e8400 Overclocked For First Time on Gigabyte MB, Failed, Help

a quick question - what should I put under the CPU block - I read somewhere that you shouldn't use silver AC5.. only ceramic paste - is that right?
 
phinix, do you have the exact same hardware as the OP? if not, please start your own thread. To answer your question, you want to use either AS5 (arctic silver 5) or some other thermal compound.

@OP: Many people OC differently, and test settings differently. Personally, I don't typically test longer than 20min if I'm not at the speed I want. If I wanted 4Ghz, and was at 3.6Ghz, I'd stress it for about 20min, and if it passed, I'd move up to 3.8Ghz, get that stable for a 20min Orthos run, then try for 4Ghz. Once I reach the speed I want, I try all kinds of settings to get it stable. The higher you go, the more voltages you'll most likely need to raise (MCH/FSB/CPU). Change one thing at a time, and write down settings that do and don't work. Read BillParish's guide for more help with your BIOS options. The BIOS Profiles will help a lot.
 
Looking at your screenshot, your only pushing 1.216 volts to the CPU (CPU-z). You'll almost certainly need to push that up that to get to 4.0 - I've had good luck with mine, and I needed 1.28-1.26 for 4.0.
 
Very odd. I lowered the voltage back down to one step above stock which is like 1.25625 and now its working fine at 3600? Can anyone shed some light on why it would not before but does now?
 
Did you change any other settings or voltages in the interim? Maybe it wasn't the CPU voltage that was holding you back. If the NB was the problem and you upped that voltage, that could explain the difference, for example.
 
you also need to monitor what speed/timings your ram is running at. if your ram is running 1/1 it may need some juice to remain stable. and btw, what difference does it make what vcore other people are running their cpus at? this is not a competition. so what if you need more juice to run at x speed than some other asshole. it doesnt mean that you are inferior. as long as nothing is on fire and/or smoke is not pouring out of your case, just give it as many volts as it needs, dammit. it takes 1.48v to run my e8400 stable at 4300mhz. i dont know how high it will go at stock voltage because i am not interested in that. i just want to know how high it will go with as much voltage as it can stand without overheating or breaking. why bite my toenails over how much less vcore i can run than the next guy? i dont get this whole stock vcore/undervolting thing. whats the point of this? does it matter that i am shortening the life span of my cpu by 10, 20, 30 or 40% if i am just going to swap it out in a few months?
 
vanilla_guerilla said:
and btw, what difference does it make what vcore other people are running their cpus at?
Click to expand...

It matters dramatically if i knew certain batches of processors are having a better run at hitting the goals I have in mind. So based on that fact that I am after, and not a competitive spirit, I would say it matters greatly. If I had to increase my vcore to the limit of 1.5 and it still didnt hit my goal with all the other settings being proper then I would say time for another processor.

I can phrase this another way... If no one sets a "standard" then you do not know where to start or finish, but at a bear minimum know the factual "hard stop" limitations of your equipment. Safety in a multitude of counselors bro.
 
bare.

anyway, in your previous posts you espoused no pre set limit of 1.5v or whatev. only that you were worried that your cpu could not run at x speed at x vcore like all the other guy's cpu's did.
 
vanilla_guerilla said:
bare.

anyway, in your previous posts you espoused no pre set limit of 1.5v or whatev. only that you were worried that your cpu could not run at x speed at x vcore like all the other guy's cpu's did.
Click to expand...

That is correct. "Like all the other guys cpu's did". They appeared to be "the standard" or "the average" that my cpu should be able to do. If not, then I had a problem, and I needed to know why.

I am suddenly puzzled that this is an issue? If I did not know what others could do, then how would I know that "on average" most peoples DO do such and such. And if that is the case, and mine did not, then heavier scrutiny of mine would be in order.

Am I the only one seeing my motives here?
 
CLR said:
Yeah, thats where I want to be. 4ghz. I dont need 4.2 or 4.5, just 4 flat and I am happy. I want to achieve that level in the most stable fashion that I can.
Click to expand...

I was hoping for the same thing but here's my story...

Out of the box, @ first boot I set it to 400x9 and it didn't work @ stock voltage of 1.2v in BIOS so I bumped it up to 1.28v and was stable for 30mins in OCCT. I had been running it like that for a week (heavy CoD4 time too) and not a single crash/issue.

Swapped out my Big Typhoon for a Ultra-120 Extreme this afternoon and started playing again. I wanted to make sure my RAM was up to the task of 445x9 (The magic 4GHz) so I set it at 445x8 (3.55GHZ) and ran OCCT stress just fine. I went for gold and tried 445x9 @ 1.3v and no go. Finally got up to 1.375v and it still wouldn't work for more that a few seconds in OCCT. I decided to do things backwards and start down clocking until it was stable. Tried 417x9 (3.75) @ 1.35v and it was no go. 412x9 (3.7GHz) didn't work either. Finally I just set it back to 400x9 (3.6GHz) and it's stable again.

Either I got an OCing dud or it just needs 1.4v or greater to become stable above 3.6GHz.
Basically, I'm pissed I got the E8400 and the P35 when my P965 and E6600 were working strong @ 3.32GHz for a long time. It wasn't worth the 'upgrade'.

I just hope you're not in the same boat.

EDIT-
Batch: Q748A223
Pack Date: 01/29/08
 
If your temps and voltage are below the thresholds, you'll be fine.

Yes, it matters what everyone else's chips can do, as it gives you an idea of what your chip should be able to do. As always, your results will vary. Some boards have a higher vDroop than others. Some chips need more voltage than others. A modest OC is easy, but squeezing 4Ghz out of the chip will require a bit more time and work -- so take your time. Anything under 1.5v is fine, as long as your load temps don't go above 60°C.
 
Yeah, this is definitely an art of combinations. And I read everything you guys post, and try to get an idea of what the average is, so far I know that 3.6 is pretty achievable by everyone. That is a good start, and currently where I am at.

I am going to keep testing, reading, and studying the input my everyone and slowly turn up the settings with what I have learned.
 
I have wanted to share a few things I have experienced trying to get my e8400 up to 4.0ghz. let me first state that I havent succeeded yet but ive reached 3.8 stable and my thermal paste should be burned in good now so ill be tweaking it more the next few days to hit 4.0

#1 make sure you turn off stepping. I could not break 3.6 with the multiplier reduction no matter what my vcore was

#2 as others have said here bump your nb voltage one step i could not break 3.65 until I did.

#3 make sure your rams not where the real issue lies. It couldn't hurt to loosen you timings or even drop your ram multiplier below normal in order to see if yuo can squeeze more out of your cpu you can always tune them back up to speed later.

I would say your vcore just isnt whats causing the problem it seem though that this configuration will hit certain walls and just stop going any higher until you find that one setting to change. Good luck I really think you will be able to break 1.6 just dont give up and keep trying new things. Also I will drop back in Saturday to let everyone know if I broke 4.0 and if so what I had to change.
 
You must log in or register to reply here.
Back
Top