Apple wouldn't even offer significant video card upgrades for Mac Pro towers with simple PCIe slots, and demand was insanely huge.
There is little to zero evidence that demand was insanely huge. Severely out of balance supply/demand typically brings more suppliers. There were not any 3rd party vendors stumbling over themselves to provide solutions for the Mac Pro market.
You want to know how much Apple cares about upgrades? 2009 Mac Pros ALL have an SMC bug that causes the PCIe/PSU fans to spin up to midrange speeds with any video card upgrade, even the OEM Apple video cards sold at the Apple Store!
Again, in normal procedures bugs go through triage to be prioritized by impact. Far more likely this didn't get fixed because it did not impact that many users relative to other bugs on the queue.
So now Apple will suddenly offer a worldwide service for performing complex upgrades to old hardware expressly designed to limit upgradability?
If it was $600M/yr business then yes. That would generate about $600M more per year than fixitng that bug would. It is a revenue, rather than cost, center.
If Apple wanted to offer video card upgrades for a new Mac Pro, they would have designed it so Joe Sixpack could visit an Apple Store with $600 in his pocket and walk out with a video card he could install within five minutes without any tools. You want to know what such a design would look like?
Check it out here.
If that was a successful growth business for Apple they would have tracked that business with more timely upgrades. They didn't.
The fact is the underground hack video card market is where a signficant number of buyers were. Very few wanted to pay anywhere near $600 for a video card. The few vendors who did put in the extra work were largely undercut by folks who hijacked the work and undercut both the 3rd party vendor and Apple on the revenue. So no that wasn't a big motivator for Apple to keep the same model going.
The reality with the new Mac Pro is that the primary replacement card vendor is Apple. Unless folks present a case to Apple that it is a profitable business to be in they won't do it. This really isn't all that big a change from the current set up where the default BTO cards are the primary cards variants sold.
The "mainstream model" is a standard for good reasons: it enables users to upgrade video cards on any tower with PCIe graphics. It's a valuable feature because it enables the buyer to prolong the usefulness of an expensive computer. It's innovative in that it promotes GPU innovation among competitors offering video card upgrades.
Promoting competition is not innovative in and of itself. That is a different dimension. The mainstream model is standard is largely because that is the way things got done with the limited amount of integration available. It is the same form that was in place before could do things like laptops, handhelds , and tablets.
People vote with their dollars. Where dollars are being increasingly being poured in now is not with box-with-slots. People have higher value on the products they buy as opposed don't buy.
You seem thinking of the video card design of the iTrash as being somehow innovative, or better than the "mainstream".
That's your connotation, not mine. The video card design is following basic market trends (i.e., what people are buying). Apple is not trying to cover as many folks as possible with their product. Since the return of Jobs they have consistently gone after selected subsets of PC users. The argument that Apple "has to" offer product XYZ because all the other PC vendors are offering XYZ has not so much to do with innovation as focus.
Apple's "innovation" is in forcing users to buy new workstations instead of upgrading components.
That is your opinion. I doubt you have real broadly sampled quantitative data to back that up. The broad "flattening" sales trend in the overall PC market and even more so in the workstation market is that for increasingly larger sets of users the PC/Workstation they have is good enough for large blocks of time. The components are flattening out also. AMD and Nvidia graphics segments "print money" of high mark up pro cards.... not on overall market.
http://www.anandtech.com/show/6137/the-amd-firepro-w9000-w8000-review-part-1/3
The unified thermal core has no benefit to the user except for enabling a tinier Mac Pro.
Presuming not overloaded in most configurations, it has a few number of moving parts. If there are usage patterns were not all three are 100% loaded at once it should run cooler with lower complexity.
No merits other the origami effect is more than a bit of underevaluation.
There are numerous video card models with silent coolers that fit in a standard PCIe slot, so the iTrash isn't breaking any sound barriers.
I doubt you are equating the TDP constraints of the cards to these. Apple-to-Oranges solutions don't really hold much water. Apple's solution is not a passive one. Have no idea why you are comparing to one.
And it isn't about individual component it is about overall system noise. ( yes there are just as quiet systems but how they get to that state is also a factor. Making noise and then spending alot of effort to cover it back up isn't addressing root cause issues. ).
From the looks of it, the iTrash will have a substantially lower BOM than the Mac Pro tower.
I have no idea why you would think so. Other than not being a dual CPU package model, in a single CPU package to single CPU package comparison Apple didn't drop a single highest cost component. In fact depending upon how the GPUs are priced may have added an additional one. This is not a "goose the profit margins higher" move. Not by a long shot. The price points are probably going to around the same range as the previous single CPU package Mac Pro's and the high end BTO options go a bit higher.
Dropping empty space doesn't save a ton of BOM costs.
ODD. no. Multiple fans. no. Swapping a relatively mid range HDD for a PCI-e SSD solution, that is a BOM cost lowering move? Not even hardly.
Neither is swapping 1 FW controller and two standard PCIe socket connectors for three Thunderbolt controllers.
My guess is that it's profit margins will break new records for Apple, and so will shortness of buyers' upgrade cycles. Or maybe they'll just give up and use Windows.
Your guess without seeing the prices, or even projecting them, don't really amount to much. There is no way to get to margin unless have price.
