OpenDarwin observations
I'm running OpenDarwin 8.0.1 (corresponds to Tiger) just fine on my Xeon machine here -- full hardware support and everything. From the looks of the drivers on the CD, the new Apple macs are going to be bog-standard ICH6-based machines (there's an i915 graphics driver, and it doesn't really work on non-ICH systems). They have drivers for AC97 sound, ACPI power management, etc.
Because OpenDarwin exists, I imagine they will have a very hard time locking down the OS. They simply can't keep it from booting -- it already does. The only thing I could imagine them doing is making Aqua only run in the presence of a hardware key of some kind, but the presence of this chip could presumably be faked fairly easily with an obligingly created kext.
I strongly suspect that Apple will merely not *support* OS X on non-Apple hardware, since it has a pretty good chance of not running anyway -- the kernel was built to only run on Pentium 4s (it requires SSE2, so it *might* work on some Athlons), only really boots on ICH4 and newer Intel chipsets, has support for only three kinds of graphics card, supports exactly one family of audio chipset, etc. Aside from my machine (where it works perfectly) I've been very hard-pressed to find another where it will even boot.
Another possibility is that they're using EFI instead of PC-BIOS (which I'm sure Intel would love...). Does anyone know anything about this?
Because OpenDarwin exists, I imagine they will have a very hard time locking down the OS. They simply can't keep it from booting -- it already does. The only thing I could imagine them doing is making Aqua only run in the presence of a hardware key of some kind, but the presence of this chip could presumably be faked fairly easily with an obligingly created kext.
I strongly suspect that Apple will merely not *support* OS X on non-Apple hardware, since it has a pretty good chance of not running anyway -- the kernel was built to only run on Pentium 4s (it requires SSE2, so it *might* work on some Athlons), only really boots on ICH4 and newer Intel chipsets, has support for only three kinds of graphics card, supports exactly one family of audio chipset, etc. Aside from my machine (where it works perfectly) I've been very hard-pressed to find another where it will even boot.
Another possibility is that they're using EFI instead of PC-BIOS (which I'm sure Intel would love...). Does anyone know anything about this?
Comments
Originally posted by NWhitehorn
I'm running OpenDarwin 8.0.1 (corresponds to Tiger) just fine on my Xeon machine here -- full hardware support and everything. From the looks of the drivers on the CD, the new Apple macs are going to be bog-standard ICH6-based machines (there's an i915 graphics driver, and it doesn't really work on non-ICH systems). They have drivers for AC97 sound, ACPI power management, etc.
Because OpenDarwin exists, I imagine they will have a very hard time locking down the OS. They simply can't keep it from booting -- it already does. The only thing I could imagine them doing is making Aqua only run in the presence of a hardware key of some kind, but the presence of this chip could presumably be faked fairly easily with an obligingly created kext.
I strongly suspect that Apple will merely not *support* OS X on non-Apple hardware, since it has a pretty good chance of not running anyway -- the kernel was built to only run on Pentium 4s (it requires SSE2, so it *might* work on some Athlons), only really boots on ICH4 and newer Intel chipsets, has support for only three kinds of graphics card, supports exactly one family of audio chipset, etc. Aside from my machine (where it works perfectly) I've been very hard-pressed to find another where it will even boot.
Another possibility is that they're using EFI instead of PC-BIOS (which I'm sure Intel would love...). Does anyone know anything about this?
I agree. Actually, we may very well see Apple-sanctioned, if not supported, MacOS X for non-Apple hardware. These new Apple development systems use PC motherboards with Phoenix BIOSes replacing the standard Mac boards in G5 cases. You can't very well develop software on computers that can't run the software.
It is clear that the reports about Apple's switch prior to the announcement were part of an Apple-orchestrated leak campaign. Just a few weeks prior to the announcement were reports that several Intel-based computer manufacturers were strongly lobbying Steve Jobs for MacOS X licenses. Risks aside, I believe that Jobs has decided to acquiesce to the lobbying.
Think about it. Apple has a choice--it can either devote considerable resources in a futile effort to prevent MacOS X from running on non-Apple hardware, or it can embrace the prospect. I believe that Apple has chosen to embrace the prospect. What is left is to develop a business model that doesn't put the company out of business.
Remember how someone figured out that all you had to do was image the CD, change a few bytes, and you had a full install for free?
I am willing to bet a similar mechanism will be put in place for OS X for Intel. Hell, it's what Apple uses on all its install CD/DVDs. I found that out when trying to install GarageBand for a Mac it was not intended for - but in that case it was in an "invisible" directory.
Consider it Apple's backdoor for geeks. Let them install a warez version on their home-built box - and find out it doesn't work as well as it would on a real Mac (like I did with GarageBand - and it made me buy something that could run it.)