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

Please check my OC...

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

yamaharacer19

Member
Joined
Jan 13, 2013
Location
Jackson, Michigan
Hey guys. Can you guys please check my OC. I only ran P95 for 10 mins but I concerned about the temps that I'm getting. Below is a pic of full-load running P95 small FFTs.
 

Attachments

  • 4.7 GHz.png
    4.7 GHz.png
    115.1 KB · Views: 109
You could definitely use some tweaking on that VCore, that will help reduce temps. I would imagine 1.33-1.34v should be pretty attainable but hey each chip is different...

If you were only running for 10 minutes I would be concerned for the long term. Was this your first pass through with an OC or were you unstable at lower voltages?
 
Hey guys. Can you guys please check my OC. I only ran P95 for 10 mins but I concerned about the temps that I'm getting. Below is a pic of full-load running P95 small FFTs.

Run a blend, then report back at the 20 minute mark, where the temps reach their max. Even low 80's won't actually cause substantial harm to your part. As the last poster said, you may be able slightly lower your vCore for a degree or two. You may even be able to lower your pll without any ill effects, but that's not going to give you much of a temp reduction.
 
^ Indeed, but that is definitely less than ideal especially for a 24/7 OC.

If you cant attain stability with a lower vcore and acceptable temps, I would dial down the clock a couple hundred mhz which would allow for a significantly lower vcore with an inconsequential performance hit.
 
^ Indeed, but that is definitely less than ideal especially for a 24/7 OC.

If you cant attain stability with a lower vcore and acceptable temps, I would dial down the clock a couple hundred mhz which would allow for a significantly lower vcore with an inconsequential performance hit.

Aye, I wouldn't suggest low 80's for a 24/7, or even upper 70's for that matter. Just letting him know he needs to get his MAX load temp before we can make really good suggestions, and that comes 20m in (Also that he shouldn't freak if he gets into the low 80's)

Pwn has this down good though, tuning it down a bit and lowering voltages in any way possible maintaining stability is probably a good idea
 
I had to cut P95 short because I had to go somewhere and I didn't want to leave it running. I was afraid that something bad might happen if I wasn't around to see it. The vcore was hovering from 1.344 to 1.36 when I did the 10 min test. Right now, I'm using PLL overvoltage @ 1.709 with PLL @ Level 2. I'm also using offset vcore @ +0.075V (i think...I'll check it after this post).
 
Everything looks fine to me... temps when not stress testing will be around 10c less or more for everything else. Voltage is under our recommended 1.4 so...all good to me.
 
Everything looks fine to me... temps when not stress testing will be around 10c less or more for everything else. Voltage is under our recommended 1.4 so...all good to me.

Mm, you're right. Not by my standards, but factually this is very true. OP, fiddle around with voltages, try to lower them as much as you can, run 2 hour P95s, etc. Make sure TIM is applied well, make sure you have proper airflow..

ED, you should look at my posts over here!
 
Doing blend test as we speak. I went back into BIOS and lowered the offset down to +0.065V. If it survives that then I'll try lowering the offset again. Right now, I'm at 10 minutes into a 20 minute blend test, my max temps are 64, 67, 66, 65 with vcore ranging from 1.344V to 1.360V, but mostly staying at 1.352V. These temps are better compared to my OP.

If I have PLL overvoltage enabled, should I keep C3, C6, and Package C enabled? Or should I disable them?
 
Here's the results after 20 minutes using P95 on blend at 4.7 GHz and offset vcore +0.065V. Now I'm going to try a lower offset.
 

Attachments

  • 4.7 v2.png
    4.7 v2.png
    119.5 KB · Views: 93
Pll overvoltage and c states have nothing to do with each other.

When starting out, leave c states disabled.
 
Don't waste time...run Cinebench 11.5. If that passes without WHEA 19 / WHEA logger errors in the Windows Event Logger tool you're good to go and move on to Intel Burn Test max memory for 5 passes and then Prime95 Large FFT for a few hours if you wish.
Blend is good but only if you run it 18hs with 90% of your ram size.
 
^ Indeed, but that is definitely less than ideal especially for a 24/7 OC.

If you cant attain stability with a lower vcore and acceptable temps, I would dial down the clock a couple hundred mhz which would allow for a significantly lower vcore with an inconsequential performance hit.

Alright. Right now I'm at 4.5 GHz, level 2 LLC, offset at +0.005v. Do you think I could use a negative offset at that speed?
 
Alright. Right now I'm at 4.5 GHz, level 2 LLC, offset at +0.005v. Do you think I could use a negative offset at that speed?
only testing can give you that answer... every cpu is different!

This is why I do not like cookie cutter settings... well one or the reasons. The other is you don't actually learn what you are doing or how you got there making troubleshooting that much more difficult.
 
I'm going to leave my 4.5 OC the way it is because I'm testing a 4.7 OC at -0.055 offset with a LLC at level 1. I'm getting a core that ranges from 1.336 to 1.360v. I'm going to try and lower the offset some more and see if I can get a max core of around 1.344v.
 
Last edited:
Excellent!

Well I failed P95 12 minutes within starting the blend test. So I raised my offset up to -0.050v and I passed P95 on blend with max temps of mid 70s and normally running at high 50s low 60s, but the vcore on cpuz was ranging from 1.344v up to 1.368v. Is that a good vcore for a 4.7 overclock?
 
Forget what I just posted. Lol. I tried playing Black Ops 2 but I got a BSOD error code 0x124. So I'm going to try fixed vcore at 1.350v with llc @ level 1 instead.
 
Last edited:
Back