*Official Retail G0 Q6600 Overclocking Thread*

[TheVaLVe]

Well if that 3.86Ghz is on an Q9450 well I find it very good due to it's handicap of the 8 multi... that would be 482FSB...sounds pretty good to me... 500FSB on quads and co is only for a few CPU's under water, under higher end cooling that's another thing... like is said before await the 9x Q9650 and it will give y'all 4ghz too...

and for ya Crysis high detail experience, there's only one way to make that playable that's either an 9800X2 or a newer ATI or Nvidiot card, CPU power there isn't that much needed as GPU power is... such a badly coded game...

,sry couldnt resist, because its so true...
My quad does pretty well for a 1.3v VID. 484FSB with 6GB and 489 with 2GB RAM, with 1.6375V voltage. No 4GHz though. Still it eats everything on its way, and cheap also.
But indeed quad processors are really burdening the motherboard. They need high voltage on NB and PLL. Me runs 1.7V NB and 1.8PLL voltage, tbh also 1.5 VTT and 2.45 for RAM.
But since its stable and i got watercooling, me can be arsed with the voltages...

[Hans Gruber]

Your motherboard may catch on fire. I would be impressed with your overclock if you hit 4Ghz, but you missed it by 100Mhz. You Germans have direct current vs. alternating current here in the states. You may have a fire hazzard on your hands.

,sry couldnt resist, because its so true...
My quad does pretty well for a 1.3v VID. 484FSB with 6GB and 489 with 2GB RAM, with 1.6375V voltage. No 4GHz though. Still it eats everything on its way, and cheap also.
But indeed quad processors are really burdening the motherboard. They need high voltage on NB and PLL. Me runs 1.7V NB and 1.8PLL voltage, tbh also 1.5 VTT and 2.45 for RAM.
But since its stable and i got watercooling, me can be arsed with the voltages...

[aussie-revhead]

My mobo (Asus Maximus Formula) gives me an cpu overvoltage warning above 1.6v , can I disable this for benching ? Should I do this ? Im a little unstable under 1.6v at near 4ghz but I think its ram related , so I have more ram on the way.

[TheVaLVe]

Your motherboard may catch on fire. I would be impressed with your overclock if you hit 4Ghz, but you missed it by 100Mhz. You Germans have direct current vs. alternating current here in the states. You may have a fire hazzard on your hands.

True that , it happened once on me. Well luckily not on fire though, thanks to good PSU i had .I had P4 oced to 4.05 GHZ with an asus P5AD2-E Premium maxed on all voltages, only on RAM i had 2.10V and powered by Be-Quiet 550W. One day, after my uni, i turned on my PC, and it just did nothing, only the electricity in my apartment fell down. I inserted another coin, and it did the same. I knew this is because of my PSU, so i brought it to the store for RMA, and it blew up there in the store , with "bang" sound, luckily no fire.

I did 4GHz on benching 3dMark05, well not quite 4GHz, it was 3995MHz....
but it did help me enough to reach my goal 24k.

[TheVaLVe]

My mobo (Asus Maximus Formula) gives me an cpu overvoltage warning above 1.6v , can I disable this for benching ? Should I do this ? Im a little unstable under 1.6v at near 4ghz but I think its ram related , so I have more ram on the way.

Yes under power management in BIOS, search for you CPU voltage, enter on the number, and choose disable, and you good to go.

[Hodori58]

Hi guys quick Newbie question sorry if its the wrong place but i installed a Q6600 on Abit IP35Pro with a arctic cooling freezer with AS5 (since i smeared the factory Thermal ) and the temp in the Bios is showing 53-55 after post, its in a P182 case. I re-applied the AS5 and remounted the HSF and still the same. How accurate is the Bios Temp compared to the CoreTemp and what not? cuz 53C is pretty bad for just POST. thanks guys. I have not yet installed the OS yet since the HDD i got for the OS was defective, but when I get a new one then imma start the OC process.

[Th3MadScientist]

53C is pretty bad, BIOS should be the most accurate temperature reading.

[Hodori58]

Really ouch!!! hmm I used the arctic silver 5 that I had when I used on the AMD XP2500 back in the day. should still be good right?
did I put too little? I put it on the HSF only and spread it pretty evenly. any suggestions will be appreciated. thanks for the quick reply.

[Leeghoofd]

Arctic Freezer can't really handle the heatoutput of a quad... unless it runs at full speed and then it sounds like a tornado... maybe ya mobo doesn't let the fan spin up enough... anyway before you indulge yourself in any OC'ing buy a beefier cooler and enjoy the true power of 4 cores...

for the TIM we draw a thin line accross the cpu lie mentioned on arctic silver webbie :

http://www.arcticsilver.com/pdf/appi..._quad_wcap.pdf

[flexy]

You Germans have direct current vs. alternating current here in the states. You may have a fire hazzard on your hands.

nonsense

Sry...AC here too, 250V.

[ZenEffect]

nonsense

Sry...AC here too, 250V.

how great would it be to get dc from the wall?

[Hodori58]

Arctic Freezer can't really handle the heatoutput of a quad... unless it runs at full speed and then it sounds like a tornado... maybe ya mobo doesn't let the fan spin up enough... anyway before you indulge yourself in any OC'ing buy a beefier cooler and enjoy the true power of 4 cores...

for the TIM we draw a thin line accross the cpu lie mentioned on arctic silver webbie :

http://www.arcticsilver.com/pdf/appi..._quad_wcap.pdf

nice thanks for the link. well the arctic cooling is ok as long as im not gonna push it say past 3.2 or so which I dont really plan on. But if it still is getting those high marks then I probably will have to go with the beefy TRUE 120. thanks again for the fast replies. Still a Newbie for OCing and stuff only OC'd last on the XP1800 days . sry if all this was covered in a different topic as well.

[conzymaher]

You Germans have direct current vs. alternating current here in the states

what?! lol I have never been to or even heard of anywhere in Europe that uses DC electricity... Maybe some folks who live on Wind farms

[Hans Gruber]

nonsense

Sry...AC here too, 250V.

In the US we have 120v wall outlets. I was under the impression all of Europe was DC. My (computer programmer) father lied to me 20 years ago when he referred to European electrical outlets as DC instead of AC.

With 250v outlets, I would think you would get much better overclocks? I also think there is a much finer line drawn to when a computer system blows with the extra voltage.

[Leeghoofd]

You think too much Hans...are you an engineer maybe ? If ya analysis was correct then No_name wouldn't get so good clocks would he...

[DDTUNG]

With 250v outlets, I would think you would get much better overclocks? I also think there is a much finer line drawn to when a computer system blows with the extra voltage.

Most of us are OK since we connect our motherboard to the PSU and not directly to the wall outlet.

DDTUNG

[OVERK|LL]

In the US we have 120v wall outlets. I was under the impression all of Europe was DC. My (computer programmer) father lied to me 20 years ago when he referred to European electrical outlets as DC instead of AC.

With 250v outlets, I would think you would get much better overclocks? I also think there is a much finer line drawn to when a computer system blows with the extra voltage.

I don't think so, it's still being converted to DC by the PSU. This would also be why most PSU's have a switch on the back to go from 115 to 230V.

[Unoid]

Are you kidding me ? The worst of my quad 45nm did 3.6ghz at 1.25volts (that's loaded CPU-Z) the best at 1.13, prime stable..., Going up to 4ghz worst again at 1.39volts , best at 1.25 volts... and you dare to claim there's no big gain...

For dual core even better : 3.6Ghz at 1.15volts, 4-4.5ghz between 1.2-1.5volts, who could have dreamt that a year ago when many where struggling to get 3.6ghz stable with the E6600... no way you could have run that CPU at 3.6Ghz with a stock cooler... my E8400 runs for 2 weeks now at 3.6Ghz 1.2volts with the boxed cooler, temps stay below 60°C on load...

Compare bigger cache, less volts needed and less heatoutput ... I don't know what you expected out of the 45nm but for me it's pretty amasing stuff... some cpu's are great some just plain suck or let's call them bad clockers... but overal we got a nice improvement with 45nm...

Intel just messed up with the multies and kept them low... and why would they have to launch some new superstuff if the competition can barely keep up....

How reliable 24x7 do you believe a 45nm quad will do above 1.3V? Been hearing about instabilities and suddens deaths an awfully lot.

[Leeghoofd]

How reliable 24x7 do you believe a 45nm quad will do above 1.3V? Been hearing about instabilities and suddens deaths an awfully lot.

Stick below 1.4volts and decent cooling... CPU VTT below 1.35 and CPU PLL maxed around 1.6... then you will be fine... if ya mobo can handle it ofcourse

My older retail Qx is still running strong at 1.39Vcore at 4ghz under water (though it's only 6 months old)

[aussie-revhead]

CPU VTT below 1.35 and CPU PLL maxed around 1.6

CPU PLL Voltage: Maximum of 2.78V. Out of all the voltages the user can manipulate this one is by far the most dangerous. Maximum vCPUpll, as established by Intel, is 1.60V (default for this board) making 2.78V a whopping ~75% over specification! (As an aside, this would be the equivalent of subjecting your 65nm CPU to a core voltage of over 2.5V). Exercise extreme caution when utilizing higher values as setting this value too high can result in the CPUs "losing cores" after being subjected to voltage in excess of ~2.0V. The good news is that we did not see an increase in overclocking potential with voltages above 1.68V.

Can you tell me what this actually means ? What should I set it to for benching (Q6600 and maximus Formula) ?

[Leeghoofd]

Vcore I would max out 1.6V for short benches... Regarding ya question I'm not a technical engineer and I get completely lost in reading Anandtech's and others specs and analysis of settings... regarding PLL, Loadline calibration and FSB termination voltage... I gathered the last one was the most dangerous as it killed the Anandtech CPU...

I got it from the big boys that even 1.7V PLL is completely safe with CPU's... so I stick to 1.6 PLL ; might be one of the reasons I cannot breach 475FSB...
But I have enough with 450FSB on all my quad CPU's, just the wolfdale has to undergo 500FSB torture...

[aussie-revhead]

Thanks , I get an overvoltage warning at 1.6v , should I reduce vcore a notch ? Or ignore/change settings ?

[ZenEffect]

Most of us are OK since we connect our motherboard to the PSU and not directly to the wall outlet.

DDTUNG

[saint-francis]

Vcore I would max out 1.6V for short benches... Regarding ya question I'm not a technical engineer and I get completely lost in reading Anandtech's and others specs and analysis of settings... regarding PLL, Loadline calibration and FSB termination voltage... I gathered the last one was the most dangerous as it killed the Anandtech CPU...

I got it from the big boys that even 1.7V PLL is completely safe with CPU's... so I stick to 1.6 PLL ; might be one of the reasons I cannot breach 475FSB...
But I have enough with 450FSB on all my quad CPU's, just the wolfdale has to undergo 500FSB torture...

VTT killed the CPU we are hearing about at anandtech. And it seems to be pretty well proven that raising the PLL at all quickly lowers the Max FSB of the CPU.

[sbinh]

In the US we have 120v wall outlets. I was under the impression all of Europe was DC. My (computer programmer) father lied to me 20 years ago when he referred to European electrical outlets as DC instead of AC.

With 250v outlets, I would think you would get much better overclocks? I also think there is a much finer line drawn to when a computer system blows with the extra voltage.

So lucky that your father is/was computer programmer .. not electrical / electronic engineer .... Else .. we're all in deep **** ....