Michael S (already5chosen.delete@this.yahoo.com) on June 3, 2022 5:16 am wrote:
> Eric Fink (erik.delete@this.anon.com) on June 2, 2022 10:00 pm wrote:
> > anon2 (anon.delete@this.anon.com) on June 2, 2022 4:13 pm wrote:
> > > Then do you think it is ARMv8 instruction set with the magical properties?
> > > In that case why not compare with another ARMv8 implementation?
> >
> > Of course not. I think that Apple has a "better" CPU design and can throw more money at the problem
> > — their chips are not the actual product and don't have to be competitive on the open market.
> > Combine that with access to a better node and you'd get yourself a fairly commanding lead.
> >
> >
>
> I was under impression that Apple's most senior CPU designers left few
> years ago and after short NUVIA interlude are now working for QC.
> Should we expect Snapdragon 7c+ or 8cx or whatever its name, to be
> competitive against Apple/Intel/Amd in absolute ST performance?


That's the assumption.

Though I wonder how long they will stay at Qualcomm. Buyouts like that of Nuvia might require the founders to work for them for 2 to 3 years, but Qualcomm wants them to design CPUs for phones and PCs, not servers. That's exactly the situation that caused them to leave Apple and form their own server focused startup. If they weren't happy at Apple not being able to design server focused CPUs, why would they be happy at Qualcomm doing the same? Especially now that they are a lot richer than they were before and will have the ability to do pretty much whatever they want.