This cost minimization seems quite strange when every year people were anxious for a new Pro. How hard it is to make the trashcan bigger and put in more airflow? (Bigger fans with the same RPM.)
And sure, it might cost some, but Tim should count lost profit (opportunity costs) too.
>How hard it is to make the trashcan bigger and put in more airflow? (Bigger fans with the same RPM.)
My guess is that because they tried to manufacture and assemble in US, so they had very high automation for the TrashCan, which also meant these automation are likely not flexible enough to change the design. I don't think the Mac Pro was about cost though. Because Mac Pro unit shipment would not have made up for that investment anyway. I think it was merely an exercise for Apple's operational supply chain. And Apple's design team had lack of time and had to focus on many other more important issues. That was why it took so long. Basically the design department in Apple were not scaleable.
Design bandwidth limitation is pretty plausible. (I have no idea how hard it is to change automation, but my guess is not that hard/expensive. After all assembly robots are not custom built, they are probably programmed like the simpler CNC machines.)
Plus on top of that probably in Apple's hivemind the cost of coming out with something bad is more than the opportunity cost of missed profit.
And sure, it might cost some, but Tim should count lost profit (opportunity costs) too.