7000 * 25t / 2.5t/m^3 = 70,000m^3 of water. So about 1% of the water storage of Dinorwig, which runs dry in about 6 hours of use.

Now, I know reservoirs are ecologically pretty iffy, expensive and obviously geographically sensitive, so you can't slap them around everywhere. But all these mechanical schemes have big "look what they need to mimic a fraction of our power" vibes!

I could imagine that steel-on-steel block movements could actually be quite efficient and effective in limited scenarios, but logistically it just seems like a lot of squeeze for not a lot of juice considering how much power is required to be stored for utility-scale projects. I would like to say that that they're just delusional people truly hoping it'll work, but I think there's a core of hard-nosed scammers who smell money for a shiny PPT and a plausible-to-non-engineers Wile E. Coyote/Troll Physics contraption with big numbers in the brochure: 7000 blocks! 25 tonnes per block! Megajoules! Efficiencies! Scale! Repurposed coal mining infrastructure! They even have AI in the spiel now: https://www.energyvault.com/solutions/software

Or maybe it'll work and I'll look stupid in 30 years when there are huge fields of hundreds of kilometre-deep boreholes with 100 kilotonne masses moving up and down in them. But somehow it seems quite unlikely on a practical level considering the cost of boring gigantic holes that you'd have to do to make it scale. Onsies-twosie installations in a few mines here and there may work for lucky outlying towns, but they aren't civilisational scale solutions.

Gpt says that would require about 275 million steps on a magic rigid weightless stairway. Roughly 4.1 million calories. So at Phelps level energy expenditures you are still talking over a year of climbing every day.

That doesn't really escape the gravity well, does it?