• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Think I nearly killed my Q6600

Asus Striker II extreme 790i Ultra. When I've got Loadline Calibration Enabled in Bios, it does eliminate Vdroop, so when I specify 1.4875 in bios, then with older cpu-z it reports 1.49. But with the newer cpu-z it reports 1.58. Even when loadline calibration is disabled.

I'm afraid to clock it again incase I actually kill it. I think it was temperature related though.

It is very hard to kill a Q6600 (infact I have not herd of anyone killing one...) I have seen people running 1.7v+ at high temperatures... and myself have run 1.65v+ for suiside runs at 4.3ghz+

I believe Jonny Bravo ran at 2v+ (on LN2 Mind) for his Q6600 WR
 
Thats good to know then Pneumonic. I guess that my mobo or ram was unstable, I've set ram to 1280, upped the ram volts, and it's been stable for a while now. I also ran Chkdsk and it corrected a few errors. Prime and IBT were passing fine but COD5 multiplayer kept crashing after 10 mins.

I'll try raising the multi back up one at a time, see if she'll give me 3.6.


Anyone know why my reported voltages is higher than what I set them to?
 
I would try setting your PCI-E Frequency manually... start at 101 and raise it up until you become stable.. anything up to 115 should be fine... you might have issues going above 105 or so if you are running a lot of or pain in the butt HDD :)
 
I can't understand this, Intel Burn test passes 25 tests no problem. But if I fire up Call of duty 5 MP, it'll BSOD within 10-15 minutes. I also ran prime small FFTs for an hour and that produced no errors. Whats going on?
 
I can't understand this, Intel Burn test passes 25 tests no problem. But if I fire up Call of duty 5 MP, it'll BSOD within 10-15 minutes. I also ran prime small FFTs for an hour and that produced no errors. Whats going on?

Just like mentioned above, each test stresses the component in different ways. Just because IBT, OCCT, Prime95 or any "burning" tests passed does not mean it is 100% stable. It just states that the current system has a higher probability of being stable.

Also, running a game and having a BSOD could also mean a lot of things -- maybe GPU, powersupply, RAM, mobo, HDD or even drivers. May I suggest that you try to eliminate all but one variables and start from there. Have everything at absolute stock and then go from there, and test/stress each component one at a time.
 
Yeah, its fine on stock, the overclock is affecting something. I also updated drivers lately. Should I revert to stock settings in order to install the display drivers?
 
Yeah, its fine on stock, the overclock is affecting something. I also updated drivers lately. Should I revert to stock settings in order to install the display drivers?

What I would do is --

if the CPU-Z is reporting incorrect CPU vcore,

I'd make sure that the CPU and heatsink/fan is seated correctly, dust is off the mobo and contacts,
Then I'd flash/upgrade BIOS to a 'tried-and-tested' version, fresh OS install and update all drivers to the newest ones.

Then I'll make sure that all/most settings are on manual. Options that affect the vcore directly, such as Loadline Calibration, would be disabled or automatic.

Then i'd run memtest. And then if I decide to overclock, I'd o/c each component in little steps, this way I have control over the variables. If not, I'd be happy at stock. ;)
 
Well, I see no point in doing an OS reinstall, just yet. Because hardware monitor in the BIOS also reports higher voltages. The voltages reported are consistently 0.5 Volts higher than what I set. I could try the new BETA BIOS, that may help it.
 
im betting its the memory. run prime blend mode and see how it goes. also intel burntest is only good if you use the 64bit version. 32bit version puts on less cpu load.
 
Well, I see no point in doing an OS reinstall, just yet. Because hardware monitor in the BIOS also reports higher voltages. The voltages reported are consistently 0.5 Volts higher than what I set. I could try the new BETA BIOS, that may help it.

yeah, go about the whole process systematically. Usually, wanting to save time and skip a step thinking that certain 'steps' are just too obvious sometimes end up being the problem and waste a lot of time and effort, well at least for me. :D:D

Also, newer BIOS revision isn't always the best, I find. Some mobos I end up downgrading.
 
Well, I don't really suspect RAM, but as I'm pumping so much volts into Vcorem it must be ram. It took me 1 hour 15mins to complete a pass of memtest earlier, no problems. How many passes should you guys think it to be stable, I'm using memtest86 ver 3.5.
 
I can't understand this, Intel Burn test passes 25 tests no problem. But if I fire up Call of duty 5 MP, it'll BSOD within 10-15 minutes. I also ran prime small FFTs for an hour and that produced no errors. Whats going on?

How long should I run it?

Also make sure when you run Prime95 that you run an instance for each of your cores (you have to create a shortcut to the exe with -A1 -A2 parameters etc IIRC, there's instructions in the help file). If your game uses several cores but you only stress test one at a time, that would obviously explain it!
Run it for at least two hours; overnight is better. I've had errors crop up after several hours of running it. I don't know why you should get an error at one point in time when the CPU's been under the same voltage, clock speed, temperature etc for the previous two hours, but that's the way it is.
 
This may seem wierd to you guys, but it looks like a RAM error. I upped ram voltage to 1.68 in BIOS, in windows it reports it as 1.74, but so far, so stable. I've played COD5 for nearly 45 mins, no crashes.

I've left ram at stock 1333, even lowered to 1280, but I raise voltage and all is fine. Is this normal?

EDIT - Also, max temps during gaming hit 56 degrees is this ok? I'll try some GTA 4 later. That'll stress my cpu...
 
told you its to do with the ram. your not even listening to the advice which is provided to you here. memtest is a load of crap, iv ran memtest for 54 hours and it found no errors on a machine which goldmemory picked up errors in within 15 mins.

since goldmemory costs money the best option for you is to run prime95 blend mode, and run multiple instances of prime blend to max out ram usage without overspilling into virtual memory.

also what operating system are you using? proper stress testing is best done using 64bit os, and 64bit intel burntest is going to be far more stressful than any game is, what i do is set intel burntest to use max ram and 99999 cycles and just let it go overnight and i manually check the residual norms to see if there is any errors generated. 12hour run is a good minimum length to run it for.
 
Dude, I do listen to the advice given here, just trying out new setting takes time, I mehtodically tried everything the previous posters have said. I will be honest, I did think RAM wasn't an issue at first, hence thats why I left it till last.

I only started this thread a couple of days ago, so it takes a hell of a lot of time stress testing with different settings, as I'm sure you know yourself. Again, I'm not 100% sure if it is stable, as I only upped the ram voltages last night. So it'll take me a while to find if its unstable. I still havent got around to setting ram at 1280 and do some extensive tesiting, as a previous poster suggested. My actions are going to be lagged therefore.

Anyway, I just don't understand why I would need more voltage to my ram, when I leave it at stock. It's rated at 1.6, but it getting 1.72 now. Is this ok?

And a great thanks to all those who have helped me so far. Much appreciated.
 
Last edited:
Back
Top Bottom