The Official OC (OverClocking) Thread!

And aren't they also high?

 

well i guess i give up. i cant get it stable no matter what i try.
set it back to 3.7ghz and leave it alone. must not be ment for meto have any luck with this build. guess i need to go to AMD phenom and see what i can do.

 

No luck? You've got an extra 700mhz out of the CPU, which is far more than a Phenom will give you...

 

true but others are getting a whole gig more plus some.....shakes me head..

 

Silly high voltages, and watercooling, plus luck of the draw. My Q6600 can only make an extra 840mhz on a safe voltage, normally they make at least a gig no problem... I'm content though.

 

There is no chance that someone is running an E8600 stable at 4.5 and 5 GHz even with water cooling, in fact I don't think that a working stable system at 5 GHz is possible with Phase Change cooling. They clocked it, jammed it with as much voltage as they could muster, benched it, and then turn it down to prevent permanent damage to it. Note that none of the Benchmarks used lasted more than a few minutes. This is the kind of bull that RMA's are made of and newbies are left feeling inadequate because of.

 

I suspected as much, but I wasn't entirely sure. I know the E8600s are clocked higher to start with and have a good stepping, but Rob has got the sort of result I'd normally expect from an 8400, and the 8600s are only clocked 11% higher, so on a normal basis, while I'd expect to see 4ghz commonly from the 8600s, much higher than that and I'd start questioning the validity of the tests. 5Ghz I certainly think is beyond the reach of anyone on a daily basis.

 

The E8600 will hit 4 GHz easier with its higher multiplier, but as far as processors are concerned the E8400 and the E8600 are otherwise the exact same chip. That means that they both have relatively (not all chips are exact) the same ceiling. In the 2008 Tomshardware world overclock championships most teams maxed out at 5.4 and 5.6 GHz, but they weren't truly stable and the processors were chilled with liquid nitrogen. Most experts tend to agree that the maximum range for the Wolfdale core wit excellent cooling is about 4.2 GHz. Now if some members just want to try and see how far they can get it to snap a quick benchmark then I think that's great, but if they expect to keep it there and use it then I suspect disappointment won't be far behind.

 

right well then i hope to prove you all wrong, and as i have stated before i dont bench, i OC for my pleasure, if it fails, then its my loss not anyone elses, and i like to learn the hard way.

so long as i can get 4.7 24/7 use i will be content, but 5GHz is my aim.

to any newbie reading, please dont do what i do as you will not like the results when they turn bad. i am the Guinea-Pig, so just a warning to you lot. wait a month and a half, and let my chip die trying to get to 5GHz, not yours.

 

I just have these horrible visions about hearing a loud pop, your screen going blank and then seeing this:
http://homepage2.nifty.com/NewBeetle/P5K-E_Explosion.jpg

 

HAHAHAHAHAHAHAHAHAHAHAHAHA


thats amazing, but fortuantly ill be doing it on a 16phase P5Q deluxe, aka sophs one.

 

Ouch, not a pretty sight I don't expect. I was pretty mad about a cooked hard disk...

 

Spamual

Go for it! Clock it as high as you can get it. I still pop a few high overclocks every now and then, but once I'm done I turn it back down to be sensible and for stability. I always mention that it's probably not stable when I do it, but in the end I turn them all into work horses. Imagine doing a 16 hour encode and having it fail after 14 hours because of instability. When I encode Blu-Ray movies my CPU is at 98% to 100% all the way. In fact I might consider Blu-Ray encoding a new processor and memory stability test. That's also why I'm finally going to have to move to quad core.

 

Are you considering something like a 9650 soph? I would be interested to see what sort of speeds one of those could make 'genuinely stable' on water...

 

Yes, I'm just having a hard time trying to justify an extra $200 orver the Q9550 for a .5 gain in clock multiplier (117 MHz). It's tempting the daylights out of me because I feel confident that 3.7 to 3.8 GHz stable is a real possibility, and maybe even more with the new stepping. The Q9650 is exactly the same as the QX9650 but without the unlocked multiplier. I've been hoping that Nehalem will result in reduced prices but I suspect that won't happen any time soon.

 

Very understandable, after all I chose the cheapest quad of all because I couldn't justify the extra cost...

I'm really glad I got those expensive bits when I did, the dollar is rising fast, which will hurt PC parts.
At one stage, $200 would set you back about £97. It's been hovering around £102-£103 until the last month or so, it's now a whopping £114...

 

the most CPU intensive thing i do is play games, so if it can play cod4/do 3dmark vantage/06 (just for fragsoc ) them ill be happy with its stability.

 

speaking of stabilty. check this out.



with a little help from my friends...lol

the only thing i dont see in my bios is a place to adjust volts to fsb/nb

my idle temps before this OC(which was at 3.6ghz) was in the range of 39-42c now its 41-44c. which is great i think with the increase in volts.

in cpu-z it shows my volts as 1.392v. in everest and OCCT it is 1.36v. so anywhere in there im happy with and as high as im takeing it.

 

Good job!