[Dr_T]

I think so. I asked Vedder once how much he thinks his 60k ERPM 'rule' can be generalized outside of the ~60mm 14 pole outrunners he has most experience with, but he did not really address that yet. Measuring the Io for different RPMs, should give some insight in beyond what (E)RPM range the part of the non-copper losses caused by the rapid EM switching becomes dominant. I posted some simple test I did on my TP5660 once in my GT2 thread, but without reference, it does not say so much, other than that the amount of no-load losses, compared to my TP4070, did worry me a bit already.

That table you psted is a very nice overview; it does show the motors with higher Io also have higher delta T (which I guess represents temperature increase after certain time running unloaded). It is hard though to translate these figures to loaded behavior, as then the (load dependent) copper losses (indicated by winding resistance) also start playing a role. My Turnigy motor for instance had fried windings, suggesting it was the load dependent losses that fried it, and not the RPM depentent losses.

So big challenge I think is finding the right balance between the copper losses (high at start-up, high load; and higher for low kv motors, because more turns means higher winding resistance) and non-copper losses (high at high (E?)RPM, low-load).