True that they never mention which Intel processor,
but the development boxes are P-4, and I'm not aware
of any Itanium laptops. If they are going to have fat
binaries for PPC/x86, then the binary for Itanium
would be fatter: PPC/x86/PA-ish. Otherwise
performance would suffer.
Andy
--- Ken Elliott <kelliott4@tampabay.rr.com> wrote:
> >>I may be mistaken, but my understanding was that
> the Itanium was not
> really x86. It is RISC that has a hardware
> instruction emulator for x86,
> which was proven at one time to be much slower than
> a similar speed Pentium.
>
> You are correct. The Itanium is based on HP's
> PA-RISC long instruction word
> architecture. X86 was emulated in microcode. Had
> it been emulated in
> hardware, it might have been accepted better. But
> since the Pentium runs
> x86 code faster, and was cheaper, most folks took a
> wait-and-see attitude.
> Intel put less effort into making the Pentium fast,
> expecting it would have
> a short life as we all embraced the Itanium, and
> that allowed AMD to
> leapfrog them.
>
> But Apple's OS/X is CPU independent, so I believe
> they would find it easier
> to move to Itanium than Microsoft. In theory, NT
> (W2K, XP) is also
> CPU-independent. But someone more knowledgeable
> than me suggested that they
> have added x86 code to XP in order to better run
> older apps.
>
> Apple's Steve Jobs says the move is about processor
> power per watt, and that
> makes sense from a laptop standpoint. But I suspect
> there is far more to it
> than that.
>
> Secure64 is on record saying the Itanium can be made
> to run about 20x faster
> than x86, and far more securely. And Jobs never
> once said "x86", "Pentium"
> or anything else to indicate what Intel chip. Since
> the Itanium can run
> big-endiam and little-endiam code, it should emulate
> the PowerPC far better
> than the Pentium could. HP moved from Motorola CPUs
> to PA-RISC, then to
> Itanium - all big-endiam CPUs. Apple moved from the
> same Motorola CPUs to
> PowerPC - also big-endiam. But Intel uses
> little-endiam CPUs, so the
> Itanium had to be able to run both.
>
> A page on little-endiam vs big-endiam:
>
> http://www.noveltheory.com/techpapers/endian.asp
>
> Ken Elliott
>
-----------------------------------------------------------------------
This list is provided as an unmoderated internet service by Networked
Knowledge Systems (NKS). Views and opinions expressed in messages
posted are those of the author and do not necessarily reflect the
official policy or position of NKS or any of its employees.
This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 20:07:52 EDT