Comment by digdigdag

I'm a EE and had no problem finding a job and neither did any of my classmates in my EE program (early 2010s). I also didn't exactly go to anything approaching MIT, but it was an engineering school and I had a decent GPA. Particularly, there are a lot of well paying jobs in power systems with good work life balance. We have an energy transition going on, so that helps. Having an internship probably helped me too. I acknowledge that things might have broadly changed.

Is there a somber write-up anywhere as to the future of EE in the West?

Software jobs are more plentiful, sure, but you’re discounting the extremely high quantity of EE/CE jobs available at semiconductor companies (Intel, AMD, and many smaller ones) and companies like Apple. They don’t pay as well, but they can pay quite good over time and tend to be more stable than software jobs.

That is still the case and no where else is even close.

https://www.statista.com/chart/30815/top-destination-countri...

https://news.gallup.com/poll/468218/nearly-900-million-world...

Today's Einstein ARE immigrating to US for such positions as finance, adtech and management, ones that explicitly produce no physical artifact.

Einstein didn't emigrate to get rich, he emigrated because the Nazi's took over Germany. Germany had the best universities in the world before they took the path of self-destruction. So that was a second, separate event that helped America.

America stills gets a lot of immigrants.

For nearly a century Europe incinerated itself twice over.

> That is no longer the case.

For all I shit relentlessly on this country and its culture, it's still an extremely attractive place to live if you're well-situated to make money. (Most people are not—hence my contempt for how the society functions. This presumably DOES apply to an "Einstein", if indeed this Einstein wants money.) China still has a way to go in catering to and granting citizenship (or some amenable equivalent) to foreigners.

The US didn't win World War 2, break the sound barrier, or put a man on the Moon only or primarily due to immigrant workers. We scoured the country's public school system for the sharpest young minds, sent them to institutions of higher learning with rigorous curricula, and found them positions in industry, government, or the military which made good use of their talent. Fetishizing the "nation of immigrants" narrative at the expensive of the native-born Americans who actually built most of this country's prosperity is, at best, ahistorical.

why would someone pursue a route that's harder AND pays less AND has far fewer jobs available?

Hard and well paid gets a flood of people pursing it so difficulty can't be the only explanation. Finance, actuarial science, medicine, and law get plenty of applicants. I think it's that CS is an office job that pays well and is in-demand.

I studied both, can't say for sure EE was harder. Some courses in computer science were extremely hard for me (complexity, discrete math) and some courses in EE engineering were equally hard (most of the physics courses, analog circuits and more)

Both degrees can be made super hard, as hard as the school desires them to be...

Nah I did EE and then CompE (which was just replacing some later EE classes with hardware design stuff) and EE is not "actually really hard" - although people like to put it on a pedastel.

The jobs needed for chip manufacturing aren’t primarily EE. It’s largely chemical engineering with specializations related to semiconductor tech. EEs use the tools developed by fabs to make their products, but those are typically separate companies (or, in the case of in-house fabs like Intel, basically run as separate companies).

I suspect the kinds of salaries that's possible in Silicon Valley only happens because:

(A) Skills are fairly transferable. (B) There is a lot of employers competing for workers. (C) An awful lot of value is created along the way.

If you specialize in some tiny part of chip manufacturing, there aren't many places you can transfer your skills.

Even if, in the future, you have multiple chip vendors. They won't all use the same processes, and you might only fit into one role at each of these businesses.

Maybe it's not that simple. But few chip companies have to compete against startups for workers. And that probably won't change.

Not saying the jobs can't be well paid, just that it's not unlikely that it won't be absurd SV level salaries.

It's also highly-skilled, yet very boring work. The way it was described to me is that every major piece of equipment has a PhD assigned to it and their job is basically to babysit the machine and troubleshoot when things go wrong.

US PhDs typically have other options and would consider this sort of work a waste of their time.

Even China has ruled 996 illegal in 2021: https://en.wikipedia.org/wiki/996_working_hour_system#Legal_...

No one should be forced to work those kinds of hours. It's unreasonable to call Westerners/Americans lazy if they refuse to work 996.

Not just long hours right? Speaking to Taiwanese friends involved in semiconductor work (not TSMC employees though) it's the shift work that's really hard to manage in the US.

50k is/was recently a decent salary (not SF). In the last 5 years, not so much anywhere outside the absolute lowest CoL areas.

But yes, most Americans do not want to work on a death march. And employers don't want to pay it. I doubt they can argue 50k as exempt so that's a lot of overtime. They may as well be salaried 6 figures at that point.

996..? doesn't fit into weeks, months or years

McDonalds in Sunnyvale CA starts at 20/h, so 41k/year for the lowest role

Perhaps - in California.

Median US Salary is $59,384. Half of workers make less.

I think that's obviously a major part of it but it ignores other stuff like lax environmental and safety standards.

It would be interesting to see how much of the economic advantage of off-shoring is due to lower wages due intrinsic to lower cost of living vs stuff like ignoring/bribing foreign officials or non-existent environmenta/safety standards that objectively should exist.

Personally I won't mind paying more to buy manufactured goods. My mom told me that a pair of sneakers before the offshoring back in the late 80s usually cost more than $300 in today's dollars. Yes, it was expensive, but I would just buy fewer and use the one for longer time. The reason is that in the long run the manufacturing cost would get lower due to increased efficiency, and loss of supply chain is detrimental to the entire country - and our living expenses will increase overall. Case in point, how much tax do we have to pay and how much inflation do we have to suffer in order to build those super expensive weapons? Part of the reasons that we had $20K toilet and $100 screws is that we simply don't have large enough supply chain to offset the cost of customized manufacturing.

Besides, the US loses know-how on manufacturing, eliminating potentially hundreds of thousands of high-paying engineering jobs - it will also be a pipe dream that we can keep the so-called high-end jobs by sitting in an office drawing boxes all day. Sooner or later, those who work with the actual manufacturing processes on the factor floor will out compete us and grab our the cushy "design" jobs.

Easy to get better value on wages when you get to pay under the minimum wage of your home country. And/or aren't required to offer benefits, vacation. And are able to work them twice as long without overtime pay. And don't need to care about child labor laws.

To be blunt: yes, slavery is cheap, isn't it?

> Unfortunately housing is super overpriced, due to the Asian mentality resulting in high property ownership.

I have no clue what this means and in countries like Japan, housing is a depreciating asset vs. an investment, so…?

> Real estate is always the monkey wrench in the gears of capitalism because of high necessity yet limited supply.

This only happens when the government becomes captured by land owners to constrain the supply, since otherwise you can build up. But governments getting captured by land owners happens a lot.

Packaging facilities are delayed but in progress.

> Shame they did not match each other in completion time.

Why?

If the packaging facility was ready early it would have sat idle losing money.

If it's ready late, products from the fab can obviously easily be shipped off to be packaged.

Tight coordination of timelines adds needless cost when there is an easy alternative.

Yeah definitely unfortunate. That said, I'm guessing the overall cost of overseas packaging is really tiny, otherwise Intel would've made a great customer since they are already packaging TSMC N6, N5, and N3 in New Mexico for their Arrow Lake CPUs.

It's parasitic, not productive.

A tick can contain a lot of blood, doesn't mean it produced that blood.

Because is fucking undproductive, useless and detrimental to society. Advertising is a cancer, an immoral activity.

a profitable market can still be unproductive if the overall result is a nuisance to society on almost every level

Stealing is raking billions every year as well, yet I wouldn't call it productive.

It doesn’t produce any things.

How do you feel about online gambling?

Imo, profits != productive or to a benefit of society.

By that definition, war is extremely productive

What makes you think pay is necessarily correlated to productivity?

Taken to the extreme, literal theft can pay well, and produces absolutely nothing.

Pay indicates the transfer of wealth -- it can be a heuristic for productivity, sure, but productivity is clearly not its only source.

I think about this quite often. What I'd really like to study at some point is: How much more does the receptionist at JP Morgan's head quarters make than the receptionist at Walmart's headquarters?

Because fundamentally I think there is an effect where the people in proximity to lots of money earn more. Obviously the Walmart receptionist and the JP Morgan receptionist are doing basically the same job. But the JP Morgan receptionist is surrounded by people who wouldn't think twice about doubling the receptionists pay and I would imagine that has a significant effect.

These companies hire all of these exemplary graduates and pay them so well because (1) they are flush with cash because businesses are essentially held hostage to adtech; and (2) so that they won't go out into the world and build systems that make them irrelevant, as smart people are wont to do from time to time. Someone on your payroll doesn't have the time nor the inclination to knock you from your pedestal.

Why else would Google need 182,000 employees? Or how about Facebook with 67,000? Microsoft clocks in at a whopping 228,000, and Apple at 161,000.

These are staggering numbers of employees. So many, in fact, that it would be an exercise in futility to try and manage so many for the number of products they offer, especially Google and Meta.

It's cheaper to make busywork than risk the cash cow.

Because there are costs that are externalized.

Options traders are paid well. It's still unproductive.

You're just shifting around a bunch of numbers temporarily to make a bunch of money for someone and lose a bunch for someone else.

Lots of shit we do is well-paid and unproductive.

If, as a species, we eliminated all bullshit jobs, there's a good chance only 20-30% of the species would be working. Here in America, only around 50% of people are actually working. Everyone else is in school, or retired.

I'd think otherwise and imagine these kinds of high-tech chip factory jobs are quite desirable.

Such a great victory for American industry... the future is to bring workers from Taiwan with skills and willingness to receive a fraction of US salaries.

How much does salary contribute to the overall cost of operating TSMC? Perplexity said that the average salary of a TSMC employee is $76K a year, and TSMC had about 80K people. So it cost them around $6B a year on salary. In the meantime, their operational cost was about $46B a year, so that's 13%. TSMC shipped about 16 million 12-in wafers. Each 12-inch wafer can make about 300 to 400 chips. Let's say 200 to stay on the conservative side. That will be 3.2B chips a year. That means the cost per chip on salary will be less than $2 a year. It looks HC cost is not that dominant?

We don’t. We expect some, but you’re right it won’t be transferred easily.

Surely there is a lot of software involved in the design / operation of these fabs, it's not just designing the chip directly. Another commenter mentioned EDA so maybe I'll look into that.

It might be possible but domain knowledge might give some candidates a leg up on the competition, going in blind just seems suboptimal, though most of the relevant EE undergraduate classes were in sophomore and junior level for me in the late 1980's and I only got to use EDA software when working a couple of semesters for AMD as a junior.

That’s very niche. Very few people in the general population will have heard of them.

Apple is well known. If they say the new iPhone SE 7 has a Made In America chip, people will know about it to buy if they care about that.

And it's also a pile of shit compared to an iPhone or a Galaxy S device.

There's your real issue right there. People are already paying $1199 for new phones. According to this article: https://www.vox.com/technology/2018/9/13/17851052/apple-ipho...

Another $100. That's a little over six years old now though, so bump it up to $200.

Would I pay $1399 for an American made iPhone with American made internals, as the article suggests it would cost ($100, but I doubled it for inflation, because, why not?)? You bet your sweet ass I would.

We just need to make sure that we never fight directly with another regional power, e.g. China or Russia. IMO, neither of them wants a fight with the US too, because you don't want to push a super power to the corner, EVEN if you think you are good enough to win.

In the mean time, the situation in EU and Asia is going to deteriorate and North America can absorb more talents as it sees fit. The last two times it was mostly EU but this time Asia might be the new talent pool we can draw from.

China invading Taiwan seems a ton more likely than China lobbing missiles into Arizona.

e.g. you can get 572 a15 dies per 300mm wafer at 90% yield. These likely weigh a few hundred grams.

By my rough calculations a million iPhones of a15s is about 200kg of silicon. excluding packaging, which would dwarf this mass entirely.