My Q6600 with vid of 1.325....

M

MaStaFLaSH

2[H]4U
Joined
Sep 11, 2007
Messages
2,224
is stable at 3.6GHz :D

Q6600 9 X 400 1.504v under load, temps max out at 73 but usually stay around 65c or less :eek:
OCZ Vedetta 2
P5Q Pro Bios 1306 (non-modded)



Sorry no better screenshot atm but it is 24 hours primestable :D

So even if you get a vid of 1.325 there is hope for you, my last Q6600 (vid 1.3) took 1.48v to reach 3.6GHz (450 X 8) but ran hotter (Tuniq Tower btw), it was pretty much 72-73c all the time. This one will get up to 73c but usually the hottes two cores are in the low 60's :D
 
Sweet!

Mine is running at 3.2 stable at the moment and loads at around 55c after 2 hours of Prime95. I just built this system, so I just left everything on auto and kicked the fsb up to 356. When I get a chance, I think I'll try to reach 3.6 on my Q6600.

My VID is 1.3125v, so do you reckon my chances of getting 3.6 are pretty good too? I'm using a Zalman 9700NT right now, so do you think it would be adequate?

I guess I'll probably have to find out most of this stuff for myself, but at least now I know that it's not entirely impossible to hit 3.6.
 
I am in the same boat right now with a VID of 1.325V... I finally had my computer running so I was hoping for something lower and I got pissed when I saw the number....Let's hope I can real 3.6GHz at 1.45V !!!
 
I tried for 3.6 on my 1.325... Put 1.6v into it and it wasn't stable @ 3.6. You're lucky...
 
I'm running Prime95 at 3.6ghz right now...

1.5v, 1:1 ram ratio for 1066, and 6-6-6-16 timings

15mins into stress test...so far so good

All four cores are around 63c right now...


Wish me luck...

Edit:

Well, no BSOD...but a hard lock... poo...


Going for 3.5ghz...

Stable for 30min, gonna try some suggested settings and try 3.6 again...
 
CompuG##K said:
I'm running Prime95 at 3.6ghz right now...

1.5v, 1:1 ram ratio for 1066, and 6-6-6-16 timings

15mins into stress test...so far so good

All four cores are around 63c right now...


Wish me luck...

Edit:

Well, no BSOD...but a hard lock... poo...


Going for 3.5ghz...

Stable for 30min, gonna try some suggested settings and try 3.6 again...
Click to expand...


Your ram ratio is wrong, at 3.6 (400 X 9) it should just be PC 6400 aka 800. What motherboard are you using?

With my P5Q Pro literally all the settings I changed was mem timings, vcore, vdimm, GTL ref and load line calibration. everything else (or almost everything else) on auto and FSB at 400. If I changed other settings it'd get unstable, now that I have it stable at 3.6 I will probably tweak it further so i can get the stuff off auto (I'm a bit worried about over volting).
 
I'm on a Rampage Formula. Maybe I read wrong when I was looking at the divider, but I was sure the divider said 1:1 in cpu-z whenever I had it running at 3.6. Would that be correct for DDR2 1066?


I got prime95 stable for an hour at 3.6, closed out of that and went to play some Crysis. Everything ran fine there, but I noticed one thing when I rebooted and went into the bios. My NB was running hot at 64c. I was using the settings that another guy used to get to 3.6 on his Q6600 with the same board, and he had his NB voltage cranked up a good bit, so it's no surprise that it was running hot I guess... I'm still pretty much a noob at this stuff, so I'm guessing I would get better results if I actually knew what I was doing :D
 
Bah... My G0 Q6600 (1.2VID) hits 4.1 on a friend's X48 board, but alas she's stuck at 3.4 in this p5b-deluxe.

Grats OP :)
 
CompuG##K said:
I'm on a Rampage Formula. Maybe I read wrong when I was looking at the divider, but I was sure the divider said 1:1 in cpu-z whenever I had it running at 3.6. Would that be correct for DDR2 1066?
Click to expand...


No, if you're running your ram 1:1 then for your ram to be at 1066 your FSB would have to be at 533 :eek:
Remember your FSB get's multipled by four and your ram gets mutiplied by two. FSB of 1066 is really 266 X 4, thus for ram to run at 1:1 you'd only need DDR2 533, 266 X 2. If you're at 3.6 (400 X 9) then your FSB is 400 X 4, so if you're at 1:1 then the memory would be running at 400 X 2 aka DDR2 800.
 
CompuG##K said:
I'm on a Rampage Formula. Maybe I read wrong when I was looking at the divider, but I was sure the divider said 1:1 in cpu-z whenever I had it running at 3.6. Would that be correct for DDR2 1066?


I got prime95 stable for an hour at 3.6, closed out of that and went to play some Crysis. Everything ran fine there, but I noticed one thing when I rebooted and went into the bios. My NB was running hot at 64c. I was using the settings that another guy used to get to 3.6 on his Q6600 with the same board, and he had his NB voltage cranked up a good bit, so it's no surprise that it was running hot I guess... I'm still pretty much a noob at this stuff, so I'm guessing I would get better results if I actually knew what I was doing :D
Click to expand...

Make sure your RAM voltage is set to the top end of the manufacturer's recommended voltage range. If your motherboard has the RAM voltage set to auto, it's likely undervolting your RAM, causing instabilities.
 
Ah, you are right... It was 3:4 ratio that it was using.

Memory is running at factory spec 2.1v at 1066mhz, cpu is running 3.6 with 1.472v, nb is running 1.5v...

Been priming for 30min now and it's stable so far... but I know that's no indication that it won't bsod after a few hours... so I hope for the best on this one...

Cpu is hitting 70c under full load, and I could probably turn up my case fans to full power to help the temps out..


Is 60c too hot for the NB? So far the highest it has hit is 60 under load, I'm just wondering if that is within it's tolerance...
 
60c for the NB is fine. You might want to try setting some voltages on auto, literally the only voltage settings I touched were vdimm, vcore and GTL ref. Or if it becomes unstable you might just need to bump up the vcore a notch or two. I've found with my overclocks that if I can boot into windows and be prime stable for around 30min and then it BSODs or fails or something usually it only needs one little bump up on the vcore then it's rock stable.
 
Ok, I took the nb voltage down to 1.45v, and the fsbt voltage down to 1.36. (Auto seemed to have been overvolting them quite a lot). Left everything the same. Currently my GTL. ref is on auto. What should I be aiming for on the GTL. ref for a 3.6ghz oc?

Anyways, I Primed for 2 hours with no crash. Max temps hit 69c on the cpu and 56c on the nb. Played Crysis for 2 hours with no crash. Gonna prime some more tomorrow when I get time.


One thing I noticed though, was when I tried Intel Burn test. Now I know that it's suppose to really stress the cores hard, and run the temps up higher than anything, so I was expecting higher temps, but not as high as they got. The temps hit the upper 80's before I called it quits on that.

Now I've seen some people that are able to run Burn Test without getting such high temps. Maybe they have better cooling than me? I don't know how a Zalman 9700NT could not cut it, but it seems that way. I applied the thermal paste as per the AS5 instructions for a quad core, so I know that it couldn't be that it wasn't installed correctly.

Should I not worry about the high temps from Burn test since normal usage won't ever stress the cores that hard?



Anyways, thanks for the help. Each time I read one of your posts, I learn a little more about overclocking. :)
 
CompuG##K said:
Ok, I took the nb voltage down to 1.45v, and the fsbt voltage down to 1.36. (Auto seemed to have been overvolting them quite a lot). Left everything the same. Currently my GTL. ref is on auto. What should I be aiming for on the GTL. ref for a 3.6ghz oc?

Anyways, I Primed for 2 hours with no crash. Max temps hit 69c on the cpu and 56c on the nb. Played Crysis for 2 hours with no crash. Gonna prime some more tomorrow when I get time.


One thing I noticed though, was when I tried Intel Burn test. Now I know that it's suppose to really stress the cores hard, and run the temps up higher than anything, so I was expecting higher temps, but not as high as they got. The temps hit the upper 80's before I called it quits on that.

Now I've seen some people that are able to run Burn Test without getting such high temps. Maybe they have better cooling than me? I don't know how a Zalman 9700NT could not cut it, but it seems that way. I applied the thermal paste as per the AS5 instructions for a quad core, so I know that it couldn't be that it wasn't installed correctly.

Should I not worry about the high temps from Burn test since normal usage won't ever stress the cores that hard?



Anyways, thanks for the help. Each time I read one of your posts, I learn a little more about overclocking. :)
Click to expand...

CPU GTL ref voltages vary per each individual chip, so we can't tell you what value to use to make your OC work better. Only thing I can say is set the CPU GTL to .630x to start and test it, then move it 1 notch up if you still get stress test failures. If you're not getting stress test failures, don't worry about the GTL's.

As far as IBT goes, yes it will heat up your cores more than anything else you could ever possibly conceive, short of a blow torch. No application will ever heat it up that much, so if Prime/Orthos leave your temps at a reasonable number, don't worry about IBT. I wouldn't recommend using IBT if it gets into the 70's/80's. Just stick to Prime/Orthos.

With that said, you shouldn't be gettings temps that high, anyway, unless your ambient temps are in the 90's all the time. I'd recommend checking to see if your AS5 application went well by removing the heatsink and checking the coverage area/thickness (too much AS5 is as bad, or even worse, than not enough). If it looks ok, I'd consider getting a new cooler altogether.
 
Mine can boot at 3.6GHz with 1.325V(BIOS) but at load it will drop to around 1.27V (CPU-Z). I've never pushed the voltage in the BIOS to more than 1.35V so I can't get it stable under load at 3.6GHz. With 1.325V I can get it to 3.2GHz stable 24/7.
 
jimbo6846 said:
CPU GTL ref voltages vary per each individual chip, so we can't tell you what value to use to make your OC work better. Only thing I can say is set the CPU GTL to .630x to start and test it, then move it 1 notch up if you still get stress test failures. If you're not getting stress test failures, don't worry about the GTL's.

As far as IBT goes, yes it will heat up your cores more than anything else you could ever possibly conceive, short of a blow torch. No application will ever heat it up that much, so if Prime/Orthos leave your temps at a reasonable number, don't worry about IBT. I wouldn't recommend using IBT if it gets into the 70's/80's. Just stick to Prime/Orthos.

With that said, you shouldn't be gettings temps that high, anyway, unless your ambient temps are in the 90's all the time. I'd recommend checking to see if your AS5 application went well by removing the heatsink and checking the coverage area/thickness (too much AS5 is as bad, or even worse, than not enough). If it looks ok, I'd consider getting a new cooler altogether.
Click to expand...

Well, I reapplied the thermal compound to the exact AS5 spec for quad cores, and my temps weren't affected. Still 100% loading at around 69c in prime95. So I guess this is as good as my 9700NT can do with this chip.

I think I'll play it safe for now and just run the chip at 3.2ghz. I don't want to risk stressing the chip/board too hard before I get a Q9650 in the coming months. I just bought a Q6600 to have for the time being with this system until I got enough cash saved up for a Q9650. Would it be true that hitting 4ghz on a Q9650 would be less stressful on the motherboard than trying to hit 3.6ghz with a Q6600? My reason being that maybe a 1.2ghz oc on a 65nm chip is more stressful than a 1ghz oc with a 45nm chip...
 
CompuG##K said:
Well, I reapplied the thermal compound to the exact AS5 spec for quad cores, and my temps weren't affected. Still 100% loading at around 69c in prime95. So I guess this is as good as my 9700NT can do with this chip.

I think I'll play it safe for now and just run the chip at 3.2ghz. I don't want to risk stressing the chip/board too hard before I get a Q9650 in the coming months. I just bought a Q6600 to have for the time being with this system until I got enough cash saved up for a Q9650. Would it be true that hitting 4ghz on a Q9650 would be less stressful on the motherboard than trying to hit 3.6ghz with a Q6600? My reason being that maybe a 1.2ghz oc on a 65nm chip is more stressful than a 1ghz oc with a 45nm chip...
Click to expand...

It will be more stressful on the board itself to run at 444 FSB (4.0Ghz on 9650) than to run at 400 FSB (3.6Ghz on 6600). The difference in stress is nearly none, though. 444 FSB for your board will be a walk in the park. You will have an easier time clocking a 9650 to 4.0Ghz, I'd think, then a 6600 to 3.6Ghz.
 
CompuG##K said:
Well, I reapplied the thermal compound to the exact AS5 spec for quad cores, and my temps weren't affected. Still 100% loading at around 69c in prime95. So I guess this is as good as my 9700NT can do with this chip.

I think I'll play it safe for now and just run the chip at 3.2ghz. I don't want to risk stressing the chip/board too hard before I get a Q9650 in the coming months. I just bought a Q6600 to have for the time being with this system until I got enough cash saved up for a Q9650. Would it be true that hitting 4ghz on a Q9650 would be less stressful on the motherboard than trying to hit 3.6ghz with a Q6600? My reason being that maybe a 1.2ghz oc on a 65nm chip is more stressful than a 1ghz oc with a 45nm chip...
Click to expand...

The 9700NT isn't that great of a cooler, looks cool but not that great performance wise.

One thing you could try is running 450 X 8, for whatever reason my last Q6600 didn't want to do 9 X 400 but was fine at 450 X 8 @ 1.48v.
 
jimbo6846 said:
It will be more stressful on the board itself to run at 444 FSB (4.0Ghz on 9650) than to run at 400 FSB (3.6Ghz on 6600). The difference in stress is nearly none, though. 444 FSB for your board will be a walk in the park. You will have an easier time clocking a 9650 to 4.0Ghz, I'd think, then a 6600 to 3.6Ghz.
Click to expand...

That's what I was thinking. I've seen a lot of reports of Q9650s hitting 4ghz with ease. I really want to hit the 4ghz mark to really make sure my 4870x2 isn't very cpu limited at 1600x1200 on any of my games, most notably Crysis. Going from 3.2 to 3.6 really helped the framerate (much less dips into the 20's on Very High) and I'd imagine that 3.6 to 4ghz would help even more.

What kind of cooler would you recommend for a Q6600/Q9650 Mastaflash?

I'll try 450x8 when I get home and see how that does. I haven't really had any time to do any long period prime95 tests, so I really don't even know if the settings I'm at right now are completely stable.
 
CompuG##K said:
That's what I was thinking. I've seen a lot of reports of Q9650s hitting 4ghz with ease. I really want to hit the 4ghz mark to really make sure my 4870x2 isn't very cpu limited at 1600x1200 on any of my games, most notably Crysis. Going from 3.2 to 3.6 really helped the framerate (much less dips into the 20's on Very High) and I'd imagine that 3.6 to 4ghz would help even more.

What kind of cooler would you recommend for a Q6600/Q9650 Mastaflash?

I'll try 450x8 when I get home and see how that does. I haven't really had any time to do any long period prime95 tests, so I really don't even know if the settings I'm at right now are completely stable.
Click to expand...

S1283 or one if it's many rebranded (kingwin, OCZ, etc.) variants. What I'm using right now is an OCZ Vendetta 2 which is basically a slightly modified rebranded Xigmatech S1283. My OCZ Vendetta 2 keeps temps around 65c @ 3.6GHz 1.5v :D

I'd love to be able to hit 4.0GHz but Q6600's max out around 3.6GHz and to get a CPU that'll hit 4.0GHz you really have to shell out some cash whereas you can find a Q6600 for around $180 or less. I'm sure the extra 400MHz boost would help some but to me it's just not worth it price to performance wise.
 
MaStaFLaSH said:
S1283 or one if it's many rebranded (kingwin, OCZ, etc.) variants. What I'm using right now is an OCZ Vendetta 2 which is basically a slightly modified rebranded Xigmatech S1283. My OCZ Vendetta 2 keeps temps around 65c @ 3.6GHz 1.5v :D

I'd love to be able to hit 4.0GHz but Q6600's max out around 3.6GHz and to get a CPU that'll hit 4.0GHz you really have to shell out some cash whereas you can find a Q6600 for around $180 or less. I'm sure the extra 400MHz boost would help some but to me it's just not worth it price to performance wise.
Click to expand...

Which is why I hope the Q9650s will drop at least some when the Nehalems come out. :D
If not, oh well. I'll scrape up enough to get one anyway. This is likely to be the last system I build for a while, so I want to make sure it's as future proof as I can get it.

I wonder how that new Thermaltake V14 would do...
 
Bump, sorry for double posting.

Ok, I just primed for an hour at 3.6 before bsod, with these settings:

Ai Overclock Tuner : Manual
OC From CPU Level Up : Auto
CPU Ratio Setting : 9
FSB Frequency : 400
FSB Strap to North Bridge : 400
PCI-E Frequency: 100
DRAM Frequency: DDR2-1066
DRAM Command Rate : 2N
DRAM Timing Control: Auto
CAS# Latency : 5
RAS# to CAS# Delay : 5
RAS# Precharge : 5
RAS# ActivateTime : 15
RAS# to RAS# Delay :
REF Cycle Time :
Write Recovery Time :
Read to PRE Time :

Read to Write Delay (S/D) : Auto
Write to Read Delay (S) : Auto
Write to Read Delay (D) : Auto
Read to Read Delay (S) : Auto
Read to Read Delay (D) : Auto
Write to Write Delay (S) : Auto
Write to Write Delay (D) : Auto

Write to PRE Delay : Auto
Read to PRE Delay : Auto
PRE to PRE Delay : Auto
All PRE to Act Delay : Auto
All PRE to REF Delay : Auto

DRAM Static Read Control: Disabled
AI Clock Twister : Auto
AI Transaction Booster : Manual
Common performance level : 07
Pull-In of CHA PH1 - Enabled
Pull-In of CHA PH2 - Enabled
Pull-In of CHA PH3 - Enabled
Pull-In of CHA PH4 - Enabled
Pull-In of CHA PH5 - Enabled
Pull-In of CHB PH1 - Enabled
Pull-In of CHB PH2 - Enabled
Pull-In of CHB PH3 - Enabled
Pull-In of CHB PH4 - Enabled
Pull-In of CHB PH5 - Enabled

CPU Voltage : 1.46875v
CPU PLL Voltage : 1.50v
North Bridge Voltage : 1.25v
DRAM Voltage : 2.0v
FSB Termination Voltage : 1.20v
South Bridge Voltage : 1.050v
South Bridge 1.5 Voltage : 1.50v
Loadline Calibration : Enabled
CPU GTL Reference : .63x
North Bridge GTL Reference : .67x
DDR2 Controller REF Voltage : Auto
DDR2 Channel A REF Voltage : Auto
DDR2 Channel B REF Voltage : Auto

CPU Led Selection : CPU Volt
North Bridge Led Selection : NB Volt
South Bridge Led Selection : SB Volt
Voltiminder Led : Disabled

CPU Spread Spectrum : Disabled
PCIE Spread Spectrum : Disabled


These are the exact settings that another person from another forum used to get 3.6 stable at 24/7 on a Rampage Formula. Now I know that all systems are different, so that's why I've come here again. Since I got an hour of prime stable with these settings, with reasonable temps, I must be pretty close. Like you said Mastaflash, a small voltage bump might do the trick. So, with that said, using these settings as a base, what voltage(s) do you think I would need to bump up a bit?
 
Well, I bumped the vcore up to 1.4725 and I played Crysis for a solid 3 hours with no issues whatsoever. Difference between 3.2 and 3.6 is pretty huge. Now the fps hardly ever dips into the 20s where it did quite frequently before. It must be pretty close to stable or it would have crapped out on me in the middle of my play session and net browsing. Of course I won't know for sure until the cpu has been heavily stressed, but I haven't had a chance to prime it yet, though. Don't really want to prime it until I get a better cooler. Crysis only stressed it to 59c where prime got it up to 73c during the last run I did. All the other temps seem fine though. The NB got up to a max of 55c with 1.25v, which is a far cry from the 66c it was hitting at 1.45v.
 
73C won't hurt your CPU. It's hot, to be sure, but your CPU can handle that temp. A better cooler certainly would help keep the CPU's lifespan high, though.
 
73c is fine but is about the max I'd be comfortable with. Remember that while it gets up to 73c during prime it's probably not going to even get close to that hot during normal usage, you're probably never going to put near as much stress on it as prime does.
 
You must log in or register to reply here.
Back
Top