How far can your E6300/E6400 O/C on stock volts?

[GuitarDaddy]

Just curious because I'm seeing tons of reports of people hitting 3.2ghz or higher on stock volts, and mine gets nowhere close. It scales really nice with volts and tops out around 3.6ghz on stock cooling, but it won't even pass orthos or TAT at 2.8ghz unless I bump it to 1.375

I'm wondering if alot of these reports are from people leaving the vcore on "auto" and not understanding that the auto setting increases volts as you raise the FSB. I'm not interested in what CPU-Z or other windows based utilities report as none of them seem to be very accurate and include vdrop and vdroop.

 

[idiotekniQues]

i can get to 3.1 orthos stable at 1.325v i can easily run windows at 1.325 and 3.2ghz but it will fail orthos within 30 minutes.

right now i am running my 24/7 oc as i am working on my temps at 1.337v at 3.2ghz.

if i get a different heatsink i will probably try to run at 3.4 24/7. right now i have a ninja that i am not too happy with.

 

[GuitarDaddy]

Nice CPU!

By the way the photos in the gallery in your sig are fantastic? Are you the photog?

 

[idiotekniQues]

Originally posted by: GuitarDaddy
Nice CPU!

By the way the photos in the gallery in your sig are fantastic? Are you the photog?

i am pretty happy with my chip.

the second i start going over 400mhz fsb, or over 3.2ghz basically, i start needing more voltage. even running at like 402mhz fsb i need 1.35v - i ran super.pi at 3.6ghz at 1.41v but needed more juice to go orthos stable and i didnt like the temps. plus my ocz ram is promos chips, but they do me good until 450 fsb when i have to really loosen timings.

yes i am the guy behind the camera, i got into photography about last march when the 30d came out and i spent some loan money on it best debt money i ever spent though

thanks for the kind words.

if you like photography, my gf who got me into it, has her stuff at www.jenniferfrances.com it is a pleasant way to pass the time

 

[Diogenes2]

Opps.. I voted before I read the post ..

How do you know what the default voltage is? Even according to Intel specs, the voltage will vary with the CPU frequency, not the FSB ..

<a target=_blank class=ftalternatingbarlinklarge href="ftp://download.intel.com/design/processor/datashts/31327803.pdf"><a target=_blank class=ftalternatingbarlinklarge href="ftp://download.intel.com/design/processor/datashts/31327803.pdf">ftp://download.intel.com/design/processor/datashts/31327803.pdf</a></a>

2.3 " Individual processor VID values may be calibrated during manufacturing such that two
devices at the same core speed may have different default VID settings. "

2.6.2 Note 4. " 4. These voltages are targets only. A variable voltage source should exist on systems in the event that a different
voltage is required. See Section 2.3 and Table 2 for more information."

In other words : Intel may adjust the VID on any given CPU, so it will run at the speed that they will set the multiplier for..

So, a 6800 will have a higher vcore than a 6600 ..

So setting it at auto, will in effect be the default vcore for whatever frequency you are running.. Anything below 1.5 should be safe .. ( and keeping the HSF below thermal max 61c )

 

[Hulk]

Originally posted by: Diogenes2
Opps.. I voted before I read the post ..

How do you know what the default voltage is? Even according to Intel specs, the voltage will vary with the CPU frequency, not the FSB ..

<a target=_blank class=ftalternatingbarlinklarge href="ftp://download.intel.com/design/processor/datashts/31327803.pdf"><a target=_blank class=ftalternatingbarlinklarge href="ftp://download.intel.com/design/processor/datashts/31327803.pdf"><a target=_blank class=ftalternatingbarlinklarge href="ftp://download.intel.com/design/processor/datashts/31327803.pdf">ftp://download.intel.com/design/processor/datashts/31327803.pdf</a></a></a>

2.3 " Individual processor VID values may be calibrated during manufacturing such that two
devices at the same core speed may have different default VID settings. "

2.6.2 Note 4. " 4. These voltages are targets only. A variable voltage source should exist on systems in the event that a different
voltage is required. See Section 2.3 and Table 2 for more information."

In other words : Intel may adjust the VID on any given CPU, so it will run at the speed that they will set the multiplier for..

So, a 6800 will have a higher vcore than a 6600 ..

So setting it at auto, will in effect be the default vcore for whatever frequency you are running.. Anything below 1.5 should be safe .. ( and keeping the HSF below thermal max 61c )

I wonder if there is anyway to determine the VID for a particular chip?

 

[GuitarDaddy]

I agree that different model CPU's may have different default vcore values. And I will conceed that in some cases different chips of the same model can have different default values(usually EE versions) but its pretty rare IMO. All the E6300/E6400's I've seen default to 1.325 per bios.

But it seems that as you O/C using "auto" different boards overvolt at different rates and this is controlled by the mobo manufacturers, not intel. And once the board starts overvolting it is no longer "stock" or "default" voltage IMO. To me "stock" means the voltage specified by Intel for operating at "stock" speed. For example my Asus P5B-E overvolts like crazy on auto, it ramps vcore up above 1.5v with just a moderate overclock.

I guess my point in making this thread is that it's widely spread here and on other forums that if you overclock using "stock" volts that your in no danger of harming your CPU, and some will go as far as saying it's OK to RMA an overclocked CPU as long as you never went above "stock" volts. I just want everyone to understand that leaving the voltage on "auto" is NOT providing you any level of safety, and in many cases it is putting your CPU at more risk than setting your vcore manually.

 

[GuitarDaddy]

Originally posted by: Hulk


I wonder if there is anyway to determine the VID for a particular chip?

Some boards show the default value next to the manually set value as shown
Here

Don't know if there is a utility that will show it, maybe someone will post one

 

[Diogenes2]

Originally posted by: GuitarDaddy
I agree that different model CPU's may have different default vcore values. And I will conceed that in some cases different chips of the same model can have different default values(usually EE versions) but its pretty rare IMO. All the E6300/E6400's I've seen default to 1.325 per bios.

But it seems that as you O/C using "auto" different boards overvolt at different rates and this is controlled by the mobo manufacturers, not intel. And once the board starts overvolting it is no longer "stock" or "default" voltage IMO. To me "stock" means the voltage specified by Intel for operating at "stock" speed. For example my Asus P5B-E overvolts like crazy on auto, it ramps vcore up above 1.5v with just a moderate overclock.

I guess my point in making this thread is that it's widely spread here and on other forums that if you overclock using "stock" volts that your in no danger of harming your CPU, and some will go as far as saying it's OK to RMA an overclocked CPU as long as you never went above "stock" volts. I just want everyone to understand that leaving the voltage on "auto" is NOT providing you any level of safety, and in many cases it is putting your CPU at more risk than setting your vcore manually.

I agree that ' auto ' is not necessarily safe. You make a good point for making sure ' auto ' does not send you past 1.5 ... I need to change my sig to say ' Auto ' , instead of ' Default ' I guess ..



That seems like a design flaw, if it does...

 

[StopSign]

I can get to 3.2 while undervolted. I've gong as low as 1.20 but I'm running it at 1.25 right now.

 

[n7]

Originally posted by: GuitarDaddy
Just curious because I'm seeing tons of reports of people hitting 3.2ghz or higher on stock volts, and mine gets nowhere close. It scales really nice with volts and tops out around 3.6ghz on stock cooling, but it won't even pass orthos or TAT at 2.8ghz unless I bump it to 1.375

I'm wondering if alot of these reports are from people leaving the vcore on "auto" and not understanding that the auto setting increases volts as you raise the FSB. I'm not interested in what CPU-Z or other windows based utilities report as none of them seem to be very accurate and include vdrop and vdroop.

The reason you can't get anywhere w/ 1.325V is vdroop.

Your mobo should have terrible vdroop if it's anything like the P5B-E my buddy has, which is why setting vcore in bios & telling people to test for max will not provide consistent results.

Some people will end up with vcore @ 1.3V, some 1.325V, others 1.28V, all depending on vdroop, which makes for very unfair comparisons.


I don't know how far i can go with 1.325V in bios, but with in Windows vcore readings, i can do:

E6300 @ 3276 MHz 1.35V idle/1.33V load Orthos 10 hours blend
E6300 @ 3003 MHz 1.275 bios/1.25V idle/1.23V load Orthos 10 hours blend

This is with case side open & who knows what ambient temps...

 

[imported_Husky55]

I have an Asus P5B-E and I can confirm that on AUTO it overvolts like crazy (Guitar Daddy) with small OC i.e. 3000. On the other hand, with bios at 1.325 V it undervolts, Speedfan, Asus Prove2 and TAT all show voltage to be UNDER 1.325 set in bios.

 

[goingmerry]

on my 4300 , 1.325v i can only get teh fsb to 250...what could i be doing wrong?

 

[Roguestar]

Sig.

Also, StopSign, I'm impressed. Some people here have gotten really good overclocks while undervolting.

 

[GuitarDaddy]

Originally posted by: n7

The reason you can't get anywhere w/ 1.325V is vdroop.

Your mobo should have terrible vdroop if it's anything like the P5B-E my buddy has, which is why setting vcore in bios & telling people to test for max will not provide consistent results.

Some people will end up with vcore @ 1.3V, some 1.325V, others 1.28V, all depending on vdroop, which makes for very unfair comparisons.


I don't know how far i can go with 1.325V in bios, but with in Windows vcore readings, i can do:

E6300 @ 3276 MHz 1.35V idle/1.33V load Orthos 10 hours blend
E6300 @ 3003 MHz 1.275 bios/1.25V idle/1.23V load Orthos 10 hours blend

This is with case side open & who knows what ambient temps...

Good point!

And you are correct about this board the vdrop(diff between bios and windows) and the vdroop(diff between idle and load) are huge on this board and seems to get worse as you increase the volts. For example if I set vcore in bios to 1.375 I get 1.30 at idle in windows and it drops to 1.26-1.28 under load. If I up it to 1.475 it shows 1.40 at idle and drops to 1.35-1.36 under load:shocked:

I'm considering getting a volt meter to check the actual at the board. If a volt meter confirms this I may try and do the vdroop mod, although I'm hesitant because I'm quite a rookie with a soldering iron

 

[idiotekniQues]

while in bios my voltage is at 1.337 in speedfan i show idling at 1.30v and under small fft's load in orthos it droops to 1.28v and even 1.26v - at 3.2ghz stable

 

[GuitarDaddy]

Wow! That is an amazing chip

 

[bobsmith1492]

3.15 stock volts here but limited by my ram so I really don't know.

 

[GuitarDaddy]

I hope you mean 1.35 not 3.15

 

[bobsmith1492]

Originally posted by: GuitarDaddy
I hope you mean 1.35 not 3.15

That is, 3.15 GHz using stock volts.

 

[GuitarDaddy]

:laugh:

I read it wrong thought you were saying 3.15v

 

[judasmachine]

still working on mine, and deciding if i want to push it. however, right now i'm at 2.53ghz on stock voltage. it's at home folding as we speak, so i'll see how that goes before pushing it further.

 

[Hemsky]

I've got a E6600, i know this is for 6300/6400 but I managed to get it to 3.0 stable using stock cooling and a 1.225 voltage. I believe my vcore was defaulted at 1.2.

 

[gerwen]

My E6400 is orthos blend stable for 20+ hours @ 3.2GHz with Vcore in bios set to 1.275V. CPUz reports the voltage under load as fluctuating between 1.248V and 1.264V.

My ram limits my overclock to 400FSB, so i can't test my cpu any higher.

 

[idiotekniQues]

Originally posted by: gerwen
My E6400 is orthos blend stable for 20+ hours @ 3.2GHz with Vcore in bios set to 1.275V. CPUz reports the voltage under load as fluctuating between 1.248V and 1.264V.

My ram limits my overclock to 400FSB, so i can't test my cpu any higher.

are you an L stepping chip owner?

i really cant complain about 3.2 stable at 1.337volts - but if i had one of the L chips i could be running at 3.4 24/7 stable at lower volts.