[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: ARAnyM speed



Followup-To: poster

On Sat, 11 May 2013, Thorsten Glaser wrote:

> For the record: the new CPU is
> Intel(R) Core(TM) i7 CPU         950  @ 3.07GHz
> and the old one was an AMD X4 II or something like that.

Oh, interesting. This seems to really be something on the
setup of the machine – I installed BOINC on three boxen,
and the benchmark results look like this:

• the slow box:
  GenuineIntel  Intel(R) Core(TM) i7 CPU 950 @ 3.07GHz [Family 6 Model 26 Stepping 5]
  Measured floating point speed  613.16 million ops/sec
  Measured integer speed  3304.48 million ops/sec

• another i7 with same specs:
  GenuineIntel  Intel(R) Core(TM) i7 CPU 950 @ 3.07GHz [Family 6 Model 26 Stepping 5]
  Measured floating point speed  3090.63 million ops/sec
  Measured integer speed  12634.06 million ops/sec

• an AMD, for comparison:
  AuthenticAMD  AMD Phenom(tm) II X4 965 Processor [Family 16 Model 4 Stepping 3]
  Measured floating point speed  3175.31 million ops/sec
  Measured integer speed  15516.88 million ops/sec

So I wasn't exactlt wrong when saying that AMD is still
faster (except this one has 3.4 GHz and 6830.50 BogoMIPS)
but I would expect the two i7 to have similar performance,
not about ONE FIFTH (in float) and ONE QUARTER (in int).

The /proc/cpuinfo of the two i7 systems only have a grand
total of eight *lines* differ, one per CPU core, this one:
-bogomips       : 6128.75
+bogomips       : 6206.92
(the - is my slow desktop, the + is the fast box).

So this is probably some kind of hardware/BIOS issue.
I have no idea where to look but now I at least know,
somewhat, how to bench (pybench would probably also
work). Ideas welcome, but off-list, since this is no
ARAnyM or Debian issue.

bye,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-314
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Boris Esser, Sebastian Mancke


Reply to: