• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Need help oc'ing e8400

S

s75h

n00b
Joined
Jun 3, 2006
Messages
21
From everything I've heard about the e8400's is that they oc well but I can't get past 3.45 booted and that is with a fair amount of adding voltages (see below). Am I doing something wrong or could it be a bad cpu to oc or could one of my other parts be bad?

One unusally thing is that the cpu temp never goes up or down from idle that shows 43 C. I'm using CPUID hardware monitor.

my specs:
e8400
evga 750i mobo
2 x 2g g.skill 1000
corsair tx750 ps

The best I've been able to even get it to post was using the following settings:

linked and sync
fsb at 1535
mem at 767

voltages:
cpu core 1.35
cpu fsb 1.3
mem 1.9
spp 1.4
nf200 1.2

memory settings 5-5-5-15-2T

Any help would be appreciated.
 
For memory that is pretty low voltage. My Gskill 1066 (same as yours) needed at least 2.0 before i could break theexact SAME barrier as you. The manfact specs on those are 2.0-2.1 so try adding some more voltage. Of course every time i tried to run Prime 95 i would fail instantly no matter my voltage so i knew something was up. Once i added some voltage to my memory and changed it to 2T i was able to OC to 3.8GHz on my 780i. Your 750i should be able to do this easily.

Download realtemp from the link below for temp monitoring: That is unusual but try this program
http://www.techpowerup.com/realtemp/

Your situation seems to be the same as mine before i added voltage and changed the clockrate on my memory. Hope this helps! :)
 
Thanks for the reply. I tried raising the voltage to the 2.0 to 2.1 range in my early testing but had no luck up to a certain point. I then read a post from a guy who said he had to lower his voltage for his memory to get better overclocking. So I tried that and it actually did allow me overclock the fsb a little more.

Since my last post I've run into a new problem. I lowered my system back down to. 3.3 and upped the memory to 1000 mhz and things were working okay as I got into windows to check on some things. I tried prime and it locked up immediately and now I'm getting the following error message:

"A Disk Read Error Occurted. Press CTRL+ALT+DEL to restart"

I've booted back into the default bios settings but still getting this message. Did I just mess up my hard drive?
 
I would try making sure that all the correct boot orders are set in the bios. Trying making the HD the primary boot drive, then put your HD there. Make sure all ur connections are tight. I dont know much about this error, but it seems like it cant recognize the boot sector so something could be messed up in the boot order.

When does this error occur exactly?
 
It does it right after it does the IDE search.

I am running the repair utility on the vista dvd as I type this. Hopefully that does the trick. I would hate to have to reinstall vista.
 
The repair utility couldn't fix the error that came up so I reinstalled vista. Everything is back up and running so now I'm back to trying to figure out my overclock problem. Any suggestions?
 
Ahaha, you're having the same issue I did. Unlinking fixed it for me.

we almost have the same setup too, only difference is my ram is 1066 instead of 1000 :D
 
I tried realtemp and it shows cpu temp at 33 C for min and max, and that's even if I start up a game.
 
I started from scratch and I keep getting to a wall at 9 x 375 regardless of what I try for voltages. Prime95 locks up in a few seconds.

I also tried MemTest and if I got error after error when I left everything at default and just set the ram to run at 1000 mhz with the mem voltage set at 2.0. I bumped it up to 2.1 and it wouldn't even boot into windows and I had one bsod. I have MemTest running with it right now at 1.9 and so far so good. Could I have some bad ram? Both sticks gave error after error at 2.0.
 
Sounds like the temp sensor is stuck, not sure if there's a fix for it.

Is the ram actually getting errors in memtest or is it just crashing? If it's not actually saying it has errors, just turn down the voltage a bit. In all honesty, I'd set the ram down to 800 while you work on the cpu, then work the ram back up to 1:1.

btw, are you running unlinked now?
 
unlink the ram before OC'ing the cpu. keep the speed and timing at stock. once you get to a satisfactory and stable OC for your cpu, then tweak your ram. you want to eliminate any ram instability issues while you are OC'ing the cpu.
 
This last time trying to oc from scratch I unlinked and left the ram at 800 and went up in 10 increments for the fsb. After hitting the wall at 375 I decided to test out just the ram by setting it at 1000 mhz and everything else at the default voltages except for memory at 2.0 and 2.1 while running memtest.

At 2.0 it was error after error with both sticks. I'd say an error occurred ever second or less. At 2.1 I got the bsod and never could get into windows. This ram is rated for those voltages. At the default 1.9 I see no errors. I guess I should RMA this ram.

Before going to bed last night I left prime95 run and I saw no errors. I also left realtemp on to see if my cpu would ever move and it actually did increase by 1 C for one of the cores. The max hit 34 C approximately 5 hours into the test. I'm using the Thermalright IFX-14 Heatsink without the back side heatsink.
 
Well, I sent back the ram thinking it might be defective because it won't boot run okay with the ram's voltage settings, but I'm still seeing the same problem with the new ram. It won't boot all the way to windows if I go up with the voltage.

Does this sound like a mobo problem?
 
WiLLiSTER said:
Sounds like the temp sensor is stuck, not sure if there's a fix for it.

Is the ram actually getting errors in memtest or is it just crashing? If it's not actually saying it has errors, just turn down the voltage a bit. In all honesty, I'd set the ram down to 800 while you work on the cpu, then work the ram back up to 1:1.

btw, are you running unlinked now?
Click to expand...

I'm not quite clear on how the OP is describing it - is he actually running 1:1? Or is he running the RAM faster than FSB? For RAM rated for 1000Mhz, assuming his CPU was up for it, 500MHz FSB would be no problem.

Because running it at 1:1, at only 375 FSB, is 750Mhz - which well below the rated 1000Mhz.

*Doh - the OP is already running at CAS5...
 
I've tried running it linked and unlinked but keep hitting a wall at 9 x 375, even when unlink and set the ram to stay at 800 mhz.
 
Here's what I notice immediately:

Your mobo is rated to run at 1333 MHz FSB, you are currently pushing it beyond that. You may need to up the voltage and cooling on the FSB to continue OCing the FSB.

Other tips:

Try more temp programs. Not all programs are calibrated for every CPU properly.

Grab a voltage monitoring program.

Vdroop. Is the voltage for your CPU changing when you put your CPU under full load? That can indicate faulty mobo.
 
I'\ve been using realtemp, cpuz and hwmonitor.

My evga 750i mobo came with a small fan to connect to the fsb and I've done that. I've upped the voltage to 1.3v for the cpu fsb. I'm not sure how high I can go but from reading other posts from folks with similar mobo's and cpu's this seems to be the norm. Many don't even have to got that high. This mobo was reviewed on Hardocp and overclocked well.

I'll check on the cpu voltage to see if its moving when under full load and report back.
 
While running under full load the cpu voltage was at 1.18 most of the time with the occasional 1.19 here and there.
 
You must log in or register to reply here.
Back
Top