*Official Retail G0 Q6600 Overclocking Thread*

[Mobsters]

i got Q6600 GO 1,3125 VID, L727a839, E10342-001. Any idea how high this is generally going on air? as right now its 3GHZ(9x333) with 1.340V

[LennyRhys]

Your CPU has a VID over 1.3v so don't expect anything great. My 1.3000 CPU needs 1.375 to keep 3.2GHz stable, maybe 1.5v (too high) for 3.6GHz.

[OVERK|LL]

Your CPU has a VID over 1.3v so don't expect anything great. My 1.3000 CPU needs 1.375 to keep 3.2GHz stable, maybe 1.5v (too high) for 3.6GHz.

Mine is at 1.425V right now at 3.45Ghz. Might be a number to try?

[mikeyakame]

If you raise VTT accordingly with vCC (vCPU) you should be able to get away with lesser of a voltage, just remember VTT is used as a termination voltage for VCC and is also there to minimize crosstalk and electrical noise from other signals, on top of preventing it from overvolting too high on first risingedge of clock during boot. vTT does little to raise CPU temps if you keep it slightly higher than vCC only. Only needs to be 0.01-0.02 higher minimum. If you raise it higher only reason would be to raise GTL_REF voltage to create headroom.

My G0 has 1.2875 VID, VID is just Intel's way of saying this chip will be 110% stable at this Voltage with in the min and max boundaries. VID doesn't really do much other than that. Of course a really low VID might be a good sign, but there is always the chance that the chip's voltage vs efficiency curve might fall off early, much like highish VID might cause it to fall off much later.

Difference with mine is at 3.57GHz I only need 1.4350v in bios, which is 1.41-1.42 real. Step up to 3.6GHz and need to add larger increments to keep it stable. 58-59C at full load with 3.57GHz is fine by me, the extra 3-4C at 3.6GHz seems like its nearing the end of its efficiency.

[fatguy1992]

YAY i finally got past 3.6GHz on my Q6600, i got 3.73GHz



I knows its not that much higher, but i feel it could go a lot higher, i'm just waiting for a really cold morning Then i'll try for over 3.8GHz.

My Settings:

Vcore: 1.425 in BIOS (1.392 on desktop)
CPU PLL Voltage: 1.8v (is that safe?)
FSB: 415
Multiplier: 9
FSB Strap To NB: 333
RAM: 830MHz @ 4-4-4-12 2T

Everything else on Auto

[fatguy1992]

Update: I just made it to 3.85GHz.



Same settings as before, just a higher FSB (425)

[LennyRhys]

Of course a really low VID might be a good sign, but there is always the chance that the chip's voltage vs efficiency curve might fall off early, much like highish VID might cause it to fall off much later.

Very true - one guy I was speaking to yesterday also has a 1.3000 VID Q6600 and he can get 3.85GHz Prime stable at 1.49v... I needed 1.6v to get 3.8GHz futuremark stable, so it probably needs over 1.65v to get it prime stable. In my personal experience, and from what I see in this forum especially, the lower VID chips 99% of the time are the best overclockers/the least voltage hungry.

[mikeyakame]

Very true - one guy I was speaking to yesterday also has a 1.3000 VID Q6600 and he can get 3.85GHz Prime stable at 1.49v... I needed 1.6v to get 3.8GHz futuremark stable, so it probably needs over 1.65v to get it prime stable. In my personal experience, and from what I see in this forum especially, the lower VID chips 99% of the time are the best overclockers/the least voltage hungry.

These days that definitely seems the case. Guess the low voltage batches are the highest binned of the current 65Nm desktop processors still in production. I believe Intel has end of lined the rest of the 65Nm quad core desktop derivatives. The first 2 or 3 months of the G0 stepping brought many low and high VID processors of which were all similarly binned, the higher VID might be due to the chip refusing to be 110% stable at a lesser voltage and every chip has some kind of manufacturing flaws.

Some might make the chip unstable at a lower than nominal voltage but still a very flexible high clocking chip with sufficient temperature management to reach its peak efficiency range. While others might leave you with a chip that works with any minimal voltage u can throw at it have a great voltage vs efficiency curve but simply not be able to withstand 20-30% higher clocks over design specifications. Then you've just got some chips that finished up with very minimal faults and since Intel arent pushing all the high end 65Nm chips they were last year round this time, the best binned quad cores probably end up as Q6600s. Good for joe consumer.

[LennyRhys]

Well I sure hope my next Q6600 (L737B460) is better than the two I've tried so far; I'll find out on Tuesday hopefully if I receive it then. Wizard's useful list of L737 chips shows that most of the early 737Bs have VIDs at the 1.2 mark and ALL of them reach 3.6GHz with crazy low voltages (the best being 1.2000 VID making 3.6GHz @ 1.272v), which is precisely what I'm after (who isn't).

[S1mon-]

Well I sure hope my next Q6600 (L737B460) is better than the two I've tried so far; I'll find out on Tuesday hopefully if I receive it then. Wizard's useful list of L737 chips shows that most of the early 737Bs have VIDs at the 1.2 mark and ALL of them reach 3.6GHz with crazy low voltages (the best being 1.2000 VID making 3.6GHz @ 1.272v), which is precisely what I'm after (who isn't).

My L737B does 3,6Ghz (8x450) at 1,288-1,296V and I'm at 3,8Ghz atm with 1,392V

[LennyRhys]

Man that's a sweet chip... I now have three other Q6600s arriving early this week along with my 737B (one of them is a 738A, might get lucky) so let's hope I get a nice one - gimme 3.6GHz @ 1.3-1.35v and 4GHz benchable and I'll be happy. My E6750 did 3.6GHz stable at 1.375 (BIOS) so to get that from a quad would be immense. And the reason I'm buying so many quads is that I want it to be a great overclocker that I'm gonna keep for a long time, so it's worth the effort to get a good un, especially after playing with two pretty crappy L800s.

[S1mon-]

Yep I can easily bench at 4GHz on aircooling and max valid I got when I was pushing on aircooling was 4,15GHz. Good times

[fatguy1992]

This morning i overclock my Q6600 up to 3.9GHz



I also tried for 4.0GHz, just before but the amount of volts i need is making it get quiet hot. So in the morning its only going to be like 5C, so it should work

[LennyRhys]

3.9GHz with 1.424v is extremely nice for a 1.23 VID chip - you've got yourself a very good overclocker there.

I got three Q6600s in the post today and have established the following:

L737B460 has 1.3250 VID
L738A887 has 1.3250 VID
L804B266 has 1.3250 VID

Disappointed. I now have five Q6600s in my posession, all of them poor overclockers with VID greater than 1.3v.

I even thought for a moment that Core Temp is not giving me a correct VID reading, but my first Q6600 is 1.3000v, so it seems to be reading the CPUs rather than just churning out the max VID like it used to. :|

But I have also bought another two Q6600s which will hopefully arrive tomorrow, and one of them reads "1.12v" in CPU-Z at stock 2.4GHz. That's promising...

[Loser777]

I've never seen a L737 1.3250. That is some major rotten luck there. My first chip was 1.3125v, took it back to the store, and I got a 1.2125v .

Are you going to sell those other chips?
Seriously, I have never seen anyone get such rotten luck...

[fatguy1992]

3.9GHz with 1.424v is extremely nice for a 1.23 VID chip - you've got yourself a very good overclocker there.

I got three Q6600s in the post today and have established the following:

L737B460 has 1.3250 VID
L738A887 has 1.3250 VID
L804B266 has 1.3250 VID

Disappointed. I now have five Q6600s in my posession, all of them poor overclockers with VID greater than 1.3v.

I even thought for a moment that Core Temp is not giving me a correct VID reading, but my first Q6600 is 1.3000v, so it seems to be reading the CPUs rather than just churning out the max VID like it used to. :|

But I have also bought another two Q6600s which will hopefully arrive tomorrow, and one of them reads "1.12v" in CPU-Z at stock 2.4GHz. That's promising...

Thx, I hope u can finally get a good one. I got it to 4.0GHz this morning







It was meant to be like 5C room temp, but it was only 10C to 13C.

[sabrewulf165]

Where do you guys get all these lower VID Q6600s? All 3 that I have used (2 for myself, one for my brother) were 1.325vid, unless CoreTemp is reporting the VID incorrectly. I don't mind so much, both of mine do 3ghz on 1.300(1.288 observed) and 3.2 on 1.325 (1.312 observed)

[Loser777]

When did you buy these Q6600s? Most from L746 onward have been bad, with the exception of the L808s.

[FabricioGS]

Thx, I hope u can finally get a good one. I got it to 4.0GHz this morning

Is this Prime95 stable?

[jeeka]

Where do you guys get all these lower VID Q6600s? All 3 that I have used (2 for myself, one for my brother) were 1.325vid, unless CoreTemp is reporting the VID incorrectly. I don't mind so much, both of mine do 3ghz on 1.300(1.288 observed) and 3.2 on 1.325 (1.312 observed)

Bought mine from ClubIt right when the G0's came out. Mine (L721A814) has a VID of 1.275v but I need 1.5875v BIOS (1.56v real) to get to 3800Mhz where it has been sitting comfortably since I reached that clock....

Medium VID w/ a high tolerance for voltage, but not as good voltage-to-clock performance like some of the golden CPU's that other dudes have here.

[LennyRhys]

both of mine do 3ghz on 1.300

Yeah I heard about a few 1.325s that do that, but my attempts at 3GHz with the L811 in my sig stopped just short of 1.34 vcore - it was so pathetic that I simply gave up. My 1.300 is stable at 3.2GHz with just over 1.35 vcore but that's not good enough for me lol.

Are you going to sell those other chips?
Seriously, I have never seen anyone get such rotten luck...

Yeah I'm gonna sell them, so it's not too bad. But I agree that my luck is pitiful just now - a total of five chips from a range of years/weeks/batches, and not one of them with a VID lower than 1.3000. Let's hope that changes tomorrow.

[sabrewulf165]

When did you buy these Q6600s? Most from L746 onward have been bad, with the exception of the L808s.

Both of the NIB ones I bought are L737A832, ordered them from Newegg on 3/27/08

I couldn't tell you what the used one I had last year was, though obviously older. I could've sworn I had an IHS pic around here somewhere but I can't find it.

All three were/are G0

[fatguy1992]

Is this Prime95 stable?

LOL no, not enough volts, later one day i'll try to make it stable, but for now i'm just trying for a ultra high overclock

[FabricioGS]

I can do 4.10GHz PI Stable on air




But anytime I try to Prime above 3.65GHz my system restarts. Dunno if it's a PSU problem or maybe high motherboard temps

I'm certain my 1.25v VID can do better then 3.6GHz Prime stable on air, but my knowledge limitations are not helping.

[fatguy1992]

Nice What your mobo's temps, if its that, maybe u could get a TR HR-05 etc.

I'll try for 4.1GHz (460x9) next, i'll use a 1.525vcore.