• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

OC'ing my i5 750 - again!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

-Ice

Member
Joined
Nov 26, 2011
I've OC'ed this CPU 2+ years ago and managed to get to 4.1GHz with miahallen's guide. However, I started playing a different flight sim and had good performance without the overclock so after testing, I returned the settings to "optimized defaults" and have been playing at those settings ever since.

I've recently upgraded my setup to 3x27" 1920x1080 res screens and with my new hardware, I was getting an average of 32.9fps on my flight sim (Falcon BMS) and I knew that if things got "busy," my frames would definitely be lower. After playing around with Eyefinity and upgrading graphics drivers, the next step for more performance seemed to be the CPU overclock since MSI Afterburner showed that I was only using max of 62% of the GPU.

Step 1: I managed to get the bclock stable at 210Mhz, x15 clock ratio, 6 SPD with idle temps of 33-36*C and max temps of 70-74*C. I have left all power savings options on [Auto] and only disabled Intel Turbo Boost. Testing is done with Intel Burn Test (max) with a 3- or 10-run setup, temps are shown via RealTemp.

Step 2: I could only manage to get stable results at bclock 200, x15 clock ratio, 8 SPD, QPI/VTT 1.21v, DRAM 1.660v with idle temps of 29-33*C and max temps of 66-70*C. Attempting to raise bclock to 210 would not be stable even when going up to QPI/VTT 1.47v (max temps 72-77*C).

A memtest86+ at this point showed that my RAM is error-free after 1 pass/1hr 5min of testing.

Step 3: I had to raise vCore to 1.4v (from 1.2v) to be stable at bclock 200, x20 clock ratio, 8 SPD, QPI/VTT 1.21v, DRAM 1.660v. My RAM setup is all on auto and at this settings it is at 9-9-9-24 1T. However, my temps were 31-38*C on idle and 81-88*C on max and this high temp worried me so I've only done a 3-run test. A 10-run test would run for almost an hour and I wonder if I should do this considering the high temps generated. I did do two 3-run IBTs and the system was stable with the whole test taking 15min. Testing with Falcon BMS gave me an average of 41.5fps but I have failed to note the temps generated, so I need to do this again. I am happy with the FPS gain but I need to verify my temps before I can be at ease with the current settings.

Is a vCore of 1.4v too high to get 4GHz? Are my temps too high? Looking at my notes 2+ years ago, I could get to 4.1GHz with bclock at 205, vCore at 1.375, RAM timings at 10-10-10-28, and temps of 74-79*C max.

I tried upping the clock ratio to x21 but it wouldn't stick, it goes back to x20 after reboot and CPU-Z confirms this.

Any help appreciated!
 
Last edited:
I ran 2 more IBT 3-run tests and the system was stable although the temps would vary from 80-86*C on one test and 73-77*C on another. One thing I also noticed was that on the high-temp test, IBT would show the speed at 57GFlops and on the low-temp test, the speed was 37GFlops. Testing BMS after each test still gave me an average of 41fps though.

I also went ahead and did a 10-run test which took 49 minutes with idle temps at 34-38*C and max temps at 81-86*C, but the system was stable.

On a "just browsing" usage of the PC, temps would go up to 58-60*C and after testing/playing Falcon BMS for 20 minutes, temps would range from 58-65*C.

While this relieves me of the worry of day-to-day use of the PC as the current tests show that I don't approach 70*C when using it, I am still worried about the vCore of 1.4. Current setup is bclock - 200, ratio - x20, SPD - 8, vCore - 1.4v, QPI/VTT - 1.21v, DRAM - 1.66v, LLC on, RAM timings is 9-9-9-24 1T.
 
Yes, and I don't know why I can't get the same now, especially as I've not OC'ed the CPU since that time.
 
I assume you mean QPI/VTT? Why bump it? What is it supposed to achieve? My system has been stable with 1.21v all the time, the earlier OC test only needed 1.25v.
 
You said you needed 1.40v for a 4.0Ghz clock (200 BCLK x 20). By increasing the VTT (yes, QPI/VTT/FSB Termination voltage), I was suggesting a lower Vcore of ~1.35v like in your previous thread.
 
Are you saying that by increasing the QPI/VTT, I could lower the vcore?

Looking at my notes, I was getting "unstable" messages from IBT and blue screens at 1.35v and 1.375v vcore. Only at 1.4v was it able to pass IBT.
 
Odd... I started off by lowering my vcore back to 1.375v and ran IBT (max) for 3 runs, and it was stable with temps of 73-79*C. I was surprised as the last test at this voltage didn't even finish IBT and blue-screened. I rebooted, ran IBT (max) for 10 runs which took 49 minutes and was stable, giving me temps of 74-80*C. I think I'm more comfortable with this now, but I will test again at 1.35v and see if I can get stability by raising QPI/VTT instead of vcore.
 
Back