[Pxtl]

Which seems like a reasonable concept - if the big cost of subways is wide-diameter tunnels, experimenting with narrow subways is worthwhile. A Subway with a limousine form-factor, for example. But it's absurd how many permutations will have to happen before we land there.

[ComradeTaco]

If you have the kind of nice geotechnical conditions that permit the use of tunnel boring machines then tunneling isn't actually very expensive. What's killer is the cost of subway stations, which need to be dug out manually.

[WalterBright]

Unless the stations are put on the surface.

[Pxtl]

I thought cost-savings from a system that used narrower bores was the key to Musk's whole strategy? I thought that was the whole thing - by designing smaller hardware he could make underground transport affordable?

[ComradeTaco]

To clarify, I was talking about traditional subway construction and why changes in boring costs would not change that.

[jcranmer]

> if the big cost of subways is wide-diameter tunnels, experimenting with narrow subways is worthwhile.

It's not. The tunnels generally account for at best 1/3 the cost of a subway project, and even then, the cost of the tunnel that's variable per diameter is an even smaller fraction (since you're sticking a lot of other stuff in the tunnels). The big money sinks tend to be the underground stations. Overall, moving to a smaller diameter tunnel might save you around 10%, and you'd probably eat up much of those savings with rolling stock incompatibility issues.