HOW TO TURN OFF HYPERTHREADING

On Mon, 23 Jun 2008 19:17:53 -0700 (PDT), [email protected] wrote:

>Hi, as you all know I'm having trouble with clicks and pops. I'm
>running a delta 10/10 on an Intel 2.6. I was given a list of things to
>do from Steve, the first of which was to turn off hyperthreading. I'm
>looking through device manager now but don't see anything specific
>about hyperthreading. Does anyone know how to turn off hyperthreading?

(To rip off somebody genuinely funny, in the Amazon reviews thread):
I once turned on hyperthreading, and then I couldn't find my cat.

But semi-seriously, is this done in BIOS or in Winders?

Much thanks,
Chris

It's a setting in your bios, so you won't find it in Windows.

<[email protected]> wrote in message
news:a372b9bd-bc16-41fa-99f6-bd267a264c2e@f1g2000prb.googlegroups.com...
> Hi, as you all know I'm having trouble with clicks and pops. I'm
> running a delta 10/10 on an Intel 2.6. I was given a list of things to
> do from Steve, the first of which was to turn off hyperthreading. I'm
> looking through device manager now but don't see anything specific
> about hyperthreading. Does anyone know how to turn off hyperthreading?

On Jun 24, 12:46 pm, "Steve" <[email protected]> wrote:
> It's a setting in your bios, so you won't find it in Windows.
>
> <[email protected]> wrote in message
>
> news:a372b9bd-bc16-41fa-99f6-bd267a264c2e@f1g2000prb.googlegroups.com...
>
>
>
> > Hi, as you all know I'm having trouble with clicks and pops. I'm
> > running a delta 10/10 on an Intel 2.6. I was given a list of things to
> > do from Steve, the first of which was to turn off hyperthreading. I'm
> > looking through device manager now but don't see anything specific
> > about hyperthreading. Does anyone know how to turn off hyperthreading?- Hide quoted text -
>
> - Show quoted text -

Just looking through your list Steve. How do I turn off "enable write
combining?"

In Windows:

Start --> Settings --> Control Panel --> Display --> Settings Tab -->
Advanced Button --> Troubleshoot Tab:

This will allow you to turn down accelleration and disble write combining.

Make sure you change one setting at a time and test it (I know it's
painful!!!). Your best bet is to record or play back "silence" to make sure
you aren't getting the same dropouts (pops, clicks) as before.

Steve

"jacksmith" <[email protected]> wrote in message
news:2b8944d3-2468-4b5b-ae5f-047ff4c23b4a@a32g2000prf.googlegroups.com...
On Jun 24, 12:46 pm, "Steve" <[email protected]> wrote:
> It's a setting in your bios, so you won't find it in Windows.
>
> <[email protected]> wrote in message
>
> news:a372b9bd-bc16-41fa-99f6-bd267a264c2e@f1g2000prb.googlegroups.com...
>
>
>
> > Hi, as you all know I'm having trouble with clicks and pops. I'm
> > running a delta 10/10 on an Intel 2.6. I was given a list of things to
> > do from Steve, the first of which was to turn off hyperthreading. I'm
> > looking through device manager now but don't see anything specific
> > about hyperthreading. Does anyone know how to turn off hyperthreading?-
> > Hide quoted text -
>
> - Show quoted text -

Just looking through your list Steve. How do I turn off "enable write
combining?"

On Jun 24, 2:39 pm, "Steve" <[email protected]> wrote:
> In Windows:
>
> Start --> Settings --> Control Panel --> Display --> Settings Tab -->
> Advanced Button --> Troubleshoot Tab:
>
> This will allow you to turn down accelleration and disble write combining.
>
> Make sure you change one setting at a time and test it (I know it's
> painful!!!).  Your best bet is to record or play back "silence" to make sure
> you aren't getting the same dropouts (pops, clicks) as before.
>
> Steve
>
> "jacksmith" <[email protected]> wrote in message
>
> news:2b8944d3-2468-4b5b-ae5f-047ff4c23b4a@a32g2000prf.googlegroups.com...
> On Jun 24, 12:46 pm, "Steve" <[email protected]> wrote:
>
>
>
>
>
> > It's a setting in your bios, so you won't find it in Windows.
>
> > <[email protected]> wrote in message
>
> >news:a372b9bd-bc16-41fa-99f6-bd267a264c2e@f1g2000prb.googlegroups.com...
>
> > > Hi, as you all know I'm having trouble with clicks and pops. I'm
> > > running a delta 10/10 on an Intel 2.6. I was given a list of things to
> > > do from Steve, the first of which was to turn off hyperthreading. I'm
> > > looking through device manager now but don't see anything specific
> > > about hyperthreading. Does anyone know how to turn off hyperthreading?-
> > > Hide quoted text -
>
> > - Show quoted text -
>
> Just looking through your list Steve. How do I turn off "enable write
> combining?"- Hide quoted text -
>
> - Show quoted text -

Thanks Steve, I've made all the mods that you listed, I'll also keep
the dma buffer size in mind. Fingers crossed, I'll keep you posted.

> Make sure you change one setting at a time and test it (I know it's
> painful!!!). Your best bet is to record or play back "silence" to make sure
> you aren't getting the same dropouts (pops, clicks) as before.

I should think that an steady sinewave would be best. I would actually
guess that the dropouts would be the least noticeable recording 'silence'.
The clicks are due to immediate discontinuity in the waveform. If all
you have in the file is a string of zeros, missing a few of them from
time to time will not be audible. I would try 2000Hz or so at full
amplitude.


** Posted from http://www.teranews.com **

"Tobiah" <[email protected]> wrote in message
news:[email protected]
>> Make sure you change one setting at a time and test it
>> (I know it's painful!!!). Your best bet is to record or
>> play back "silence" to make sure you aren't getting the
>> same dropouts (pops, clicks) as before.
>
> I should think that an steady sinewave would be best. I
> would actually guess that the dropouts would be the least
> noticeable recording 'silence'. The clicks are due to
> immediate discontinuity in the waveform. If all you have
> in the file is a string of zeros, missing a few of them
> from time to time will not be audible. I would try
> 2000Hz or so at full amplitude.

I've found that your general approach is very good, but that a test signal
of say 100-200 Hz is a more sensitive test. Dropouts have a spectral
signature that is vastly different from the test signal. IME a 2 KHz test
signal masks dropouts to a greater degree.

On Jun 25, 11:03 am, "Arny Krueger" <[email protected]> wrote:
> "Tobiah" <[email protected]> wrote in message
>
> news:[email protected]
>
> >> Make sure you change one setting at a time and test it
> >> (I know it's painful!!!).  Your best bet is to record or
> >> play back "silence" to make sure you aren't getting the
> >> same dropouts (pops, clicks) as before.
>
> > I should think that an steady sinewave would be best.  I
> > would actually guess that the dropouts would be the least
> > noticeable recording 'silence'. The clicks are due to
> > immediate discontinuity in the waveform.  If all you have
> > in the file is a string of zeros, missing a few of them
> > from time to time will not be audible.  I would try
> > 2000Hz or so at full amplitude.
>
> I've found that your general approach is very good, but that a test signal
> of say 100-200 Hz is a more sensitive test. Dropouts have a spectral
> signature that is vastly different from the test signal. IME a 2 KHz test
> signal masks dropouts to a greater degree.

I just recorded a three and a half minute bass line after having made
the mods and there are no audible dropouts. Shouldn't this suffice?

"jacksmith" <[email protected]> wrote in message
news:[email protected]
> On Jun 25, 11:03 am, "Arny Krueger" <[email protected]>
> wrote:
>> "Tobiah" <[email protected]> wrote in message
>>
>> news:[email protected]
>>
>>>> Make sure you change one setting at a time and test it
>>>> (I know it's painful!!!). Your best bet is to record or
>>>> play back "silence" to make sure you aren't getting the
>>>> same dropouts (pops, clicks) as before.
>>
>>> I should think that an steady sinewave would be best. I
>>> would actually guess that the dropouts would be the
>>> least noticeable recording 'silence'. The clicks are
>>> due to immediate discontinuity in the waveform. If all
>>> you have in the file is a string of zeros, missing a
>>> few of them from time to time will not be audible. I
>>> would try 2000Hz or so at full amplitude.

>> I've found that your general approach is very good, but
>> that a test signal of say 100-200 Hz is a more sensitive
>> test. Dropouts have a spectral signature that is vastly
>> different from the test signal. IME a 2 KHz test signal
>> masks dropouts to a greater degree.

> I just recorded a three and a half minute bass line after
> having made the mods and there are no audible dropouts.
> Shouldn't this suffice?

Seems like that recording could fit the spectral profile that I suggested,
above.

I can tell you from sad experience that if you make a long recording (like
an hour or more) drop-outs can start appearing towards the end of the
recording.