• 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***

You're volts are fine. Do you have a 3570k or a 3770k? (guessing the 3570k since you're temps are maxing at 70) LLC is basically something to combat vdroop its full name is Load Line Calibration. I don't fully understand how LLC works however.
 
3570k
I will leave LLC if this overclock turns out to be stable looking good tho :). I see no need to go higher than multiplier x44. I tend to only run prime for about 3 hours on small fft. Normally I have found if unstable on prime will crash within the first hour. After this I goto other benchmarking/stress testing programs. (this is the way I done with my past amd chips assume same with intel). Very little experience with overclocking only done overclocking once before and that was with a 965be turned out to be great chip ran at 4.2 with stock volts so to overclock only had to up the multi.

Edit: grr failed. Got windows error service has crashed viewed windows error log dozens of the yellow errors error prime was still running. I have now set LLC to exstream and upped volts to 1.25.

Max temp since volt increase and setting llc to exstream 71.
 
Last edited:
Hey guys and gals,

Just set up my new Ivy PC yesterday! All going well now, had some problems initially as Id put the mobo stands in the wrong places so it was shorting when I moved the case.. stupid I know.

Ive got a 3570k running at stock with a coolermaster 212 evo and it's idling about 25-30c which I'm happy with. I'll get overclocking later though.

Is it normal for one core to always be higher temperature than the others? For me its the second core runs at least 5 degrees higher than the others. Any reason for this?

Thanks
 
I think i ve got the crappiest i-7 3770k. I set the multiplier to 42 and the Vcore to 1.12 v and LLC to turbo which seems to be stable under burn test. Tho for some reason CPUZ shows 1.104v voltage during the stress test. However i cant seem to increase the multiplier to anything more than 42... i tried 43 starting from 1.175 to 1.265 using 0.005/0.010 increases at a time and it simply wont boot to windows. At 43x @ 1.12 v computer freezes while loading bios...
 
Ran Prime for an hour after the above changes that I posted. No windows error logs this time still going strong. Now running IBT max temp 78. Going strong. i will run more prime tonight while football is on but I think if passes IBT and few game benches should be stable. BTW how many times people recommend running IBT

Just finished its first run successfully in 311 seconds on high.
 
I have had a couple of different things happen with Prime - BSOD, 'Prime95 has stopped working' and even the odd error within Prime where 1 'worker' has errored and then stopped after completing x tests in x minutes. Also, my chill-control (Antec 920) and Nvidia settings have occasionally stopped working. The BSOD is a bad overclock but what about the others - do they all point to the same thing?

In all case except the BSOD I have been able to carry on and have tweaked and tried again, sometimes with a further error, sometimes not - or not until a couple of days later.

Currently sitting at multiplier 46, volts at 1.265 (But that gives WHEA errors in Prime) although I haven't changed any other settings. Temps after an hour of Prime are a max of 84c. 1.275 volts clears the WHEA errors but puts the temps up to a max of 89c
 
iv got my 3770 at 4.2ghz with stock volts
id like to get it up to 4.7-4.8 but im not sure what i need to adjust to what
is 1.3v safe?

ps its on a custom water loop which helps
 
Try going to 1.3V and see how that goes the safe vcore distance for IB is 1.45V (intel suggests you don't go over 1.52V but in all fairness OCuk suggests not to go over 1.45V and I'd take their word over intel any day)
 
not quite so much stability as such, more of chip life span decreases == to voltage increase. After 1.52V of voltage a chips life span drastically drops in time.
 
I have had a couple of different things happen with Prime - BSOD, 'Prime95 has stopped working' and even the odd error within Prime where 1 'worker' has errored and then stopped after completing x tests in x minutes. Also, my chill-control (Antec 920) and Nvidia settings have occasionally stopped working. The BSOD is a bad overclock but what about the others - do they all point to the same thing?

Yes, those errors are due to instability. I got the prime95 stopped working due to lack of Vcore, The one worker stopping happened a couple of times for lack of Vciore but also when I was overclocking memory. In your case thuogh i'd say it all points to a bad overclock.

iv got my 3770 at 4.2ghz with stock volts
id like to get it up to 4.7-4.8 but im not sure what i need to adjust to what
is 1.3v safe?

ps its on a custom water loop which helps

1.3v should be fine on custom water provided it keeps temps manageable. I'd start off at 1.25v and 4.4Ghz and work your way up slowly. What motherboard are you using?

I think i ve got the crappiest i-7 3770k. I set the multiplier to 42 and the Vcore to 1.12 v and LLC to turbo which seems to be stable under burn test. Tho for some reason CPUZ shows 1.104v voltage during the stress test. However i cant seem to increase the multiplier to anything more than 42... i tried 43 starting from 1.175 to 1.265 using 0.005/0.010 increases at a time and it simply wont boot to windows. At 43x @ 1.12 v computer freezes while loading bios...

What type of cooling do you have an which motherboard?
 
Managed to get mine sorted now and I am fairly pleased with the results.

Using 3570K, Gigabyte Z77-D3H and an Alpenföhn K2 I am seeing the following results when running prime95 small FFTs:

4.2ghz with 1.125v and 64c
4.3ghz with 1.15v and 67c
4.4ghz with 1.19v and 73c
4.5ghz with 1.25v and 77c
4.6ghz with 1.3v and 83c

4.7ghz would not boot with 1.3v.

The above were all stable with 30 minutes of testing. I think I am going to stick with 4.4ghz as my daily set up and will run prime95 overnight tonight to check it is stable.
 
Managed to get mine sorted now and I am fairly pleased with the results.

Using 3570K, Gigabyte Z77-D3H and an Alpenföhn K2 I am seeing the following results when running prime95 small FFTs:

4.2ghz with 1.125v and 64c
4.3ghz with 1.15v and 67c
4.4ghz with 1.19v and 73c
4.5ghz with 1.25v and 77c
4.6ghz with 1.3v and 83c

4.7ghz would not boot with 1.3v.

The above were all stable with 30 minutes of testing. I think I am going to stick with 4.4ghz as my daily set up and will run prime95 overnight tonight to check it is stable.

Very nice voltages on them. Cant get mine stable under 1.25 for 4.4. Any chance you able to list yer settings?
 
Very nice voltages on them. Cant get mine stable under 1.25 for 4.4. Any chance you able to list yer settings?

I set the following:
base clock to 100.00
Set the CPU Clock Ratio to 44
Enable Turbo and set each core ratio to 44
Disable CPU Enhance Halt, C3/C6 State Support, CPU Thermal Monitor and CPU EIST Function
Set cpu voltage to 1.19v
Set LLC to Turbo
 
What type of cooling do you have an which motherboard?

Gigabyte Z77-UD5H latest bios with Phanteks PH-TC14PE in a silverstone Raven 1 case. I do need to sort a signature with these details at one point.

But back to the point, Computer not booting to windows with a a 43x multiplier even when increasing the voltage up to 1.265 simply doesnt make any sense. Getting to windows and finding it to be unstable is one thing but not even reaching that far is to laugh about. Could it be that my motherboard is ****/faulty?
 
Gigabyte Z77-UD5H latest bios with Phanteks PH-TC14PE in a silverstone Raven 1 case. I do need to sort a signature with these details at one point.

But back to the point, Computer not booting to windows with a a 43x multiplier even when increasing the voltage up to 1.265 simply doesnt make any sense. Getting to windows and finding it to be unstable is one thing but not even reaching that far is to laugh about. Could it be that my motherboard is ****/faulty?

Using a 3570k? check all you're power settings and turn them off to begin with including speedstep. make sure you're RAM is also 1600Mhz (just change the RAM profile to XMP)
 
Gigabyte Z77-UD5H latest bios with Phanteks PH-TC14PE in a silverstone Raven 1 case. I do need to sort a signature with these details at one point.

But back to the point, Computer not booting to windows with a a 43x multiplier even when increasing the voltage up to 1.265 simply doesnt make any sense. Getting to windows and finding it to be unstable is one thing but not even reaching that far is to laugh about. Could it be that my motherboard is ****/faulty?

hmm, that doesn't seem right. As long as RAM's at normal timings/frequency then you should try these settings:

revert to default then change to manual/xmp mode and:
Multiplier 43
turbo mode off
LLC 75%
CPU PLL overvoltage enabled
EPU disabled
Speedstep disabled
Vcore: fixed 1.26v
All current capability settings at 100% except CPU at 120%
CPU PLL: 1.8v
BLCK frequency 100.0
(If applicable) the equivalent of ASUS Multicore Enhancement Enabled
VCCSA voltage 1.1v
All C states disabled

I'll add some things in a min, just about to bsod most likely as I'm still sorting out my offset mode OC.
 
Using a 3570k? check all you're power settings and turn them off to begin with including speedstep. make sure you're RAM is also 1600Mhz (just change the RAM profile to XMP)

Its an i-7 3770k. Now that i disabled the power saving **** it did manage to load to windows. Am at 4.4 Ghz/ 1.212 v now. Gona test for stability while am studying for My Applied Mathematical Models exam on Monday :D
Thanx for the advice, i have Gskill 2400mhz memory and i ll try to overclock that as well after am done with cpu
 
Back
Top Bottom