Running E6600 at 3.6ghz now takes less voltage?

"Paul" <nospam DeleteThis @needed.com> wrote in message news:fa2i8k$45t$1@aioe.org...
> Ed Medlin wrote:
>> I decided to keep my E6600 on a Striker Extreme 680i MB at 3.6ghz at
>> 1.6v and after a week or so running perfectly stable, I decided to try
>> and lower the voltage a bit. I was able to first drop it to 1.55v and now
>> to 1.52v and it is staying stable. When I first got it up to this speed,
>> nothing under 1.6v was at all stable. It would not even boot into XP at
>> 1.55v. My idle temps have dropped from 32-34c to 29-31c too. Load temps
>> never changed a lot and are still in the 50-55c range with liquid
>> cooling. I have been overclocking for a long time and this is something I
>> have never ran into other than in a couple cases where it was due
>> possibly to heat drop from the thermal compound setting up better. The
>> Swiftech liquid cooling system is pretty efficient and my temps at any
>> voltages have not raised drastically at any voltage staying in a +/- 5-6
>> deg C range at either end of the spectrum from stock speeds to 3.6ghz. I
>> was just wondering if anyone else has experienced anything like this with
>> the C2D processors and I am also pondering as to why the processor would
>> react this way. I am not complaining....., I am just looking for some
>> theory to explain it. I am sure it is not temperature related due to the
>> 100+F daily temperatures and my AC straining to keep my house at liveable
>> temperatures......
>>
>> Ed
>
> The theory is referred to as "burn-in", but I haven't read a technically
> sound
> description of why it works. You're free to make up your own theory...
>
> The opposite effect, is electromigration, and it works against your
> overclock.
>
> http://en.wikipedia.org/wiki/Electromigration
>
> Paul

"Burn in" is what I usually call testing the stability of a new system
or new overclock by stressing all pertinent components as much as possible
and I have been doing it at every drop in voltage. Electromigration is found
at too high of voltage and can vary from processor to processor even if they
are identical in every way (it is also about the only way to kill a present
Intel processor). In the past, if a processor needed a certain voltage to
stay stable, that is what it needed and I have never seen it change unless
the cooling was changed when heat was an issue. Since my temperatures have
stayed fairly constant from stock 2.4Ghz to 3.6Ghz I am kind of baffled
about this change in the amount of voltage required to keep it stable. I
have now lowered the voltage to 1.5v, down from 1.55v and it is still
stable. Hell, I might be down to stock 1.4v before this is over........,
although I doubt it. This is not a really big issue, it is just an
interesting little quandry that I have been pondering. Silicon works in
mysterious ways sometimes..... 3.6Ghz is the max this processor will do
at any safe voltage on my rig. My major issue now is finding some good
memory timings for my Mushkin Extreme PC9200 memory to get it past 933 or
so. It is rated for 1100+Mhz but I can't get it anywhere near that. While
overclocking the CPU I have been running it at "Linked" and "Sync Mode" at
800Mhz just to eliminate it from the project at this time. 2.2v is where the
memory seems the most stable even though it is rated at 2.35v. There are not
many folks running my memory and the Striker Extreme NV680i MB so it is
tough finding anyone with some timings that work well. Timings seem to vary
a lot between different NV680i boards so that is my issue now.....

Ed
 >> Stay informed about: Running E6600 at 3.6ghz now takes less voltage? 

This message is not archived
 >> Stay informed about: Running E6600 at 3.6ghz now takes less voltage? 

On Aug 17, 9:16 pm, "Ed Medlin" <e... DeleteThis @edmedlin.com> wrote:
> "Paul" <nos... DeleteThis @needed.com> wrote in messagenews:fa2i8k$45t$1@aioe.org...
> > Ed Medlin wrote:
> >> I decided to keep my E6600 on a Striker Extreme 680i MB at 3.6ghz at
> >> 1.6v and after a week or so running perfectly stable, I decided to try
> >> and lower the voltage a bit. I was able to first drop it to 1.55v and now
> >> to 1.52v and it is staying stable. When I first got it up to this speed,
> >> nothing under 1.6v was at all stable. It would not even boot into XP at
> >> 1.55v. My idle temps have dropped from 32-34c to 29-31c too. Load temps
> >> never changed a lot and are still in the 50-55c range with liquid
> >> cooling. I have been overclocking for a long time and this is something I
> >> have never ran into other than in a couple cases where it was due
> >> possibly to heat drop from the thermal compound setting up better. The
> >> Swiftech liquid cooling system is pretty efficient and my temps at any
> >> voltages have not raised drastically at any voltage staying in a +/- 5-6
> >> deg C range at either end of the spectrum from stock speeds to 3.6ghz. I
> >> was just wondering if anyone else has experienced anything like this with
> >> the C2D processors and I am also pondering as to why the processor would
> >> react this way. I am not complaining....., I am just looking for some
> >> theory to explain it. I am sure it is not temperature related due to the
> >> 100+F daily temperatures and my AC straining to keep my house at liveable
> >> temperatures......
>
> >> Ed
>
> > The theory is referred to as "burn-in", but I haven't read a technically
> > sound
> > description of why it works. You're free to make up your own theory...
>
> > The opposite effect, is electromigration, and it works against your
> > overclock.
>
> >http://en.wikipedia.org/wiki/Electromigration
>
> > Paul
>
> "Burn in" is what I usually call testing the stability of a new system
> or new overclock by stressing all pertinent components as much as possible
> and I have been doing it at every drop in voltage. Electromigration is found
> at too high of voltage and can vary from processor to processor even if they
> are identical in every way (it is also about the only way to kill a present
> Intel processor). In the past, if a processor needed a certain voltage to
> stay stable, that is what it needed and I have never seen it change unless
> the cooling was changed when heat was an issue. Since my temperatures have
> stayed fairly constant from stock 2.4Ghz to 3.6Ghz I am kind of baffled
> about this change in the amount of voltage required to keep it stable. I
> have now lowered the voltage to 1.5v, down from 1.55v and it is still
> stable. Hell, I might be down to stock 1.4v before this is over........,
> although I doubt it. This is not a really big issue, it is just an
> interesting little quandry that I have been pondering. Silicon works in
> mysterious ways sometimes..... 3.6Ghz is the max this processor will do
> at any safe voltage on my rig. My major issue now is finding some good
> memory timings for my Mushkin Extreme PC9200 memory to get it past 933 or
> so. It is rated for 1100+Mhz but I can't get it anywhere near that. While
> overclocking the CPU I have been running it at "Linked" and "Sync Mode" at
> 800Mhz just to eliminate it from the project at this time. 2.2v is where the
> memory seems the most stable even though it is rated at 2.35v. There are not
> many folks running my memory and the Striker Extreme NV680i MB so it is
> tough finding anyone with some timings that work well. Timings seem to vary
> a lot between different NV680i boards so that is my issue now.....
>
> Ed

I am having the same issues with my Corsair PC8500 Dominator and e6750
on the and eVGA 680i SLI board. I have settle on 916MHz 4-4-4-12 2.2v
and 3.66Ghz (7x458) linked and synched. This gives me my best Everest
Memory scroes. If I unlink the ram and increase the speed I get
instability despite the fact that I have tested the ram to run at
higher speeds in different configurations. My advise is not to worry
about the ram sped to much as the straps on the 680i baord can
actually hurt performance at higher ram speeds.
 >> Stay informed about: Running E6600 at 3.6ghz now takes less voltage?