Samsung Foundry: New $17B Fab in the USA by Late 2023 | Hacker News Reader

Samsung Foundry: New $17B Fab in the USA by Late 2023

anandtech.com

623 pointsmanojkr376 comments

I see a lot of people in this thread who don't really know much about Samsung's semiconductor presence in the USA. This expansion has been in the works for years. Samsung has a 100k wafer per month fab in austin texas. They are negotiating for tax abatements for expansion and using alternative sites as leverage.

They are not building a new fab they are expanding their S2 fab in austin texas. There was a permit application that was released a few years back that shows the true scale of their expansion plans.

https://www.swf.usace.army.mil/Portals/47/docs/regulatory/pu...

If this goes through it could made Samsung's S2 fab larger than even TSMC's monster gigafabs.

If anyone has seen the Austin S2 campus before, you'd know that it has a massive blacktop to provide parking for its thousands of personnel.

It would be nice if as part of the conditions for application approval, Samsung were asked to commit to erect awnings for solar collectors integrated into covered parking and strongly encouraged to invest in making solar more cost effective.

As a side note. SAS recently bought 100s of acres around their property probably to expand parking.

Sorry for the ignorance, but what's SAS? DDGing that word might not get the appropriate result.

assuming SAS software, an analytics software company

https://en.wikipedia.org/wiki/SAS_Institute

Thank you! It says it's the largest privately held software business. Why is it that it doesn't go public?

Why go public if you already have money?

Why _should_ they go public?

I once visited the SAS campus in North Carolina, when I was working for another well-known North Carolina headquartered software company. Very nice place. Made me quite jealous.

Samsung austin semiconductor (the name of the fab)

Idk why I see downvotes here for a legit comment and concern. Parking lots are absolutely awful.

First, cars. Enough said.

Second, they take up a lot of space with low utilization.

Third, heat.

> Idk why I see downvotes here

discussion derailment

turning every thread into a debate about your pet sociopolitical issue is unhelpful.

>First, cars. Enough said.

No not really. Cars are wonderful, not everyone who lives on this planet wants to share space with strangers.

Cars are wonderful and useful when moving, less so, when stationary. Allowing a second way to use the area taken by the cars is reasonable — a car parking does not need access to the sky, until we have the fabled flying cars.

Nobody said cars have no benefits. But they do have very real drawbacks that are obvious at this point, they just aren't borne by the owner exclusively.

Cars as the exist today for the most part are the best method we have for increasing mobility and economic independence for those that don't live near or can't leverage public transportation (and sometimes in those cases too). They are still objectively awful in many other aspects though.

Not everyone wants to pay taxes either, but sometimes sacrifices will need to be made so that we don't ruin this planet. Sure having personal space and full autonomy in movement is nice, but it may have to be realized in different ways as we start to see the damage we are doing to this planet.

A planet with only mass transit is not a planet I want to live on. Many will fight you to the bitter end on this.

I think the crux of the argument here is that we will not have a planet to live on anymore.

> A planet with only mass transit is not a planet I want to live on

Pitiful strawman. I've never heard someone argue to have "only mass transit". More focus on common transports, cycling cities? Yes, please.

Of course, many like you will still fight it "to the bitter end", whatever that means.

Most people lived and died within 20 miles of where they were born until extremely recently. In that context, this sort of idea is kind of amusing given we have many different parameters (such as the viability of life on earth) to consider in solving social problems.

Let’s follow the science.

Cruise ships pollute more than cars. But you won’t see any green politician discuss that. Why?

https://www.thedrive.com/news/28469/carnival-cruise-ship-fle...

Because cruise ships do their polluting primarily in international waters and are registered under flags of convenience. They have separate fuel tanks for in western countries' jurisdictions to fulfill the regulations that these green politicians have already passed.

This sort of non-compliance compliance infuriates me. Any ships with dual tanks should be banned from western waters.

Then they'll invent one with triple tanks ;)

But you won’t see any green politician discuss that

Do you live in a city that gets a lot of cruise ship traffic? Because in my experience a lot of green politicians are very keen on banning cruise ships.

Source?

That article says nothing about CO2, which is by far the biggest man-made contributor to global warming (and probably what the person above you was referring to)

> Cruise ships pollute more than cars

Sulfur oxides (SOx), yes. Carbon dioxide (CO2), no.

I recall during hurricane Harvey in 2017, the news reported that part of the reason for the innundation was due to so much green space and dirt being covered up by impermeable artificial surfaces, which caused the water to have no where to go except to accumulate and flow into the city.

If you choose to live in an over crowded area like a Coastal City, this is true. Texas is huge with a lot of space; so while yes, gas powered vehicles might not great, consider for a second there a much larger world than the small bubble you live in.

If the long-term return from solar cells were higher than the cost involved, Samsung would do it themselves. Since they don't do it, that means it is not efficient enough. So why force them?

There are plenty of requirements in Conditional Use Permits that force entities to do things they wouldn't otherwise do themselves. These are generally to mitigate the negative impacts on others. For example, onsite stormwater retention and traffic mitigation.

> Since they don't do it, that means it is not efficient enough.

This assumes perfect market efficiency -- that everyone has perfect information and is perfectly rational. In my experience, that isn't the case.

Additionally, Samsung is asking for a 25 year property tax waiver. It's a negotiation.

> There are plenty of requirements in Conditional Use Permits that force entities to do things they wouldn't otherwise do themselves.

Ah, let's remove those requirements.

> For example, onsite stormwater retention and traffic mitigation.

Why do you think "entities" would not do these things without being forced? Do these "entities" not benefit from better traffic and managed stormwater?

> This assumes perfect market efficiency -- that everyone has perfect information and is perfectly rational. In my experience, that isn't the case.

OK, then let's fix that.

> Additionally, Samsung is asking for a 25 year property tax waiver. It's a negotiation.

No, it's not a negotiation. Negotiation can occur between similarly empowered entities. There is no negotiation when one side has the power to lock the other up at gunpoint.

"Why do you think "entities" would not do these things without being forced? Do these "entities" not benefit from better traffic and managed stormwater?"

Before you advocate muh free market, read about history of urban panning - we tried it, most houses didn't even have sewers. I don't want poop on the streets.

More recently in UK over 100 people burned to death because a massive skyscraper was built with violations of fire safety, and then plasteres with more flamable material.

Do these entities not benefit from not having the building they manage burn to the ground?

> I don't want poop on the streets.

I doubt Samsung fab employees are going to do that. I don't get the relevance of this to our discussion.

> More recently in UK over 100 people burned to death because a massive skyscraper was built with violations of fire safety, and then plasteres with more flamable material.

Again, this is irrelevant. A parking lot, just by existing is not a fire risk. And even if it does pose some fire risk, putting solar panels and the required electrical equipment on the parking lot will only increase that risk.

> Do these entities not benefit from not having the building they manage burn to the ground?

No, they don't. Ask the owners of that skyscraper. What's your point again?

Negative externalities should be regulated to some degree. Otherwise we could have a tragedy of the commons situation.

I'm usually in favor of free market-influenced methods such as taxing negative externalities and subsidizing positive ones. So instead of forcing a company to install solar panels, the city could provide a tax incentive for everyone in the city to do so. The overhead of managing that could become large though, so I understand why sometimes more small scale and direct individual negotiations can be good.

>No, it's not a negotiation. Negotiation can occur between similarly empowered entities. There is no negotiation when one side has the power to lock the other up at gunpoint.

They wouldn't say do this or you'll go to jail. They would say do this or we won't give you a giant tax cut. It would be unconstitutional to throw people in one company in jail for something that's fine for others to do.[1]

[1] https://en.wikipedia.org/wiki/Bill_of_attainder#Constitution...

> They wouldn't say do this or you'll go to jail. They would say do this or we won't give you a giant tax cut.

"Won't give you a tax cut" translates to "pay us a percentage of your earnings, otherwise we will throw you in jail." So, I'd like to disagree here and say my original assertion was accurate.

> Negative externalities should be regulated to some degree. Otherwise we could have a tragedy of the commons situation.

I am not convinced this applies here. Solar panels over the parking lot sounds like a solution to an invented problem. Parking lots are not harming the environment. If Samsung is paying for the land, they should be able to do whatever they like on that land as long as it doesn't harm someone else.

This sounds like the typical anti-car communist mindset that I've seen many times over here on hn, which I find highly weird. Cars are probably the best innovation humanity has done in the entire history of the species, and I guess people are triggered by that fact?

>"Won't give you a tax cut" translates to "pay us a percentage of your earnings, otherwise we will throw you in jail." So, I'd like to disagree here and say my original assertion was accurate.

Your worst case scenario happens to all companies in the absence of the deal. So your argument seems to be more that taxes themselves are bad rather than the deal is bad.

>I am not convinced this applies here.

Sure, I might agree with you there. I was more replying to the part of your comment about removing all requirements that force companies to do things they wouldn't do themselves.

the urban heat island is real, and austin is hot enough already, so thanks for the very humane suggestion! the screen on my phone used to turn black if i forgot it in my car at work. sometimes i left frozen tamales on my dashboard and came back for them at lunchtime.

in my student coop, i once looked up from my studies to see that my clock/thermometer said it was 110F on my desk. so yeah, if you're going to cut down all the shade, can a bro at least get some awnings?

I have only been to Austin once to consult on an OpenStack install - Cray was in the same building as my client...

But when I was packing for the trip - I was like "I am going to texas - so I don't need a jacket and only packed short sleeve shirts and brought a pair of shorts!

When I arrived in Austin, IT WAS FUCKING SNOWING

I was so damn cold.

Me too! I got stuck there because taxis weren't running to the airport due to snow. Had a rad time... everything was covered in snow, not a car on the road. Apparently the locals hadn't seen anything like it, so maybe a freak one. This was late 2000's, can't remember exactly when

We had a good snow last month - about 6 inches. But yes that kind of thing is so exceedingly rare (about once every 10 years) that the entire city economy basically grinds to a halt so everyone can go outside and play in the snow. The last time that happened was probably the incident you are describing.

Such vast, paved, surface-level parking lots are a distinctively American thing.

In Europe, for example, you’re much more likely to find covered, multi-level parking structures once a parking lot exceeds a given size.

If Europe had the space that exists in and around Austin, they would do the same thing. It's simply far less expensive to build surface lots.

Yes, this is commonly done in areas where land is cheap, I'm thinking of rural train station parking lots, regional and small airport parking lots.

Exactly. It’s not like you see massive flat lots in Manhattan

There is this fairly efficient 10-level parking garage in Philadelphia, Arch Street. It is essentially two 10-floor towers connected at the top. They saved room by making driveways one-way. So you enter, you have to drive 10 floors up, drive across the roof, and then drive 10 floors down to the exit. You can park wherever along the way, but you always end up driving all the way up and down.

That sounds terrible. I feel like I'd get nausea or dizzy. Thats essentially driving in a circle 20 times.

Itd be okay if there were 2 mid level cross connections

The Sydney Opera House is like this too, it's a 12 story double helix excavated from sandstone.

I'm sure it was fascinating to design, but it's an absolute nightmare to park in.

Multilevel parking is about $10-15k per space. An asphalt lot is $1.5k.

You also don't pay land tax and stormwater drainage levies on an undeveloped parking lot.

In Europe, people are more likely to engage in active transportation. You´ll find more public transit, cycling and even walking. Public transit is just a normal way for anybody to get around, not a mobility fallback for poor people.

That is changing though. I've not seen the inside of a train for years and I'm happy with that TBH.

”A developed country is not a place where the poor have cars. It's where the rich use public transportation.”

Gustavo Petro

Personal anecdote is not "change"

Some parking places in Europe are also covered in gravel or permeable pavement and have trees between rows rather than being made of tarmac.

They are trying to strong arm the locals into basically saying they (Samsung) pays zero property taxes to the next 25 years. https://www.statesman.com/story/business/2021/02/04/samsung-... . I hope Austin passes on it if Samsung can't do any better than that.

I hope then you can face the desperate residents who don't have the luxury of having a cushy data collection yuppie job like most of us here do then for missing out on a decent wage like and tell them "no it was for your own good, trust me!"

No it sets a terrible precedent. Businesses need to support their community. I would willing to bet that the city council will not let it happen. They may cut a deal but it ain't gonna be 100%. More like on the order of 10-25% like other companies get. Corporations are too greedy. I contract for a living and through local contacts never go more than a month between contracts. The job market here is still pretty great for people who have solid backend/embedded skills, in general it's solid for most people. Those other people can go work for the other big guys or the new Tesla plant who actually cut a reasonable tax deal.

doesnt samsung create thousands of jobs in the community with this? not sure I share your outrage. this isnt about some one trick pony YC startup hiring a dozen backend devs

Why should labor pay for all the externalities created by capital?

Did you not read the title? It's $17 billion dollar investment creating thousands of high paying jobs which in turn will create more jobs and create more capital in community. Why are you so against this?

This isn't some VC backed zombie healthcare software company hiring a couple of developers per year and not getting to pay taxes. This is a major corporation investing billions with profound impact on the community.

> Why are you so against this?

Why should the people of Austin, Texas gift millions of dollars to a company that made a profit of over $8 billions in the last quarter alone?

I've seen places refuse to budge and lose out the investments

I've also seen Intel refuse to negotiate price for chips with Apple for the original iPhone and look how great that turned out for them and TSMC

The number of cities losing out does not change. If there are X competing cities X-1 cities will always lose out.

This is a company asking a city, that will see no income taxes from the at most 2,000 employees, permission to pay zero taxes at all to the city in exchange for the honor of housing a future EPA Super Fund site. Sure, the city will get some sales tax revenue from the 2,000 who will almost assuredly live outside of the city. So they'll get what, 6.25% of a $10 lunch from maybe 30% of workers per day? So they'll expect about $100k/yr in tax revenue in exchange for more road utilization and a future EPA Super Fund site. That doesn't seem like a good deal for the city. And that's assuming 100% of the sales tax goes to the city which it doesn't. Most of that goes to the county and state.

Why no income taxes on the employees?, they are part of the community, why arent they paying in? How does the city even fund itself then?

Texas does not have an income tax. Texas has property tax and sales tax.

Who said I’m against it? I merely raised the question.

> This is a major corporation investing billions with profound impact on the community.

Sure. Is that impact the same for everyone in the community, though? Or is it ultimately better for some, and not others? Could it also be that it may be a net negative for some members of the community?

Also, as noted by others, Samsung already has strong incentives to build this facility in Texas, and has wanted/needed to build it for a while. This latest gambit is an effort to force local government to sweeten the deal.

The investment is creating the jobs. The tax discount isn't creating jobs. It's just moving them around. Do you really think Samsung will give up if no city gives them a tax break?

without investment there would be no jobs or taxes to collect in the first place. With this single move, texas gov will be able to collect more taxes from thousands of people making 100k+ and downstream suppliers also hiring and making more money.

Samsung can easily find other cities to produce in at lower cost outside America. Why is being competitive for once such a big issue here? Without the tax benefit Texans won't have thousands of high paying high skill jobs.

The tax break adds up to about $40 million a year which pales in comparison to the benefits Texans will receive.

> Samsung can easily find other cities to produce in at lower cost outside America.

No they can't. Austin and the surrounding area are key silicon fabrication regions. There are very few places in the world that can compete, and it is possible that they are forced to produce within the USA due to import/export restrictions on what they are manufacturing.

This is a situation where the decision is already made, but it can't hurt to ask for tax break. Because hey, free money.

Kind of like how Amazon "passed" on building out an NYC headquarters because the city wouldn't give them tax breaks (after locals protested), only to come back six months later and begin construction of their NYC HQ.

Sure, the tax thing is probably one of the maor component here. In addition to local taxes, my understanding that SAS's current plant is in free trade zone where they also receive exemption for import/sales taxes, etc.

The whole US semi industry is worried that the lion share of semi production is stlil occuring in Asia, not in the US -- which is probably why Trump admin started twisting TSMC and Samsung's arms few years ago:

https://www.semiconductors.org/wp-content/uploads/2020/06/20...

It's an interesting question: incentives. Samsung can't negotiate away wages like they can land taxes.

Look at it from the local government prospective, they could get 0 in tax income or they can settle from the tax income of thousands of new high income jobs.

Do gigafab manufactures like that really create high income jobs? I assumed it was mostly low income close to minimum wage kind of thing.

On a similar tangeant, how are these able to compete with similar manufacturing done in Asia or Africa?

I don't know anything about how they are run, but I would assume you need a large number of on-site engineers and process overseers. What are the low-income jobs? Security?

I thought it was mostly assembly line workers putting parts together, with machines doing the rest. Kinda like what this photo shows: https://bit.ly/3a8uRFM

That’s board assembly, not chip fabrication.

So you admit you were ignorant. That's a rarity on here. People will defend their wrong views here because they don't want to hear they aren't a polymath.

This is how it looks like now: https://www.extremetech.com/wp-content/uploads/2018/12/TSMC-...

Machines feed each other via conveyor belts, everything automated, tight control loops check/move things around.

Notice the difference in human presence.

Average salary is probably around $100k. Even the techs with high school educations can make in the range of $40-60 an hour and generous overtime.

>On a similar tangeant, how are these able to compete with similar manufacturing done in Asia or Africa?

Because it is all high skill labor. Techs that are able to strip down a 10 MM machine and identify the source of sub micron particule contamination are worth their weight in gold.

>>Average salary is probably around $100k. Even the techs with high school educations can make in the range of $40-60 an hour and generous overtime.

Do you have a source for that? I'm surprised as when I contracted at the Intel fab in PHX, the workers in the actual fab were not making anything close to that. Management maybe, but not actual line workers. The standing joke was "they were working for the stock"...

Were they working for the stock?

I found this which has a manufacturing plant section. No idea how accurate: https://www.zippia.com/intel-careers-6055/salary/#by-job-tit...

Hmm - that's interesting. If the "Intel Plant/Manufacturing Salaries" section is true, process technician and production operator (which I guess are the people that run the line?) make quite a bit below $100K. Oddly enough, it seems like the software jobs are better paid then the hardware ones?

Thanks for the link!

How is that odd? Software salaries have been outpacing hardware ones for at least the past 1-2 decades. When the big tech companies are paying over $100k for software developers fresh out of college, supply and demand is probably forcing this state of affairs.

I just would have figured a hardware company would be focusing on the hardware personnel.

After all, there are plenty of places that don't pay FAANG level salaries for software developers. I still regularly see < $60K software jobs advertised from smaller companies.

They may be costing the company the equivalent of $40–60/hr, but more realistically, techs are starting at < $20/hr and have to spend years on the payroll of a third-party staffing company.

That's a good point, $40-60/hr is the pay for seniors level techs. Entry level techs are making much less and are employed through staffing companies.

In Taiwan, an average semi technician salary is $38-$40k a year

> or they can settle from the tax income of thousands of new high income jobs.

There is no income tax in Texas. Even if there was, it would go to the state, not the county or city.

Property and sales taxes, then.

Dude, the whole thread started with a comment about Samsung trying to pay no property taxes.

OP means Property tax from employees. Because it creates a ton of jobs

There are plenty of businesses that create tons of jobs and pay their taxes.

Tax deductions aren't meaningful for most companies. Samsung is pretty limited on where they can actually build this plant. It's not like Hattiesburg, Mississippi is awash with people experienced in silicon fabrication. So the company would have to pay a lot of money to import people from other regions. But Hattiesburg isn't exactly top on the list of places where experienced technical professionals wish to live.

The city counsel should play hardball with Samsung, since they will cave. This is one of those situations where it doesn't hurt to ask, but the decision is already made.

But what says those people will live within Austin city limits (ba dum tiss) and pay property taxes to the government that gave out the tax break?

Surely a sizeable number would commute?

Also, collecting new property taxes from a corporation would allow the city to keep rates lower for residents.

texas doesn't have income tax.

The local government can lose money on this deal even if it leads to more jobs.

Someone has to build the roads, run the schools, pump the water, etc.

A tax cut for Samsung is effectively a tax hike for everyone else.

> A tax cut for Samsung is effectively a tax hike for everyone else.

That is a remarkably zero sum way to think about it.

zero sum implies everyone loses in the long run.

Samsung benefits. Texas state benefits. Workers benefit. Community benefits.

Please enumerate all the ways the community benefits. Be mindful to include externalities in your ledger.

These business can act as anchors to attract skilled talent to the region. If that happens, then a small no-name town can really prosper.

But the reality is that companies don't really want to be the ones anchoring a small town. The sacrifice is too great in terms of limited talent pool, and lack of amenities necessary to attract talented people. Maybe after 20 years, the town grows into something significant, but at that point, the market has shifted so much that it probably doesn't matter.

In this situation, the decision is made, regardless of tax incentives. Samsung can't just plop one of these plants anywhere. The plant needs to be in a place that has a large enough pool of talent, the amenities (schools, restaurants, housing, etc) that attract new talent (ideally without Samsung having to pay for relocation), and have the logistics infrastructure to support business.

I've seen plenty of large companies negotiate for aggressive tax breaks locally, only to up and move to places like NYC, LA, or another high-tax locale a few years later because they either can't attract talent, or senior executives don't like living in cheap midwestern cities.

Negative sum. Corruption destroys wealth. The vast majority loses a lot so that a few can benefit a little.

The difference between creating wealth and merely transferring wealth (aka theft, grift).

Companies can and do build their own roads -- we aren't exactly talking about building interstate highways. Schools would be paid for by SAS employees and contractors' paycheck and additional local property taxes (for those who have to relocate from another place). I highly doubt that SAS is going to get free water.

Parmer Lane, the road that SAS sits off of, is a parking lot in the morning and evening. I have to think that Austin is planning on expansion eventually if yet more people are driving on it, because that's what Texas does - build wider and wider roads, public transportation be damned.

The precedent was set long ago, and if Austin does not take it I can assure you some other city will..

Why, it is easy... a new factory (or a factory expansion) means more jobs, more jobs means more people, more people means more Home and more secondary businesses all of which the local government will get taxes from.

In principle I agree, but in reality Austin will make more tax revenue from these other properties and business than the amount of the abatement

Source? Just like this comment in the story - " Nonetheless, the deal still makes a lot of sense for the state". Numbers, please?

Samsung has been around Austin, TX since the late 1990's and invested well over $15B. I think the state and county officials have enough dataset to evaluate whether Samsung's new expansion would benefit their community or not.

In this case, it seems like a viable investment by the state. We're talking about a leading node fab, not a nebulous "LCD manufacturing center."

That's a bit harder to vaporware, and more importantly, why would you?

That worked out great for Wisconsin residents recently

For every one deal that went sour, there are hundreds of successes.

Additionally, comparing a rushed Foxconn plant to a Samsung fab facility, in an area where Samsung already has roots, is dishonest to say the least.

Source? How does one even begin to measure the opportunity costs of pitting politicians with short term incentives against each other? Many of the supposed benefits aren’t even able to be calculated.

The source is the hundreds of large corporations that have came to a remote location away the already larger and established metro areas with international airports nearby, so your usual suspects of LA, SF, NYC and ATL.

There almost always is some incentive package to attract the businesses. My city, Charlotte, has 10 examples alone. Neighboring Charleston and Columbia in SC have quite a few examples of theirs, namely in Aerospace and Auto Manufacturing.

How can one figure out if they would not have come anyway due to lower wages/land prices?

What benefits? Every fab becomes an ecological disaster over time. They should be assessed penalty rates not given discount rates to cover the future cost of cleanup.

Wisconsin's tax break was conditional on Foxconn meeting key hiring/production metrics. Foxconn had a long trail of failures building viable manufacturing business outside China where the CCP provided everything from free land, to free buildings, to unlimited supply of young, obedient, unskilled labor pool for rural China.

Samsung on the other hand has been in Austin, Texas for almost two decades, investing well over $15+B so far.

I'd like to sell you a car, since you're not the negotiating type.

It’s a prisoners dilemma situation - cities would be better off if across the board they just applied the rules evenly, but there’s always a place that’s willing to sacrifice that for its own benefit.

Its not just residents, its a complete violation of free market.

How are startups and family businesses meant to compete when the megacorps get special deals, pay no taxes and get bailouts? This is how you get monopolies and oligopolies.

I worked at SAS a couple of years ago. The majority of jobs they will be offering are contractor positions with no benefits, and pay that in no way compensates for the crazy schedules they run. Starting pay was about $18/hr in 2018.

They aim for ~2/3 of their technicians being contracted positions. A tech might work there for YEARS before a conversion spot finally opens up. It's an absolute travesty and should not be allowed, IMO. Why should a company be allowed to have people doing the exact same jobs but only offering benefits to a select few?

Isn't that like exactly the sort of thing the interstate commerce clause was originally meant to stop? There shouldn't be such competition between states or local governments, whether real or just negotiation bluff.

No, states compete on tax structures all the time. Look at the Amazon headquarters boondoggle a while back, states were falling all over themselves to offer the best tax/infrastructure packages.

Then just have places like San Diego always win because they have nature on their side?

But really I think the taxes here would technically count as within-state commerce, as they would be taxing a resident business if it is built.

Eventually San Diego fills up. Why not let efficiency dictate these things?

Filling up is winning. The population will stop growing soon enough and it becomes a zero-sum game.

There are still people moving. Internal migration is bigger than external. Cities develop even if the population is roughly stable. Generations are not the same size, etc.

Also the economy changes.

It cannot really become zero-sum in my view. Could you explain how you think about this?

How is that different from the deals that pretty much every other large company looks for when locating a new facility? Amazon pretty much had local governments humiliating themselves in their HQ 2 'contest' not too long ago.

It's not. And while there's opportunity to make a big construction more efficient by handling permitting/approval in a bundle (combined with some of the necessary developments in traffic/transportation, housing), that doesn't mean cutting corners and just crippling the state/city budget.

One of my favorite things to do when I am depressed due to negativity is to read Hacker News comments when Dropbox launched and the NYTimes story on space flight being impossible.

I am going to add this submission to that list.

That fab in Texas was the one used to manufacture the early Apple A-series processors. The media and most pundits generally assumed they were made in South Korea, because Samsung.

It's kind of odd that Anandtech seems to make a lot of fuss about Samsung's tax break whereas Anandtech's recent article on TSMC's Arizon plant mentions virtually none of the same tax break highlighted for Samsung there. Is Apple still hiring away Anandtech's key editors?

https://www.anandtech.com/show/15803/tsmc-build-5nm-fab-in-a...

Comparing an article by a different author from a year ago. gg

> New $17B Fab in the USA by Late 2023

Impossible.

Leads times on fab hardware are on all time high.

Not to say that construction, and setup itself would be extremely challenging in such timeframe.

And not saying that US has no supply chain locally for modern fabs. Intel famously flies a lot of its consumables from abroad.

Samsung will have similarly to transport its consumables by air if they go with this plan.

doesn't samsung have the capability to create its own fab hardware? if they are the supplier to themselves, wouldn't it be reasonable that they can prioritize delivering to themsleves

ASML is the only supplier for EUV machines. Samsung can’t just create those easily themselves (or they would).

Is there any reason they couldn’t just start with multipatterning 193nm immersion? Maybe even ship that equipment out from a facility being upgraded.

Also, ASML’s EUV light source division was fairly recently (2013) acquired from... California. Cymer, LLC.

Retooling of anymuch modern fabs is not economical.

I don't see much retooling for anything below 45nm, if not 65nm.

The type of equipment ASML produces generally gets brought in at a specific time during the construction. They kinda build the cleanrooms around ASML's machines instead of bringing them in afterwards.

Retooling for EUV (or any new generation of such equipment) is very hard. Most (if not all) foundries build new factories rather than retooling old ones.

> They kinda build the cleanrooms around ASML's machines instead of bringing them in afterwards.

I know they're massive, but do you have a source for this? I was under the impression that construction and tool installation are two distinctly separate phases of construction due to the requirements for an immaculate environment. Also, my understanding is that companies like to bring the tools in as late as possible to shorten the delay between capex outlay (tools are expensive) and first production.

My source for this is my professors at the TU/e in Eindhoven and guest lecturers from ASML (TU/e is the closest university to the ASML campus in Veldhoven).

Unfortunately I don't have a more reliable source on that though.

multipatterning DUV will get you up to a point if thats what you were asking. You cant reach 5nm with that. (if that were possible you wouldnt see everyone jumping on ASMLs euv bandwagon)

As for lightsource, sure, that was/is a US company, but there's a lot more to a machine then just the light source.

> Is there any reason they couldn’t just start with multipatterning 193nm immersion?

Time, yields, line size.

Quadruple patterning quadrupples the lithography steps counts. And quadrupples thermal cycles. Which adds more than 4x impact on the number of bad wafers.

100% - 2% * 4 != 100% * 0.98⁴

You need 4x the equipment to complete the same volume of wafers in the same time, which means 4x the line length.

not that I am aware of (read: no unless someone proves me dumb)

Major supppliers include:

Nikon

ASML

KLA-Tencor

Applied Materials (AMAT)

Lam Research

the timeline they are suggesting isn't literally impossible as a matter of phyics but is, to me, laughably unrealistic and likely done to try and extract tax incentives.

Unless they have already broken ground in Texas...and are just shopping for a better deal

> Major supppliers include:

Add like 100 more companies to the list.

A fab is not only a lithographic line, but a ton of other very unique, industry specific equipment.

Air handling for fabs stuff is up to 24 months lead time, and has no use in any other industry.

Ultra clean water plants — 9 to 18 months. Just two vendors on the planet.

Gas plants — take few years to construct by themselves

Chem plants — again years long lead time equipment from only one vendor in the world

On site mask shops — equipment, and processes not much cheaper, or less scarce than the last gen chip making gear itself.

Cleaning stations, degassing stations, industry specific SCADA hardware, AMHS... (https://youtu.be/XKXZT-BBUEE)

Post-backend stuff...

The list will go on for a few screens.

Even a Pencil (I, Pencil) takes thousands and thousands of people, I'm guessing it's on the order of millions for a big fab.

I’m honestly surprised we have not seen a blank wafer shortage. I remember the craziness post 2011 earthquake...”well how many wafers do we have”

There is a blank wafer shortage, but - strangely - for 200mm wafers (not the state-of-the-art 300mm ones)

https://www.extremetech.com/computing/318554-a-massive-chip-...

Interesting. Tells us something about the maturity of the products seeing unanticipated demand.

I wouldn't even have thought to look at 200mm for a shortage. That is an (in hindsight) obvious corner of death for these processes/products I wouldn't even have thought of.

Samsung has a long-term capital and capacity planning process. Isn't it possible (if not likely) that they would pre-order equipment as part of capacity planning without having already decided where it's going to be deployed?

it's possible but it's going to take at least a year to build the building, another year or so to install, and another year to qual (these are ballparks).

Not to mention they have to staff the thing, train the staff, etc.

All of those things can be parallelized. It's one thing to say "opening by the end of 2023 is unlikely", but to say it's impossible is unnecessarily dismissive.

While I'm also optimistic on Samsung's ability to execute some version of this plan, those things can't be parallelized. You can't set up a complex water delivery mechanism until after the building is completed and the cleanroom has been built to spec and tested, and you can't calibrate the gigantic ASML machines until after that.

But you can train the staff in Korea (or at any other running fab) and have them ready to go the day it's qualified.

You can’t parallelism those... you need a building to install tools in it. You need the tools installed to qual them. You need the tools quaked to qual a process...

From personal experience I would posit scaling semiconductor manufacturing as 10x to 50x harder than getting a process flow to work in a development fab. Ramps are absolutely killer.

That seems like a reasonable timeline, which means the fab will be online by 2023, just like the article says. Mass production won't start until 2024, though.

>Samsung has a long-term capital and capacity planning process.

This. Samsung is another company like Amazon that takes long view and flywheel to another level. And to answer question to comment below. Yes at one point Samsung was trying to make their own EUV Scanner, if you consider their NAND and DRAM scale it should comes as no surprise. And Samsung has their own Chemical Subsidiary along with dozen of others for their Foundry. They are taking vertical integration to an extreme.

And that is speaking as someone who dont particularly like Samsung. But Credit where Credit's due.

Let's also not forget their digital imaging line, specifically their foray into Mirrorless cameras

Sure, they shelved it (Hard to compete with established players who can spread R&D over volume), but people were pretty amazed at how well Samsung did entering that market from a quality standpoint.

Why does that matter? Because that indicates they have plenty of skill in lens manufacture, which is another critical component to a good fab.

Makes you wonder why one of the major semiconductor equipment companies doesn't team up with Canon or Nikon and compete with ASML/Zeiss.

Canon and Nikon already makes their own lithography machines:

https://global.canon/en/product/indtech/semicon/

https://www.nikon.com/products/semi/

That's right. Deep UV. They all but threw in the towel on EUV.

But the difference is that ASML + Carl Zeiss absolutely run the market.

Nikon has about 10% market share, and Canon 5%. The rest is basically ASML.

The thought on my mind is that any one of the major American semiconductor equipment manufacturers could partner with either of the two Japanese optics houses. Canon is worth about US $25B, and its semiconductor business generates about a fifth of its revenue. It's big enough to be useful, but small enough to be acquirable.

comparative data point:

Intel's last fab build took 9 years to go from green field to producing wafers (note...faster is possible)

https://www.tomshardware.com/news/intels-long-awaited-fab-42...

Some more realistic notes because I sound a little hyperbolic:

2011 - break ground

2013 - finish basic construction

2014 - stop progress

2017 - start install of tooling

2020 - production

Intel does not run at Korean speed.

Samsung's Pyeongtaek Line 3 fab broke ground in 09/2020, construction expected to be complete by 2H/2021, mass production by 2023. This is not an unreasonable speed for Samsung.

Of course, the slower rate of work by American construction companies could slow this down.

I would bet money on this one. No first complete wafers in 2023.

New fab in 4 years is manageable. It may have been in plans for some time, they just chose a new location now.

They build it in Arizona where Intel has several fabs and where TSMC is building their medium-sized 25K/month 5nm Megafab.

25k starts a month is not a megafab

It's a megafab. It's not a gigafab.

megafab means medium capacity in TSMC's terminology.

> There is an intrigue about the new fab though: Samsung hasn't stated which process node it will be designed for.

Let's restate the question. Which node can they get up by 2023 and how much of security critical U.S. consumption can that cover?

> how much of security critical U.S. consumption can that cover?

What US can't source for F-35 locally are Xilinx FPGAs at either 65nm, or 40nm nodes.

The question is not in node size as such as having an FPGA process capability.

FPGAs on generic CMOS don't work great. The ability to tweak, and optimize the process specifically for individual design is huge for FPGAs.

Intel with their CPUs are very much in the same situation, but in reverse. They cannot move from "hand polished" proprietary processes they've been developing for decades to anything else.

What process is Xilinx using; not CMOS?

CMOS, but likelly with much more differences from CMOS for mainstream logic.

I doubt that. FPGA's are just digital circuits. They don't use any onboard "ROM" the vast majority load their programming from an external chip. Where are you getting they don't use standard CMOS tech?

Typical SRAM FPGAs use a conventional process but there are architectures that use floating gates and OTP antifuses that need specialized processing.

And? Xilinx+Intel is 90% of the market. They're the only ones who actually matter. And I can tell you from experience that defense is considering almost no on else.

> [Intel] cannot move from "hand polished" proprietary processes they've been developing for decades to anything else.

Later this year TSMC will be producing Intel CPUs on 5nm. Clearly contradicts your assertion.

That's the rumor, but it's still unsubstantiated.

True, not confirmed by Intel, but backed by fairly substantial sources: http://www.trendforce.com/presscenter/news/20210113-10651.ht...

And: Intel CEO Bob Swan said the company had identified a "defect mode" in its 7nm process that caused yield degradation issues. As a result, Intel has invested in "contingency plans," which Swan later defined as including using third-party foundries.

Prior art: https://www.google.com/amp/s/amp.theguardian.com/us-news/202...

Foxconn promised a giant fab in exchange for massive tax breaks, and then just took the tax breaks and did nothing. Any reason to expect this is different?

Foxconn was an obvious con from the start. If you go back and look at contemporaneous reporting, even the "MSM" reporting was a bit incredulous - it just was really obvious, with only certain politicians claiming otherwise.

Which is not to say this won't become a cesspool, too. But at the very least, I don't see scummy pols waving wads of cash at Samsung right now.

>then just took the tax breaks

They didn't get the tax breaks because they didn't meet the targets.

From your article:

After Foxconn failed to meet its job creation targets, Wisconsin’s governor, Tony Evers, last month pulled a deal that would have handed the company nearly $4.5bn in incentives for completing its plans.

I absolutely don't understand Foxconn's game there. Was this just a scam from the start? What was the upside for them? Or did they actually intend to follow through and just failed?

Pretend, delay, outlast.

My read on the situation is they saw the opportunity to get subsidies and so negotiated to maximize what they were given. I imagine it comes from some maximalist position that says that if you plan to do X, you need to say you'll do 10X so you never 'lose out' on subsidies you might have been able to take advantage of in the good years.

A lot of the reporting at the time emphasized that Foxconn had a history of doing this in other countries. It seems like a negotiating tactic that evolved in an age of cynical political environments where everyone's proximate goal is to announce a 'deal' with the full expectation that they won't be in a position to take credit (or blame) for the quality of that deal.

I don't think Foxconn saw this as screwing anyone. I think they saw it as doing business the way they typically do in China.

What would have happened if this deal had been done in China?

No one would have reported on the failure. The local jobs numbers would have "jumped." Infrastructure would have been "built." Everyone would have gotten a pat on the back and/or promotion for hitting their economic targets.

The weakness of command-style political systems have always been that they incentize systemic lying.

Foxconn is an international company that operates in many countries outside of China. Suggesting they are too simple to realize the business climate is different outside of China seems insulting and infantilizing. Their operating income is ~$180 billion USD a year. They are aware that business conditions are different in different countries.

The Foxconn deal had $3-4.8B worth of subsidies attached, and investigative reporting says the branch of the company executing on the opportunity couldn't even accomplish basic tasks like "Decide what to build" or "Decide what skills were needed in hired staff."

The takeaway is that either (a) everyone who touched the Wisconsin project is an idiot, but an outlier in the company, or (b) the failure to execute in Wisconsin is indicative of the general state of the company.

(a) seems like a stretch, given the project's size and visibility. Ergo, (b).

Leaving a subsidy on the table is not exactly like leaving money on the table for either party. They lower costs and lower the tax revenue from a business, but they don't represent a cost (outside of the cost to negotiate and write the deal). It seems like calling the people involved idiots (or suggesting the entire company is...setting itself up to fail to execute?) requires knowing internal details of Foxconn's financials.

A $1bn increase in revenue would be less than a 1% increase for Foxconn overall. That suggests they're probably willing to bear extremely high costs to establish new sites, as long as those sites will eventually yield significant revenue. Their negotiating tactics here seem short sighted from my point of view, but when you're as large as they are you create your own conditions.

Edit: Like...this deal seems to allow them to operate any of their businesses in Wisconsin. If they decide not to, they burn bridges in Wisconsin, but they wouldn't care about burning those bridges because burning them means they don't want to do business in the state. It seems like all they lose is their negotiators' salaries, small (for them) facilities costs, and whatever general goodwill they lose.

This may be too political for Hacker News, but...

Hope that the Governor of Wisconsin stayed a Republican? Now, that's not to say Democrats can't engage in grift, but, the deal was started under a Republican Governor in a notoriously corrupt state where the promise of Foxconn jobs was lauded by legislators, the governor, and the former President as a win for the people of the state, for our country, and a blow to China (that we were taking jobs back).

Foxconn's incentives seemed aligned with promising the moon and if those people were reelected, well, wouldn't compensating Foxconn for those promises and their outcome (the election outcome, not the factories) be fair?

Why would Foxconn care who governs Wisconsin?

If the deal never really made sense and was just a ploy by the Governor to drum up support, then a change in administration would mean that the new administration has no interest in paying off Foxconn for their part in the charade.

Exactly.

Foxconn negotiated a deal with members of a political party that was in power, that deal was an empty promise, and those power brokers used that promise from Foxconn as part of their political messaging with the hope of being reelected.

Had those politicians stayed in office, it's possible they would reward Foxconn for participating in the charade that helped them obtain reelection. Is it certain? No, I won't say it is.

But, what else was Foxconn's goal if not to actually realize those incentives without having to make the investment to justify them? They knew they were promised an enormous incentive by the state, and they knew that they would not fulfill their end of the bargain. If they also knew that they wouldn't realize those incentives, why did they pursue them? The simplest answer is that they fully intended to realize them.

Surely at least some of the negotiators of that deal legitimately believed that the other party - the Republican Governor and majority legislature in the state - would play ball?

Foxconn leaders seemed to be trying to create a big project that appeared to create lots of jobs to appeal to Trump (perhaps so the prez could point to it as jobs created, and then maybe Foxconn would get something out of it?) and less caused by the gov of Wisconsin. I agree it looked obviously like a silly deal that made no sense.

But what the hell does Foxconn get out of it? The deal was for tax breaks which only pays off if you have taxes to pay on profits or real estate and they don't have either. They didn't come close to meeting the employment goals for the deal to kick in either. So, they puff up some local Republicans which reflects a bit on the president. Then they embarrass themselves and everyone who promoted the deal and end up in the exact same financial position as before. This just feels like a big screw up more than a conspiracy.

Perhaps you should ask Foxconn why they are so strenuously objecting to the new leadership not holding up their end of the deal?

Many people's gut reaction to a whiff of partisanship here or suggestion of corruption or conspiracy is to push back hard. Well, dammit, what did Foxconn want to get out of it except for billions of dollars in ill-gotten tax incentives?

I'm responding so incredulously to you because everyone's motivations here was advertised loudly. It's not a conspiracy if it happens in broad daylight.

It seems clear to me that they wanted to not have to pay taxes in Wisconsin for many, many years to come.

https://apnews.com/article/technology-wisconsin-d3f025e43ab0...

The tax breaks to Foxconn were just one aspect of the deal, and unless you're specifically interested in Foxconn, it makes sense to look at everything that happened.

Hundreds of millions were spent on variously infra improvements, eminent-domaining people out of the way, lawyers, etc.

https://www.chicagotribune.com/suburbs/lake-county-news-sun/...

I actually don't know how much Foxconn did get out of it. They've apparently run similar scams elsewhere, it appears to be a Foxconn side gig.

There's a fun book in here for a scrappy investigative journalist.

Well, as a Wisconsinite I'd be remiss in not pointing out that this is the Chicago Tribune. Chicago and Illinois are always gleeful about the various piles of excrement we land ourselves in here in Wisconsin.

That said, yeah, I also have to be honest. We're down about 1 billion dollars on that Foxconn thing. That's what we're out of pocket. The new governor, wisely, took what remained of our chips and left that table. You get mad, but you just have to chalk it up to a learning experience.

The Illinois people are right on this one by the way. Not even in Illinois would their leaders get away with a 5.5 billion dollar con. The people in Illinois are so used to dealing with corruption that they would have started asking questions and raising a fuss long before it got to that point. Our problem is that we're so accustomed to our corruption that we never raised a fuss until it was way too late.

All that said, in Wisconsin our leaders didn't get away with 5.5 billion dollar con either technically speaking. Scott Walker and his boys got us for about a billion, but they didn't get as much as they wanted. So fibs shouldn't gloat too much, because while I can't see anyone in Illinois getting away with a 5.5 billion dollar pure con, I could see someone getting away with maybe a billion. (It's just that they're far more wealthy than we are, so that billion doesn't hurt as much.)

> you just have to chalk it up to a learning experience

Everyone knew from the start this was a BS deal and never going to happen or make ANY sense. This was pure incompetence at a political level that was paid for by tax payers.

I'm definitely taken aback with how you speak of this like its two sports team.

Hundreds of millions were spent on variously infra improvements, eminent-domaining people out of the way, lawyers, etc.

This is potentially wasted money, but it did not go to Foxconn, and presumably created infrastructure jobs. Not defending the deal, but it is often mischaracterized. The benefits to the companies usually are tax rebates that are only valuable if the company makes real investments that are taxable. They are not direct payments.

Purpose-built infra is almost always wasted if the purpose is unfulfilled; sure, maybe someone else can find a way to use it, but it was built to do things nobody needed.

> and presumably created infrastructure jobs

Usually you hear people complain about dig-a-hole-then-fill-it-up jobs. But if we're going to start supporting those, it would be best not to tie them to scams, I think.

> and presumably created infrastructure jobs

Yeah. Sounds a lot like the dig-up-all-the-carbon-burn-then-bury-it-again civilizational project going on the past 100, and next 100 years.

What exactly was the purpose to "re-arranging small bits of matter near the surface of the earth" again? [1]

Basically I wanted to make a comment about greedy exploitation-exploration trade-off strategies here but will stop short at this kind of eliptical passing remark.

[1] paraphrasing Russell in https://en.wikipedia.org/wiki/In_Praise_of_Idleness_and_Othe... it is appropriate given current moment

>What exactly was the purpose to "re-arranging small bits of matter near the surface of the earth" again?

From the same man that proposed the idea of "re-arranging small bits of matter near the surface of the earth".

>It would, indeed, be more sensible to build houses and the like; but if there are political and practical difficulties in the way of this, the above would be better than nothing.

It's a social hack to get around idiotic situations. It's like moving to mars because you are worried about a nuclear war or climate change. Dumb games require dumb players.

>This is potentially wasted money, but it did not go to Foxconn, and presumably created infrastructure jobs.

If Foxconn tricked a local government into providing free "useful" infrastructure it could just rent that "useful" infrastructure out to a different company and thus capture the entire tax benefit. However, they didn't even do that... The money was fully wasted.

There was an interesting episode of Reply All on this https://gimletmedia.com/shows/reply-all/2ohgza

> eminent-domaining people out of the way

On top of everything else, abuses of eminent-domain, civil forfeiture, and so forth really, really bother me.

In principle, I unconditionally support the right of the polis (the government) to do what needs to be done. My own city basically used extralegal means to wrestle numerous blighted, condemned properties away from a trust. The creators of the trust were dead and no one was left to negotiate.

But hot damn these levers of power get abused.

Please recommend any books, articles, etc advocating reforms.

> Please recommend any books, articles, etc advocating reforms.

Ilya Somin is a law professor who writes a lot on ED-reform.

As far as action, the Institute for Justice looks for cases they think they can make with, and lobby for legal reform.

https://ij.org/issues/private-property/eminent-domain/

You'll find they both come at things from a libertarian direction, but neither are rabid.

Samsung already has a fab in the U.S. That's reason to expect different.

I'm not going to say Foxconn was doing business the way Chinese do business, but maybe there's a culture thing there.

Foxconn is a Taiwanese company.

Every HN’er here promotes for the freedom and democracy loving people of Taiwan. They even claim it’s a country.

So get your “country” correct.

Another thing is the type of manufacturing Foxconn does is extremely labor intensive, it's essentially assembly shops. No way paying US wages would be competitive for them. But things like semiconductor manufacturing requires way less people so salary differences are negligible.

Nit: factories are not fabs, "fabs" refers to semiconductor manufacturing factories.

Well Samsung has already built a fab in Austin, Texas, and it has run for many years, so that's one reason.

Sounds like a market opportunity to produce more of those consumables locally

Probably not at 100% capacity.

Pretty sure that the Global Foundries fab in New York didn’t take much longer once funded.

It's an expansion of an existing fab

baybal2's comment is incorrect. this is merely an expansion on top of their existing capacity in Texas.

If it is, than it makes more sense.

But still, if it was a greenfield project, than it would've been nonsense.

Since they are expanding an existing facility, many complex things might be already available on the site, and building the rest in 3 years may be realistic.

Really happy to see this. It's really frightening how dependent we are on other countries to manufacture our chips. I hope that more American chip companies choose to build fabs here as well, though I recognize how expensive that would be...

Also... is every single tech company investing heavily in Austin? It sure feels like it!

Frankly, I'm surprised that the Austin region's notable decline in quality of life hasn't already begun to act as a feedback mechanism limiting investment and population growth.

Austin was plagued by decades of nimbyism and poor urban planning that prevented sufficient infrastructure development before it was a tech hotspot. ~10 years ago, things like traffic and public transport had already become borderline unbearable relative to comparable cities. It has only become much, much worse in recent years. I imagine that it might still seem livable to tech folks arriving from west/east coast megacities, but to us locals...not so much.

Nearby San Antonio has done a much, much better job in terms of infrastructure development, but of course nobody wants to set up shop there because it doesn't have the (fading, imo) cultural cachet that Austin has.

Decline is relative. Despite being a crappy deal Austin is still a better deal than the options where a lot of the new people are coming from.

Yeah, comparing with San Francisco/CA (massive income tax, possible wealth tax, absurd real estate, dysfunctional schools, etc) and Seattle (merely very expensive real estate, insane city council, famously unfriendly people, depressing weather, and possibly way more taxes and soon as our legislature can figure out how to get around them being unconstitutional)...

Austin is definitely still a good deal. Sure, the weather isn't as good as California (but better than southern Nevada), and the scenery isn't as pretty or geographically interesting as the Pacific Northwest (but better than Oklahoma), but there are jobs, a good university and Texas doesn't appear to have lobotomy patients as local government.

The main problem with Austin is that it's in Texas. Which means you're still subject to Texas laws and Texas schoolbooks. Some of those schoolbooks still teach that, "slaves were treated well because they were expensive property".

Which makes me think, maybe there is a business opportunity to open a private school that uses California or Massachusetts/New York text books.

Is that particularly common/exist at all in good Austin area schools? Or are you just generalising based on some Texas schools (genuinely curious as I have no idea myself)

My understanding is that the Texas state requirements used to be borderline insane, but that they have improved over time. Skimming http://ritter.tea.state.tx.us/rules/tac/chapter113/index.htm... , I actually found much less objectionable than I expected to.

This is a meme that may require updating

Possibly, I haven't looked in a few years.

In general I don't disagree about bass-ackwards public school curriculums in TX. However, for those who are willing and able to pay for private school, that's not really an issue in Austin, especially considering how liberal it is compared to literally any other city in Texas.

So basically just screw the poor by letting them suffer through Republican pushed indoctrination and lies?

> better than Oklahoma

That's a broad brush.

Currently sitting in a log cabin in an old growth forest on a stunning lake near Broken Bow, OK, with pretty hills / small mountains all around.

Interestingly, one of the few (maybe only?) place in the world that has both black bears and alligators.

Mountains, lakes, swamp, pasture... McCurtain County is a treat.

caveat to enjoying a wondrous life in OK of course is don't be a BIPOC

Of Oklahoma's five Congressional districts, one is represented by a Chickasaw, another is represented by a Cherokee, and a third is represented by an Iranian.

I’m closing on 20 acres near talahina in a couple weeks. SE Oklahoma is very underrated.

That's a great place to invest right now.

I'm buying a cabin in Broken Bow, the owner is selling and reinvesting in some land very close to where you are buying.

Seems to me plenty of places have both gators and black bears. GA and FL for starters...

> nobody wants to set up shop there because it doesn't have the (fading, imo) cultural cachet that Austin has

.. and has a public ivy college, the Texas legislature, and decades of semiconductors in the area. There's still NIMBY & traffic in the city, but to sum Austin up as simply a cultural hub feels dishonest

I'd say Phoenix gives it a good run for its money :-)

Austin is the "new" SF and it's happening on a shorter timescale which is probably going to create an even more dysfunctional local political system.

I know so many friends and colleagues that have moved to Austin from SF over the years to escape the issues they have with SF (mostly around trying to raise a family, poor schools, high coast of homes with any real space, etc) but it seems like Austin might just slingshot past SF in terms of issues.

There's always Denver and RTP.

Shhhhh, no there's not. These are not the metro areas you're looking for. Nothing to see here. Move along.

RTP?

Research Triangle Park, NC

I absolutely love the Triangle, but public transit there is a joke. My favorite tinfoil hat conspiracy theory is that Duke Parking killed the light rail system.

Can you comment a bit on why you love the triangle?

It functions much like a very large university town, except it has a much more diverse economy.

The sum of its parts make it a particularly nice part of the world to live in. Housing is reasonably affordable (although this is becoming a problem), there's an interesting and competitive restaurant scene, good public schools, some of the best coffee shops in the world, great art museums and functioning greenways. Lots of good places to support local agriculture and artists.

The local parks are lovely, and the Appalachians are a day-trip away. Same with the Atlantic coast.

A few local businesses that I deeply miss: Boulted Bread (bread), Jubala (coffee), Foundation (cocktails), Garland (food), Cocoa Cinnamon (coffee, churros, hot chocolate), Guasaca (arepas), Nuvo Taco (tacos), Wine Authorities (wine), Saltbox (seafood), Locals Seafood (seafood) ...

>it seems like Austin might just slingshot past SF in terms of issues.

As a former San Franciscan that has lived in Austin for 10 years what do you think are these issues?

Sure there is a lot of old school residents that don't want any change and fight every code change and infrastructure upgrade but the new residents are doing stuff like passing the new rail props. The city has built more housing in the last year than SF has in 5 years.

Another thing is the exploding homeless population, the city is so angry that we got enough signatures to put the camping ban on the ballot. A certain district actually voted for a far right wing candidate whose major platform was to get rid of the homeless. That is not something that would ever happen in SF.

> As a former San Franciscan that has lived in Austin for 10 years what do you think are these issues?

The biggest as always is public transportation. SF is a "small" city geographically but it's a nightmare to get around and I see Austin going down the same route. You have busses and MetroRail but frankly both of them operate at street level and that will just not cut it. The subway plans are good and a step in the right direction but by the time they actually finish they won't be anywhere near enough which will be viewed as a failure and then the next subway expansion will be too small and so on an so forth.

I just don't see the city getting out ahead of it's growth. Too little too late kind of thing. Obviously I don't live there so my knowledge is filtered though a set of peers with there own bias but that's my current view.

As a Michigander, I find it funny to see people argue over growing cities and whether their cultural cachet is growing or fading. People here would kill to have the problems that Austin has.

Having just lived there for a couple years, I would argue that Ann Arbor has all the same issues as Austin: Strong nimbyism; general under-development of public infrastructure; a university that keeps the local government on a leash; absurd housing prices subsidized by student loans and tech workers; trademark 'hippie'/counterculture mentality that's increasingly marginalized; increasing presence of cookie-cutter chain establishments...

But does Ann Arbor have a massive influx of homeless?

Who is we?

Can someone smarter than me explain why larger companies (e.g., samsung, amazon) can get 50-100% tax breaks from various governments, and we are all seemingly ok with that? What about mid-size businesses? Should they band together to come up with the same "we'll create X jobs in your town claim"?

Whatever happened to taxing the rich?

I'm not sure many people are particularly happy with it, but it's a fundamental problem. If you give different levels of government the ability to tax people, then private corporations are going to have the ability to shop around tax jurisdictions to get the best deal, and atleast some of those jurisdictions are going to include places that are deeply republican and believe it's good to have low tax revenue, and even better to attract businesses.

So what do you do? Create federal laws to control state aid and watch as libertarians/states rights people spontaneously combust.

Yeah I'm not sure what a policy-solution would be here. I'm definitely for allowing states to do whatever they want. It's the great A/B test.

Just seems like the smallest businesses and the largest businesses get the most deals and those in the middle get hosed. Maybe that's working as intended.

The reasoning, as I understand it: if we taxed rich people too much, it would discourage other people from wanting to be rich, which is bad for the economy.

Wow it would stun me if people were this stupid but I guess it is possible.

“Socialism never took root in America because the poor see themselves not as an exploited proletariat, but as temporarily embarrassed millionaires.“ -disputed John Steinbeck quote

Economics is never about absolutes.

It's always about the right response to the right situation. Taxes should be raised in an inflationary environment and lowered in a deflationary environment. The reasoning is quite intuitive. Inflation is a boom, deflation is a bust. It's easier to make money in a boom so the economy can support a greater tax burden. In a bust it's difficult to make money which means the economy can only support a lesser tax burden.

So the questions are "What is currently booming?" "What is currently undergoing a bust?". Well, obviously housing and stocks are booming. Wages are stagnating and thus are undergoing a (relative) bust. Just blindly reducing taxes on one side can be just as harmful as it can be good.

>Just blindly reducing taxes on one side can be just as harmful as it can be good.

Name any right-wing politician who has ever said something like this. Your explanation is nice in theory, but it's never been part of the national discussion about taxation.

Economy is literally the measurement of the movement of money. aka Velocity of money.

People hoard money. Rich people manage to hoard a lot. All the hoarding is slowing the money down. Depressing the economy.

> Taxes should be raised in an inflationary environment...

Did you mean spending? Because then you'd be touching on the philosophical difference between Keynesians and Modern Monetary Theorists (MMT). Whereas as most classical Liberal economists emphasized deficits, MMT emphasizes inflation. (Or maybe you're nodding towards the Laffer Curve.)

Regardless. Taxes are how governments pull money thru the economy. If we want to keep the money moving, thereby grow the economy, we need to tax the hoards.

In the case of more tax breaks, abatements for Samsung, things are a little different.

Samsung's just exfiltrating money. From Austin, from Texas, from the USA.

Since our polis can no longer have a rationale discussion about taxes, "omgherd taxes are theft", I propose a novel solution:

Austin should become a Samsung shareholder. Holdings equal to the value of the tax breaks. Like for like. Mutual cross holding investments, just like a true keiretsu.

No, that is not it. =)

If you are rich enough you can pick what country gets to tax you, if one country raises the tax too much, rich people will just move their taxable assets to another country.

Because people/companies with bargaining power can get things more easily.

USA wants chips manufactured in the USA for jobs and probably more importantly for national security. Too many chips made in/near China + Intel going down the toilet.

When US universities sold out to china $ the national security argument went out the window as labs were willingly giving the r&d away.

I don't know much about semi plants but what sort of jobs are these? Why aren't these plants all automated?

50% of a big number can be bigger than 100% of a smaller number.

It is super difficult to actually do that when tax laws are national but corporations are global.

When a corporation is large enough to be able to move around in the world they can start shopping around for the most advantageous place to set up shop.

If you’re wondering like me why they would bother building in an expensive country with fewer qualified workers:

“To build a leading-edge manufacturing facility, Samsung needs rather huge incentives from authorities. In particular, Samsung is requesting combined tax abatements of $805.5 million over 20 years”

So 40 million per year? That does not sound like much..

"which essentially means that Samsung demands a 100% tax abatement from the county and 50% from the city."

So the figure couldn't really be much larger.

50% tax abatement for 5 years from the City (which has a ~50% higher tax rate than the county). I can think of a lot of ways it could be much larger. How about 100% abatement for 5 years? 50% abatement for 20 years? 100% abatement for 100 years?

"50% abatement for 20 years?"

The article reads like that's what they are asking for from the City.

They are asking for 50% for 5 years for the city (which has a higher tax rate), and 100% for 20 years for the county (which is a smaller govt entity with a lower tax rate).

17B investment for 805M tax rebates... over 20 years... 5% of invested ammount... seems just like a way to get government involved and avoid them blocking investment.

Only a small fraction of that $17B would be spent on building the plant and hiring workers (1800 jobs is what I've read). Most is for the extremely expensive equipment used to produce the chips and that will be bought and shipped in from elsewhere.

First, those 805M are tax rebates, they need to generate taxable profits to get anything back.

Second, even the 1800 jobs, say 100k avg - 180M a year vs 42M reduced taxable profits?

Apart from that, sure, you have huge equipment costs, where they come from now means its an opportunity for local companies to get in the supply chain.

I'm not sure what you are implying but this seems like a win anyway.

It's also interesting to see companies investing in the US, it means there are opportunities long term and US hasnt lost the chip sector.

Presumably the government values US-based factories for reasons beyond just domestic employment (even if that's the biggest factor). Being able to produce chips in the US is good for national security.

Yes is is good for the US, but not so much for the Manor Independent School District that is being ask to not collect property (school) taxes on the plant despite the massive increase in attendance that it will generate.

wouldn't income tax on employees or sales tax fix that?

School districts get some federal funding, but it's mostly from property taxes.

I have no idea how american school funding works, but property taxes? Is that really true?

That would mean that schools in richer areas with higher property values got more funding?

I think you're right. Semiconductor manufacturing is not a particularly employee-dense business, and the US must sustain power projection in the South China Sea and on the Korean Peninsula in the status quo, all the time, rain or shine. Even if you don't think about the product and supply chain, but only look at externalities, the US is less trapped in a battle of brinkmanship over Taiwan and South Korea if it has an alternative source of semiconductors in case of emergency. Keeping critical tech manufacturing in the Americas actually makes it easier for us to protect our allies in the Asia-Pacific region, by giving us less reason to make aggressive-looking moves that can be misinterpreted as offensive by China.

Still, there is a clear economic benefit from such a move closer to home, even in times of peace and predictability. The US economy is all about outsourcing the basement and keeping the penthouse. But the hosts of our outsourced industry can "tech up," so to speak. If we let Asia-Pacific totally dominate the semiconductor fabrication industry, it's only a matter of time before they dominate the semiconductor design industry as well.

While true that's more of a federal government interest while the tax breaks are coming from the local government. Maybe the local government will ask for some funding from the federal government.

It's called Automation. Samsung can move senior people around to teach new people. It's not rocket science. Also they will never get that tax abatement from the current Austin city council. There's no way. They might get 20-25%

It is strategically important for the US govt to have fabs in the US rather than next door to China.

I know SV wasn't in the running, but I can't imagine the Bay Area or California even addressing the permits by 2023... On another note, does anyone know if the supply chain/parts/materials/machinery can be sourced from outside of Chinese suppliers?

> does anyone know if the supply chain/parts/materials/machinery can be sourced from outside of Chinese suppliers?

I don't believe there's a single part of the semiconductor supply chain that is exclusively available from China.

I don't believe there's a single part of the semiconductor supply chain that is exclusively available from China.

I agree with you. But the parent comment reminds me of when Apple started making computers in Austin and HN was full of people saying, "No, you can't build computers in America because America doesn't have any screws, and can't possibly figure out how to make its own!"

Time and money solve all problems. Even supply chains.

The nice thing about China at its present stage of development is that everything made in China will perpetually get more expensive to make there and that will provide an ongoing advantage gain to every other location around the world that has until recently been at a steep cost disadvantage to China. Five to seven years ago it was already nearing cost-even to begin making things in the US again vs keeping that manufacturing in China, that will continue to tilt in the favor of the US by the year.

I doubt

If it was only cost... Cost of manufacturing labour is already lower in most of US territorially than in China, and is dramatically lower in other poorer countries.

The saying is that only $2 of iPhone is manual labour.

Samsung smartphones shows that assembly in China is no longer required: no Samsung smartphone is made in China anymore. They're mostly made in Vietnam and now India (with some small scale manufacturing in a bunch of protectionist countries like Brazil for domestic demand but not for export)

The age of China being the factory of the world is ending. And that was before Trump's trade war.

You need to know difference in between assembly, and manufacturing.

Just screwing parts, and subassemblies coming from China is assembly.

Manufacturing implies integration not on just parts, and subassemblies level, and having much bigger part of the supply chain under one roof.

Samsung still has some ODM devices assembled in China.

https://www.reuters.com/article/us-samsung-elec-china-focus/...

Thanks for the correction. A few years ago I did read that they had low-volume ODM devices made in China for the Chinese domestic market. 80% of Samsung's smartphones being made outside China is still great though.

>The saying is that only $2 of iPhone is manual labour.

$2 under which wage? Like, is it $2 at $1/hr or $2 at $20/hr?

If the latter, it's negligible. If the former, it's an extra $39 which is apparently huge in the context of manufacturing.

I heard recently that that Apple factory’s output was limited by the number of screws their local supplier could make per week. Can this be called a solved problem? They just stopped depending on the US supplier. https://www.nytimes.com/2019/01/28/technology/iphones-apple-...

I've been meaning to do a deep dive on semiconductor supply chains for a while now. I'd start by aggregating annual reports for the companies named in [1], identify their major categories of capex, gather information about their suppliers, and recurse.

[1] https://www2.deloitte.com/content/dam/Deloitte/cn/Documents/...

I am doing the same. I started at the SMH etf and have been working downwards. It's interesting stuff!

Don't miss Ajinomoto - the company that makes msg (monosodium glutamate). They also make ABF (Ajinomoto Build-up Film) that is used in circuit fabrication and is currently in short supply. https://www.ajinomoto.com/innovation/action/buildupfilm

Probably one advantage of the proposed Austin site is that it's only a 10-minute drive from the Austin location of Applied Materials (https://www.appliedmaterials.com/), a really large manufacturer of equipment for fabs.

I wonder if a large fab is a "if you build it, they will come" type situation? When I worked at Micron in Boise, the surrounding area had many fab suppliers.

Hmm. And that makes me wonder if the city might be right to give all those tax breaks to the foundry. They get to charge full tax on all the businesses that come there as suppliers to the foundry.

China cannot fabricate semiconductors exclusively on Chinese machinery. In fact, there is very little viable Chinese-made equipment at state of the art technology. They must source machinery primarily from the US [AMAT, LRCX, KLAC], Japan [TEL, Screen] and Europe [ASML].

Maybe earthquake risk (though they are still in earthquake prone areas elsewhere)? https://semiengineering.com/recent-earthquakes-highlight-ris...

I worked at Tesla on the CapEx Team, so not chip fabrication but we definitely bought similar things since it is still industrial automation. We mainly sourced from the country we were building in with an emphasis on vertical integration. We even designed our own chips and have them fabbed in the COO. People seem to act like only China makes things. America still makes things, it's just sometimes they are more expensive when you aren't looking at the project holistically.

For Tesla it was simple. Lower lead times means lower safety stock, better quality and usually US suppliers are willing to push payment terms out. So it was a better deal for Tesla. However, some companies use different metrics for their 'deals' and those companies tend to source from China.

Sheesh. I can't believe all the comments here that say, more or less "It's an expansion of an existing fab".

It's most emphatically NOT that!

Directly from the article:

The plans for the new fab are called Project Silicon Silver (PSS), and it will be located adjacent to S2. It will not use any of the existing buildings of S2, but will be a completely new fab constructed from the ground up. It will have its own operations support, central utility building, industrial waste treatment, air separation plans, storage for inert gases, and other constructions.

The only currently-contemplated interconnections between the new facility and the surrounding existing property may be a pedestrian and/or material bridge or walkway constructed between the existing improvements on the site and the new construction

That couldn't be more clear as far as I'm concerned!!!

Woo, go team!

Love that the US is bringing critical technology supply chains back onshore.

Its very important for the US strategically to build out its foundry capabilities, so this plus the TSMC plant are big steps forward there.

did the TSMC project get mothballed as a boondoggle for Wisconsin?

You’re thinking of a Foxconn project.

Interesting if it is related to the possibility of manufacturing smart phones in Mexico [1].

In that sense if you were to take the output of an Austin based fab to supply a smart phone manufacturing supply chain in Mexico versus China, that would be a significant reduction in the length of a smart phone supply chain.

[1] https://www.reuters.com/article/us-mexico-china-factories-ex...

Chips are super value dense so the cost to fly them to the other side of the world is teeny tiny. Heck, the weight is so low you could just about afford to put the fab on the Moon (for chip logistics costs), if there was any real advantage to it.

Shipping costs from the moon to anywhere on earth is probably pretty minimal, since you could do it with a catapult and a shell+parachute so the goods survive re-entry. I doubt mass would make a big difference there so much as volume.

Shipping costs from earth to the moon is $10 000/KG IIRC though, which, if we assume that a processor weighs 10 grams and costs $100 means the value of the stuff you're selling is $10 000/KG, i.e. you can't make a profit on them.

But, suppose you ship 50% of the processor's materials up (with a convenient 100% efficiency of raw materials usage), and you mine the other 50% locally. Which, does the moon have silicon? Yes, it has plenty. https://en.wikipedia.org/wiki/Lunar_resources#Resources

I suspect lunar silicon would be more expensive to process than terran silicon, primarily due to the lacking skilled labour market in the area. It would likely need to be extensively automated, and likely re-invented from the ground up around the different atmosphere (which while not breathable, is possibly still not clean enough for a cleanroom), gravity, potential power sources (metallurgical coal is a no-go, mainly because there's no easy oxygen but also because coal is quite expensive there).

While it's possible that some raw materials are cheaper on the moon than elsewhere (e.g. platinum from asteroids), I doubt this is the case due to the previously mentioned potential for cheap moon-to-earth shipping.

From what I can tell, a high end chip goes for almost $1/mm^2, which given typical wafer thicknesses and density, is about $1/mg, or $1,000,000/kg.

Great post. Though current stated costs to the moon are like $1.2M/kg via Astrobotic and that's a bit speculative as they haven't landed. Fingers crossed SpaceX gets Starship online and pulls that number down!

If you're interested in reading about proposals to use the Moon for manufacturing, here's a fun blog post on potential use of lunar resources for manufacturing:

https://denniswingo.wordpress.com/2018/08/20/lunar-metals-an...

Lots of interesting proposals and people to follow on Twitter too for both lunar and astroid ISRU (in situ resource utilization): https://twitter.com/drphiltill https://twitter.com/george_sowers https://twitter.com/joelsercel https://twitter.com/wingod

I'm not a supply chain expert, but just envisioning that export controls have been and will be getting a lot more challenging going forward. Not to mention higher risk at stuff getting held up due to other logistical challenges (i.e. bad weather, etc.) while doing the full roundtrip to Asia and back.

If there was chip production (CPU, memory, etc.) out of an Austin fab and LCD screen, battery and other material production in a special economic region in Mexico, seems like there could be a full 3-5 day reduction in the production.

Apple sells about 43 million iPhones in the US per year. Shipping 43 million of anything (i.e chips) to Asia only for it to be shipped back (as a finished product) to the US seems counter-productive?

Almost all ICs are packaged in Asia at this time anyhow. Main countries are phillipines, malaysia, and taiwan.

In the scheme of things shipping around the world doesnt take all that much time. (Lead times for ICs are 20 weeks at the bare minimum)

We really need to package ICs in America. It's all automated now, anyway, so the low labor costs (the original reason for sending packaging to Asia) is MUCH less important.

Don't forget about the Intel assembly & packaging plant in Costa Rica! Ref: https://en.wikipedia.org/wiki/Costa_Rica#Economy

To quote Wiki: "An Intel microprocessor facility in Costa Rica that was, at one time, responsible for 20% of Costa Rican exports and 5% of the country's GDP."

As I understand, for a long time, assembly and packaging was labour-intensive, so ICs were packaged in places with low labour costs. However, with the advances made in automation, I wonder if this still holds true. Maybe assembly and packaging will move back to high income countries in the next generation.

regarding job: modern fab is pretty much automated.

however, people who run the fab are mostly engineers. TSMC recruit from Taiwan's top University like National Taiwan University. you've to pay these professionals a high salary regardless where the fab is. (each new gen fab is more automated than pervious.) just check out the salary + bonus for TSMC engineers in Taiwan.

Hi. I checked salary and bonus for TSMC engineers in Taiwan. Here is what I found:

Engineer (1.3M NTD -> 46.5K USD): https://www.glassdoor.com.hk/Salary/TSMC-Engineer-Taiwan-Sal...

Senior Engineer (2.0M NTD -> 71.5K USD): https://www.glassdoor.com.hk/Salary/TSMC-Senior-Engineer-Tai...

Unless I misunderstood your comment, you seem to be hinting that these pay packages are very high. They don't look very competitive when compared to other high income countries.

I cannot prove it, but my sense about TSMC is that their most senior researchers, sales people, and operations people are paid silly money. The remaining (below them) are paid good, but not insane, wages. That said, most of the TSMC R&D labs and fabs are in cheap locations, like Hsinchu and Tainan, so these pay packages are still quite good.

Why Austin? I thought you needed a lot of water for fabs.

I'm no expert, but my understanding is that Fabs like stable weather patterns and climates so they can more easily regulate the cleanrooms.

Knowing nothing about Austin's climate, of course, I'm assuming that it's reliably hot and dry?

There's a joke about how quickly Austin's weather can change. It's not unheard of to see 30 degree swings over the course of a few hours. Austin has no protection from polar winds coming down from the north, nor is it near water, which has a moderating effect on climate. So it does have erratic weather. It's possible to see an inch of snow every other year, and yet, in the summers, it's likely to see 100+ degree (F) weather for two months straight.

It's more likely how Austin is still attractive to high-tech personnel - a good public university nearby, a reputation for being a lively city, somewhat reasonable real estate prices (relative to the Bay, but still 2X what they were 10 years ago), and excellent schools if you're in the right area, a decent number of tech options to jump to if you want to switch, a spritely startup scene, enough outdoorsy things to do, etc etc.

Texas is quite large... you're probably thinking of West Texas (as shown in innumerable western movies). That is indeed very dry.

East Texas, particularly south-eastern Texas (near the Gulf) is very humid and can experience torrential rainfall. It's more like Louisiana than El Paso, climate-wise.

Austin is sort of in the transitional zone between the two.

By comparison with California, Austin gets an average of 34.32 inches (872 mm) of rain per year. San Jose gets 15.82 inches (402 mm). San Francisco proper gets 23.65 inches (601 mm).

Austin is considerably more water-rich than the Bay Area.

I think there's some consideration of how prone an area is to natural disaster, which makes Intel's picks of places like Arizona and New Mexico ideal. But I think other benefits like a strong labor market or a business friendly jurisdiction will probably have an overall bigger impact on where a business chooses to build.

Arizona has a long tradition in and worker base for semiconductor manufacturing.

Knowing nothing about Austin's climate, of course, I'm assuming that it's reliably hot and dry?

Austin isn't in the desert. It gets ice storms, tornados, hurricanes, and the epic Texas thunderstorms that have various names like Blue Norther, Possum Stomper, and Toad Strangler.

I spent five years in Texas, near Austin, and my college meteorology classes came back to me.

One of the moments I'll always remember growing up in Texas was seeing the newscaster ask the very well dressed and always presentable local weatherman "what's the weather look like tomorrow?" and when the camera turned to him we saw a disheveled weatherman, tie not done, hair messed up turn to the camera and deadpan "I have no idea"

Many areas have the joke "if you don't like the weather, wait 5 minutes". That was less a joke and more a warning in large areas of Texas.

As a lifelong TX resident, to say Austin gets those three phenomenons in any regular, meaningful sense is inaccurate. They can upon occasion get any of those, but not like Oklahoma gets tornadoes, Houston gets hurricanes, or Denver gets ice storms.

Thunderstorms and wild swings in weather during non-Summer months are definitely mainstream, though.

It is neither reliably hot nor reliably dry. The only think I think you can say about “consistency” regarding Austin’s weather is that it is hot in the summer and doesn’t snow very often.

They already have a fab in Austin.

And I believe they already own the land where they'd put it as they'd been planning expansion. I don't see the city/county council going for their plans on 100% tax free infrastructure and property taxes though as Texas pretty much substitutes property tax for income tax. I think it will actually end up being built in Arizona as Austin is doing fine without Samsung trying to bully it in to ridiculous tax incentives.

Texas business-friendly laws with access to technical Austin workforce and university.

Austin has a long established semiconductor workforce, and there are already multiple fabs there, as well as a lot of semiconductor design. Right now, Freescale/NXP has two fabs, Spansion has a flash fab, and Samsung already has a fab in Austin.

Spansion’s fab was purchased by Cypress, who were acquired by Infineon recently.

The reason is that Samsung already has a huge logistical presence in the city supporting the A2/S2 lines. Adding deliveries & pickups for across the street is much easier than building out a new facility in Arizona and replicating all of that over there.

Additionally, if there is some technical fire in the new PSS fab, it would be expected that employees from the existing fabs could walk across the way and assist. They would probably do something where the existing engineering & manufacturing teams are divided up and shared across the facilities to get everything bootstrapped. Their systems are very strictly standardized, so staff from one factory can very easily support any other in the fleet.

When fab hardware depends on chips made by fabs.

Assuming proposed US semi fund will be chipping in for Samsung asking 1B in tax subsidies like TSMC.

Finally - this is great news - I hope they put a lot of design staff here too, not just fab workers.

They layed off a lot of their Austin-based Exynos CPU design staff two years ago, save for those working on the AMD/Samsung GPU partnership work - still, will be interesting to see how/if Exynos comes back, know they are aggressively hiring for GPU work.

This will be good for the US chip supply chain, hopefully more chip makers follow suit

Of course it's in Texas, not in SV. And TSMC going to Arizona.

This is awesome! The US needs domestic chip manufacturing, in order to be able to effectively compete, take business from, and vanquish China on all fronts. This includes AI, automobiles, the cloud, etc. The coming technology arms race is looking to be like a second Cold War, so we need to remember the lessons of the first (like containment) and apply them again.

Yes! Agree! I was happy to read this, too much reliance on TSMC

TSMC is also building a fab in the US. [1]

1. https://www.cnn.com/2020/05/15/tech/tsmc-arizona-chip-factor...

They already have fabs in the US. e.g. https://www.wafertech.com

>This is awesome! The US needs domestic chip manufacturing, in order to be able to effectively compete, take business from, and vanquish China on all fronts.

Have you forgotten all the intel/GF fabs?

https://en.wikipedia.org/wiki/List_of_Intel_manufacturing_si...

https://en.wikipedia.org/wiki/GlobalFoundries#Fabrication_fa...

from TFA:

>... “leading edge” ...

Does Intel have some leading edge fabs? I read they have some coming in the future, but I think the point is it’s good to have more!

Yes, of course. Intel has been shipping 10nm devices for a while now.

If the question is about the definition of the term, this document from a faceless consultancy in Aug 2020 describes leading-edge nodes as 14nm and newer.

https://www.mckinsey.com/industries/advanced-electronics/our...

Shipping for a while doesn’t mean much when technology has moved on.

I doubt McKinsey would stand behind that definition now. If they did, it wouldn’t do them credit.

Others are shipping 5nm. I mean I’m typing on a laptop that has a 5nm process cpu right now, and Intel is nowhere inside.

Intel is even having trouble with 7nm.

Yep, and it's wonderful that we have important fabs that haven't moved production out of the US. However, other parts of the chip supply chain have been moved out of the US, such as iPhone chip manufacturing and stuff related to data centers & servers. While this can help lower prices, it also increases the risk of the US losing control of IP and having its technology compromised.

Remember when in 2018 SuperMicro servers sold to Apple were compromised by a small chip inserted by a Chinese sub-manufacturer? [1] Or how iPhone designs sometimes get leaked by employees of Foxconn in Taiwan? [2] We've got fabs in the US, but I believe domestic chip manufacture should become central policy to prevent incidents like this.

[1] https://www.bloomberg.com/news/features/2018-10-04/the-big-h...

[2] https://bgr.com/2019/08/14/iphone-11-release-date-soon-foxco...

[1] is generally considered an incorrect story. It was denied by every party involved, presented no evidence, and not confirmed by anybody else.

It is possible in theory, but this particular story is considered a BS.

And two years later there has been zero development on this bombshell story which makes the story look even worse.

Did anything ever come of that bloomberg article? Supermicro denied all allegations, but also began moving production out of China from what I can find online. Outside of a tech blog or two I haven't heard anything of it again

The Bloomberg article was a dud.

They exist, but Intel doesn't do contract manufacturing and GlobalFoundries is behind the other major contract manufacturers (TSMC and Samsung) by many process node generations.

> Intel doesn't do contract manufacturing

What about Intel Custom Foundry?

Practically nil.

Devil's advocste: why can't the US do that via Taiwan as it does currently?

They're an ally, they're totally reliant on the US and afraid of the PRC.

Won't we also need "rare" earth metal supplies to of it is a vital industry we can't cope with being overseas? And then there is the fact the fab machines all come from Belgium.

No horse in the race on either side here but just to answer the devil's advocate: it's pretty easy to imagine a future where China could establish a trade blockade of Taiwan, putting the US in the position of "potentially start a war, or go without Taiwanese goods" -- that's like the diplomatic equivalent of a Fork[1]. If you depend on it, and you can do it at home, you should.

And "but what about ______, then?" (e.g. rare earths, fab machines) does not mean that all other "______" (e.g. chips) should follow suit. You control what you can.

1 - https://en.wikipedia.org/wiki/Fork_(chess)

Taiwan is a great ally. But if a war breaks out between US and China, Taiwan will likely be the first casualty.

Nothing against Taiwan or anything. Its just the bloody obvious "first strike" move. China will almost certainly attack Taiwan as part of an offensive-strategy vs the USA.

Having fabs on this side of the Pacific Ocean is important for strategic defense reasons. Of course, the US should move to defend Taiwan in that situation, but if worst-comes-to-worst, we'll need a way to produce chips under such a scenario.

-------

This story is about Samsung however, which is South Korean. North Korea now has nuclear weapons, as well as tons of artillery pointing at South Korea.

For the Samsung-side of things: same thing. South Korea are great allies, but if the North Korean war ever gets started again, Samsung's production inside the US-proper would be a strategic advantage.

True. It's also important to remember that Taiwan is located in a seismically active region ...

> Taiwan is a great ally.

Should I remind of the fact that Taiwan was very glad to sell any amount of semiconductors to mainland China.

If anybody had an option to strangle them economically, it would've been Taiwanese themselves.

China is world's biggest semiconductor market. If the tap will close on their semiconductor imports, so would've their economy at large.

Now, it's them, and not US who has to regret their prior indecisiveness the most.

Now of course attitudes have changed in Taipei, but only now.

I bet, in the next 5-10 years, Japan, and Korea will have their own time regretting being soft with Beijing just like that.

> Should I remind of the fact that Taiwan was very glad to sell any amount of semiconductors to mainland China.

And US companies moved production to China. Apple, TSLA, etc. etc. Yeah, we're in peacetime. These things happen. We're not planning to go to war with China, but we need to be realistic about potential threats in the future.

There's something to be said about the ability for trade to reduce tensions. I don't think trade with China is necessarily bad.

> Should I remind of the fact that Taiwan was very glad to sell any amount of semiconductors to mainland China.

> If anybody had an option to strangle them economically, it would've been Taiwanese themselves

> China is world's biggest semiconductor market. If the tap will close on their semiconductor imports, so would've their economy at large.

Why would anyone (least of all Taiwan) want this? Trade isn't a zero-sum game, and both parties benefit from it. Trade wars are very unpredictable, as the way the US lost the Trump-era trade war with China shows.

It's true that some in the US see China as a rival that must be "strangled". Views outside the US are much more nuanced - China isn't going to disappear and is way too big to just strangle. There are major issues dealing with China, though.

US politics have had a "big bad" for 80 straight years now, it's basically a reflex.

>Should I remind of the fact that Taiwan was very glad to sell any amount of semiconductors to mainland China.

US sell chip to China as well until the Huawei ban. TSMC stop making chip for Huawei when the US hammer came down.

how can you tell Taiwan to stop selling when your own companies is selling to China? how's that fair?

Which fab machines are you referencing? Atmospheric modules? Vacuum modules? Or maybe not talking about the process tools at all, and more along the lines of inter-fab equipment?

I won't pretend I know, fabrication seems very technical and I get the vague idea but I'm also an idiot.

I was referring to ASML Holdings.

https://en.wikipedia.org/wiki/ASML_Holding

The equipment industry is global. US, European and Japanese companies make up most of the largest equipment suppliers. Regardless of where they’re headquartered, most of those companies also have global R&D footprints.

A couple of strategic reasons: Taiwan's proximity to the PRC would knock out fab production should a war occur between the two. Also, the same proximity could lead to unwanted IP transfer, which could compromise US national security if China gets its hands on any new chip tech the US develops.

Taiwan is a great ally to the US and having it manufacture chips is good for both economies, but if the US wants to lower risk and keep control of its technological edge it is in its interests to keep manufacturing domestic.

> Devil's advocste: why can't the US do that via Taiwan as it does currently?

I think TSMC was working on a similar deal.

> Won't we also need "rare" earth metal supplies to of it is a vital industry we can't cope with being overseas?

Rare earth materials are available in the US but due to cost competitiveness, all the mining happens overseas. Also the mining process has environmental impacts some countries choose to ignore.

> And then there is the fact the fab machines all come from Belgium.

Which company in Belgium make the fab machines. The biggest manufacturer of semiconductor fabrication equipment are Applied Materials and Lam Research, both in the bay area. ASML from Netherlands make the critical Photolithography equipment however. Beyond this, there are hundred of mid and smaller companies involved all over the world.

https://en.wikipedia.org/wiki/Semiconductor_equipment_sales_...

There are also rare earth separation plants being built in Texas.

"DOD Awards $30.4 Million for Rare Earth Minerals Facility in Texas"

The Department of Defense and Lynas Rare Earths Ltd. will each contribute $30 million to establish a rare earths processing facility in Texas.

https://thetexan.news/dod-awards-30-4-million-for-rare-earth...

But that doesn't actually have much to do with advanced microprocessor fabs. The "tech" uses for rare earth elements are primarily in making magnets, lasers, and phosphors.

Geography puts us into a perpetual contortionist act. Taiwan is very close to China so our looming presence is perpetually necessary to ward off the dragon, so to speak.

For example, what happens if China decides to take one of the Matsu Islands?

The US is obligated to respond with aggression, and Europe too. Not out of alliance, but out of strategic and economic necessity to keep the second domino from falling.

It's foolish to leave that kind of pressure point exposed to an adversary. Sooner or later, each side will feel it is defending itself against an aggressor and both will engage in what they see as a righteous war.

We can make Taiwan more secure by making it less of a pressure point and freeing our hands strategically. If we are going to react to Chinese aggression with American aggression to protect an ally, it should be decisive not reflexive.

I'm all for securing domestic chip manufacturing but can we stop with the vanquishing? It sounds like you're talking about an army of orcs.

Reporting on Foxconn’s Wisconsin “manufacturing plant” at the time.

https://www.theatlantic.com/business/archive/2017/07/foxconn...

Reply on news.ycombinator.com