Overclocking my Q6600 to 4 ghz if possible

[jonkb]

So I got my new thermalright chipset cooler today as I thought that the heat from the NB limited my 4.0 ghz OC, making it unstable. But to my big surprise, Prime95 still crashes at 4 Ghz after 5 secs, and my PC reboots and sais "Overclocking failed" It went fast from 34 C to 68 C in Prime95 before crash.

Asus P5Q Pro and Q6600 with Prolima Megahalems CPU Cooler.

I tried this:

9x 440 FSB, it did boot, but crashed in prime. 8x500 didn't boot at all.
CPU Voltage: 1.6 V also i tried 1.5 and 1.55.
CPU PLL Voltage. 1.6 V. tried 1.7 and 1.8.
Ram: 2.2 V
NB Voltage: 1.5 (still cool on my thermalright)
Load Line: Enabled
FSB Termation Voltage: 1.3. Tried 1.4 and 1.5 to.
SB: Auto. Did try 1.4.
Ram runs 5-5-5-15 timings, so it shouldn't limit the OC.

All that CPU Skew and NB Skew Idon't know that it is...

Cheers

 

[IINexusII]

leave southbridge alone. that wont make any effects in overclocking

 

[MRCL]

Heh, the q6600... I got mine to 3.3GHz on water, altho I didn't go above 1.45v. Might try more for the heck of it.

Now. That Skew stuff, set it to normal, that should do it. And don't forget the FSB strap to NB (set 333 or even 400). That added stability for me above FSB300. I dunno if it really did, but I felt that way.

 

[jonkb]

leave southbridge alone. that wont make any effects in overclocking

I did leave it on auto as you can see.

 

[IINexusII]

just done a quick google and seems 4ghz is almost impossible. the voltages are too high anyway. i suggest keeping a lower clock.

 

[jonkb]

Heh, the q6600... I got mine to 3.3GHz on water, altho I didn't go above 1.45v. Might try more for the heck of it.

Now. That Skew stuff, set it to normal, that should do it. And don't forget the FSB strap to NB (set 333 or even 400). That added stability for me above FSB300. I dunno if it really did, but I felt that way.

Oh... there is still hope... ha.

I didn't adjust the settings for the FSB strap. I will give it a try.
Thanks

 

[jonkb]

just done a quick google and seems 4ghz is almost impossible. the voltages are too high anyway. i suggest keeping a lower clock.

Doesn't surprise me I had it once at 3.7 Ghz, and that's pretty high too, so I guess I should be more than happy with that. But I still wanna give it a try

 

[3870x2]

the NB is what will affect it

 

[jonkb]

the NB is what will affect it

Weird, cause my Thermalright chipset cooler keeps it damn cool.

 

[IINexusII]

Weird, cause my Thermalright chipset cooler keeps it damn cool.

what temps are you getting for the nb?

 

[jonkb]

what temps are you getting for the nb?

I don't know any program that reads the NB temp. Please tell me

All I know is that the stock cooler was burning hot, and the thermalright keeps it very cool.

 

[MRCL]

Everest tells it. Oh, and a simple 120mm fan aimed at the NB drops temps considerably. And I mean considerably.

 

[jonkb]

Where is the NB temp then?

 

[4x4n]

Doesn't look like your board has a sensor for the nb. As for getting to 4.0, it's very hard to find a Q6600 that will be stable at 4ghz. I have had well over a dozen of them and found a few that would bench at 4.0, but only one that was stable for 24/7. Temps were very high also. I ran that chip at 9x430 at about 1.45vcore for about a year 24/7 running WCG. Getting above 3.7, you're doing good.

 

[angelkiller]

Doesn't look like your board has a sensor for the nb. As for getting to 4.0, it's very hard to find a Q6600 that will be stable at 4ghz. I have had well over a dozen of them and found a few that would bench at 4.0, but only one that was stable for 24/7. Temps were very high also. I ran that chip at 9x430 at about 1.45vcore for about a year 24/7 running WCG. Getting above 3.7, you're doing good.

Were you on air? Do you think 4GHz is attainable on water?

 

[a_ump]

Were you on air? Do you think 4GHz is attainable on water?

when your going that high, the limitations of your board and chip itself or more of what to look at, if your temps are fine then they are fine no matter the cooling. Though if i were to run my q6600 at 3.4ghz 24/7, i'd get a better cooler, 3.6ghz+ i'd get water cooling. Nice temps btw, what are they at load? and how much did you pay for your water cooling setup?

 

[MRCL]

Were you on air? Do you think 4GHz is attainable on water?

I tried yesterday because of jonkb But it needs insane volts. Looking at the temps, its attendable on water, but I couldn't boot above 1.6v, because my PSU couldn't take it anymore. Maybe I try today with just one graphics card. It would be cool to reach 4ghz, just to make a CPUz screenie and be in the 4ghz club again

 

[jamesrt2004]

he uses the Megahalems not watercooling a_ump

+ everest is out by 10 degrees compared to realtemp/core temp/occt etc ...

Everest uses TJMAX of 90... whereas it should be 100 just so you all know

 

[Deleted member 24505]

You can change the tjmax on everest to whatever your chips tjmax is.

 

[Mussels]

very few Q6600's ever reach 4Ghz stable. i've owned three, my brother had one, and both my current housemates have one each. NONE of them can go past 3.6Ghz stable on air. That took upto 1.45v to achieve on two of them, two others took 1.5v+ (and 85C load temps on a TRUE120)

All those reports of 4GHz Q6600's were from people who did nothing but post screenshots proving they'd hit windows - very, VERY few of those people could prove their systems were stable at those clocks.

 

[4x4n]

Were you on air? Do you think 4GHz is attainable on water?

Yes, air cooling with a TRUE, but for 4ghz you really need water. I found this shot of 32m at 4ghz. It was prime stable at this setting but temps would shoot into the 80's after only 10 minutes.

very few Q6600's ever reach 4Ghz stable. i've owned three, my brother had one, and both my current housemates have one each. NONE of them can go past 3.6Ghz stable on air. That took upto 1.45v to achieve on two of them, two others took 1.5v+ (and 85C load temps on a TRUE120)

All those reports of 4GHz Q6600's were from people who did nothing but post screenshots proving they'd hit windows - very, VERY few of those people could prove their systems were stable at those clocks.

4ghz stable is rare, but I've had a few chips that were stable in the 3.7-3.8 range.

 

[Mussels]

pi stable isnt what i'd call stable, fwiw.

as has been discussed over the forum a few times, people who have a 'benchmark' stable system find that they get random crashes, and they dont last long in regular stress tests (OCCT, prime95, orthos)

And even the people who find themselves stable in those, tend to be completely unstable in some of the more modern tests like the linpack based tests (try the latest OCCT's linpack tests on a chip like that, and see it go up another 10 degrees)

 

[oily_17]

he uses the Megahalems not watercooling a_ump

+ everest is out by 10 degrees compared to realtemp/core temp/occt etc ...

Everest uses TJMAX of 90... whereas it should be 100 just so you all know

According to Intel DTS specs it should be 90C, so Everest is right.

http://www.tomshardware.co.uk/intel-dts-specs,news-29460.html

And I agree you may be able to bench at 4GHz but it probably wont be stable for running 24/7

EDIT:Inserting link is not working

 

[Mussels]

Q6600 G0 has a TJmax of 90. the guesses of 100 and 95 were wrong.

 

[Drizzt5]

I have my q6600 watercooled and I have it 3.87 ghz stable.... Going to 4ghz 24/7 stable is near impossible due to the amount of voltage I need to put into the fsb/nb/vcore.... It's just not safe for any of my hardware and not worth it. But benching @ 4ghz+ is easily done.

If temps are a problem, a quick and easy solution is to lap your processor. If done right it can lower temps a lot. It lowered mine anywhere from 6-10c's load.



http://valid.canardpc.com/show_oc.php?id=577553

Look.... I got here with only 1.48vcore...Temps were fantastic.

I even ran 3dmark06 with it and it didn't crash. It just will shut down if using prime95 though.

I have 1.5 in now 24/7 and never go over 69 load with pretty high ambient temps.