If thats the case Apple would have canned the Mac Pro and left the Xserve. Why would they turn the Mac Pro into a rackmount when the Xserve already exists?
That makes zero sense. The XServe sold in far fewer numbers than the Mac Pro. The machine that will survive is the one that sells in larger numbers. As of today the XServe doesn't exist as a product in large part because it couldn't "make the numbers". However, these threads are filled with folks moaning about the gap in Apple's product line up for a more easily "rackable" Mac that has the chops to more load than a Mac Mini.
A rackmountable tower can be both. For example:
http://h10010.www1.hp.com/wwpc/us/en/sm/WF06a/12454-12454-296719-307907-4270224-3718645.html
Form factor
Rackable minitower
The primary reason the current Mac Pro isn't more easily rackmountable is the somewhat gratuitous handles (at least from the racking perspective). That's it. Minor case redesign and it will fit with some optional attachments (e.g., handles or rackframe attach to core base unit. Alternatively a deatchable stand for when floor mounted and handles intregrated into the core "rectangle" like the z800. Or ... about 3 other options that spring to mind that have been implemented. )
It is not like there aren't several vendors who have
already solved this configuration flexibility issue. Apple doesn't have to directly copy them but to assert it is difficult or not cost effective (at Apple's price points) is bizarre.
No. It won't be optimized to take absolute minimal space, but it doesn't have to be. If you look around lots of folks only need 1-4 boxes. The rest of their equipment is something else. Sure there are folks who want some "feature 42" that the XServe had. However, if that wasn't a standard feature that a very high percentage of the customers leveraged on a daily basis , it is not a mandatory feature in order to capture some of the former XServe market.
The objective of making the Mac Pro more easily rackable is to increase the number of Mac Pros sold. If they go too low and don't show healthy growth that box will get "end of lifed" also. A modified box doesn't have to draw in all of the old XServe sales, but even 25% would be very helpful in sustaining the Mac Pro (and Mac OS X Server) as a product into the future.
While the XServe existed the "too tall to mount" Mac Pro improved product differentiation for Apple. They drove some folks to the XServe. (just like they drove more than a few folks to XServe when didn't have a sub $1,500 server offering). Without an XServe in the line up there is zero need for that anymore.
As for case redesign not needed or "costly". WTF... the next generation Mac Pro needs a new motherboard anyway ( whole new Xeon family and support chipset being introduced, etc. ) . If there was an opportune time window for a case modification this cycle, due late Summer/ early Fall, is it. It was the
wrong time to toss it out as a "just in time" for the XServe retirement. The Mac Pro had just came off an update and it was a "no new motherboard, just a 'tock' Xeon with shrink and minor updates". Mostly a minor firmware bump and a new chip that plugs into a socket anyway.
Apple did exactly what was low cost to do. They put a new SKU in the product line up with the current Mac Pro design. If your argument is the "cheapest" solution; that's it. Furthermore, Apple has a fixed set of enigineers to do Mac development with. With the addition of more MBA models it isn't very surprising that a product that doesn't have "Mac" in its name and sells in lowest volume got the axe. By axing the XServe they should be able to put some of the engineering resources that were going into XServe into the Mac Pro team. If anything there is a net decrease in costs. Having "another" design team to do a separate product increases costs. Also if Apple puts stuff like lockable, but removable front mounted drives or LOM on the Mac Pro that will distribution those design costs over a product with a more customers. Again making cost recovery more reachable.
Hiding improvements in low volume products is not a good strategy of reducing costs.
.