What I think is quite interesting about these era of machines is where the power goes. First off, there aren't actually many heatsinks in the machine, and those few that are used, are really tiny. This means that the power used must be distributed rather evenly among all components, even on the CPU boards. The CPUs can at most pull 10-15 W with those heatsinks, meanwhile the big ASICs spread around everywhere can probably sink perhaps 3-5 W with no extra heatsink. Compare this to modern systems, where the (single chip) chipset has a TDP of ~5 W, while the CPU(s) run anywhere from 50 to 100+ W.

So the gap between high-power and low-power components was way, way smaller back then, and I'd wager that a huge chunk of power in these SGI machines went into their high-speed bus systems.

Another big difference is how the power system was laid out. Here are the specs of a SGI Onyx 2 PSU [1]:

    1750 W @ 230 V
    3.45 V, 375 A
    5 V, 85 A
    12 V, 22 A
    5 V, 1 A (probably standby power?)

In that day and age, most of everything in there would've run off 3.3 V. So the PSU directly, centrally provides that voltage and due to the large currents droop and such will have to be considered carefully in the entire system design (and it almost certainly uses remote sensing at the backplane, perhaps even sampling various points in the system). This makes sense, because there are many components and the difference in power between them is kinda small.

Nowadays, a computer PSU provides essentially two voltages: 12 V and 5 V standby. It still provides "legacy" power, like 5 V and 3.3 V rails, but these are derived with simple buck-converters from the 12 V rail(s) in modern designs. Because a modern computer has few components requiring a lot of power, each of them has their own local converter sourced by 12 V to reduce losses.

[1] https://web.archive.org/web/20170314114558/http://sgistuff.n... https://web.archive.org/web/20170314160011/http://sgistuff.n...