From: glen herrmannsfeldt on
Michael Plante <michael.plante(a)gmail.com> wrote:
(snip, someone wrote)

>>> But I don't know if the Intel chips do this.

(then I wrote)
>>If they do, it doesn't seem to be used. Windows has an idle
>>process, and I believe that IA32 unix/linux do also.

>>I don't remember if the 8080 does.

> No idea about the 8080, but the 8086+ has the HLT instruction. As far as
> more recent manuals go, the P4 manual states that any enabled interrupts
> will wake it.

I am sure that VMWare knows. I thought I remembered reading how
VMWare detects that the virtual machine is in an idle loop and
reduces the waste of CPU cyles. I don't have a reference, though.

> Also, re the idle process, that may just be a placeholder to
> make the times add up; I wouldn't assume anything there. Windows does some
> weird stuff with accounting (e.g., the memory in task manager doesn't
> necessarily add up the way one would expect).

I remember running NTBACKUP on a dual processor machine and at the
end having it indicate the elapsed time for the backup ... twice as
long as it actually took.

-- glen
From: Rick Lyons on
On Wed, 24 Feb 2010 12:30:56 -0700, Eric Jacobsen
<eric.jacobsen(a)ieee.org> wrote:

(Snipped by Lyons)
>
>I doubt they'll be able to seriously approach anything remotely close to
>the technology that Intel, et al, use with a $27M program spread across
>multiple institutions. The equipment needed to do anything substantial
>at those geometries is way beyond the entire cost of the program.
>
>They might publish something, but I'm skeptical that it'll be relevant.

Hi Eric,
The whole idea is a shame. A shame of government
waste at tax payer expense.

[-Rick-]