Fair enough. You can call it the highest - value part of the chain or the least commoditized part, but it's still very far from meaningful supply chain Independence and a long ways from "chips made in America"

They don't even make chips in Arizona actually.

Arizona only performs middle-of-line wafer processing. It's a tiny slice of the value chain. Almost the entire semiconductor production chain is in Asia still, which is why the TSMC Arizona fabs have no technological or economic point to exist. Their only possible point is geopolitical or strategic.

All the actual wafers (substrates) are made elsewhere and shipped into America. Along with most of the equipment, a lot of the labor, and many of the chemicals (sulfuric acid in particular is still being shipped to the US from Taiwan because there is no suitable US supply). Then, after shipping all that shit into the middle of the desert Southwest, they perform wafer processing in Arizona, and the complete wafers are shipped back overseas again and sawed up, tested, and packaged, potentially assembled onto PCBs, all overseas.

Why do you want the transistors to be smaller, anyway?

The historical reason for shrinking transistors was cost and yield (almost the same thing). By having higher transistor density, you get more chips per wafer geometrically, and you also get more yielding chips per wafer because every killer particle only kills 1 chip, so more chips per wafer translates into higher yield even with the same number of killer particles on the wafer.

There are other benefits to advanced process technology, like lower power consumption (in some cases) and higher speeds (in some cases). But there are also problems, like other yield problems because the process is so difficult, and higher leakage current, and poorer data retention in the case of memory, etc. So it's very possible that shrinking transistors INCREASES costs, then there's no point to doing it anymore.

Furthermore, shrinking transistors only improves density if you can also shrink your metal 1 pitch, and M1 pitch has been a density bottleneck for a long time.

For a long time, the key to lowering costs was to shrink transistors. But this is 2026. The reason to shrink was and is cost. There's no intrinsic technological value to shrinking; it's only a single (important) way to unlock performance per $ and per Watt.

No. You are getting confused by communism. This is fascism.

When the government takes over industry, it's called communism. When industry takes over government, or collaborates closely with government, it's called fascism.

Maybe cars shouldn't drive in the fucking city where the people are?

Or really, maybe sometimes roads are used by a variety of users including people riding bikes, crossing the street, getting into and out of cars, working on broken cars, doing road construction and maintenance, making deliveries, going from cars to school, or any number of thousands of reasons a cars might (and do) collide with people in the actual real world...and cars have no choice but to leave their highways and drive directly to people's homes, businesses, schools, parks, supermarkets, and other places people live their lives...so making sure that car drivers can see pedestrians makes a lot of sense?

Couldn't we solve this by simply creating visibility standards?

There's no forward visibility standards at all. Right now, you could build a car literally with an opaque windshield and it would be OK to sell. Why shouldn't there be a minimum standard for forward visibility blind spots, maximum distance you should be able to see a traffic cone or 5-yo, etc?

You will need to fix the tax code and CAFE for that. They will continue buying F350s as long as they can write them off on their taxes.

Isn't interesting to ponder what the market would look like if tax policy encouraged them buying efficient vehicles instead of encouraging them to buy the least efficient and most dangerous ones?

And in the bigger competitive picture, you don't have to choose between a cheap car that's missing basic features, and an expensive car that has modern features. You can get a cheap car that ALSO has basic modern features like power windows, as long as you buy from a capable, modern car company i.e. a Chinese company.

This is the same dynamic that happened in the 1980s with Japanese cars and the 90s'/00s with Korean cars...American companies loved to lock basic features like power locks, power windows, and then-advanced "climate control" behind expensive trim paywalls. The other companies realized that power locks cost a $5 solenoid and some wiring, and power windows are actually cheaper than crank windows in volume, and climate control consists of a $0.50 thermistor in the rearview mirror and a few lines of code, and all of it was getting cheaper by the minute because of efficiency of scale, so they could offer all these things in the cheap cars just as well, and guess what, people decided they liked driving cheap cars that had most of the key functions they wanted.

This $25,000 Slate truck is only interesting because you aren't allowed to buy competing $25,000 Chinese trucks that blow it out of the water by offering all the normal, modern car features at the same price.

How about congestion pricing, which 1) provably works better than anything else and 2) actually generates revenue instead of costing money?

Nah, let's just blow 30 million on a system that will, if anything, make traffic worse.

Your puny 120kW is fine for you and it's fine for most people, but you have to understand that flash charging isn't about the consumer demand, and it's not about road trips. It's happening way faster for that to be the case. Flash charging is about making more money for charge point operators and the car companies.

There are two reasons for hyper-fast charging, and neither one is for road trips:

1) People who live in apartments. If your car has 5-minute charging, you can sell EVs to more of these people.
2) Charge point operators. Charge point operators make a lot more money off of faster chargers than they do off of slow chargers. The economics of running a charging site strongly favor faster charging. Faster charging means fewer stalls needed, less land and queuing space needed, more people stopping because it's more convenient, more people stopping because there's less chance of a queue. Since you charge money per kWh, basically the rate you can make money off of charging is directly proportional to how many Watts you can push out. And it's more economical to push more Watts with flash charging.

In the early days, I expected companies to charge more money for faster chargers. I expected there would be slower chargers, fast chargers, and "premium" hyper chargers that cost more. But in fact, it's going to be the opposite. We will probably see lower rates for faster charging. In fact, this already exists now...some chargers begin charging idle fees once your car gets to 80% and starts charging slower. You pay more, if you want to charge slower!

This is why flash charging is being pushed so hard. It's profit incentive both for the car companies and for the charging companies, not value added to most customers.

Hyper charging is emphatically not needed for road trips. This is why Teslas still charge slow as fuck. Tesla charging speed has actually gotten worse over time. It's because their customers live in America, they tend to own houses and charge at home, and even "slow" fast charging is totally fine for road trips already. That's why Tesla "super"chargers are not even fast. Because for road trips, which is the Supercharger use-case, it doesn't matter that much.

I have two gas cars and an old Kia Niro that charges at 80kW on a good day. I need to drive between Boise and Portland every so often. Most people assume I drive the electric around town, and take one of the gas cars to Portland, but it's the opposite. I only drive my gas cars around town, because the miles are low and gas consumption is basically insignificant. I take the Kia on road trips because it's slightly cheaper (I get the same reimbursement from work) and electric cars are nicer to drive. I just put it on cruise, and it climbs all the mountains with no shifting, no noise, and it goes down the other side perfectly in cruise control, no brakes needed, just put it in cruise and drive.

Even though it charges at a pathetic 80kW, I never really wait on the car. I just plug in when I stop for pee/coffee/cleaning off bugs from the windshield breaks. I probably spend less time charging the Kia than I would filling up a gas car once or twice, because I don't have to stay with the car while filling gas. And that's at a "pathetic" 80kW car. I now understand why American car companies keep putting out cars like the Equinox, that only charges at 150kW, or the Toyotas, that also charge at less than 200kW. I used to think the 250+kW charging speed like the Hyundai's was going to be some big unlock. But now I realize it doesn't matter because most Americans that can afford newer cars live in houses, and we are already at the zero-time-cost point for road-trip charging. The next bottleneck will be when enough people start using EVs that queues start to form at the highway fast chargers, and since land is cheap in America, the response will probably more like what Tesla already does...just put in a fuck ton of stalls so people can charge while they pee, rather than a few hyperchargers that people have to keep moving their cars through.

Which is very similar to how SSDs work. They all include buffer capacity so as the NAND blocks die off, the user never notices. So the 1TB SSD may contain 1.5TB of NAND inside, but you don't know, and you don't need to know, as long as it keeps delivering 1TB of storage.

Different companies have different levels of transparency about battery state-of-health, and there's no standard right now to make sure you can compare between manufacturers. Tesla tends to report state-of-health that drops from 100% relatively quickly into the 90s in the car's life, but then stabilizes at a slower degradation rate over the long term, which is close to the truth about how batteries actually behave. Ford tends to report 100% state of health for a long time. It's easy to assume that Ford thinks the consumer expects their "new" car to stay new for a long time, so they want to show 100% state-of-health to avoid spooking customers and reduce warranty or trade-ins. In reality, Ford just programmed in a certain amount of buffer capacity into the life-o-meter so the customer THINKS the battery is at 100%* health (where 100% is defined as e.g. "above 90%"). Ford doesn't have better battery technology than Tesla, they just have better marketing.

Lightnings are super popular around here with contractors. They spend most of their time driving around to sites doing estimates or doing supervision. Occasionally they need to haul or tow something. Lightning is pretty much perfect for that use case. The ones with diesel trucks often leave them sitting around idling with the air conditioning on all too much.

Not to mention, "maximising share holder profit" is subject to judgement and interpretation, especially with respect to the Buxton Index (the time horizon over which an entity makes plans).

It's very common for companies, even big tech companies like Amazon, to say "we are deliberately going to show lower profits this quarter because we are planning to maximize profits over a longer period of time". This is so common as to be universal. You can just as easily say "We value our company's reputation and believe in our strategy, so we are not going to engage in (unethical practice X) because we legitimately believe it's in the best interest of the company long-term".

This is not to say that moral hazards exist, and most of those moral hazards can be blamed on regulatory environment that creates them. But whenever somebody says "we have no choice but to maximize quarterly profit", it normally translates to "I get a bonus next quarter if quarterly profit is up, and I don't care what happens to the company in 5 years because I'll be gone by then".

So then you keep using diesel trucks on those types of routes then?

I'll never understand why, on a site supposedly full of tech people, the constant framework of technology adoption is that any new technology must be capable of completely displacing all other technologies in all other uses cases, or else it's worthless. That's not how literally any technology adoption has ever progressed in human history.

European trucks are significantly more capable than American ones. American trucks really haven't technologically advanced at all. The last advance was when Detroit Diesel introduced electronic engine management in the 90s.

The US maximum gross weight is 80,000 lb, while the European standard is 88,000 lb, with categories up to over 200,000 lb.

Higher-power European (diesel) trucks also have more power, with up to 770hp, whereas US trucks rarely go over 600hp.

European trucks are also more aerodynamic (counterintutively), and also have more advanced tires and axles, so they get better mileage on a per-truck basis. But this is even more important on a per-ton basis, because weight doesn't make a huge impact on efficiency compared to aerodynamics. A European truck might get 8mpg hauling 88k pounds, vs. an American truck getting 7mpg hauling 80k pounds, for 30-50% more efficient per ton. All of this with lower emissions too.