Re:Intel delays Cannonlake 10nm processor for the third time
by djsumdog on Wednesday September 20, 2017 @10:48AM
(#15294326)
I remember the Rosetta era when they switched from PPC to Intel and had their VM layer inbetween. You could go to your process list and see what was using x86 vs PPC via Rosetta and later even x86 64-bit. Eventually Rosetta was removed from the default install and I think it's gone entirely now, so you can't run really old PPC apps without an older version of MacOS and an older Mac.
[ Parent
| on HN ]
Re:Intel delays Cannonlake 10nm processor for the third time
by jsjohnst on Wednesday September 20, 2017 @11:04AM
(#15294498)
I've heard of people using Rosetta up till 10.9 (but they had to rig the install). Not sure if anyone has tried it past 10.9.
Re:Intel delays Cannonlake 10nm processor for the third time
by monocasa on Wednesday September 20, 2017 @11:45AM
(#15295038)
A Rosetta equivalent that runs x86 on ARM is way harder from a pure technical perspective than running PPC on x86. Going from a weak memory model to a strong one is waaaaay easier than the reverse.
Re:Intel delays Cannonlake 10nm processor for the third time
by jsjohnst on Wednesday September 20, 2017 @12:47PM
(#15295795)
Unicorn / QEMU have done it already, no?
Re:Intel delays Cannonlake 10nm processor for the third time
by monocasa on Wednesday September 20, 2017 @12:53PM
(#15295871)
Nope, they ignore memory model.
Re:Intel delays Cannonlake 10nm processor for the third time
by striking on Wednesday September 20, 2017 @11:01AM
(#15294474)
Yeah, if Apple tried to emulate Intel's ISA they'd very likely get sued. https://www.forbes.com/sites/davealtavilla/2017/06/09/intel-...
Re:Intel delays Cannonlake 10nm processor for the third time
by jsjohnst on Wednesday September 20, 2017 @11:05AM
(#15294516)
Except isn't X86_64 ISA technically property of AMD, not Intel?
I know some of the later stuff (SIMD / AVX / SSE / etc) is Intel, but technically I think they could leave that off and still be successful assuming that would be all that's needed to not be infringing (which I don't know if that's the case or not).
Re:Intel delays Cannonlake 10nm processor for the third time
by monocasa on Wednesday September 20, 2017 @11:42AM
(#15295009)
It's x86_64 is so obviously inspired by 32/16-bit x86 that practically speaking, it's patented heavily by both.
Re:Intel delays Cannonlake 10nm processor for the third time
by Vogtinator on Wednesday September 20, 2017 @11:07AM
(#15294548)
Software emulation is fine (otherwise QEMU would be pretty dead), it's just hardware accelerated emulation and actual HW implementations of the ISA that Intel does not like.
Re:Intel delays Cannonlake 10nm processor for the third time
by monocasa on Wednesday September 20, 2017 @11:44AM
(#15295026)
There's no real case law on that, unfortunately.
And Intel has been making claims that they could start to litigate. https://www.theregister.co.uk/2017/06/09/intel_sends_arm_a_s...
Re:Intel delays Cannonlake 10nm processor for the third time
by lightedman on Wednesday September 20, 2017 @01:45PM
(#15296388)
I know the x86 instruction set up to at least the 486 has had its patents expire. Intel wouldn't win on every claim.