Well look at it as reading internal temperatures then. In that case, the new model internal temperature is lower.I'm not entirely convinced of taking a measurement of backplate for the CPU side though... and rather skeptical about this too on claiming "better" either. It seems at least CPU part is just as inaccurate as Austins.
The mounting plate may have different thickness due to lighter/smaller heatsink design, or made with different thermal conductive material. Too many variables to consider. 10 degree difference is a big telling sign too, even though I'd be happy to admit and congratulate the new consumers with cool-running chip compared to my old PS5.
As far as I know, the PS5's backplate is part of the general cooling - the ram modules with small rear heat sink/pipe that runs around. Other places (RAM etc), I can generally understand the thermals may be more likely in the ball park as those are at least directly connected to the back plate via thermal putty - but the big important CPU part is a big question... when you consider at least 2-3 layers of different materials that could have different thermal conductivity and all.
At least we know that it uses same fan running at same RPM. Variables are reduced - so kudo's to them for confirming it.
Hope GN gets the unit soon from DF, although it may take a little while till he finishes testing/shooting and editing etc..