you’re right @CreekWalker it would be helpful. i think there’s a number of people who’d like to know that as well.
i think it’ll be possible to have a decent answer once a number of things are know. things like how many pools (1,000? 2,000?) is optimum for the network to run efficiently and securely and profitably/sustainably. that would then affect the pledge amount as well as the costs cause these two aspects closely affect each other. then lastly - the level of hardware/knowledge needed.
if raspberry pi will be capable, and technical knowledge to start a pool would be at a level that many could start one following a tutorial - this would also mean more pools which affect the rewards, paying out less to more pools - so even though the hardware in this scenario would be cheap, the rewards would be low as well.
the opposite (expensive hardware and high technical knowledge resulting in fewer pools with higher rewards for fewer people) would also be true.
Personally i see it landing somewhere in the middle.
what we do know (from what Duncan has said) is that if you have a lot of Ada (in the millions) it would most likely make sense for that person to delegate than to start a stake pool themselves cause it won’t be profitable for them and worth their time.
either way good question @CreekWalker , would make for a good video topic to make, which i might do at some point!