2500k 4.4GHz limit

fishingcat

fishingcat

fishingcat

Associate
Joined
10 Jan 2013
Posts
284
I've had an i5 2500k with a Gigabyte Z68X-UD3P-B3 motherboard for the past 18 months, which has never been good at overclocking. The maximum overclock I got on my Hyper 212+ was 4.4GHz, which required 1.35V VCore and 5 step LLC.

Yesterday I got a Phanteks PH-TC14PE in anticipation of getting a Haswell chip in a few months, and figured I'd see if I could squeeze some more out of my 2500k.

I set the VCore to 1.38V with 6 step LLC and put the turbo multiplier to 45x, and my PC failed to POST entirely. It got stuck at the BIOS screen and clearing the CMOS via the jumpers or a battery pull did nothing. Eventually after repeatedly attempting to clear the CMOS via both methods (probably took half a dozen tries and around half an hour) it booted again at stock settings, and works fine at my old overclock settings. Temps are fine, as I can't break 60 degrees when stress testing at 4.4GHz 1.35V.

Any ideas what's going on?
 
What is your VTT set at?

You got any C states on?

Have you got your memory over clocked? How does it fair in mem test?

That may just be your chips limit.
 
VTT, PLL etc. are all on their stock/auto values.

C states are enabled (I didn't want to lose downclocking to 1.6GHz at idle), but never seemed to affect stability in the past.

Memory is at stock (1600MHz, 1.5V, stock timings). I haven't run memtest.

I once booted at 4.5GHz at 1.4V VCore + 6 step LLC, so I'm aware that the chip can probably handle it with sufficient voltage.

The thing that I was really confused about was the repeated failures to clear the CMOS though. That's something that I'd never seen before, and it's pretty worrying. It makes me wonder if the motherboard is alright.

Thanks for the advice though.
 
Mine is stable at 1.35v with Extreme LLC at 4.4ghz, any higher requires 1.41v. I can get 4.8ghz at 1.46v. But I run 4.4ghz stable at the moment. My RAM is 1866MHz too. So to be honest don't be surprised if that is the chips limit at less than 1.4v.
 
I must be very lucky, I've got mine at 4.5GHz @ 1.288v and that's with overclocked samsung green memory @ 1866
 
How have you got on with this op? Go any higher what voltage did u settle at?
I'm currently debating taking my oc further currently @[email protected] I'm running under water with gpu on a sr-1 560 rad.
Was wondering what safe 24/7 volts are under water and what safe max volts for benching is anyone know?
Also what's safe max temps I'm guessing under 70-75c
Sorry to hijack your thread.
 
chris-82 said:
How have you got on with this op? Go any higher what voltage did u settle at?
I'm currently debating taking my oc further currently @[email protected] I'm running under water with gpu on a sr-1 560 rad.
Was wondering what safe 24/7 volts are under water and what safe max volts for benching is anyone know?
Also what's safe max temps I'm guessing under 70-75c
Sorry to hijack your thread.
Click to expand...

For safe temps you want to stay under 80c when stress testing.

If you're under water you should be able to use 1.4-1.5V without breaking a sweat.
 
I tried to boot yesterday with 1.42V VCore, 6 step LLC, PLL overvoltage and all power saving features turned off and it wouldn't even POST.

At this point I'm going to assume that my chip just hates me and leave it at 4.4GHz until Haswell comes out in a few months.
 
Then use a higher voltage? i use 1.48v with my 2600k, been on it 18 months. My cpu is terrible so i use high volts. I couldnt care less what people say is safe or isnt safe, clearly it's fine, had no issues. Just use more, problem solved.
 
Honestly I'm only keeping the chip for another few months so it hardly seems worth using 1.45V+ just to get up to 4.5-4.6GHz. LLC would put idle voltage over 1.5V if I set it that high anyway.
 
You must log in or register to reply here.
Back
Top Bottom