Beastian (noemail.delete@this.aol.com) on April 19, 2019 11:10 am wrote:
> Doug S (foo.delete@this.bar.bar) on April 19, 2019 10:40 am wrote:
> > Beastian (noemail.delete@this.aol.com) on April 19, 2019 9:05 am wrote:
> > > Does anyone think that the upcoming expiry of patents on the core x86-64 instruction set
> > > which includes SSE2, maybe sooner than 2023 (as the Athlon 64 was introduced in 2003) would
> > > open the flood gates for third party implementations of compatible processors?
> > >
> > > The basic architecture including importantly the memory model could freely be implemented in hardware
> > > once that core instruction set falls out of patent; I'm unclear about the amount of traction that later
> > > Intel instructions have gotten in common software, but emulation could at least ensure compatibility.
> > > Competition in x86 implementation has languished for long
> > > stretches of time, though it has seen a resurgence
> > > as of late; compare that to the ARM ecosystem which has seen vast improvements due to a larger number
> > > of implementers: Apple, Qualcomm, nVidia, Samsung, ARM, Applied Micro and Cavium.
> >
> >
> > Who would be interested in making an x86-64 compatible CPU?
> > I know, I know, someone is going to suggest "Apple"
> > so they can have a dual x64/ARM64 SoC for the Mac to ease the transition, but ignoring them, who?
> >
> > Some may point to China, but AMD already sells some pretty cheap CPUs so unless they can do a good
> > enough design to match higher end AMD/Intel CPUs what's the point? And if they were going to do
> > it, they wouldn't wait until the patents expire - they would have started the design years ago so
> > it could be ready when the patents expire. Surely someone would have caught wind of it by now.
>
> That's a good lead, and you were right:
>
> https://fuse.wikichip.org/news/733/zhaoxin-launches-their-highest-performance-chinese-x86-chips/

The answer would be VIA.

I think they still have x86 IP from Cyrix? They developed Zhaoxin for China.