Prev: A link to news on help of tapping power of GPU
Next: Prime Number Fluid model data and graphs
From: Maaartin on 23 May 2010 12:23
On May 21, 8:07 pm, Clark Smith <noaddr...(a)nowhere.net> wrote:
>         Thanks for your feedback. This makes me think that the number
> reported in the article I mentioned was obtained with -multi 4 (or -multi
> 3, or however many cores are available in that CPU.) Still, I can't help
> but feeling that something is wrong with your platform: With those specs
> I would have expected way more than 574 1024-bit signatures per second
> per core.

I see that my results are too low, here are results from slightly
better computer running Linux with DDR3 (which is IMHO irrelevant
here), AMD Phenom(tm) II X4 945, 3013 MHz, single core:

sign verify sign/s verify/s
rsa 512 bits 0.000087s 0.000008s 11526.5 121034.9
rsa 1024 bits 0.000415s 0.000023s 2407.1 43036.6
rsa 2048 bits 0.002624s 0.000075s 381.1 13339.7
rsa 4096 bits 0.017500s 0.000271s 57.1 3695.7

It's more than 3 times faster then mine. However, I'd say, the
performance reported by the OP was definitively for all three cores,
unless he uses liquid nitrogen or helium cooling.
From: Clark Smith on 23 May 2010 12:31
On Sun, 23 May 2010 09:23:21 -0700, Maaartin wrote:

> On May 21, 8:07 pm, Clark Smith <noaddr...(a)nowhere.net> wrote:
>>         Thanks for your feedback. This makes me think that the
>>         number
>> reported in the article I mentioned was obtained with -multi 4 (or
>> -multi 3, or however many cores are available in that CPU.) Still, I
>> can't help but feeling that something is wrong with your platform: With
>> those specs I would have expected way more than 574 1024-bit signatures
>> per second per core.
>
> I see that my results are too low, here are results from slightly better
> computer running Linux with DDR3 (which is IMHO irrelevant here), AMD
> Phenom(tm) II X4 945, 3013 MHz, single core:
>
> sign verify sign/s verify/s
> rsa 512 bits 0.000087s 0.000008s 11526.5 121034.9 rsa 1024 bits
> 0.000415s 0.000023s 2407.1 43036.6 rsa 2048 bits 0.002624s 0.000075s
> 381.1 13339.7 rsa 4096 bits 0.017500s 0.000271s 57.1 3695.7
>
> It's more than 3 times faster then mine. However, I'd say, the
> performance reported by the OP was definitively for all three cores,
> unless he uses liquid nitrogen or helium cooling.

I am the OP and I agree with you in that the performance reported
in the link I provided must be using the three cores in parallel. I just
wish the guys who ran the benchmark specified such details. As for your
new figures, they definitely make more sense.

From: Mok-Kong Shen on 24 May 2010 05:47
Clark Smith wrote:

> I am the OP and I agree with you in that the performance reported
> in the link I provided must be using the three cores in parallel. I just
> wish the guys who ran the benchmark specified such details. As for your
> new figures, they definitely make more sense.

There is one additional reasoning supporting that. They gave benchmark
figures for processors having different numbers of cores, if I don't
err. So, if the comparison is on the basis of one single core, it could
be 'reasonably' expected that they would have mentioned that fact.

M. K. Shen

First  |  Prev  | 
Pages: 1 2 3
Prev: A link to news on help of tapping power of GPU
Next: Prime Number Fluid model data and graphs