While I'd suspect the design is still in flux, the current design is for a 120kw satellite with 110 square meters of radiators. Scaling to hundreds of gigawatts is intended to be by repeatedly launching smaller designs.

piva00 2 hours ago | parent [-]

300GW / 120kW = 2.5 million satellites, I don't think SpaceX can launch 2.5 million satellites. Even less keep replenishing all the ones needing decommissioning after 3-5 years, no maintenance can be made, so on and so forth.

It's ridiculous anyway you cut it, it's a pipe dream.

▲

ben_w 2 hours ago | parent [-]

Indeed; though one thing I've found researching this is that the exact numbers are all over the place, so for some it's "let's make a single giant DC" and others are "let's make one million small ones", with mass estimates for each bird in the bigger constellations varying from 1200 kg (at 120 kW, which is an absurd ratio) to about 8000 kg.

Like I said higher up, at this scale, if you want to make your own fantasy plan you can draw a contiguous ring filling a single orbit.

	▲	Ekaros an hour ago \| parent [-]
		Pretty much. There isn't really question could you build satellite with GPUs at certain scale and launch it to space. It is certainly already doable. No special unsolved engineering challenges when you have say dozen or hundred... Now if we are talking about thousands or millions you have some real questions. Well cost was question to start with. But at millions satellites yeah there is clear issues. Simple rule for numbers 1 gigawatt is 1 000 megawatts or 1 000 000 kilowatts... So math is pretty simple there in estimations.