Sign in with
Sign up | Sign in
Your question

2500k, faulty core?

Last response: in Overclocking
Share
April 30, 2012 10:15:32 PM

Hello,
I have a 2500k oc'd to 4.5. Under prime95, one core is always 10 degrees cooler than the other 3. This same core always fails prime within 4-5 hours. Under stock clock I haven't had this core fail, but it is still 10 degrees cooler. Then today I had a bsod while updating podcasts in iTunes, nothing demanding obviously. Error was "Clock interrupt not received from secondary processor" Never had any problems playing games, benchmarks, etc. considering this bsod, the fact the one core is always cooler, etc. could this one core be faulty? Rest of specs:
Asrock E3 G3
Gskill 1600 8gb
Corsair hx750
Hyper 212+
ocz agility 3 120gb
W7
Sapphire 6870

PS no hardware changes since being built in January.

Thanks

More about : 2500k faulty core

a b K Overclocking
April 30, 2012 11:07:03 PM

The tempeture being uneven is indicative of improper cooler installation...Take the cooler off, clean off the existing thermal paste, reapply new thermal paste, and reseat the cooler .....Make sure all of the push pins are seated completely....
m
0
l
April 30, 2012 11:09:39 PM

Have done this twice already. Currently on my 3rd seating.
m
0
l
Related resources
a b K Overclocking
April 30, 2012 11:15:04 PM

Well, then the top of the CPU chip, the HSF, or Both are not fitting flush and will require lapping them to assure an even fit
m
0
l
April 30, 2012 11:25:38 PM

Ok thanks, I see what you are saying there... But the other 3 cores aren't getting above 62 degrees ever, so I don't think too much heat is the problem. Are you trying to say the uneven temperature will cause the same core to fail prime95, and possibly cause my bsod?
m
0
l
a b K Overclocking
April 30, 2012 11:32:52 PM

Heat can cause the core to fail and bsod....Try lowering your clockspeed a smidge, or upping your vcore a touch.... It looks like your OC is not as stable as you may think...
m
0
l
April 30, 2012 11:34:26 PM

I get that, but I get the same temp difference at stock speeds too.
m
0
l
a b K Overclocking
April 30, 2012 11:45:40 PM

Well, the chip, HSF, or both need to be lapped....
m
0
l
April 30, 2012 11:46:45 PM

Thanks for your input...
m
0
l
a b K Overclocking
April 30, 2012 11:50:46 PM

I hope it helped...Good Luck

m
0
l
a c 110 K Overclocking
April 30, 2012 11:59:57 PM

That one core being that much cooler than the rest IS strange, but your other core temps are fine, and having a core run cooler than the rest wouldn't necessarily be a problem, by itself. It just sounds like a simple unstable OC to me. For what it's worth, I've had OC's that passed 12+ hours in Prime, but crashed while watching YouTube vids (if I'm not using SpeedStep and the Balanced power profile in Windows to let it clock down). A little extra voltage fixed it. Also, my core temps aren't as out of line as yours, but I have 1 core that runs 4-5C hotter than the rest, and always has, no matter what I did to fix it.

It COULD still be a bad core, I suppose, but I haven't heard of that with any SB CPU's, and logic would dictate that it should cause problems all the time, at any speed.
m
0
l
a b K Overclocking
May 1, 2012 12:10:45 AM

Its for sure not a bad core...Unless, Aeris pushed it too hard and broke it with voltage or heat...OP never mentioned what the maximum voltage and temps were while stressing the machine....So, Aeris...What were your max temps and voltages reached while running prime95? Which test are you running in prime95?
m
0
l
May 1, 2012 12:20:11 AM

Max temps 62, 61, ,62, 52 on the 4 cores. Vcore 1.312 max using hwmonitor and CPU-z.
m
0
l
May 1, 2012 12:22:12 AM

And prime blend test
m
0
l
a c 110 K Overclocking
May 1, 2012 12:30:05 AM

Aeris said:
Max temps 62, 61, ,62, 52 on the 4 cores. Vcore 1.312 max using hwmonitor and CPU-z.


Yeah, nothing wrong with those temps and that voltage at all. With core temps, you shouldn't worry until you get to 80C+ (70C+ is a bit high, but not devastatingly so) and as long as you stay at or below 1.45V, you're good there too. I'll stick with my original assumption that it's just an unstable OC. Try a little higher voltage (1.33V or so).
m
0
l
May 1, 2012 12:35:19 AM

Thank you for your reply. At work now, I'll tinker some more later tonight.
m
0
l
a b K Overclocking
May 1, 2012 12:40:45 AM

@DJ---> I have already addressed the instability issue up in the thread...It is an unstable OC for sure
m
0
l
!