[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20120712171345.GB19748@openwall.com>
Date: Thu, 12 Jul 2012 21:13:45 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: RAdmin, SIP speedup
On Thu, Jul 12, 2012 at 10:14:02PM +0530, Dhiru Kholia wrote:
> On Sun, Jul 8, 2012 at 11:00 AM, Solar Designer <solar@...nwall.com> wrote:
> > BTW, note that this is not a true 8-core machine (it's Bulldozer with
> > its 4 modules and 2 cores per module; also it reduces clock rate from
> > 4.0 GHz to 3.1 GHz or maybe slightly higher when under full load), so
> > the perfect speedup would be about 6x. Thus, 3x is "only" twice worse
> > than perfect.
>
> 4 modules * 2 cores / module = 8 cores. I don't understand why the
> perfect speedup is only 6x.
3.1 * 8 / 4.0 = 6.2
Well, sometimes I get 3.2 GHz at 8 threads (with non-optimal code that
does not bump into 125 W TDP at 3.1 GHz), then:
3.2 * 8 / 4.0 = 6.4
This is with FX-8120's stock settings, which are: 3.1 GHz base, 4.0 GHz
max turbo, APM enabled.
One thing that I find puzzling is that overclocking to 3.9 GHz base /
4.5 GHz turbo did provide some speedup at 8 threads as well, even though
the TDP limit remained the same.
> (In practice with OpenMP and JtR, I am seeing a speedup of ~ 4.7x on bull).
There are cases of over 6x speedup as well (try the SSH format), but not
much more.
Alexander
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
