And yet, the cost of a kWh from renewables is less than from nuclear, for all the supposed advantage of nuclear's energy density. This should tell you that energy density is not actually a good metric.

Comparing a source controllable uninterrupted source of energy with an unreliable intermittent one doesn't make sense. Having power when you need it is very valuable. And in the case of renewables, you'd need to account for storage, after which the prices explode.

What would you say if you could only turn on AC in the 3 months of winter?

But there is another ugly side with regards to price (which are heavily subsidised): it is mostly cheap because offer far exceeds demand when power is available, because when it's sunny, it's sunny everywhere nearby so there's nobody wants to buy. Peak demand are at 7-8pm on during week days in winter. Good luck with solar or wind.

On a bigger scale, see Denmark (invested heavily on wind) vs Norway (almost entirely hydro). When wind is up, Denmark, which _has_ to sell this energy, sells it to Norway, which will buy it at the effective operating cost of hydro, which is ~$0. And when there is demand, Norway will happily sell it to Denmark at a normal price. In effect, Denmark pays twice.

Intermittency and power density are two entirely different things. I was addressing the former there. But elsewhere in this thread, I address the intermittency issue. Constancy is an advantage, but not an unlimited one. If renewables have a sufficiently large levelized cost advantage they can push nuclear entirely out of the picture, even though they are intermittent.