I just ran my 3570K with prime for about an hour at stock setting, 3.8GHz turbo. But my temps were maxing out at 65C on one core with the rest about 5-10C lower, is this normal or are they a little high? Is it safe to try to Oc it or should i re-sit my cooler, a PH-TC14PE?
-
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.
***Official Ivybridge Overclock Thread***
- Thread starter MonsterMoshi
- Start date
That is normal dont worry about it. 1 core tends to be higher than others.
After having my CPU prime stable at 4.4ghz @ 1.24v for 2 hours using the small FFT format, i was happy. But i started getting BSOD's in STALKER which was odd. So i ran a blend test and it BSOD within a couple of minutes which was annoying.
I have backed off the overclock to 4.3ghz @ 1.2v and it is stable (8 and half hours of prime95 on blend) and about 10 degrees cooler so i think i am going to stick to this now as my clock, the difference in speed is marginal yet the 10 degree drop in temperature is a big plus.
Core 1: 60c
Core 2: 66c
Core 3: 64c
Core 4: 61c
Voltage: 1.20v
Multiplyer: x43
BCLK: 100mhz
Core Speed: 4300mhz
I have backed off the overclock to 4.3ghz @ 1.2v and it is stable (8 and half hours of prime95 on blend) and about 10 degrees cooler so i think i am going to stick to this now as my clock, the difference in speed is marginal yet the 10 degree drop in temperature is a big plus.
Core 1: 60c
Core 2: 66c
Core 3: 64c
Core 4: 61c
Voltage: 1.20v
Multiplyer: x43
BCLK: 100mhz
Core Speed: 4300mhz
Permabanned
- Joined
- 9 Jun 2009
- Posts
- 11,925
- Location
- London, McLaren or Radical
Blend test + gaming crash... sounds more like memory instability to me - try memtest.
Increase volts/relax timings on the mem to see if that helps...
Increase volts/relax timings on the mem to see if that helps...
Strange that you had so much voltage for 4.4Ghz and were unstable. Im stable 4.3Ghz @ 1.12V and 4.5Ghz at 1.24V I think it is. Guess you got unlucky. Nice temps though
As far as I am aware the 3770k needs less volts to get stable, which may explain your success, the problem is if I raise the volts much over 1.24v I get a huge temperature increase; 80+.
I am using Noctua NH-D14 for the record, but yeah I am pleased with the temps in the 60's and 4.3ghz is still a good clock, in the real world 100-200mhz will make negligible difference.
I did memtest before I did the blend test and it made 2 hours without any errors, so I would like to think I can rule that out.
One word of warning on stability ... Check your windows events log ...
I thought I was stable @ 1.275v for 4.6
Prime95 ran for 6 hrs
Ran LINX for 20 runs
Etc
Did a test handbrake encode and failed at 99%
Checked my windows event logs to find while encoding WHEA error code 19s (internal crc processor errors which are correctable) every 20 mins or so ..
Upped vcore to 1.295 and none in event log since
NOTE too low vcore can be why you can get stable but slightly lower LINX results processor will internally be recalculating on errors ... Transparent to you as end user other than LINX results slower
I thought I was stable @ 1.275v for 4.6
Prime95 ran for 6 hrs
Ran LINX for 20 runs
Etc
Did a test handbrake encode and failed at 99%
Checked my windows event logs to find while encoding WHEA error code 19s (internal crc processor errors which are correctable) every 20 mins or so ..
Upped vcore to 1.295 and none in event log since
NOTE too low vcore can be why you can get stable but slightly lower LINX results processor will internally be recalculating on errors ... Transparent to you as end user other than LINX results slower
So I ran IBT at 4.5Ghz just to experiment and it's interesting that it actually ran somewhat cooler than prime using AVX (by about 2-5C)
IBT temps:
prime95 temps:
I realise that the prime run was a bit short but the temps would have probably gone up by about 2-3C whilst running it a bit longer. Also the HWMonitor readings on the IBT are more likely to be correct than the realtemp readings for a constant temp range.
Run CPU-z so you can see your actual voltages. Those temps are high if the voltage is as low as you say it is!
BTW - You are running IBT on standard stress level which isn't really good enough for the task at hand!
Associate
- Joined
- 28 Apr 2011
- Posts
- 2,157
- Location
- North West
Hello guys. Just got a 3570k and would like to know where should I start with in overclocking this chip?
Just turn the multiplier up until it becomes unstable, then add steps of vcore ?
Do I need to adjust anything else? In my AMD chip I increased the northbridge as well, which improved memory performance. Can you do this here?
Just turn the multiplier up until it becomes unstable, then add steps of vcore ?
Do I need to adjust anything else? In my AMD chip I increased the northbridge as well, which improved memory performance. Can you do this here?
Only have to change to multiplier Id suggest going up in steps of about 200mhz you'll probably be stable at stock volts at around 4.2/4.3 you don't have to change any other setting other than the multiplier until you become unstable at which time you start adding little increments of vcore.
Associate
- Joined
- 28 Apr 2011
- Posts
- 2,157
- Location
- North West
Crashed after about 25 mins using prime95 4.0 Ghz @ stock volts
woahhhhh, i see guys tempretures are reading around 86degrees C. i would freak out if it got that high. my 3770k overclocked to 4.2ghz under 100% CPU load performs around 58. and at no use it peaks at around 30. here are some results n pictures of my system
Soldato
- Joined
- 15 Nov 2008
- Posts
- 2,535
- Location
- Reading
Watercooled and only 4.2Ghz
Ikr ahah.