juanrga (nospam.delete@this.juanrga.com) on August 8, 2014 11:00 am wrote:
> Michael S (already5chosen.delete@this.yahoo.com) on August 8, 2014 4:58 am wrote:
> > juanrga (nospam.delete@this.juanrga.com) on August 5, 2014 10:27 am wrote:
> > >
> > > I was considering a 100W SoC, when I mentioned "Xeon-class". Several companies have announced
> > > 90W ARM SoCs with throughput superior to 140W Xeons.
> >
> > On paper. Somehow, no one of those "several companies" published industry standard benchmark
> > results supporting their claims. SPECpower_ssj2008 would be the simplest and cheapest to setup,
> > but I somehow don't expect that any of "several companies" will submit the score in the next
> > year or 3. Claiming superior energy efficiency on paper is so much more convenient!
> >
>
> But some of those companies were showing benchmarks behind the doors. And Nvidia has already admitted
> in public that those ARM64 CPUs can feed its faster GPGPUs the same than with traditional Xeons.
>
> If you really believe that those companies will be hiding numbers during next three years expecting people
> to pay them hundred of thousand of dollars in the base of paper claims then I have very little to add.

All of that is totally unverifiable. There is significant doubt that it is reality and not just extremely narrow case for marketing.

With Xeons we know, there are tests. With Opterons we know because of tests. With ARM we know only some samples from Caldexa (no longer exists) and from NAS. Not really good showing as the only really advantage of ARM is cheap, but performance is missing.

Evidence is still mostly missing.