Q6600 Overclock

Soldato
Joined
1 Jun 2010
Posts
7,058
Location
Bedford
Hi guys

I need your opinion and advice on this please.

I have been planning to overclock my Q6600 G0 to 3.4GHz. For now I overclocked it to 3.0GHz on stock voltage, ran Prime95 small FFTs test for 8 hours and it is solid:). My cpu vid spec is as follows:

Q6600 G0 1.2625v VID (stock voltage)

My mobo is Gigabyte GA-EP45-UD3LR and the current cpu cooler is 'Thermaltake Big Typhoon' (which I am planning to replace soon). The thermal paste I am using atm is 'Zalman ZM-STG2' which will also be soon replaced by MX-3.

-So basically I upped the cpu speed to 334MHz x 9 = 3006MHz= 3.0GHz
-Changed the Vcore in bios from auto to 1.2625v (stock voltage)
-PCI-E is set 100MHz
-Disabled EIST and C1E
-System Memory Multiplier (SPD) is set to 2.4B for 800MHz DDR2 giving me slight overclock of 802MHz DDR2

Here are the results

[email protected] Idle
q6600idle30ghz.png



[email protected] Load
q6600load30ghz.png



As you can see Bios Vcore is 1.2625v, idle voltage is 1.216v and load voltage is 1.184v. The load voltage actually periodically fluctuated from 1.168v to 1.184v and vice versa. So I seem to experience a vdrop/vdroop of about 0.0785v-0.0945v which seems quite large and yet it is still stable.

-For now my question is if I were to get Thermalright IFX-14 with 2x Akasa Viper 120mm fans and MX-3, will I be able to get much lower temps for the same speed? Also with these will I be able to reach 3.4GHz comfortably?

-Also which is the actual voltage cpu uses? Is it the bios voltage which is 1.2625v or the cpu-z voltage which was 1.168-1.184v during load? When some people say not to use more than 1.45v, do they mean the bios voltage or the cpu-z voltage?

My current cpu cooler is pretty old and I was reading in a review that it is suitable for dual core cpus but not for quads overclocking yet it seems to give good temps at 3.0Ghz lol:p.

Many thanks for your time guys :)
 
Last edited:
Thanks guys for the info. My case is coolermaster HAF 922 which is very large for a mid tower with excellent airflow and it seems to be even slightly bigger than cooler master 690 ii advanced.
So the cpuz core voltage is the actual voltage that a cpu uses during idle and load.In that case shouldn't it be 1.216v during idle instead of 1.2625v as hairybudda is suggesting?

Marine what vcore you needed in bios for 3.4GHz & 3.6GHz and what were the idle and load voltages in cpuz for these speeds? Is the 1.48v bios vcore for 3.6GHz ? Also what is your Q6600 VID? Your temps are very impressive for 3.6GHz:).

Thanks again
 
Something you can consider:
http://forums.overclockers.co.uk/showpost.php?p=17566743&postcount=8


As for your Q6600 with lower VID, I would suggest you start from 1.4V work your way up, or 1.45v up or down.

Yeah I have thought of placing a fan inside the front of my case. So I should set vcore in bios to 1.4v for 3.4Ghz and test the overclock stability from there or is it for 3.6GHz?

you are very lucky. i got a q6600 too and i hit a wall at 3ghz. it refuses to go any higher unless i go into very dangerous volts and even then it is unstable. i don't remember the vid but mine is G0 too.
for 3ghz i have 333x9 and i am using 1.3750v. any lower and it is unstable. anything more than 3ghz and i need to hit over 1.4750v to be half stable. the temps are off the roof at that point and i am using a TRUE heatsink. :(

I was gonna say the same thing. It could be your mobo with poor chipsets that would prevent even moderate overclocking. Also what case are you using?
 
Since you have a P45 chipset board, the Q6600 should hit 3.6GHz no problem...so I would suggest to go straight for 3.6GHz (that's what I did). I would suggest you start with 1.45V (and a little bump for the Northbridge as well), and then run prime to check stability to determine if you want to go up or down from the 1.45V.

Wow you are brave:p. It will be interesting to try to overclock it to 3.6GHz on air even though I originally intended to go up to 3.4GHz. I know at these speeds, my cpu cooler won't be able to handle the temps and I will definetly get BSOD :(. Now all I need is Thermalright IFX-14 with 2 x Akasa Viper 120mm fans and MX-3 thermal paste :) .
 
You have all right parts for an easy 3.6GHz, a cooler is your only real option.

your board by all rights is better than a P5Q deluxe/premium for overclocking the Q6600.

I would seriously consider going water as the rest of your stuff really is as good as it gets.

(well faster RAM and maybe a lucky Q6600 would be better but thats only being said to stop me getting flamed by the pedantic)

Haha redshadows the man with [email protected]:
http://forums.overclockers.co.uk/showthread.php?t=18202102

I was seeking your opinion aswell as you did ask me in your thread what mobo I had and seemed as you have similar VID of your Q6600.:)

The thing is I don't know anything about watercooling:(.For now I am sticking with air cooling. Maybe any more hints you can give me with regards to bios settings @3.4-3.6GHz.:)

Many thanks
 
Don't worry too much about it. 3.6GHz is not going to cause a meltdown, provided you are not using crazy voltage :D

Before I got my Silver Arrow, I was only using a Arctic Cooling Freezer 7 Pro, and even that kept my Q6600 on the same settings under 70C in prime.

Above 3.6GHz though, it would require lots of extra voltage, it is simply not worth the extra hassle voltage, heat, power honest for the sake of extra 200MHz that's barely noticable in real world experience.

Like redshadows has said in his thread, 4.0GHz would most likely need 1.6V for keep the Q6600 stable (if at all)...I would definitely not consider going 1.6V unless I was watercooling the CPU (real watercooling, not the Corsair kit which's cooling performance is only as good as the high-end air-cooler).

True:).
 
Thanks guys for all the help and info:). Once I get better cooling I will definitely report back.Meanwhile if any other suggestions, please do keep them coming:).
 
Just another question.

Do Core i series in general run hotter than Core 2 Duo/Quads? I was reading somewhere that i7 runs hotter than Core2 Quad. Is it true?:)

Thanks
 
Where are you planning on buying a new heatsink? I also have a Q6600 and want to upgrade (still running stock cooler!), but from what I can tell all the heatsinks for these previous generation of CPUs have been pulled from the market. Only thing I can find is auctions, and I'm not too happy about using second hand heatsinks, I'd rather my CPU use a nice fresh unused one...

Hi

I am planning on buying either of the coolers (Thermalright IFX-14 or Akasa Venom) most probably from overclockers although I am looking at other sites aswell.

-Thermalright IFX-14 coupled with 2x Akasa Viper 120mm fan and MX-3 thermal paste:

Thermalright IFX-14
http://www.overclockers.co.uk/showproduct.php?prodid=HS-069-TR

Akasa Viper 120mm fan
http://www.overclockers.co.uk/showproduct.php?prodid=FG-059-AK

MX-3 Thermal Paste
http://www.overclockers.co.uk/showproduct.php?prodid=TH-002-AR

or

-Akasa Venom Cpu cooler

http://www.overclockers.co.uk/showproduct.php?prodid=HS-072-AK

I am gearing 95% towards the IFX-14 :)
 
Oh cool, thanks.

Also I just realised how dumb I sounded. 775 coolers are still very much very easy to find. Don't know why I thought otherwise.

Yeah the modern coolers still cater for 775 socket :).

These two are also top end air coolers:

Prolimatech Megahalems:
http://www.overclockers.co.uk/showproduct.php?prodid=HS-005-PL

Noctua NH-D14:
http://www.overclockers.co.uk/showproduct.php?prodid=HS-011-NC

Here is another good one though you can't attach 2nd fan on it:

Titan Fenrir Evo
http://www.overclockers.co.uk/showproduct.php?prodid=HS-010-TI

All the coolers I have listed including what I am planning to buy are excellent for overclocking Q6600. :)
 
hi guys back again.

I have been reading on power phases and mofsets though I am not really familiar with this topic. It seems more mofsets mean the voltage regulator of mobo runs cooler and provides more efficient transfer of voltage to cpu and this results in more stable and higher overclocking. Here is th link:

http://www.techarp.com/showarticle.aspx?artno=71

From this it would seem that Asus P5Q Deluxe has 16 phase voltage regulator or 16 mofsets whereas my mobo Gigabyte GA-EP45-UD3LR only has 4 mofsets around the cpu area and thus has a 4 phase voltage regulator:(.

So I am guessing P5Q Deluxe would be the superior board in terms of overclocking Q6600 :( although I still have yet to over clock my Q6600 to 3.4-3.6GHz .
 
Wow:eek:. Thats insane. Q6600 @ almost 4.9GHz on Gigabyte GA-EP45-UD3P. That's insane. How did he achieve it? Did he use liquid nitrogen? Yeah afaik the GA-EP45-UD3LR is same as GA-EP45-UD3P in core functionality with few differences in features. :).
 
Hi guys back again :).

I have just recently bought Thermalright IFX-14 along with 2x Akasa Viper 120mm fans and MX-4 thermal paste as stated in my previous posts. Somehow the cooler was slightly narrower at one end compared to the other end. Here are the pics of the cooler in my HAF 922 case:

ifx3.png



ifx1.png



ifx2.png



ifx4.png



Pics of [email protected] (prime95 stable) to follow very shortly:).
 
Although it has been very cold and snowing outside but inside the home it has been very nice and warm with radiators running hence the idle temps reaching generally in mid to high 30s degree celcius already:). The cooler has proved to be excellent with akasa vipers fans attached to it for [email protected]. However the only downside is that my Gigabyte P45 mobo has a high collective 'Vdrop/Vdroop' of 0.1v:(. I ran the Prime 95 small FFTs stress test for 10.5 hours and it was rock stable. These were the following voltages for [email protected]:

Bios Vcore: 1.41250v
Idle voltage: 1.360v
Load voltage: 1.312v


[email protected] Idle

q660034ghzidlebiosvcore.png



[email protected] Load

q660034ghzloadbiosvcore.png



The load temps were very good although I have noticed that core 2 and 3 run few degrees cooler than core0 and 1 and there is generally always 6-8C difference between core 0 and core 2. I suspect that core 0 + core 1 form one chip and core2+core3 form the 2nd chip as Q6600 consists of 2 E6600s joined together rather than being a 'true' quadcore so one chip may well run few degrees hotter. This is again my speculation and I could be wrong.

Coming back to the issue of large vdrop/vdroop; I have been reading easyrider's Q6600 overclocking thread:

http://forums.overclockers.co.uk/showthread.php?t=17757702&page=69

In the post 2065, setter's Q6600 had a slightly higher VID of 1.2750v compared to mine's 1.2625v. However for the same bios vcore of 1.4125v he had [email protected] prime stable as his load voltage was 1.416v compared to mine's 1.312v:( although he is using a different mobo. Is there any way of reducing vdrop/vdroop? I have heard of pencil mod but not sure how to apply it exactly.

All the comments/opinions are welcome:). I have achieved my original goal of [email protected]:). However I have been running Gigabyte 'Energy Saver' utility and noticed that for [email protected] and 3.4GHz at idle, the cpu power usage is 17w and 34w respectively. So cpu power consumption seems to have doubled :eek:. Would this have much effect on my electricity bills?

Lastly I am planning to push my Q6600 to 3.6GHz although I don't plan to run at such speed for a forseeable future though it would be interesting what voltages and temps I get considering my mobo has large vdrop/vdroop. Any more suggestions guys?

Many thanks :).
 
Looking good Wingman :D

Any reason why you aren't going for 3.6Ghz?

Haha bainbridge aka Gareth to who I sent the link to akasa viper fans:). I looked at your thread about meghalems with akasa viper fans and I am glad you reached 3.6GHz on your Q6600. What were your voltages including bios vcore and did you up the northbridge (MCH) voltage?:)

Expect an easy 3.6GHz+ under that cooler, 1.350v under load will see around 65c per core at a guess, which is still more than safe :)

How much does bios vcore matter if the load voltage is around 1.350v as you mentioned? So hypothetically if bios vcore was 1.55v and load voltage was 1.35-1.37v, will bios vcore degrade the cpu life/performance?
 
I'd be suprised if you need anywhere near 1.500v BIOS for 3.6ghz stable. To give you an idea my former Q6600 managed 3.7ghz with around 1.4v under load, and brothers Q6600 is 3.6ghz stable with only 1.360v.

I see what you mean. It's just that I always experience around 0.1v of collective vdrop/vdroop on my mobo. So whatever bios vcore I enter, my load voltage is always around 0.1v less than bios vcore. So if I set bios vcore to 1.40v, my load voltage will most likely be around 1.30v.

For a load voltage of 1.350v, I am pretty sure I will end up with bios vcore of 1.45+v unless I can reduce it somehow with pencil mod.
 
So I take it that out of the three voltages: bios vcore, idle voltage and load voltage, the load voltage is the most impotant one to look at?
 
Yeah thanks for your help mate, purring along at 3.6Ghz as I type :)

Had to go right up to 1.5v (1.5075 in BIOS) to run stable at 400x9 under load, anything below this would result in a crash, though I was able to boot to windows @ 3.6Ghz from 1.45v, anything below this would BSOD after posting.

Regarding the Northbridge all that was needed was a 1 increment nudge up in voltage from the default 1.25v to 1.33v. Before I did this I was unable to get stable under load with the clockspeed above 370mhz on a x9 multiplier.

Looking at your '£100 overclock thread', your VID is 1.2875v which is not very far from my 1.2625v. As you are on Abit and I am on Gigabyte, I think my results are likely to be different for [email protected].
Is the 1.5v the load voltage in cpu-z?
 
Back
Top Bottom