C2D Overclocking Guide for Beginners

[Clunk]

Right, this makes more sense now.

Keep the same settings and change the ram voltage to 2.0v and try again.

Did you also disable legacy usb in the bios?

Also change the "write to precharge delay" to 15 (motherboard default should be 10) this is located in the memory subtimings.

[redsky009]

E6600 + Asus p5B deluxe wifi/ap + corsair twinX2048 800 c4 + Arctic cooler 7 pro

I made a moderate overclock to 2.7 back when I first got the PC and it has been running fine on that since then... after seeing this thread I decided to push it up a little

Got it to 3.6 (9 x 400 @ 1.4v RAM at 2.1v) but it wasn't very stable so I throttled back to 3.3 (9 x 366 @ 1.365v) and ran 1x Orthos on blend and and 1 on CPU stress for 14 hours - temps were:

idle:42 oC
full load: 60-63oC

The idle seems a bit high don't ya think?

Anyway, changed all voltages back to Auto (except the RAM) at 3.294 Ghz and have it running 2 x Orthos while I am at work..
will see how it fares temp wise, but idle was still around 40oC. Does that look like bad application of thermal paste to you or normal for a 33% overclock?

Also, Speedfan was recording an AUX temp of around 80-100oC most of the time - not sure what this is measuring, is this likely to be the the Northbridge heatpipe not seated properly?

Ta for the guide, very helpful

[Areku]

Originally Posted by Clunk

Right, this makes more sense now.

Keep the same settings and change the ram voltage to 2.0v and try again.

Did you also disable legacy usb in the bios?

Also change the "write to precharge delay" to 15 (motherboard default should be 10) this is located in the memory subtimings.

Ok so it ran all night with the ram settings at 5-5-5-15 1.95v. I cant remember if I set the Write to Precharge Delay to 15 or not I will have to double check, as well as disabling the legacy usb support.

Ill run it again tonight and see how it pans out. Hopefully find out whats going on. Will try out what you said Clunk, cheers.

[chuckskull]

Originally Posted by redsky009

The idle seems a bit high don't ya think?

Anyway, changed all voltages back to Auto (except the RAM) at 3.294 Ghz and have it running 2 x Orthos while I am at work..
will see how it fares temp wise, but idle was still around 40oC. Does that look like bad application of thermal paste to you or normal for a 33% overclock?

Also, Speedfan was recording an AUX temp of around 80-100oC most of the time - not sure what this is measuring, is this likely to be the the Northbridge heatpipe not seated properly?

Ta for the guide, very helpful

The idle temp will be quite high because the AC7 will slow it's fan down. mine idles just over 40 at 3.3. Unless the fan is going fast dont worry. it's fine.

Theres 2 possibilities for the AUX temp one being speedfan reading wrong (very common, my last CPU was constantly at -150c on air according to it) If the temp is correct and you're not crashing it's the PWM temp (They're usually rated to 110c, though you wouldn't want it that high 24/7).

Install ASUS probe(should be on the mobo CD) that should give you some better reading for the motherboard temps.

[redsky009]

Originally Posted by chuckskull

The idle temp will be quite high because the AC7 will slow it's fan down. mine idles just over 40 at 3.3. Unless the fan is going fast dont worry. it's fine.

Theres 2 possibilities for the AUX temp one being speedfan reading wrong (very common, my last CPU was constantly at -150c on air according to it) If the temp is correct and you're not crashing it's the PWM temp (They're usually rated to 110c, though you wouldn't want it that high 24/7).

Install ASUS probe(should be on the mobo CD) that should give you some better reading for the motherboard temps.

Ah thanks Chuckskull, good idea. Will see on the Asus site if they have Probe for vista yet.

Hopefully all will be well when I get back home tonight, then i can check if my SupCom Mark has improved

[Areku]

Ok so I had the machine running Memtest with the following:

RAM settings - 5-5-5-15, write to precharge delay 15
Volts - 2.0

Ran fine over night. So im guessing now I should overclock the cpu with the current ram settings and then run memtest again and go from there.

Nothing is ever simple is it

[Clunk]

Yep, go for it

[Clunk]

Did it work?

[Areku]

Well ran memtest with the cpu overclocked and the memory timings as said above. Was fine, no errors.

So I decided to run Orthos last night with the same settings. It got to about 35mins again and then stopped, tried again, same thing this time pc restarted, tried a third time and it gave me a similar error to what I had last time.

I was keeping an eye on the temps however and it seemed to go badly once it reached about 66c on Orthos speedfan and the other temp apps were reporting abour 50c.

So im going to have a play tonight to see if I can get it running cooler to see if that solves the problem, or at least takes care of the temps.

The fun ensues!

[C2Dan]

This is the most easiest guide I have read. A lot of other guides tend to put a lot of technical info in which isn't explained well, but this guide is very simple and straight to the point for a n00b to understand.

I have successfully overclocked my E6400 to 2.6GHz with a 325MHz FSB, I first went with the guide and set it to 300MHz.

At first I didn't get why what I set in the BIOS keep getting reset when I restarted the computer. I found the problem and I was pressing the wrong button when exiting the BIOS. I pressed ESC instead of F10, F10 saves changes whereas ESC discards changes

[Areku]

Ok so I have narrowed down the problem to cooling. I re-positioned my extra 80mm fan and made it blow directly onto the AC7 pro. This made the temps make its way to around 63c and no higher and Orthos ran for 50mins and then crashed.

Im going to see if I can grab some Arctic Silver 5 as suggested earlier. I had a look at the cpu cooler and it seemed to be snug on the cpu.

Does anyone have any other suggestions at which I can try?

[chuckskull]

Sometimes CPU coolers will look fine and you'll never find anything wrong with the way their mounted, but just wont cool as well as they should. 'Course re-mounting will fix that.

Only other thing that would cause temps to jump up would be setting the Vcore offset in the BIOS. It should be left at auto/disabled unless you have decet cooling(water).

[Clunk]

I wouldnt have thought 63c would be enought to make it unstable.

Did you take the NB heatsink off and check if its contacting? Did you update to the latest bios?

You need to do these things forst before going any further, or you are pissing in the wind

[chuckskull]

I'm pretty sure that 65c is the magic number, I've noticed mine will fail orthos the second it hits it. I played with some more powerful cooling and the cores were fine up to over 80c(couldn't get them hotter), but as soon as the CPU temp hit 65 it all went tits up.

Not sure if it triggers the throttling or what though. Something definetly happens.

[Clunk]

Nah, Ive had a c2d with 1.8v and 94c over the weekend with orthos, probe was reporting 105c 1.9v and it ran for a few hours and then switched off, ran for another few and then switched off again.

I think 75c is the point that throttling is supposed to start. I suppose, the cooler the better, but if the system has other problems then an overheating cpu isnt going to help.

What he needs is phase change cooling.

[chuckskull]

Originally Posted by Clunk

What he needs is phase change cooling.

Dont we all

Is that 95c read from the main CPU temp or a core reading?

If it isn't a function of the chip I wanna know why mine gives up at 65