Wilco (Wilco.Dijkstra.delete@this.ntlworld.com) on August 9, 2014 3:10 pm wrote:
> Alberto (git.delete@this.git.it) on August 9, 2014 2:13 pm wrote:
> > Wilco (Wilco.Dijkstra.delete@this.ntlworld.com) on August 9, 2014 1:49 pm wrote:
> > > Michael S (already5chosen.delete@this.yahoo.com) on August 9, 2014 12:38 pm wrote:
> > > > anon (no.delete@this.thank.you) on August 9, 2014 10:17 am wrote:
> > > > > juanrga (nospam.delete@this.juanrga.com) on August 9, 2014 6:43 am wrote:
> > > > > > Aaron Spink (aaronspink.delete@this.notearthlink.net) on August 9, 2014 3:44 am wrote:
> > > > > > > juanrga (nospam.delete@this.juanrga.com) on August 8, 2014 10:49 am wrote:
> > > > > > >
> > > > > > > > Take a modern A57 core. According to AMD the A57 Opteron is faster than jaguar based Opteron but
> > > > > > > > consumes less power. The ARM core performance is ~40% faster, and consumes roughly one half.
> > > > > > > >
> > > > > > > Faster at what? Its a bunch of market point pointing to nothingness.
> > >
> > > Jaguar has about the same IPC as Cortex-A15, so is it really that hard to believe A57 is much faster?
> >
> > I don't see proof behind your claim. Looking how performs Jaguar vs.
> > Silvermont, i can say Jaguar is pretty neck to neck with A57.
>
> See this, that's a significant win on integer for A15, although Jaguar is a bit faster on FP when you normalize
> frequencies. So AMD would already make better servers if they replaced Jaguar in X2150 with A15.
>
> And since you mentiond Silvermont, this is Tegra K1 vs Z3770. Enough said.
>
> But I guess you're going to claim next that A57 is slower than Silvermont, am I right? :-)
>
> Wilco

Unfortunately i hate Geekshi***bench, so i can't see :). To be fair i cannot say what is faster.
Honestly i am a little bored to be constrained to compare different architectures (core+cache+memory subsystem) with spare data from conference calls or short papers just to say something, with strange not verified and heavily debatable benches mostly running in the L1 or in the L2 at the best.
You can't compare two cpus without taking in account the cache hierarchy closely joint with the memory subsystem, both Intel and Amd have spent thousand million dollars to refine this particular of their cpus in past years; without a fast, low latency and reliable memory access your core is nothing more than a glue of transistors theoretically capable of........of what??
Be honest!, you might judge Power8 with Dhrystone ? or Geekbench?? i think not.
I hope in numerous and exaustive submissions on Spec site from both Intel and Amd for their little cores.....i have no hopes form other companies but i am pretty confident that as Amd will ship the first Seamicro ARM server, finally some light will be done as the competition will grow and serious customers will need to know what is best.