Getting a new CPU instruction set going is hard work - the original AMD64 instruction set took 3-4 years to get mainstream Linux support, and without any hardware or software emulators out there to aide development...
Re:Chips of convergence - Apple and the ARM switch
by wriq on Saturday May 07, 2011 @10:18AM
(#2523783)
You think it would require much more work than just a recompile? With the similarities between iOS/OSX and Apple's investment in the LLVM toolchain I see the switch going smoother than PPC -> x86.
Re:Chips of convergence - Apple and the ARM switch
by ianferrel on Saturday May 07, 2011 @11:30AM
(#2523885)
I wonder how important the ability to virtualize Windows is to Apple converting Windows users?
My impression is that it's quite important. Even if people don't end up using Windows for much, they know that they can, and that significantly lowers the risks and cost of transition.
If future versions of Windows run on ARM and Microsoft does a good job of handling compatibility with older software, this might not be a major issue. But I'd bet that it's still a major consideration.
Re:Chips of convergence - Apple and the ARM switch
by nupark2 on Saturday May 07, 2011 @11:34AM
(#2523890)
I think it matters more for mid-range business market, less so for the home consumer that may find an iPad already meets most of their needs.
Personally, Windows virtualization is an unfortunate requirement I have due to Windows-only hardware development tools.
I'm still pissed at Atmel for basing their modern AVR Studio 5 on Visual Studio, rather than on the Eclipse-based work they did for the previous version of AVR32 Studio.
Re:Chips of convergence - Apple and the ARM switch
by nupark2 on Saturday May 07, 2011 @11:33AM
(#2523888)
Apple will require CPU emulation to support legacy software releases (think Creative Suite, MS Office, ...); whether this can be done with reasonable performance on ARM is the first real question.
After that, why not make a switch? We (developers) can support ARM just as easily as x86-64, x86-32, and PPC. There are some specific areas that will require rework (games!), but investments are already being made there to support iOS.
Getting a new CPU instruction set going is hard work - the original AMD64 instruction set took 3-4 years to get mainstream Linux support, and without any hardware or software emulators out there to aide development...
That said, the entire computer industry is going ARM (Windows 8, all mobile platforms, etc.) so I'm guessing that the transition will be easier.
There's much less tight x86 assembly for critical performance areas than was commonplace 10 years ago.
My impression is that it's quite important. Even if people don't end up using Windows for much, they know that they can, and that significantly lowers the risks and cost of transition.
If future versions of Windows run on ARM and Microsoft does a good job of handling compatibility with older software, this might not be a major issue. But I'd bet that it's still a major consideration.
Personally, Windows virtualization is an unfortunate requirement I have due to Windows-only hardware development tools.
I'm still pissed at Atmel for basing their modern AVR Studio 5 on Visual Studio, rather than on the Eclipse-based work they did for the previous version of AVR32 Studio.
After that, why not make a switch? We (developers) can support ARM just as easily as x86-64, x86-32, and PPC. There are some specific areas that will require rework (games!), but investments are already being made there to support iOS.