 2y agoThe multiplier effect is well known and understood. Nit sure why you're arguing against it.
- philosopher1234 2y agoI don’t know or understand this multiplier effect you’re referring to. If you’d like to persuade me (and I assume other readers) explaining your argument might be more effective. Instead I get a sense of “don’t argue against me” as opposed to “this is why I’m right”
- TeMPOraL 2y agoStrange, because put this way, it should be entirely positive - widespread benefits and concentrated drawbacks are what we want to happen, as it benefits more people and concentrated problems are much easier to manage. What's very bad is when benefits are concentrated (often in the hands of a small group), and drawbacks are widespread, and therefore near-impossible to manage. See e.g. pollution, emissions... ... and outsourcing. The benefits are concentrated: profits captured by the companies doing the outsourcing. Sure, they may sometimes trickle down to the consumer, but the costs - the distributed drawbacks - are inferior quality of goods, elimination of local jobs, high ecological footprint, abusive business practices, lack of effective customer support. And the extra magic here is, it spreads direct responsibility over national borders, so it's near-impossible to hold anyone to account.
- moneywoes 2y ago> ancilliary companies any pertinent examples? r.g. schools for workers families ?
- gjkood 2y agoThe show was talking about chip packaging companies to create the end usable chips from the silicon produced. Just for the construction work alone, they mentioned that the pipefitters local union membership has doubled since 2020. Refinery level complexity on the specialized piping needs for the plants. Special training programs geared towards the semiconductor industry being offered in the local Community and Trade schools training people to be the skilled and semi-skilled workforce for these companies. People who were teachers now making four times the income working on the construction project.
- everybodyknows 2y ago> People who were teachers now making four times the income working on the construction project. Targeted investment! Toward future generations of chips, rather than of people.
- tacocataco 2y agoI guess the people should hire better lobbyists.
- ijidak 2y agoSo, is Intel buying ASML EUV machines for their build out in Arizona? It seems like Intel is skipping ASML EUV entirely. If that's the case, I'm trying to understand how Intel ever gets decent yields at 7nm to 5nm. It's definitely not coming from High-NA, which seems like a short-term distraction.
- imperialdrive 2y agoAbsolutely fascinating, thank you for sharing! I just want to order new phones computers laptops and servers left and right now.
- louthy 2y agoStep 1. Acquire $20b Step 2. Build the $20b semiconductor fab https://knowyourmeme.com/memes/how-to-draw-an-owl https://knowyourmeme.com/memes/how-to-draw-an-owl
- schmorptron 2y agowe can collaborate on this, you handle step 1, i'll do step 2?
- ghghgfdfgh 2y agoSomewhat related: a high school student creating his own semiconductor fav in his parent’s garage on a "budget". https://youtube.com/watch?v=IS5ycm7VfXg https://youtube.com/watch?v=IS5ycm7VfXg It’s a thoroughly interesting video, but I’m a bit disappointed he never took his idea any further than he did. I’d really love to see something like a 6502 being made at home.
- krishadi 2y agoSam is now building Atomic Semi [0] with Jim Keller (ex AMD) [0] https://atomicsemi.com/about/#:~:text=general%20fabrication%20equipment.-,Team,-Atomic%20was%20founded https://atomicsemi.com/about/#:~:text=general%20fabrication%...
- ghghgfdfgh 2y agoThat's good for him, but it means we'll never find out in the near future whether it's possible for the average person to create useful ICs in their garage.
- trueismywork 2y agoIt's possible yes, but not really..
- nwiswell 2y agoIt's very similar to creating clothes by hand at home. Can you do it, given the right tools, training, and patience? Yes. Will they be any good? No. Will they be cost-competitive? No. Getting chips made on a shuttle run for an old node is very affordable. There's really no need for it. (According to one MPW supplier, 10mm2 of 340nm, up to 10 dies, costs 6400 euros, and it's unlikely 340nm is achievable in a garage anyway)
- ghghgfdfgh 2y agoFrom a business perspective, I agree that it’s a fool’s errand. But imagine being able to design and tangibly build your own computer, at home. 6400 euros is pennies for a business, but exorbitant for an individual. I believe the way Sam Zeloof circumvents the enormous amount of capital needed for a chip fab by relying on modern technology to create 1970’s technology. He simply mounts a cheap digital projector onto a cheap microscope - they didn’t have that advantage in the 70s, and thus it cost millions to start a chip fab. My point is that it could conceivably be doable for an individual to create old computing technology with the advantages of living in the modern world. I certainly don’t have the drive to do it, but I wish someone did.
- narrator 2y agoI've been to a number of big construction sites for business purposes. When I hear, "this is a $40 million dollar project" and when I see all these workmen, equipment and materials moving around, I think, so this is what $40 million dollars in motion looks like.
- tnmom 2y agoMuch of that money goes to permitting, admin, and insurance. What you’re seeing is probably the $10-20 million that wasn’t successfully siphoned off by bureaucrats.
- bmiller2 2y agoI don't know why you're being downvoted. There is a reason why development volume across states is asymmetric.
- stevofolife 2y agoDownvotes help people skip the noise and focus on the signal. Ain’t nothing wrong with that.
- tnmom 2y agoWho knows - it’s just fake internet points. Hard to get worked up about.
- zarathustreal 2y agoKinda easy to guess, there’s really only one possibility and answer: people felt a certain way after reading and voted accordingly. The idea that anyone votes based on the content of what’s been said is false. For example, you’re not gonna upvote a true statement of a statistical fact if you think there’s a terrible reason for it. You’re gonna downvote because you feel bad about what you perceive as the reason regardless
- BryantD 2y ago
- deanmen 2y agoV
- xyst 2y agoAt what point will current fabrication plants reach their limit? Is it a supply issue? Or maybe a resource constraint? > Similarly, a fab will use very large amounts of ultrapure water for wafer cleaning and CMP, along with the regular water for things like chillers for process cooling. A large fab can use millions of gallons of ultrapure water a day, as much as a town of 50,000 people, and producing it requires its own specialized plant. This is wild. All of this water so companies can create chips that will ultimately be used to … pump out “advanced” chat bots. I really hope all of this sacrifice is worth it in the end. Climate change is accelerating the loss of drinkable water around the planet. If the best we could do is a slightly better chat bot, then we are doomed.
- jiggawatts 2y agoThe water is not used up, and it doesn’t even evaporate, except for the small amount used in evaporative chillers. The big scary sounding number at the input has a big number at the output. Not to mention that up to 98 percent of the water is reused on site. It just gets cleaned and goes back into facility. It’s a big loop, not an input disappearing into a parallel universe.
- konfusinomicon 2y agoscience be damned, why oh why is all this water pouring into our universe. luckily its not 2% more or we'd all be doomed
- patmcc 2y ago"...as much as a town of 50,000 people" This is always so funny to me. Oh, something that can make enough chips for millions/billions of devices also uses as much water as a small town? That sounds perfectly reasonable. Don't build it in the middle of the desert, I guess, but otherwise it's not a problem.
- kortilla 2y agoThe desert is fine. That’s like a couple of idiotic farms worth of water for a much larger benefit.
- ijidak 2y agoSo, semi-related question. Is Intel buying any EUV machines from ASML. Not asking about High-NA machines, but regular EUV machines? I can't find the answer online although I do see that the New Ireland Fab seems to be in the regular EUV range.
- Aloisius 2y agoThere is no one else to buy EUV machines from. ASML is the sole supplier in the world. Fab 34 in Ireland got its first ASML EUV machine two years ago. I don't believe High NA EUV machines will end up in this generation of fabs. The first one was shipped to Intel's dev fab a few months back.
- langsoul-com 2y agoEveryone has to buy from ASML, and even if someone wants one, there's a wait list. The sheer complexity and costs means there's no real possibility of a competitor. Nor would a billion dollar fab want to experiment with a new company.
- kibwen 2y agoFascinating article, thanks. > (There is a Moore’s Second Law, also known as Rock’s Law, which posits that the cost of a semiconductor fab doubles every four years.) If this were to hold, then in under 30 years a single fab would cost more than three trillion dollars, which itself implies a hard upper bound on node improvements by way of economic considerations.
- meiraleal 2y agoIt's fascinating how precise this math is tho. 30 years ago, Microsoft matketcap was around the same 20bi. Today it's 3T. It would totally make sense for it to happen again.
- tim333 2y agoSam Altman was off trying to raise $7 trillion for chips the other day so maybe he's trying to get ahead to the curve.
- etaioinshrdlu 2y agoOne part really jumped out at me: Thus only a very small number of companies (currently TSMC, Samsung, and Intel) attempt to operate leading-edge nodes, and the industry has shifted to a “fabless” model where companies like Apple and Nvidia design their chips but have them manufactured by “foundries” like TSMC. By pooling the orders of many different chip companies, the foundries can achieve the scale necessary to afford cutting edge fabs. I wonder if AI training will end up being similar in the long-term (it's already partially true today).
- langsoul-com 2y agoIt is already true. The cost of training gpt jumped exponentially per model, in the realm of millions upon millions. There's a good reason why the latest gpt-4 competitor is meta or Google.
- everybodyknows 2y agoChatGPT-4 shows a certain awareness of its own training investment: If asked, it will say it "contains data up to December 2023".
- genericone 2y agoFirst, attempt to build a $10 billion dollar fab.
- jc_811 2y agoWhat a fascinating article. It truly blows my mind just how advanced humans have become. When playing angry birds on an iPhone we tend to forget the insane amount of research, resources, and specialized expertise that has made all the modern tech world possible. I’d also highly recommend the book Chip Wars by Chris Miller for anyone looking for a deeper dive into the history and current dilemmas of the semiconductor industry
- rossant 2y agoExcellent and fascinating article. I knew semiconductor fabs were incredibly complex but this article still blew my mind.
- rmu09 2y agoThe stated tolerances of typical manufacturing processes seem to be off by at least an order of magnitude. Not that it does make sense to give a tolerance number without some magnitude it applies to, nonetheless, 0.125mm stated tolerance of CNC machining is ridiculous.
- adrian_b 2y agoOff in what direction? A tolerance of 1/8 millimeter (5 mils) can certainly be achieved by CNC machining, but not always by the cheapest tools. You mean that it is easy to achieve much smaller tolerances than this by CNC machining? Googling for "CNC machining tolerances" finds at first hit a guide that says "For CNC machining, the standard tolerance limit is set around +/-.005” (0.127 mm)." The next hits show the same 5 mils standard tolerance, but they also mention the availability of better tolerances, like 2 mils or even less, but at higher costs and subject to various constraints on the features to which they are applicable. So it appears that the tolerance written in the article is indeed typical, as claimed.
- rmu09 2y agoThe discussion is somewhat moot, what did bug me is that the article claims those are typical tolerances inherent in the machining / wire edm / molding processes whereas it just may be the tolerance that is applied if nothing else is agreed upon. You can achieve much tighter tolerances if needed, even in mass-produced stuff like LEGO bricks, plastic lenses, iphone / macbook cases, PCBs, etc...
- fisheuler 2y agoCHM(computer history museum) have conducted an oral history interview with TSMC found Morris Chang in 2007. Chang told his personal Odyssey in this interview, from inner China to Taiwan, studying and working at US, PhD application refused by MIT, but luckily he choose a fast rocket track in the industry, after he accumulated enough wealth, industry insight, connections etc... He returned to Taiwan to found TMSC, else is the story. Also he compared Asian and American engineering styles: "I think the Asian engineers tend to be more methodical, they tend to be more studious, more orderly engineers; whereas, the U.S. engineers tend to be more innovative, but they tend not to be as methodical and orderly as the Asian engineers. ...... On the other hand, I also think that one group can accompany another group, you know. You can have an innovative group but not so methodical and so on, and then you can have a methodical group maybe not very innovative accompanying each other. " https://archive.computerhistory.org/resources/access/text/2013/05/102658129-05-01-acc.pdf https://archive.computerhistory.org/resources/access/text/20...
- 11thEarlOfMar 2y agoFor comparison, the cost to build the World Trade Center was $2 Billion. BTW, having spent 35 years in the semiconductor industry, this is the best 'layman's' description of what goes into chip fabrication I have seen. A perfect blend of description and illustration without over simplifying.
- simonbarker87 2y agoI was thinking the same. My PhD was in SiC adjacent stuff and tbh I could have done with this explainer back then to get me up to speed.
- mwarkentin 2y agoIt doesn't look that hard: https://youtu.be/vuvckBQ1bME?si=eCGILMLlZHZnztCR https://youtu.be/vuvckBQ1bME?si=eCGILMLlZHZnztCR