Overclock.net banner

E8400 3.0Ghz to 4 Ghz Project

2507 44
Hey all!
I have an Intel e8400 3.0 Ghz
4 Gb G skill 1066 Mhz
Asus P5Q Pro Motherboard

I successfully overclocked to 3.2 Ghz with these settings:

Multiplier - 9
FSB Frequency - 356
FSB Strap to northbridge - Auto
DRAM Frequency - 1140
DRAM timing control - auto
5-5-5-15-3-52-6-3
8-3-5-4-6-4-6
14-5-1-6-6
CPU Voltage - 1.3
North Bridge and South Bridge Voltage - Auto

I WANT 4 Ghz!...Is it possible. My Cpu is at 19c at idle.
1 - 20 of 45 Posts

·
Registered
Joined
·
673 Posts
Yes, not only is it possible, its easy.
1) Put your CPU voltage to 1.4v in your bios (1.36ish after vdrop I find).
2) Put your northbridge voltage to 1.4v
3) Raise your fsb up to 445 and see if it boots. If it does, run orthos overnight. If it doesnt, post back here and i'll see what I can do to get you stable.
 

·
Registered
Joined
·
668 Posts
yeah try to keep your RAM at 1066mhz for now. Also what exact RAM sticks do you have and what are their timings?

note: download cpuz and figure out what revision/stepping you have and also use realtemp/coretemp to figure out your VID. That will help a little.
 

·
Registered
Joined
·
1,381 Posts
Not all E8400's can easily crack 4.0GHz. Mine won't go higher than 3.825 at 1.4v in BIOS (1.36v after vdrop). Right now mine is at 3.6GHz with 1.3125v in BIOS, and +0.1v on both the northbridge and FSB (these voltage settings may look different in your BIOS).

That aside, 3.6GHz is a good starting point with 400x9. And people are going to tell you otherwise, but I only run orthos for about 10 minutes on small FFTs, if it fails in only 10 minutes, then I up the voltage 2 notches and test again. Most of the time my orthos fails, it fails in the first 3 minutes. Once I feel I'm at a stable point, I run it for a good 2-3 hours and I'm satisfied. 10 hours is overkill, I mean when are you going to run your PC at full load for 10 hours in the real world?

Now if you fold, thats another story entirely and would suggest a longer stress test to be extra safe (although judging from temps, small FFTs heat up waaay more than folding even though they both use 100% CPU).
 

·
Registered
Joined
·
69 Posts
Discussion Starter · #11 ·
It restarted and the monitor didn't recognize a signal. I pushed the power button to restart the computer and changed everything back to my first setting on top of this page.
I'll check my ram divider. I'm not well versed in RAM timings and loosing or tightening them up.
 

·
Registered
Joined
·
673 Posts
Try those settings I mentioned again, but drop your ram speed and loosen your timings by one. This will mean we are only OCing your CPU, not your ram by mistake.
 

·
Registered
Joined
·
447 Posts
Quote:

Originally Posted by Jacko87 View Post
Not all E8400's can easily crack 4.0GHz. Mine won't go higher than 3.825 at 1.4v in BIOS (1.36v after vdrop). Right now mine is at 3.6GHz with 1.3125v in BIOS, and +0.1v on both the northbridge and FSB (these voltage settings may look different in your BIOS).

That aside, 3.6GHz is a good starting point with 400x9. And people are going to tell you otherwise, but I only run orthos for about 10 minutes on small FFTs, if it fails in only 10 minutes, then I up the voltage 2 notches and test again. Most of the time my orthos fails, it fails in the first 3 minutes. Once I feel I'm at a stable point, I run it for a good 2-3 hours and I'm satisfied. 10 hours is overkill, I mean when are you going to run your PC at full load for 10 hours in the real world?

Now if you fold, thats another story entirely and would suggest a longer stress test to be extra safe (although judging from temps, small FFTs heat up waaay more than folding even though they both use 100% CPU).
That stinks. I can run mine stable 4.2 at 1.36 vcore. Gotta love the E0 stepping.
 
1 - 20 of 45 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top