the real question is at what point to fix the flow for the fixed pump power tests.
is it possible to "fix" the flow rate for your loop (regardless of number of pumps)? i'm thinking some kind of ball restricter that artificially restricts the total loop flow to that of the least flowing CPU. note: you would have to adjust each loop manually, but it shouldn't take that long, and would be pssible to "fix" any loop at a specified flow rate...that is about as apples to apples as it gets, and would allow you to test the ADP2.