Wilco (Wilco.Dijkstra@ntlworld.com) on 1/12/11 wrote:
---------------------------
>Paul A. Clayton (paaronclayton@gmail.com) on 1/12/11 wrote:
>---------------------------
[snip my previous comments]
>The Cortex-A9 above is not a netbook CPU, however it achieves almost the same performance
>as a netbook CPU at a much lower frequency. That hardly supports what Linus said:
>"And people don't understand how big the performance gap really is. ARM is absolutely
>not even close. ARM right now is struggling to compete with Atom, for chissake!".

I believe Linus is comparing the highest performance
ARM currently available against even a typical laptop x86--
not with the lowest performance non-embedded x86 available.

>Being 8% slower is not only a tiny gap, it is Atom which is struggling to be competitive
>as it needs a 60% faster clock.

Once again, frequency (or even power use) is not an
issue when comparing performance alone.

>Anyway, the netbook/laptop version of the A9 runs at 2GHz. Ie. Atom would need
>to be clocked at 3GHz to give the same performance or be radically redesigned to be more efficient.

Again, Atom is generally recognized as a relatively
mediocre x86--even for a low power design.

>You're right today there is still a big gap with the T7600, but that gap will be narrowed by Cortex-A15.

While the gap should be narrowed--it is much
easier to increase performance at a lower starting
performance (doubling issue width is much easier and has
much greater benefit going from 1 wide than from 4 wide;
doubling frequency is easier and has much greater benefit
starting from a 5-stage pipeline than from a 20-stage
pipeline)--and there are diminishing returns for the end
user as well, such does not mean that there will be a
single-core ARM breaking 10000 CoreMark (which is not
exactly a real-world benchmark) in the next few years.

>Tegra2 uses A9. The GCC score is a bit worse than the published result of 2881.
>Also, if you use a multi-threaded Atom then it is also fair to use a dual core A9.

I did not see that result--the Tegra2 result I did find
used 2 threads on 2 cores (5866.39-->2933.20). The halved
2 core result implies a speed gain of less than 11% (not
the 25% projected by some for the A9 improvement over A8
[though the original Tegra may have used an enhanced A8]).
Your figure gives less than a 9% improvement.

>Absolutely true. However the markets ARM is going after (mobile, netbook, laptops,
>low power servers) are not at all concerned with maximum performance.

There was a recent brief paper by someone at Google
indicating that single-thread performance below about half
of a middle-bin ordinary processor brought significant
additional costs. (I do not agree with that paper, but it
points to some practical and mind-share issues.)

ARM is currently only targeting such low-power systems, but
some seem to imply that ARM will take the world by storm
in a very short period of time. I think Linus is reacting
to this kind of claim.

>Indeed, you don't need to be extremely low power in a laptop, so that's why ARMs
>can be clocked at far higher speeds without worrying about power consumption.

But even a 2 GHz A9 is not comparable to a mid-range
laptop x86. I.e., ARM still has a consider performance
gap.

>In servers, performance per Watt, performance per m^3, or performance per dollar
>is often far more important than maximum single threaded performance.

But there is also some desire for a single binary to run
across a range of systems. Note also that Sun/Oracle's
T-series is not taking over the datacenters of the world.
Some server workloads demand significant single-thread
performance, even if only for shortish bursts.




Paul A. Clayton
just a technophile