Xeon Owners Club

[Zyll Goliat]

I'm honestly surprised how shitty the "Noctua nh-c12p se14" is. 14cm fan set at max speed and my x5670 @ 4.1ghz 1.275v throttles under Prime95 after 3min. And it throttles at 3.85ghz 1.225V as well after 10min. The cooler is installed correctly as when I touch its heat pipes under load they are hot as hell. I need to get rid off it and buy some proper cooler.

Just install 1 more fan on the other side of the cooler and see if that helps....

 

[Rumcajs997]

Just install 1 more fan on the other side of the cooler and see if that helps....

Nah, I will install Scythe Ninja from my lga 2011 PC and sell this Noctua. Its still expensive.

 

[Zyll Goliat]

Nah, I will install Scythe Ninja from my lga 2011 PC and sell this Noctua. Its still expensive.

Sure...Just sayin this helped me a lot it's cheap and easy to mount and it can reduce temps significantly...ahh just checked now your Noctua cooler is horizontal

 

[Rumcajs997]

Sure...Just sayin this helped me a lot it's cheap and easy to mount and it can reduce temps significantly...ahh just checked now your Noctua cooler is horizontal

But I'm honestly surprised, this cooler looks like it can handle 150w easly. This x5670 at 1.225v takes maybe 90w or so, so even box cooler should be fine.

 

[Zyll Goliat]

But I'm honestly surprised, this cooler looks like it can handle 150w easly. This x5670 at 1.225v takes maybe 90w or so, so even box cooler should be fine.

Hmm....I am sure that under Prime 95 he will start juicing a lot more than 90W...install HWInfo start prime and see how many W taking under the heavy load.......

 

[Edwired]

Is it in a case or on the bench? That what the asus p6t v2 deluxe does it throttles the vrms due to max power been hit check the power consumption in hwinfo64

I'm honestly surprised how shitty the "Noctua nh-c12p se14" is. 14cm fan set at max speed and my x5670 @ 4.1ghz 1.275v throttles under Prime95 after 3min. And it throttles at 3.85ghz 1.225V as well after 10min. The cooler is installed correctly as when I touch its heat pipes under load they are hot as hell. I need to get rid off it and buy some proper cooler.

I'm surprised that your having issues with the Noctua nh-c12p se14 I bought mine brand new a few years back and it working as it should the heatsink pipes only get luke warm to touch on mine. Unless you got the x5670 that just output more heat at a lower vcore is what I am thinking try another x5670 and see if the problem goes away.

 

[lexluthermiester]

I'm thinking it the blck that causing the sleep issue above 200. But if I have it on exactly 200 the sleep issue don't happen.

With my old MSI board, the OC was happiest and most stable at 201. This problem you're having could be a unique characteristic of the board you've got. I would say call it good right there and be happy with a stable OC at a low voltage.

 

[Edwired]

With my old MSI board, the OC was happiest and most stable at 201. This problem you're having could be a unique characteristic of the board you've got. I would say call it good right there and be happy with a stable OC at a low voltage.

Yeah it holding up at blck 200 no problems so far since since it the same voltages I had before seems to have done the trick not pushing it any further now maybe some day go higher just bumping the cpu multi one step to see if more or less vcore is needed for 4.40ghz but for now I'm happy that I got it this far in the learning curve

And the voltages fluctuation I mentioned before seems to clear up after I adjusted the 24 pin atx socket and things on the pc seems to work more fluent no wild dropping or weird glitching issues since

 

[Rumcajs997]

Gigabyte GA-X58A-UDR3 works fine with 2rx8 Samsung ECC Reg sticks I posted above too.
My second X5670 just arrived and I think it should be a better one. The first one's standard voltage is 1.22V, this one has got 1.15V only on Auto.

 

[Rumcajs997]

Ok, the second X5670 has got a promising start. 1.1v on auto under load, that is super low. I tried 1.2V 4Ghz and it seems OK so far, I've got a feeling that it can sustain 4ghz under lower V. Paired with Scythe Mugen has got nice temperature too.

Edit:

Even better, 4.07ghz with 1.176V. Over 2h Prime95 stable.

 

[Edwired]

Ok, the second X5670 has got a promising start. 1.1v on auto under load, that is super low. I tried 1.2V 4Ghz and it seems OK so far, I've got a feeling that it can sustain 4ghz under lower V. Paired with Scythe Mugen has got nice temperature too.

Edit:

Even better, 4.07ghz with 1.176V. Over 2h Prime95 stable.

How is it holding up since?

 

[pevac]

Nice, 1.35-38v for 4.5 ghz approximately ...

 

[Rumcajs997]

Nice, 1.35-38v for 4.5 ghz approximately ...

That would require Water Cooling no doubt, even Scythe Mugen seems to be getting to 4.2ghz max.

 

[Edwired]

I did abit testing with Handbrake to convert 22 videos about 43 minutes long each from mkv to m4v while using Powerplanswitcher app from Microsoft Store I switched it to power save mode which limited the cpu clocks from 4.20ghz to 2.40ghz. The workload on the cpu was average range between 50% to 60% usage to simulate the system to see would it crash during the coverting work load it never once crash or bsod. It took about 10 hours to finish. The reason I did this method to see if the 200 fsb is stable enough to ensure me that it can handle

 

[pevac]

Can BCLK over 200, only other components have to send it.
I with a 980x + p6x58 e-ws go relaxed 220.

 

[Rumcajs997]

I did abit testing with Handbrake to convert 22 videos about 43 minutes long each from mkv to m4v while using Powerplanswitcher app from Microsoft Store I switched it to power save mode which limited the cpu clocks from 4.20ghz to 2.40ghz. The workload on the cpu was average range between 50% to 60% usage to simulate the system to see would it crash during the coverting work load it never once crash or bsod. It took about 10 hours to finish. The reason I did this method to see if the 200 fsb is stable enough to ensure me that it can handle

I would try Prime95 or OCCT if I were you. They will push the procesor 100% on all cores and that's the only way to be 100% sure that OC is stable for every scenario.

 

[Zyll Goliat]

I would try Prime95 or OCCT if I were you. They will push the procesor 100% on all cores and that's the only way to be 100% sure that OC is stable for every scenario.

IntelBurnTest is also very good and it will push CPU to the MAX

 

[Edwired]

I would try Prime95 or OCCT if I were you. They will push the procesor 100% on all cores and that's the only way to be 100% sure that OC is stable for every scenario.

The weather is quite warm here in Ireland and I know I should do that but there's no need for me to do it since I know the system is stable due to the testing of handbrake which does push the system to 100% but I limited the cpu clock to 2.40ghz during the encoding which brings the usage to 50% to 60% which is what I wanted in the first place as I don't mind the system running abit slower and taking longer to finish so I can do other important things while the system is doing the work

If the weather is alot colder I will let the system in high performance mode and let it run but most of the time I leave it in power save mode for daily usage like youtube and other things

As for temps it about 15 to 20c cooler on power save mode during the encoding it peaks 51c on the cores while in high performance mode it started to climb to about 65c to 70c on the cores so yeah I pretty much favor the power save mode all the time.

 

[lexluthermiester]

IntelBurnTest is also very good and it will push CPU to the MAX

Be VERY careful with that one. It has the potential to burnout a CPU that has any weakness at all. I have personally witnessed this. It's the ultimate stability test, but with that ultimate status comes a risk.

 

[pevac]

IntelBurnTest is also very good and it will push CPU to the MAX

IBT max 90C

 

[Edwired]

I would avoid IntelBurnTest at all cost as that my opinion it isn't designed well since it haven't been updated in a long time

 

[lexluthermiester]

I would avoid IntelBurnTest at all cost as that my opinion it isn't designed well since it haven't been updated in a long time

It's still a good utility for Socket 1366 CPUs as that was the platform it was originally designed for. 2011 V1/V2 are also valid test platform CPUs. It's good for a short burst of 5 or 10 runs to test stability. More than that is risky. Intel designed it this way deliberately.

 

[Edwired]

It's still a good utility for Socket 1366 CPUs as that was the platform it was originally designed for. 2011 V1/V2 are also valid test platform CPUs. It's good for a short burst of 5 or 10 runs to test stability. More than that is risky. Intel designed it this way deliberately.

I know but an average user would keep doing the same cycle with intelburntest bringing the cpu on it knees is an unnecessary stress to the chip. In my theory it bring the life span down further than is it needed to be. But at the end of the day it everyone's choice what they want to do

 

[Zyll Goliat]

Well it's a TORTURE test it's exist to push to the maximum your CPU,Prime 95 also it's not better at all but in fact I prefer IBT as have shorter runs then prime and I will never suggest to anyone to use those test longer then 10-15 min and to always monitor CPU temps......

 

[Edwired]

Well it's a TORTURE test it's exist to push to the maximum your CPU,Prime 95 also it's not better at all but in fact I prefer IBT as have shorter runs then prime and I will never suggest to anyone to use those test longer then 10-15 min and to always monitor CPU temps......

I know but still it pointless if it going to burnout the cpu regardless. It don't matter how low or high the vcore is it still going to generate alot of heat no matter how good the heatsink/custom loop managing that