Best investments for computing.
This is fundamentally wrong if you're looking for performance per dollar.
THis doesn't mean that if your workflow is MacOS based, they're not the best you can get, but currently, if your business is dependant on pure computational power, there's nothing in the Apple lineup I can recommend. If your business requires beating the competition's time to production and time to delivery, I cannot recommend Apple products currently.
an Example:
A large engineering firm that I have consulted with was losing out on bids against their competition. Their time to delivery was too long. Suddenly, despite them using the newest apple computers, their competitors were all using standard PC's. The workers were happy, but their production was 10-40% slower and longer than their competition. Apple could offer no upgrades to individual machines, nor could they offer machiens that were comparable in performance.
The machines looked great. Their users liked MacOS. But they were no longer winning contracts due to their longer turn around times than their competition.
Why were their competition producing faster? Because they were able to update more frequently, the computers and components than Apple has managed to deliver in the last 5 or so years. in Addition the Apple computers suffered from thermal throttling further reducing overall capabilities.
Would I recommend apple computers? Depends. the Macbook is a decent (if expensive) day to day consumer machine. The MacBook Pro's are great ultrabooks. But Apple has NOT kept up in the highest of high end. their product line has become more consumer focused, which is OK for consumers, But not those who need to eek out every single minute cycle of their computer.
A desktop computer should NEVER thermal throttle. Designing a computer that is so small / thin that tthermal throttling occurs in the desktop, is a design failure.
you and I might not notice this sort of behaviour in day to day usage, or even with Apple specific software. But Apple is NOT the best investment for computational power today.