The Official OC (OverClocking) Thread!

spamual,

Considering Anandtech did some comparing between the B3 stepping and the G0 very recently (Aug. 16th) and was unable to get to 3.6GHz, stable with either the G0 or B3 stepping examples they had, I would re-think that statement if I were you! It's also not very good advice for Newbies, either. You will almost always get better results by doing a proper burn-in for 12-24 hours before you start overclocking! Fast and dirty isn't the way to overclock a new computer, especially for a complete Newbie! Time and Patience are mandatory!

Respectfully,
Russ

 

Setting it to 3.6Ghz at first isn't the best idea, especially for a Quad. Although I don't know why anadtech had problems getting 3.6Ghz stable. Probably bad chips. I've gotten two 3.6Ghz Q6600 stable.

 

abuzar1,

It all comes down to the luck of the draw! There's no way to tell what you will be able to get out of one in advance! I doubt the Chips they used were hand picked by Intel, which is good as it better illustrates what to expect in the real world from a typical Q6600, regardless of the stepping!

Russ

 

Q6600 G0, currently passive cooled (I'm no idiot, it won't be that way for long, but I have no means of reattaching a fan at present)

3.15Ghz (started at that frequency, my board kept the old clock) - stable (10 minutes of Prime95, I know that's not very long)
3.24Ghz - currently at. Stable (3dmark testing only)
3.42Ghz - unstable (reboots during xp loading)
3.60Ghz - unstable (crashes at xp loading)

Using auto volts atm, setting them manually doesn't seem to make a blind bit of difference. Getting 3.24Ghz at 1.41V is promising though. 3dmark score is now 18,054.

 

russ i wasnt adressing any newbies, i was just saying what i would do. i like my clocking thick and fast, but thats me, not advising anyone, just stating what i would do.

btw, anyone used this yet?

supposed to be better then prime, and shows results in minutes.

http://www.xtremesystems.org/Forums/showthread.php?t=197835

 

man and i thought i was the only one that went to XSF's those guys aint no joke when it comes to OC'ing.

that program is supose to be great for stress testing. it can do in a few mins that prime95 and others takes hours....

cant wait to try it on my new CPU.

 

spamual,

Anything that will raise my temps 22C above what Prime95 shows, is not welcome on my computer. I already have problems with being able to see the need to overstress an already overclocked CPU with unrealisticly high loads! Folding doesn't put that much of a load on the CPU, yet is one of the best checks for stability there is! With that other program you linked, I would be well past the start of throttling the CPU (72C), so what good is it, or is smoke the determining factor? LOL!!

BTW, I understand what you are saying about "thick and Fast", but this is a technical forum where many Newbies come to learn. I know you weren't addressing newbies, the problem is the Newbies don't know that! They do tend to take what we say as Gospel! LOL!! I've told NuckNfuts than many times, for the same reasons! This is still a beginner's Thread!

Russ

 

btw where is NuckNFuts??? I've missed his posts and he's one of the few guys that is INSANE when it comes to OC'in!!!!

The program doesn't go that far above prime95. lol.

Remember this is a B3 Q6600 and I don't have the batch number because I didn't think about writing it down before the cooler install. I set the TJMax to 85*C if you couldn't add... lol. 3.0GHz OC.....





Quick program I'll say that... lol. Didn't even have time to consider a 'live' shot of realtemp but it shows the max values so it's all good.

.....gm

 

looks good GM... guess your ready to do some burning. dvd-rebuild useing Hanks encoder to let you use all 4 cores.....

burn baby burn!!!!! thats what Al Davis said oh thats "just win baby"

lmao

 

Interesting, I stopped my testing before the cores even reached those figures... I've no doubt they'd have gone higher though.

 

I haven't read enough about the progie Sam but I'll comment anyway. lol.

I like the idea of only testing for a few minutes and if Intel uses something similar or the same "method" to test then I'm in!!! I have NEVER liked the idea of testing for HOURS upon HOURS with prime95, orthos, or what ever "burn in" test you wanna choose. Granted my temps got about 5C-10C higher with that test but it was for only a few minutes. I just like the idea better. Call me simple, it's ok.

I guess I equate it to a car's redline. I you run a car just BELOW redline for hours upon hours it's gonna take it's toll!! BUT if you run it right at redline for a few seconds it's prolly gonna be ok. IF not you know right away... lol. This is the same thing to me. The test spits out in a few minutes and let's you know if it's ok or not. btw mine passed with flying colors!! Yeah!!

....gm

 

So does the test have a finishing point, or does it just tell you to stop it after so many minutes?

 

Mine quit in about 5 minutes I think, IIRC the readme file said it should be done in 8 minutes or less. I didn't time it and all I was watching was the temps. lol. I was concerned as well, but it seemed to do just fine.

btw what do you have your OC at?? Curious about your voltages on the cpu and chipsets as well?

....gm

 

3.24Ghz (360x9) all auto, 1.408V. Kind of irrelevant though, as 5 seconds into that test I got this:
http://support.microsoft.com/kb/329284

 

Sammy back down your OC and see if it does it again. I would start at 3.0GHz and see what happens. Are you testing 1/2 RAM or FULL RAM?? I would do at least 1/2 to start with. Take into account I've only used the progie twice.

Maybe the link that spamual provided will assist you in answering some questions.

....gm

btw anyone heard from Morty81 lately?? Guess he's resting. hmmmmmm. SLACKER!!!!! lol.

 

lol Sam, I think you should lower that OC until you can get a fan into that heatsink of yours.

 

1/2 RAM, same as you.
Not heat related, the cores reached 62C highest before the BSOD sppeared, and according to Realtemp my TJMax is 95. I think it's just a case of one of those 'stable' but not really stable overclocks...

 

No, I mean you will have to up the voltage for it to get stable. And when you do that the temps will go higher.

 

Yep... that's why I suggested lowering the OC for now. Takes less voltage and should get you to stable. btw there isn't a stable "unstable" OC!!! lol. You should know better.

....gm

 

hence the inverted commas, lol. It gives the appearance of being stable because it has never crashed when I've been using it normally, but it fails a burn test, so it's not truly stable.
As for the voltage required, potentially, but my board gives the chip the same voltage on auto for any speed, seemingly.