need help increasing Q6600 OC

sw1ft

sw1ft

sw1ft

Associate
Joined
13 Aug 2008
Posts
221
Hi guys,

so I recently got an SSD and a new cooler, and decided that until I get my new build in full swing I'd try and push my clock further, it was at 3.0ghz and i've got it to 3.2ghz on less volts, but can't seem to go any higher! My specs are:

Q6600
Asus P5Q deluxe
4gb OCZ PC2-8500 (4x1gb)
Corsair TX650 PSU
XFX 8800GT
CM612S cooler

And my current clock is:

AI Overclock Tuner: Manual
CPU Ratio Setting: 8
FSB Frequency: 400
PCIE Frequency: 100
FSB Strap to North Bridge: 266mhz
DRAM Frequency: DDR2-1003mhz
DRAM CLK Skew on Channel A1: Auto
DRAM CLK Skew on Channel A2: Auto
DRAM CLK Skew on Channel B1: Auto
DRAM CLK Skew on Channel B2: Auto
(Timings are stock so 5-5-5-15)
DRAM Static Read Control: Auto
DRAM Read Training: Auto
MEM. OC Charger: Auto
AI Clock Twister: Auto
AI Transaction Booster: Auto
CPU Voltage: 1.43750
CPU GTL Voltage Reference(0/2): Auto
CPU GTL Voltage Reference(1/3): 0.670
CPU PLL Voltage: 1.50
FSB Termination Voltage: 1.30
DRAM Voltage: 2.10
NB Voltage: 1.24
MB GTL Reference: Auto
SB Voltage: Auto
PCIE SATA Voltage: Auto
Load Line Calibration: Enabled
CPU Spread Spectrum: Auto
PCIE Spread Spectrum: Auto
CPU Clock Skew: Auto
NB Clock Skew: Auto
CPU Margin Enhancement: Performance

The above is 16hrs prime stable, temps never touched 60 (max was 59 on all cores) I managed to get 3.28 by changing:

CPU Ratio Setting: 9
FSB Frequency: 368
FSB Strap to North Bridge: 333mhz
DRAM Frequency: DDR2-884mhz
NB Voltage: 1.3
CPU Voltage: 1.47500!!!

but here the temps were 62 and for the 80mhz I can't justify the CPU voltage increase really.

I feel like I could get a lot more (3.4 ideally) as the cooler is doing so well, but it just won't stabilize, I can't use FSB-NB strap of 400mhz (to get 1:1 DRAM:FSB) as it just won't post and I need to reset CMOS. Can anybody see anything I can optimize or tweak to get 3.4? I've tried 378x9 and it won't boot at all.

What steps do I take next? shall I just drop it and stick to 3.2 or do I need to do some more tweaking, I have a rev. G0 chip so I really wanted to be among the 3.4-3.6 guys.

Cheers in advance for any help, sorry for the essay!
 
Last edited:
I have 3.24GHz and I'm happy with it. I've tried 3.4 but never really could be bothered to make it stable. Good that you have a 400MHz FSB that's more important on the Q6600 than 360x9.

I think it's better to stay at 3.24GHz and keep a nice voltage, it's fast enough!

x9
360MHz
1440MHz rated FSB
1.34V (1.4V in BIOS)
Never over 60c.
DDR3 at 1440MHz (720MHz) CL8 1.5V (love my ram, it's 1600MHz 1.65v CL9 lol).

I think you should be able to get 3.4 on a P5Q Deluxe. I have a P35C-DS3R

I don't understand FSB strap to Northbridge. I can try to get to 3.6 if you like and post my settings but I think I need a lot of volts for that and my chip doesn't like the heat all that much.

TBH, I tried for a while to get stable at 3.6, but realised I didn't notice the performance loss. Getting an Ivybridge soon hopefully, but will miss this sweet chip.
 
If I was going to go for 3.6 I would just try booting with 1.5 in the bios, vdroop will usually mean you won't be anywhere near 1.5, which is the max apparently. I'm going to give it a go now, wish me luck. :P
 
Sorry to come back empty handed.

My UP key on keyboard is broken. So is the 2, END, PRTSCRN. Maybe I should buy a new one lol. Couldn't set bios higher because of the damn broken key. Can get use of another keyboard tomorrow. :P
 
Well, stupid of me to forget you can use + and - to change values in bios!

I'm running at 3.6 with 1.5 in bios, my board has terrible vdroop so it's actually 1.39 in windows. Isn't prime stable not even for 5 seconds. Going to increase voltage now will post results.
 
Ok. Testing now 1.55V in BIOS, 1.440V in CPU-Z during P95. 1.488V without. Very tempted to pencil my motherboard now because that is hefty vdroop.

Fingers crossed it won't crash but can always put more juice into it. Highest temp is 65c but average is 62c. This is with fans on full in my kinda restricted airflow case...Silencio 550 with corsair AF120 1650RPM and a D-14 fan grills cut out to improve the airflow a bit.

edit: Passed 1024K, don't understand why my cores finish the tests at such random times though. Core 3 finished first, then 4 about a minute later, then 2 then 1 much later. Bit weird or normal? lol

btw. My chip is a G0 0.825-1.288V. 3.6 is a 50% overclock. ._o

Results from SiSoftware Sandra..
 
Last edited:
which stepping and VID is that q6600 you should be doing 3.6ghz at 1.4 -1.5

I would use intel burn test too instead of prime but be careful it's normally a good 10+c in temp over prime 50 passes memory set to your free ram is rock solid stable imo
 
Last edited:
I read that IBT is too harsh and can cause damage, might not be true but not worth the risk imo, prime stable is stable enough for me. I use memtest86 for my ram.

Unfortunately I don't think my motherboard is capable of holding the overclock every boot and I've been suffering from BSOD lately but I think it's down to my 7970 so just wiped and reinstalled the drivers.

I'm sure the OP can get 3.6GHz stable on his board, he just needs to push the voltage. Make sure ram is stable, may need that extra .1v if your sticks are rated 2.2, and may need to loosen the timings/add more juice because you have 4 sticks installed.

I think mine was stable, but my board isn't amazing it has quite a low FSB wall tbh. P5Q Deluxe is awesome, good luck!
 
Cheers BrianB, really useful to know other people are having good results, just need to work out why i'm using so much vcore for ~3.2ghz. My VID is 1.3250 so I know I don't have one of the best, just hoped for a bit more really!

I'll try loosening timings on my RAM and upping vcore to see if I can get above 3.2, it just seems i've hit the wall with this chip already.

Thanks guys!
 
Hi Swift, there a few threads about regarding the relationship between clock and vid. I'm afraid that with the G0 stepping, basically anything above 1.3 and it's very difficult to get >3.4/3.6. I myself had a 1.3 vid and even with high end water could only manage 3.6. You may want to try upping the PLL slightly and as per Brian's suggesting reduce the RAM timings.
 
Hm, both of mine are 1.288 just got lucky I guess. 3.24 is still a decent clock. As long as it's cool enough try pumping more juiceeeee. Surprised mine stayed cool enough at 1.48v but that's the D-14 I guess. Pencil mod incoming lol.
 
that's a shame, I was hoping it was due to me using all 4 RAM slots and somebody would be able to tell me it was NB volts or something. I'll try upping PLL and loosen timings, maybe increase DRAM volts and then slowly see if I can stabilize.

just seems odd because I know people have been able to boot into bios, change nothing but fsb to 333 and have 3.0ghz, then easily get to 3.2/3.4 with very little voltage change, I must just be unlucky!

Cheers guys
 
That's the CPU lottery. I could probably boot into windows at 3.0GHz on near stock volts. I think it's just my motherboard that is a bit rubbish for 3.3GHz+ overclocks. Have you tried getting 3.3?
 
sw1ft said:
that's a shame, I was hoping it was due to me using all 4 RAM slots and somebody would be able to tell me it was NB volts or something. I'll try upping PLL and loosen timings, maybe increase DRAM volts and then slowly see if I can stabilize.

just seems odd because I know people have been able to boot into bios, change nothing but fsb to 333 and have 3.0ghz, then easily get to 3.2/3.4 with very little voltage change, I must just be unlucky!

Cheers guys
Click to expand...

I'm not sure if you mentioned in your previous post but have you tried removing 2 sticks?
 
I have tried pulling sticks from slots 2 and 4 but that was with an already dubious OC so not sure if I'd set that up to fail from the start. My 3.2 failed overnight yesterday after around 14 hours so it turns out its not as stable as I hoped, which sucks but I then tried just changing fsb to 333 and vcore to my VID and it booted to windows and would run prime for around 10 minutes.

What do people think about me taking out 2 sticks again and starting from scratch, as I reckon if I can get it to boot and run a bit of prime with 3.0ghz @ 1.3250 and 4 dimms I won't be far off stable with 2 dimms. I could probably get some good results (3.2 under 1.4vcore) and maybe 3.4-3.6 at higher vcore, then I can chuck the other 2 sticks in and just up the NB voltage and DRAM voltage to stabilise a known good overclock with the extra sticks? Or is that unlikely to work?

I'm tempted to look for a cheap 2x2gb kit to help with the OC but if I can do it this way without buying anything I will, as it means I can look towards new case, mobo, CPU and RAM.
 
I would definitely try 2 sticks as it should be easier to get a higher clock. If you are however CPU limited and need to purchase a 2GB x 2GB kit anyway then way not *actively contribute* to the forum so that you have enough posts to use the members marker. There are loads of second hand sandybridge bundles coming up. With a 2500K you would go straight to 4.5Ghz no problems.
 
3.24 at 1.4 is what I'm using, it's 1.34 in windows though. I would keep trying to change the voltage and tweak the motherboard volts etc. I wouldn't jump into buying more ram because it may not be an issue. If it went 14 hours I'd say it's pretty close to being 100% stable. Maybe you just need a few more volt than me.
 
I though members market was 1k posts? I'll most likely get that in a few months purely because it's a good vibe here!

for 3.2 (400x8 which sometimes goes 399.8x8) I was at 1.43 and it still failed, albeit after 14-16 hours.

I'm planning a 3570k build which the new parts i'm getting are slowly going towards, but for some reason I really want to get past this problem because I know Q6600s, especially rev. G0 ones, were brilliant clockers and on a p5q deluxe I shouldn't be struggling this hard!

2 sticks and retry is going to be my method I think, when putting the other 2 sticks back are the only changes I'll need to make the DRAM voltage and the NB voltage, or are there other things affected?
 
sw1ft said:
I though members market was 1k posts? I'll most likely get that in a few months purely because it's a good vibe here!

for 3.2 (400x8 which sometimes goes 399.8x8) I was at 1.43 and it still failed, albeit after 14-16 hours.

I'm planning a 3570k build which the new parts i'm getting are slowly going towards, but for some reason I really want to get past this problem because I know Q6600s, especially rev. G0 ones, were brilliant clockers and on a p5q deluxe I shouldn't be struggling this hard!

2 sticks and retry is going to be my method I think, when putting the other 2 sticks back are the only changes I'll need to make the DRAM voltage and the NB voltage, or are there other things affected?
Click to expand...

It really is VID and cooling dependant tbh. The 3.6Ghz + chips had <1.25. vid. I'm assuming you are in EU so the summer heat will not be helping. What are your load temps?
 
You must log in or register to reply here.
Back
Top Bottom