If that's true, how would one rank pumping supercritical liquid nitrogen at high rates through a heatsink? Super-grandmaster?

Seems like the heat flow would be substantially impeded by any boiling of the LN2.

Or, for that matter, simply using a chilled copper ingot as the heat sink? There must be some threshold at which the limiting problem is getting the heat out of the die, not getting the heat out of the chip's package.