Unfortunately I'm not that surprised by Apple's move here. Their entire manta is to cut down and simplify.
The resulting products are then dictated to the consumer as 'the new way', rather than listening to what people actually wanted in the first place.
Sometimes this works, sometimes it doesn't. Just listening to the common crowd could impede real innovative spark, but in this case there is no need for a smaller form factor machine at sacrifice of major functionality and performance.
What they have done is effectively sidestepped everyones requests for expandability and passed them onto 3rd party Thunderbolt vendors.
They created a platform where they cannot be held responsible for certain compatibility or features, because if it's not offered by Apple it's another hardware vendors fault/problem now. If it doesn't fly with the vendors or pros - or worse still, their hope for an expanding pro-sumer market, they can just kill it and say - "see that's why we hesitated to release a new one!".
In stead of trying to please everyone, they have gone the opposite direction and purposefully pleased no-one. Someone would have complained no matter what they released, but I don't think anyone would have complained that a more feature packed powerhouse wasn't 'small enough'.
I agree that the design and engineering in this machine is impressive. It's a cool idea. It's not however a professional workstation by any stretch of the imagination
in this current environment.
By 'in this current environment' I mean Apple have designed something far ahead of its time. Someone can probably point out past examples where that have done this, but I'm going to give an extreme view of what I think Apple were thinking.
If cloud storage and cloud processing was more advanced, this machine would be great. People are complaining about needing cables running everywhere
well I'm sure Apple would like you to have no cables at all in the future! Use the super fast PCIe SSD for you OS and apps only. Connect via wifi to your music and photo library either in the cloud or in a central home server in a closet somewhere. Connect via thunderbolt (adapters) to a fibre connection for heavy duty work.
Of course my assumptions are probably very flawed but you get the idea.
What is a massive failure is that TODAY you still need to compute inside the box, and their offering has absolutely no advantage what-soever over other competition.
Mac pro will have a 12 core processor
wow, I already have that today, and later you will be able to get a PC with 2x 12 core CPUs.
Mac pro will have 2 x super fast GPUs
wow, Tutor has already built a rig with 8x GTX Titans housed INTERNALLY.
Mac pro will have a super fast PCIe SSD
wow, these have been available in PCs for years already. I bought a 240GB OCZ one for $200.
Yes you can buy countless thunderbolt enclosures and jam them with as many GPUs as you want
however I think the Thunderbolt bandwidth would be a huge bottleneck on possible performance.
And to those saying you can fit more Apple units into a rack space for a render farm
well you can fit 2x or even 4x the amount of processors onto a single PC motherboard without the cost of the extra GPUs, SSD, case etc.
What is a real embarrassment for Apple is that this machine will not even have the raw computing performance of what you can do with 3-4 year old technology already.
Unfortunately most people on here are tech-heads who see the value and difference in these performance specs, hence the fierce debate. Whereas A LOT of consumers and even pros are easily convinced that newer = better and will blindly buy into it.
It makes me feel a little better that my Hackintosh could still possibly come out on top of Apple's BEST offerings for another few years. And it was the first computer I ever built. By the time it will be beaten, it could be running on five or six year old tech! (With exception of course to current GPU upgrades
but that's the advantage - I CAN upgrade easily.)
I do agree with previous statements that a big part of this move is to use the smaller size to cut down general production and also shipping costs. It's common for Apple to have stocks in very short supply after a new release. If you can fit 2x the amount of product in a shipping container, you can ease this problem somewhat.
Anyway enough of my rant, I was prepared to give Apple up to $10,000 of my money if they released a decent machine a year ago. They forced my hand and now I have learned to build my own machines so they will never see another dollar of mine again. After this release I will also be building a fleet of massively spec'd Hackintoshes for my workplace as we eventually transition over to Windows. That could be over $100,000 of revenue lost from just one person, and I'm sure I'm not alone!