I have a 2007 Mac Pro which I used for most of the experiments and a 2009 Mac Pro. Both machines represent the fundamental differences in the Apple logic board and EFI design. None of these systems will work with the ARC 1210 card to boot Windows and sleep.
It is unreasonable to use anything but the GUID partition scheme with a Mac Pro. It is Apple's approach to dual booting and deviating from that way would mean to violate the system manufacturers specification. So this will always be my first choice of partition scheme.
In a Mac Pro it is also possible to use a separate drive for Windows with MBR partitioning. This would even allow you to boot from one of the SATA ports. Unfortunately it will not help with installation or booting from an ARC 1210 port. The Windows install disk will not identify any MBR or GUID partitioned drive that is run from an ARC 1210 port. If you simply install the Windows in another non Mac Bios machine and transfer it to the Mac Pro it will not boot from the ARC 1210 port.
The simple truth is that ARECA does no firmware that allows you to run Windows in a Mac Pro from one of the card's ports. You may be able to use the card to run Windows in a Bios machine. I admit that I have no experience with that case. But to tell you the truth I'm not interested in that case because I do not consider a Hackintosh. So the case for me is closed until Areca change their firmware policy and produce something that fits the Mac Pro for booting Windows and sleeps.
I was recalling the 2008 booting from a BIOS enabled Areca, as it may have only been an array (separate single OS disk on the ICH, as that's what I usually prefer for a workstation; I like separation of the OS and data). Servers, I use a RAID 1 for the OS location.
It's quite possible that my memory's become a bit faulty since 2008 (been awhile, and I only had the MP in hand long enough to do my testing before I returned it; had other issues that I decided were to agregious to deal with). I may have also mis-read/interpreted a couple of posts here on MR. But I'll play it safe, and figure this is the case, as Areca's products are usually consistent (not a situation that the ARC-1210 won't work, while others such as the ARC-1231ML or ARC-1680ix12 I have did function under condition we're discussing).
What I do recall, is that Apple's firmware has proprietary elements to it (compared to Intel's EFI boards; I've had access to Itanium systems that were much easier to deal with to me). I've not used their more recent DP server boards (UEFI based OTB, but also offer BIOS if needed).
For example, the '09 MP memory timings were fixed, and some have issues related to BIOS based eSATA cards (BIOS Emulation; won't work properly or at all, no matter the OS). As I understand it, the different firmware revisions matter (
B07 works with eSATA cards; earlier and later <B08 = experimental as I understand it> were intermittant at best).
Given Apple's firmware isn't 100% within EFI 1.10 (or UEFI 2.0/1), the interaction problem would be on Apple (Areca designed their cards for Itanium, which are 100% EFI 1.10 compliant).
By chance have you flashed the card with EFI, and are willing to experiment?
Just to see if that would make a difference.
I assume you're earning a living with the system, I don't blame you one bit for skipping it and getting the MP.
As per GUID, the cards use their own proprietary implementations (all cards do, which is why an array created under one maker won't work on another; the offsets and whatnot are different). BIOS doesn't have native support for GUID (Windows has it, but it's a software hack to get around BIOS's limitation), which is why you can't try and force it on the Areca when loaded with BIOS. EFI can deal with GUID, as support is included.
I'm basically "chomping at the bit" per se waiting for UEFI to replace BIOS.
