This, and it's incredibly depressing not merely in its result but in its sheer needlessness. Sometimes even the best companies make choices that really don't fi

This, and it's incredibly depressing not merely in its result but in its sheer needlessness. Sometimes even the best companies make choices that really don't fit someone's use case, maybe to the extent of ending the relationship, but it's possible to step back and see the business case for it. It's not happy but it can be logical. But I always feel a special set of negative feelings when a company I like well enough does something bad and it's just doesn't even make any sense period. In this case while yes the old Mac Pro case could have used some minor refreshing to make better default use of space in a world of 2.5" drives and such, frankly I suspect most of us would have been reasonably content if Apple took the absolutely utterly laziest, cheapest possible approach, kept the exact same thing and merely kept updating the chipset and CPU and graphics card and that's it. SB-E, IB-E, etc., the same old dual socket boring workstation, just with the internals refreshed. Maybe swap the ports for USB-C eventually. They could have kept a higher profit margin, done much less engineering work, and had a better product. Instead they actively expended money making it fundamentally worse, and not just that but making it a zero-sum game. Many people correctly compared the Cylinder (it's hard to resist the more colloquial in our office "trashcan mac") MP to the old Cube in terms of basic thermal and design tradeoffs, but the Cube never inspire active dislike because there was still the MP. On the contrary the small niche of users that actually found its set of tradeoffs worth it had a soft spot for a long time. Nobody lost anything at all from the existence of the Cube, so the only people who got were those fine with it, and anyone else could just look from afar. By making the neoCube the only thing though Apple shot any hope of that. And the Cylinder was form over function in a way I can't recall from Apple in a very long time, and also seemed a fundamental misreading of what they're good at and what they want to be actively engaged in. The mirror of an easily expandable pro friendly system offering flexibility on user side is that it also provides flexibility on the manufacturer side: because 3rd parties can do more entirely on their own, the manufacturer in turn can get away with doing less because the gaps can be filled in by the market. The more control a manufacturer takes for themselves, the more responsibility falls on their shoulders in turn, which is why command-and-control economics tends to have scalability problems vs approaches that make more use of markets. At any rate it's a real bummer. I've looked at Xeon hackintosh efforts but that's definitely getting into pretty shaky territory and with Apple abandoning pro systems themselves support is probably only going to get worse. I agree it is time, probably past time, to start planning to abandon the platform but boy will it be painful (closer to 25 years in my case).