E4300 overclocking

Taskman

Taskman

Taskman

Associate
Joined
3 Apr 2007
Posts
4
I have been looking in many threads read a lot of things that helped overclock.
I have not been into this for nearly 2 years so have lost loads of things in overclocking

Atm I have a
Asus P5N-E SLI with a E4300 on it. It was stable 3.0GHz (9x333) with a vcore at 1.325 with the +0.100mW setting on. I have a scythe infinity 1000 cooling it.
I had my vmem set to ~1,9
The memory is 2x 1GB Corsair XMS2 Xtreme PC6400 800Mhz 4-4-4-12 2T

I upped it to 3,2GHz (9*355). That was not stable so upped the Vcore and atm i'm up to 1,375 and still not stable.
After a few Orthos and OCCT tests it seems my ram fails. I ran OCCT first on the CPU then on the RAM and during the ramtest it crashed.

I upped the vmem to 2,1 then moved slowly higher and are almost at 2,2 now. It is not fully stable yet and I can't seem to figure out if it is really the mem or the CPU that is not stable. Even if it is stable at 3,0 i'm not happy ;)
 
Try memtest, if that fails the RAM wont be stable

But at 355FSB the RAM is still well below spec ( assuming youre running it 1:1 )

If theres an option to run it unlinked then do that and manually set it at DDR2 800
 
Yeah forgot to say. I run it unlinked so havn't touched the mem at all except from forcing it to 4-4-4-12 2T which is the spec for it.
And the NB is set to 1.209 atm

Gonna run memtest just thought it was something strange since I ran it unlinked. Can it be the mobo/cpu causing it to fail due to OC on it?
 
What are the stock volts for you memory? Id say it needs to be at 2.1v

Its probable that you are at the limit of your cpu, but 3Ghz is a mighty good clock really.

You could try pushing it further by giving it 1.4v and upping the chipset voltage slightly
 
I have seen people running it at around 3,4 and even 3,6 with same cooling as em. But then again it might be different stepping and all.

Actually I never thought of cheking the stock voltage of my mem. Will do and up the vcore to 1,4. If it still don't work will test and up the NB a bit otherwise I will just put it down to 3,0 which is still a very good clock yes.

And I am running at 3,2 atm so it works very well and even in most games just no 100% stable in OOCT.

Will post back later when I have done more tests.
 
You need to 'isolate' the CPU, the Memory and the chipset and test them seperately.

You should be able to test the memory by running it on an upward divider (i.e set the memory to run faster than the FSB). Once you have an idea what the memory can do, then you need to set it back to normal and start testing the FSB/chipset, to do this you need to lower the divider on the CPU (i.e x6 instead of x9) and then start cranking up the FSB. You already know what the memory can do so at this point the only thing that is holding you back is the chipset, run something like Orthos to find errors and add more NB voltage if it does.

Now you have an idea what your mem and FSB can do set the whole lot back to stock and then start testing the CPU.

I've tested one e4300 and it ws happy to run 3000MHz using stock 1.325vCore, and it went up to 3300MHz using 1.425vCore and 1.45NBVolts.
 
I didn't want to up my vcore and/or NB so much so i'm settled with 3.0GHz which is a good OC. Especially for being my first in about 2 years ;)
Just didn't seem to work at 3.2 no matter what I did. And I run the mem unlinked which should make it even easier i thought.

Running it now with a 1,325 vcore and the +0,100mV at 3.0GHz

Thanks for the help though guys.
 
You must log in or register to reply here.
Back
Top Bottom