How is that reasoning specific to fusion?

Unless fusion power is dramatically more efficient than other thermal plants, like 99.9%, your bigger plant will still need massive heat removal structures and systems, which means siting them near water. All the good spots are already taken.

Alternatively you can use truly massive air heat transfer structures, driving up your construction costs again.

I neglected to mention finance costs also. With an untried technology the rate of return demanded is going to be very high, further driving up project costs.

Again this assumes fusion power will be cheap. But in reality it may cost about as much as (fission) nuclear power. So yes it may help with making energy production carbon-free (together with solar, wind, hydro), but it won't necessarily create a some sort of energy abundance.

> One of the big problems with existing power plants is that you have to build them near where people are.

I don't know how it works in the US, but this is notably not true in the UK and Europe. Gas plants are comparatively small and nestled in, but big coal (to a limited degree) and particularly fission plants are frequently in the middle of nowhere. They're somewhere near a village that can supply a workforce, but siting concerns for nukes were more based on making sure any criticality excitement could be shared with neighbours across whatever nearby border was handy than putting them anywhere near cities.

Retail electricity prices in France are €0,13/kWh, so not that cheap (€0,18/kWh pre pandemic).

https://www.statista.com/statistics/418087/electricity-price...

This doesn't matter, though, because France doesn't need many nukes anymore, therefore doesn't subsidizes new plants anymore. One new plant is built in France, and it already is hellishly expensive: "As of 2020 the project is more than five times over budget and years behind schedule. Various safety problems have been raised, including weakness in the steel used in the reactor." https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plan...

There are two problems with this reasoning. First, you are comparing the cost of written-off plants, for a technology that has huge upfront costs and much lower fuel costs.

Fission is not cheap if you build a new nuclear plant, not in the Western world. That's why almost no nuclear plants are being build. Making a safe plant is just really, really complicated.

Second, this specific argument can be used to see why fusion is a pipe dream. The primary competitor to fusion is fission. And the fuel costs of fission are pretty low, as you just said. So fusion will not be competitive unless you can built them around the same price as fission plants.

Someone else in this thread talked about S curves. Well, those kind of S curves happen for tech that gets produced in larger quantities, where it is economical to spend engineering resources making the production of the tech cheaper.

People often make the mistake of looking at the reaction energy of fusion, and comparing that with fission or burning fossil fuels.

But the majority of the reaction energy is carried away by high-speed neutrons, which are pure waste - they can't be captured by magnetic fields, they are heavy and penetrate almost any material, leaving holes behind that make the structure brittle, and when they do get absorbed, they make the atom that absorbed them unstable, turning the material radioactive.

So, at least as long as we use neutri-producing fusion (and any realistic fusion reactor has to) the actually usable energy is not that impressive compared to fission.

Please explain how it generates that much electricity that cheaply.

Direct conversion is theoretically about 60% thermally efficient, on par with combined cycle gas generators.