Too easy? is 100c too much?

[taltamir]

I decided to OC my E8400 on a gigabyte GA-EP35-DS3R.
I removed the limiters off of my two noctua fans (push pull scenario), no more silence for me. And I went into the bios.

First thing I noticed is the flash red warning telling me to leave the voltage controls on auto if I OC... it does the same if you actually change the voltage control to manual. The voltage control is either manual for everything, or auto for everything. This posed a bit of a problem for my ram OC since my ram is meant for 2.1v DDR2-1000 5-5-5-18. but I just reduced it from 2.4 multiplier to 2.0 multiplier (yea this bios measures wrong).
It seems to dynamically loosen the timings rather then provide more voltage.

Either way the CPU OC just works.. I just went and set from 333mhz to 400mhz fsb, works...
I raise it to 445 fsb, works... one pass at memetest, OCCT for 30 minutes works.
I get a little worried at the 75idle / 98-102c load temps so I lowered it to 430 fsb.
At 430fsb I was getting 65 idle / 95-102c load, been running OCCT for 2 hours now with no trouble at all.

I never even disabled speedstep, it goes down to 6x multi when idling.

I am a little leery though.
1. Is 100c too high for my E8400? Is it damaging it?

2. My voltages are high and make no sense at all. It seems to be running 1.42-1.43v when I boot, It remains at that when I am testing load. And then when I stop stressing it, it shoots uo to 1.52v and stays there at my 6x multi idling... I activate OCCT again and it drops from 1.52v to 1.488... turn it off again, back up to 1.52v.

Also, on second OCCT run with the 1.488v it goes up to 103c load temp, rather then 96... It actually seems like the load temp goes DOWN over time. When I left OCCT running for 2 hours it went from 100c solid to 95c solid.

It definitely doesn't seem to be making sense. Why is it going down when I am at load? why is it not going as low as it is when I boot up? but most importantly, why is it rock solid stable?

EDIT:
I got my answer, it IS too easy to be true. The board easily OCed the E8400 from 3ghz to 4ghz, but despite being rock solid it was massively overvolting it, resulting in 100c which too high, and 1.52v which is way too high, and would have damaged my chip, if it hasn't already degraded it. I am now manually overclocking it with much more modest voltages and fsb rates.
Lesson learned, do it yourself!

 

[JAG87]

Is this a joke? Did you forget to take your medication today?

 

[taltamir]

no i am dead serious. the bios on this motherboard is very insistant that I let it manage the voltages, and I wanted some input and sharing about what it does when I am doing it.

I really don't know if 100c is too much for this CPU. And I really am wondering if I should spend all the time to manually adjust things manually to get things working. But so far it seems to be having to trouble getting rediculous OCes like that on its own.

And btw, when I say "OCCT stable for X hours" I mean it ran stable for that long, and then I turned it off and tried something else, not that it ran that long as then crashed.

 

[schneiderguy]

Isn't the shutdown temp for a e8400 less than 100c? I don't think thats an accurate reading if your computer didnt shut down.

that's a lot of voltage though

 

[taltamir]

i am wondering about that... maybe the CPU-z is acting up?

I didn't disable any of the special features on the E8400.. so it still goes down to 6x multi when at low use, and it still has the temp protection and everything...

I do have my CPU temp warning on the mobo set to go off at 90c and it is NOT making a peep. Although, I don't have the speaker plugged in to the mobo, but I am pretty sure it still makes a beep when I turn it on.

So you think I should do it manually at a lower voltage then? Any other tool I can use to check the voltage?

 

[DarkRogue]

IMO Gigabyte flashes that big red warning sign to discourage non-technically savvy users from overclocking, breaking something, then complaining to them.

I have OC'd an E8400 on an EP35-DS3P and it does the exact same thing. I just ignored it because I was largely following the guide sticked on this forum so I figured one of us should know what we were doing. 100c is massively high in my opinion, most people don't want to even tread into 90c area. Also, what heatsink are you using? If it's the stock heatsink, I don't think that's recommended for massive overclocks.

As far as the auto-voltage management goes, I've heard that the motherboard will overvolt it if left to do its own thing. If you're going to be overclocking, you'll need to set it to manual, ignore the big flashing red warning, and manually control your voltages.

As for "why it is rock solid stable" 30 minutes is hardly any indication of stability. Try running some games or something, which is a far more 'real-world scenario' type thing and see if that crashes on you. If you can play for a good 2-3 hours without a problem, then yeah it should be stable.

I agree that is an incredibly high voltage though, especially for an E8400 which is supposedly rated for a max of 1.45v (correct me if I'm wrong here, I haven't checked in a while) so I wouldn't be surprised if some damage has not already been done.

 

[schneiderguy]

So you think I should do it manually at a lower voltage then? Any other tool I can use to check the voltage?

yes, 1.52 volts is way too much :Q

 

[taltamir]

1. Its been rock solid after 2 hours of OCCT + some hours of memtest. the 30 minutes were just a quick test @ 4ghz...

at 4ghz it was stable for 30 minutes, so I went down to 3.85 and tried it for 2 hours.

I am also playing games on it and doing other things. Nothing is having any issues.

Thanks for the info about 1.52 and 100c being too much. That is what I wanted to know. While this is an extremely easy OC, I guess I should tone it down a bit.

Also, How exactly am I going to get a push pull scenario with the stock fan?

I have the noctua U12 with two F12 fans (120mm) and an inch after that there is a 3rd 120mm fan on my case...

So it is more of a pish-pull-pull scenario.
Also, the heatsink itself is cool to the touch.

 

[DarkRogue]

I didn't mean to claim it wasn't stable, it's just that I've run 12 hours of Prime before, and when I go to launch a game, the system crashes. Those programs really need a whole lot of time to spot certain problems. As long as you're not having any issues with it, by all means go ahead and keep the overclock.

I never used the stock fan, nor have I even seen it since I ordered an OEM E8400, so I didn't know if it was still a standard fan-on-top heatsink, or if they migrated to heatpipes.
I am curious how you're managing 1-foot fans in there though..

Anyway, have you doublechecked that the heatsink is seated properly and you haven't used too much TIM? If the CPU is indeed as high as you listed, that heatsink should not be cool to the touch, although I guess having 3 fans should offload that heat pretty quickly.. I don't think it should be that fast though.

P.S.
Woot, 500th post. Hope it helps lol.

 

[taltamir]

oops. I meant 120mm fans. sorry.

I am touching it on top, so yea, I don't know how it is cool either. The video card is really hot to the touch, and it is only running @ 60c (100% fan, unchangeable, using a 7900GS atm cause i gave my brother my 8800GTS 512 and I am waiting to get a 4870).

I am impressed by how easy this is, but the only difference between my board and yours is that mine uses the ICH9R southbridge and yours uses ICH9, So if you read that it uses too much voltage when allowed to control it by itself, then I better just suck it up and do it manually.

I am still impressed at how easily it achived stability, but I would rather have it run cooler then 100c...

BTW, from reading the info about the EIST and C1E orwhatever it is called, it says that it will reduce VID and Core clock when it reaches 100c, not shut down the computer.

So maybe taht is why I am reaching 100c on both and not going over, it could just be reducing it without CPUz detecting the reduction...
Or maybe it is somehow disabled, I checked my bios settings and it was set to enable, but when I tried to run gigabyte's dynamic energy saver (i wanted to see what it reports as my voltage and temp) it refuses to run saying I need to enable EIST and C1E. (actually it just says "recommended to enable" and then doesn't turn on, weird).

 

[VirtualLarry]

Temps and voltages are WAY too high. Definately go with manual voltage control, the board is overvolting the 45nm on AUTO, and may damage the chip (if it hasn't already degraded it slightly).

You're just going to have to learn to overclock manually, or suffer the consequences of a degraded CPU.

 

[ther00kie16]

Typically, people say 1.4V and 70C are good cut-offs. Definitely don't try anything over 1.4V for extended periods of time. Anand pushed a QX9650 and it died at 1.45V after too long I think. I can't find the link as it's not in the articles apparently.

 

[taltamir]

interesting. I set it to manual, set it to 1.225 (what it claims to be the default).
And CPUz is fluctuating from 1.168 to 1.184.
I guess its vdroop causing it to be a bit lower, but why is it fluctuating like that.

Anyways, OCCT had reported an error the second i turned it on. So I am guessing either 3.6ghz is too much. Or 1.225 is too little voltage. Maybe I will just switch it back to 3ghz and try undervolting it instead (i always love paying less for electricity)...

I wonder how long the chip would have survived with the gigabyte auto settings...

Originally posted by: ther00kie16
Typically, people say 1.4V and 70C are good cut-offs. Definitely don't try anything over 1.4V for extended periods of time. Anand pushed a QX9650 and it died at 1.45V after too long I think. I can't find the link as it's not in the articles apparently.

I'll keep that in mind, thanks.

 

[DarkRogue]

Not every E8400 can run 3.6GHz at 'stock' voltage, unfortunately. It would boot but was not stable for me at 1.225v. I had to bump it up to 1.250v before it finally became stable at 3.6GHz.

As far as Gigabyte's Dynamic Energy Saver utility, if you overclock, just uninstall it. It will refuse to run until you undo your overclock and instead will just nag you every time you boot up. That thing sounds like a good idea but only if you're going to stay at stock settings.

It's funny seeing a thread go from extreme overclocking to undervolting stock speeds, but best of luck to that endeavor. The steps are essentially the same, bump voltage down, stress test, repeat until you find minimum voltage before it starts becoming unstable.

 

[BTRY B 529th FA BN]

are you sure you are reading Celsius and not Fahrenheit?

 

[ther00kie16]

Originally posted by: BTRY B 529th FA BN
are you sure you are reading Celsius and not Fahrenheit?

Yea, I thought that at first but didn't think it likely as 445 FSB = 4ghz and 100F=38C is too low for 4GHz on air. Either way, what are you using to read the temps?

 

[taltamir]

100 degrees Fahrenheit = 37.7777778 degrees Celsius

So yea, definitely C.

this thread was NEVER about extreme overclocking. It was about how easy it is to OC (even to the extreme) with the automatic voltage adjustments on the gigabyte board, and wondering if there is a price to pay... to which I was told yes, that it over voltages like crazy and will burn out my chip if I let it continue.

Anyways, aftering failing OCCT at every voltage level between 1.225 and 1.25 I decided to use larger jumps at first and then lower it, going to 1.275. Stable so far. Showin up as 1.2v on CPUz, which means that when it was detecting 1.52 it must have been closer to 1.6 (the mobo allowes up to 2v.)

Anyways, so far it is stable at at 1.275v @ 3.6ghz with the EIST and C1E turned off (i left the thermal safety thing on though).

Core temp tells me it is 105C before it engages, so i never actually hit it before, even at the highest settings I had I only reached 103c for a moment.

I am now getting 40-45cc idle and 58-60c load. Pretty nice at a solid 3.6ghz.

I am only gonna aim at 3.6 simply because 400fsb is such a nice round number, and it leaves my ram at exactly DDR2-800 v1.8 5-5-5-15.


I am using:
Coretemp 0.99 to check temp
OCCTPT to stress test CPU
Memtest+ 2.01 to test ram.
CPUz to check my voltage level.

 

[spittledip]

Auto OCing is a dangerous game to play I would think. Take your time and do it yourself to save your expensive parts and wallet.

 

[DarkRogue]

Those temps are starting to sound about right. My Scythe Ninja with a single slipstream medium fan undervolted to 5v is keeping my CPU at around 45C idle as well. It could probably be lower if I used better TIM and if I didn't put too much Ceramique, but it's good enough for me.

EIST should be disabled until you're done stability checking for a stable voltage, then people usually turn it back on for the bit of power savings during idle.

 

[aigomorla]

Originally posted by: JAG87
Is this a joke? Did you forget to take your medication today?

ROFL...

taltamir, you should know better.

 

[taltamir]

EIST actually just allowes my CPU to be controlled via software... like the gigabyte dynamic energy saver.
It is C1E that dynamically lowers my multiplier...

It seems that my C1E is bugged. I noticed that the same thing happens at the much lower voltages I use now. that is... the PC boots at voltage X, running something intensive increases the multiplier from 6 to 9 but the voltage remains at X. When I stop running it intensively the multiplier drops to 6 and voltage increases, when I run something intensive again the multi goes up to 9 and the voltage DROPS, but not as low as it was originally.

Z > Y > X

X = initial voltage during boot, and at first increase to 9x multi
Z = voltage at 6x multi after first drop down back from 9x.
Y = voltage at load and 9x multi after first drop down back from 9x.

The voltage was only 1.42v (both on load and idle) on auto before the C1E messed things up, and raised it to 1.52 to 6x idle and 1.48 on 9x load.

So now I completely disabled it, I just keep it at 9x multi at all time, and at whatever voltage I am ending up with. So far it is stable 3.6ghz at 1.28125v (1.22v measured in windows by utilities, aka: after droop). 1.275v had an OCCT error after 46 minutes.

Also, I had no idea that v2 of occtpt was out, I was using 1.2, this new version is awesome, way better then before. It has an array of internal sensores that give me temp and voltage, which match perfectly what I was getting from core temp and from cpuz. as well as other sensors as well. So now I can use a single program to get it all done. It also has much smarter testing presets, times measurements, and chart making capability.

Originally posted by: aigomorla

Originally posted by: JAG87
Is this a joke? Did you forget to take your medication today?

ROFL...

taltamir, you should know better.

what can I say. GPUs got me used to 70c idle and 100c load (what all the new ones seem to default at unless you manually increase the fanspeed).
Although, now that I think about it, I HAVE been doing just that to prevent them dying, which they do when I let them go over 90c...
DOH! I really wasn't thinking when I started doing it. I don't know what possessed me to try automatic overclocking, or to push it to 4ghz

I literally got out of bed, turned off the alarm clock, walked over to the computer, turned it on, and started ocing.

 

[BTRY B 529th FA BN]

Originally posted by: ther00kie16

Originally posted by: BTRY B 529th FA BN
are you sure you are reading Celsius and not Fahrenheit?

Yea, I thought that at first but didn't think it likely as 445 FSB = 4ghz and 100F=38C is too low for 4GHz on air. Either way, what are you using to read the temps?

real temp 2.60, hw monitor, & core temp

 

[taltamir]

ok this is odd. i thought it was the C1E (lowers the voltage and the multiplier when idle).
But i see now that this is not the case. Even when set at a solid 3.6 ghz 9x multi with no change I clearly have a voltage drop when I go from idle to max load, and then the voltage goes back up when I idle again.
it is detected as 1.213 idle and 1.211 load. Why would it do that?

 

[DarkRogue]

I could be wrong, but I was told EIST (SpeedStep) lowers multiplier and C1E (Enhanced Halt State) lowered voltage. I somewhat remember being clarified on this, but I can't remember what it was though. Try disabling both EIST and C1E and seeing if it still happens.

 

[GrJohnso]

taltamir - congrats on the setup and OC.... I have the same MB + CPU combo and one of the first things I did was test out the built in overclocking like you. I freaked when I saw the voltage being cranked through it, and immediately shut it down. I went to full manual and now enjoy a modest 3.98ghz setup at about 1.325v... temps in the 50-65c range.. It could go over 4ghz at higher voltage, but didn't want to burn the thing out too quickly...

Anyway, enjoy your setup. Now with the more modest timings it should last you for quite a while without worry..