[r-t] Hyperthreading [was Exhausted search spaces]

Simon Humphrey sh at keystrata.co.uk
Mon Feb 8 10:26:27 UTC 2010


Mark (and Graham) wrote: 

<hyperthreading stuff>

It turns out that on my machine, with hyperthreading enabled,
(a) a single instance of SMC32 produces 5.65m comps in 6 mins, a 2.7%
improvement compared with HT disabled
(b) two instances of SMC32 running the same PB6 search produce a total
of 6.12m comps produced in 6 mins, an increase of 11.3%

So as you say, disabling hyperthreading was counterproductive, if only
slightly.

Since I'm now putting a new machine on the to-do list, I guess
processor speed is the main determinant of SMC performance.  I don't
run any other CPU intensive applications these days.
Cheapest GHz on the market seems to be an AMD Phenom X4 3.4 at the
moment. Any thoughts?

SH





More information about the ringing-theory mailing list