From: Bengt Larsson on
nmm1(a)cam.ac.uk wrote:

>In article <f0bvi5lpsqpp9s7otcivtga3sqn43tknbu(a)4ax.com>,
>Experience with most forms of threading, especially SMT, is that
>whether it helps or not depends on memory accesses and not actual
>calculation.

It may have escaped you, but must people don't do supercomputing.
Especially not on a bleeding Atom.
From: nmm1 on
In article <4mgvi5hjdppsun206ci2aa79n5ac66adb6(a)4ax.com>,
Bengt Larsson <bengtl8.net(a)telia.NOSPAMcom> wrote:
>
>>Experience with most forms of threading, especially SMT, is that
>>whether it helps or not depends on memory accesses and not actual
>>calculation.
>
>It may have escaped you, but must people don't do supercomputing.
>Especially not on a bleeding Atom.

If you seriously think that most non-supercomputing applications
use large numbers of floating-point calculations and no memory
accesses, then I am impressed.


Regards,
Nick Maclaren.
From: Bengt Larsson on
nmm1(a)cam.ac.uk wrote:

>If you seriously think that most non-supercomputing applications
>use large numbers of floating-point calculations and no memory
>accesses, then I am impressed.

I think no such thing.