2600k runs HOT stock

[yumcax]

I was experiencing random crashes (reboots) when playing games. So I ran GPU and CPU stress tests (at the same time) and very quickly the CPU reached 85c, so I stopped and I assume that the CPU is the problem. GPUs never got above 65c.
Build:
ASUS P8Z68-V PRO
8GB DDR3
i7 2600k (stock cooler)
2x EVGA SC 560 Ti SLI
Antec 1200 (good cooling)

Now, this is my first build and I haven't messed around with OC at all, on any components. Strangely, though, in AI Suite BLCK is showing as 100.3 and the multiplier is 35. Isn't stock 100&34?

I checked and that TPU thing is off, is there any reason that these settings are changed? Is it likely that this is a problem with the chip and I should return it? Or is the stock cooler simply not capable of keeping the CPU cool at stock (or slightly above) speeds!?

Thanks for any help!

 

[Twoboxer]

First off, the odds are there is nothing wrong with your cpu. The cooler is most likely installed improperly, or the fan is not working.

If the fan is spinning, check the rear of your mobo to make certain that the little clips on the end of the cooler's white split-pins are hooked behind the mobo, and the black pin is down far enough to force them to stay in place.

If you can't see the back of the mobo easily, another way to test this is to grab the cpu cooler and try to rock it. If it shifts at all, it is installed improperly.

I haven't used a stock cooler in a while, but was there a sheet of material covering the paste that's pre-applied? If so, did you remove that sheet?

Asus probably tweaked the standard BCLK setting to gain a benchmarking edge on competitors. As for the multiplier, the cpu may have been in some turbo mode. But it wouldn't be a bad idea to remove/disable anything that can OC your cpu while in Windows, and maybe even reload BIOS optimized defaults. That ensures nothing is screwing with you . . . until you get the build running cool as it should.

 

[yumcax]

Okay, booted into BIOS and reset to defaults, then rooted around until I found a settings called 'Turbo Boost' which I gather is built in OC function that was set to enabled, and disabled it. Then I put it on 'Maximum power saving mode' (EPU) just to be safe . It now passes Intel Burn Test with a max temp of 84, and since Im not doing anything CPU intensive, it's good enough for now.

I did check when I installed the MOBO that the cooler was installed correctly, but I'll check again in a minute. And there was no plastic over the TIM.

 

[Twoboxer]

No, Turbo Boost should be left at its default setting. Otherwise you are losing the auto-OCing Intel built into the chip. Just use optimized defaults, let the BCLK go to 100.3 where Asus wants it, and leave it alone for now,

You may have fixed your problem, but frankly I doubt it. I can run my 2600k thru 8 threads of Prime95 at 4.43GHz and not crack 65C on any core. Admittedly, I'm not running a stock cooler, and heat buildup inside my case is not an issue..

But you are running your cpu at the flat stock frequency (turbo disabled) . . . even with the stock cooler that thing shouldn't go over eg 70C.

So my bet is you still have something wrong on the cooling side.

 

[yumcax]

No, Turbo Boost should be left at its default setting. Otherwise you are losing the auto-OCing Intel built into the chip. Just use optimized defaults, let the BCLK go to 100.3 where Asus wants it, and leave it alone for now,

You may have fixed your problem, but frankly I doubt it. I can run my 2600k thru 8 threads of Prime95 at 4.43GHz and not crack 65C on any core. Admittedly, I'm not running a stock cooler, and heat buildup inside my case is not an issue..

But you are running your cpu at the flat stock frequency (turbo disabled) . . . even with the stock cooler that thing shouldn't go over eg 70C.

So my bet is you still have something wrong on the cooling side.

Okay, but I'm not left with a plan of action, other than buying a nicer cooler. I just checked and the HS is in place properly, black pins all the way into the white. The fan is running. I don't understand what is wrong. I have 7 fans in this case! If the stock cooler is not the problem, and the case is not the problem, how can I be sure that the CPU isn't the problem? How can I check?

 

[azama]

i wonder why no one mentioned the power supply? since 84 degrees is okay with no oc then i think it isnt a the problem.
check your 12v wattage on the power supply and make sure it can feed the 2 560s AND the processor.

 

[azama]

check your 12v wattage on the power supply and make sure it can feed the 2 560s (180 EACH!) AND the processor (about 150 watts depending on your voltages).

 

[Leaps-from-Shadows]

There is no way that stressing the CPU at stock speed should send the temp up to 85ºC. The stock heat sink is crappy, but not that bad. I'd expect 70ºC at most while in Prime95, and maybe 55ºC while gaming.

yumcax:
Use the jumper on the board to reset the BIOS to defaults (using the optimized defaults setting in the BIOS doesn't actually use the default settings -- ASUS boards auto-overclock using that setting). Once at the real defaults, go into the BIOS and disable Hyper-Threading. Then run Prime95 again and report the temps.

 

[CaedenV]

Not likely to be the Proc. Your processor should be able to handle 95*+ without immediate damage (though perhaps it would take a few years off it's life cycle). Before replacing your stock HSF (heat sink/fan) go out and buy a $5-10 tube of thermal paste or grease. Artic Silver is best, but if you are running stock speeds any brand will do just fine. Remove all old paste with a cloth or Q-Tip dipped in hydrogen peroxide or other non-residue alcohol based cleaner (don't get that crap on your hands!). Put no more than a small pea sized dab of new paste on the proc and use a latex/vinyl glove to spread it evenly over the surface and replace the fan. This alone will drop you ~5*c even if the old cooler was installed properly. But my bet is that you accidently touched the cooler to the CPU and then backed off again before seating the cooler which messed up your connection.
At any rate, that should solve the heat, but the instability is likely bad voltage control (cheap/bad PS, bad power mgt. on the mobo, bad power connectors, etc.). Voltage testers only cost $10 and are handy to have around. Especially running 8GB of ram, and SLI, if your PS isn't up to snuff then you will run into all types of problems.
Then again, it could be a bad proc, but in my experience if the proc is dead then you simply don't POST, because they have too many internal checks to even initialize the chip before it lets you do anything with it. And with the reliability rate of CPUs (AMD and Intel) compared to power supplied and mobos it is hard to believe that the CPU could be the problem

 

[yumcax]

Okay, I'm not so sure that the CPU is what is causing the PC to reboot (although it still has high temps), as I was playing the JC2 demo and monitoring CPU+GPU temps (below 70c) and had another random reboot. So this makes me suspicious of the PSU. Hopefully I wont have to RMA or anything, spent way too long on cable management
The PSU is an Antec CP-850 (850w), which should be enough for my system without OC.

Also, as the stock TIM is solid before it melts, I don't think I could have messed it up in the installation. And as more thermal compound = $10 and a hyper 212+, which comes with TIM and only costs $20 more should make more of a difference, I might just go with that.

So, how can I test and make sure the PSU is/isn't the issue? Voltage tester? I may have something like that around the house, I'll have to check.

Thanks for the help so far!

 

[yumcax]

Best answer selected by yumcax.

 

[Mousemonkey]

This topic has been closed by Mousemonkey