*Official Retail G0 Q6600 Overclocking Thread*

[ultimeus]

Hi, I have my Q6600 @ 3.6ghz with 1.45v , he is lapped with a thermalright TRUE 120 lapped too (penny mod), nanoxia fan @ 1250 tr/m.

Idle : 35° c

Load : 65°c

Do you think my temps are normal ?

[EagleRock]

temps seems normal with 1.45v.
with more powerful fan u can cut of around 2-3*c
something like 1600-2000 rpm, + 60 CFM

[scottc19]

I am now playing around with my q6600 that I got from canada off ebay! L737B011 It has a vid thats 1.325, but I have had pretty good results so far (have just done a little testing-will be doing more for the next week)
Right now its priming at 3.8ghz 60C load, I might be able to even drop volts a little bit. Currently at 1.56v Anyway what is the max volts for these according to the intel spec sheet, or personal experience?

An interesting thing that I found, was that when I was using p30 bios with 680i temps read about 10C higher, but now with p32 bios temps are reading 10C lower...which is a good thing I am going to double check this as before I was using vista and now am on xp...

I was not able to boot up past about 3.7 without decreasing the multiplier, which I assume shows a fsb in the 680i... no suprize there

EDIT: update-not prime stable yet, 1 core failed rounding after 15min

[Kondik]

Imo 1.56 is too much but taste according your flavor .i personally won;t go over 1.375 because i think that my Quad doesn';t want do die young (:

[ultimeus]

I am now playing around with my q6600 that I got from canada off ebay! L737B011 It has a vid thats 1.325, but I have had pretty good results so far (have just done a little testing-will be doing more for the next week)
Right now its priming at 3.8ghz 60C load, I might be able to even drop volts a little bit. Currently at 1.56v Anyway what is the max volts for these according to the intel spec sheet, or personal experience?

An interesting thing that I found, was that when I was using p30 bios with 680i temps read about 10C higher, but now with p32 bios temps are reading 10C lower...which is a good thing I am going to double check this as before I was using vista and now am on xp...

I was not able to boot up past about 3.7 without decreasing the multiplier, which I assume shows a fsb in the 680i... no suprize there

EDIT: update-not prime stable yet, 1 core failed rounding after 15min

With your VID I think you need 1.475v - 1.5v for 3.6ghz occt stable.

[jonny_ftm]

With your VID I think you need 1.475v - 1.5v for 3.6ghz occt stable.

?? Really impossible to say, and he's talking about 3.8GHz OC, not 3.6GHz

Intel specs are 1.50v maximum voltage
http://processorfinder.intel.com/det...x?sSpec=SLACR#

But with vdroop, when you set 1.50 bios, you could shoot up to 1.6v. A good way to estimate the spikes your chip can get is to disable vdroop compensation, set bios till you get your stable vcore after droop. And, the bios set value should be an appreciation of overshooting voltage your CPU is getting.

There are no stories of intel 65nm chips dieing in few months under 1.55v, so you could run safe for few years maybe.

[scottc19]

I have done a vDroop mod, so vDroop is basically non existent now

[Elio]

Hi all! some results from ysterday :

the Vcore is not 1.3750 it's 1.425 in BIOS Vdrop is 0.03-0.05V.
Temps on load 56-59 C
if i didnt change the GTL of the CPU&NB it wont boot...
it's still not rocksolid,solid enough to play crysis though...but not OCCT.
got any ides how stablized it with less voltage?
what else can i do?

[jonny_ftm]

vSB is too high, no need unless you have 4-6 HD in raid, and even with that...
vNB is too high for that FSB, try lower
Give a screen of upper bios settings: I can see that transaction booster is on auto, should be disabled, relaxed to 3 to start. Static read control to disable to start.
vDDR at 2.20v could be not enough to run your RAM at 1066MHz, ususally you'll need 2.25-2.30v

If Prime95 small FFT is stable, your problem is not vcore
If by OCCT, you mean the RAM mode, than it is in the NB-RAM settings where you should look for, like I explained above

[Elio]

the memory is G.skill pc2-6400 4-4-3-5 (promos) and they are stable 8 hours in memtest.crashed both prime95 and OCCT after 4 or 5 hours.
i will try setting the SB NB lower and i'll update.
i'll try occt ram mode and putting the ram on 800 5-5-5-15 just to check.
Static read control is disabled. i'll post a pic later.
thanks for the help. good night.

[ultimeus]

?? Really impossible to say, and he's talking about 3.8GHz OC, not 3.6GHz

Intel specs are 1.50v maximum voltage
http://processorfinder.intel.com/det...x?sSpec=SLACR#

But with vdroop, when you set 1.50 bios, you could shoot up to 1.6v. A good way to estimate the spikes your chip can get is to disable vdroop compensation, set bios till you get your stable vcore after droop. And, the bios set value should be an appreciation of overshooting voltage your CPU is getting.

There are no stories of intel 65nm chips dieing in few months under 1.55v, so you could run safe for few years maybe.

Yes but he says that @ 3.8 one core fail after 15 min, so for me with that vid he can be stable at 3.6 with 1.5v

[jonny_ftm]

Yes but he says that @ 3.8 one core fail after 15 min, so for me with that vid he can be stable at 3.6 with 1.5v

Agree

the memory is G.skill pc2-6400 4-4-3-5 (promos) and they are stable 8 hours in memtest.crashed both prime95 and OCCT after 4 or 5 hours.
i will try setting the SB NB lower and i'll update.
i'll try occt ram mode and putting the ram on 800 5-5-5-15 just to check.
Static read control is disabled. i'll post a pic later.
thanks for the help. good night.

Memtest is not good at "benching" memory. It is rather a test for integrity of memory. It will stress it a bit, but won't stress the whole subsystem with NB like OCCT RAM or F@H, or some games

I'd try that benching with RAM at 800 5-5-5-15 like you plan, If it pass, than your memory is too much pushed for Promos chips.
Also, the SRC is less stressing than Transaction booster on high overclocks. So, I'll disable both to start relaxing the booster to 3. Once your memory is stable, go with lower CAS tests. Once stable, try to enable SRC/Booster or relax it to 0

[andressergio]

Depends, 90% of the VID quiz is correct

the more VID the more Vcore to get it stable, but sometimes is not...

I had a Q6600 L723A VID = 1,2875 and requiered 1,475 vcorebios to 3,6ghz prime stable

Now I have a Q6600 L737B VID= 1,225 and can do stable:

vcorebios= 1,325 3,6ghz
vcorebios= 1,432 3,8ghz

and benched 4234 on air at 1,625 vcorebios
11,809s SPI 1MB

Cheers
Sergio

[Psycho_eddie]

Depends, 90% of the VID quiz is correct

the more VID the more Vcore to get it stable, but sometimes is not...

I had a Q6600 L723A VID = 1,2875 and requiered 1,475 vcorebios to 3,6ghz prime stable

Now I have a Q6600 L737B VID= 1,225 and can do stable:

vcorebios= 1,325 3,6ghz
vcorebios= 1,432 3,8ghz

and benched 4234 on air at 1,625 vcorebios
11,809s SPI 1MB

Cheers
Sergio

Nice that you only need 1.325 for 450x8

I need 1.4V for 450x8 and has a VID of 1.250

I get 65C loaded temps with True 120EX after 1+ hours of prime95


What temps u getting?

[andressergio]

Nice that you only need 1.325 for 450x8

I need 1.4V for 450x8 and has a VID of 1.250

I get 65C loaded temps with True 120EX after 1+ hours of prime95


What temps u getting?

Thanks Eddie !!

At 24C ambient:
primming at 3600 i got 60C on air
primming at 3800 i got 68 to 70C on air

Let´s see on water, i'm building my first water setup hope everything turns out OK

Cheers
Sergio

[Elio]

can you explain please what does the Static read control and the transaction booster do?

[andressergio]

can you explain please what does the Static read control and the transaction booster do?

Elio in my case Static Read = enabled boost my memory performace overall

Transaction Booster has to be AUTO in my case cause if i set it ON i can't boot, happened with my E6600 and the 2 Q6600 on the P5K deluxe...

Cheers
Sergio

[Elio]

Wll guys i've hit the wall at 9X424. no boot no whether what i do.
didnt try to make solid more than i need to bench it.just wanted to know how far can i go.this as far as it booted to windows.9X423 booted too.
SRC-disabled
transaction booster disabled -3
NB - 1.70 (less it wont boot)
SB - 1.20
CPU GTL-x63 (if it's less it wont boot)
NB GTL-x67 (if it's less it wont boot)
Vpll-1.60
Vfsb-1.30
Vdimm-2.2

i think my mobo cant do with 4cores any more than that.

jonny_ftm i did test it with 800 5-5-5-15 and it was the cpu.
at 3.6 it needs 1.400V to be rocksolid . just cant get it to be less.

[scottc19]

Wll guys i've hit the wall at 9X424. no boot no whether what i do.
didnt try to make solid more than i need to bench it.just wanted to know how far can i go.this as far as it booted to windows.9X423 booted too.
SRC-disabled
transaction booster disabled -3
NB - 1.70 (less it wont boot)
SB - 1.20
CPU GTL-x63 (if it's less it wont boot)
NB GTL-x67 (if it's less it wont boot)
Vpll-1.60
Vfsb-1.30
Vdimm-2.2

i think my mobo cant do with 4cores any more than that.

jonny_ftm i did test it with 800 5-5-5-15 and it was the cpu.
at 3.6 it needs 1.400V to be rocksolid . just cant get it to be less.

Did you try more vfsb?
Did you try more vCore
Try a lower multiplier and see if you go higher.
Or maybe there is a fsb hole, try 50ish higher with lower multiplier

[aaronl]

I am running ver. 1101, with my current set up, P5B D,with voltage mods,E6600 cruising on water at 3.4, PCP&C 750, Swiftec water on cpu, NB, & video. Antec P180 modded for water
I just got a notice from micro center. Q6600 for $199.00. Im ready to jump in the car. Any recommendations

[starlon]

I was going for 3.6ghz. I took it up to 1.5v, but it's still unstable. Is it my board perhaps?

[mikesm]

I just picked up 3 Q6600's from Fry's by pricematching the Microcenter $199 price. All G0 SLACR's.

The one I targeted for my desktop (the other two will go into servers as upgrades) is a L739B890, and has a VID of 1.2250v. I can overclock it to 3.420Ghz (380x9 FSB) with 1.35v with no issues.

One weird thing though is that according to core temp (0.96.1), core #1 and 2 always run hotter than cores 3 and 4, usually by 3-4 degrees at idle, but 6-7 degrees under load. I am using a zalman CNPS9700, and the motherboard is an asus commando. Is this normal, or is something screwed with my thermal paste or something else?

I have two other Q6600's, batch numbers L745B154 and L745A675. Does anyone else out there have those? Do you think I would do better with them than with the L739B890?

One of them will not be OC'd, but the other will go into a server that does a lot of transcoding, so I plan to overclock it as well, but the best overclocker I want in my desktop. ;-)

Thanks!

[jeeka]

I just picked up 3 Q6600's from Fry's by pricematching the Microcenter $199 price. All G0 SLACR's.

The one I targeted for my desktop (the other two will go into servers as upgrades) is a L739B890, and has a VID of 1.2250v. I can overclock it to 3.420Ghz (380x9 FSB) with 1.35v with no issues.

One weird thing though is that according to core temp (0.96.1), core #1 and 2 always run hotter than cores 3 and 4, usually by 3-4 degrees at idle, but 6-7 degrees under load. I am using a zalman CNPS9700, and the motherboard is an asus commando. Is this normal, or is something screwed with my thermal paste or something else?

I have two other Q6600's, batch numbers L745B154 and L745A675. Does anyone else out there have those? Do you think I would do better with them than with the L739B890?

One of them will not be OC'd, but the other will go into a server that does a lot of transcoding, so I plan to overclock it as well, but the best overclocker I want in my desktop. ;-)

Thanks!

My same 2 cores run hotter when idle as well as under load, exactly the same cores, and about the same temp diff. as yours. I wonder if it's a designed thing, or if it is a byproduct of the design.

[starlon]

It's your heatsink's mount. Check out the below thread.

http://www.xtremesystems.org/forums/...ight=quad+core

[jonny_ftm]

Elio in my case Static Read = enabled boost my memory performace overall

Transaction Booster has to be AUTO in my case cause if i set it ON i can't boot, happened with my E6600 and the 2 Q6600 on the P5K deluxe...

Cheers
Sergio

They both act the same, but booster is more agressive and can be manually set. Ususally you can't activate both if you OC memory. Booster should not be auto, as it will be relaxed to 3. Better disable and relax it less so you get more performance if your memory handles it. Relaxed to 0 will give a good emory bandwidth improvement on benchmarks