Are theses temps alright? i7 4790k @4.7GHz @1.312V

ali13245

Reputable
Oct 14, 2014
108
0
4,690
Hey guys So I'm having some issues with my Overclock. So I was playing around with some settings in the bios, and I found some weird things happening when I booted into windows. I set my VCore in bios to 1.260V, however when I'm stress testing using Aida64, cpu-z shows the voltage @1.312V, Is there something wrong there or is this supposed to be normal? Also are the temperatures in the safer zone? I wil have a pic below

This pic was taken 5 min into stress testing, however an 1 hour later, Aida64 showed that the avg cpu temp was 71.6 degrees.
http://postimg.org/image/f2om5yxwj/

Another thing I seem to be having an issue with is my idle and load temps.

at idle (when not browsing or anything): 34-36 degrees
at idle (when browsing, youtube, streaming videos, etc...): 38-40 degrees

load: 60 degrees when playing bf4, However I must note that I have 2 x GTX 980's in SLI @stock

Are these temps within safe zone for this OC?

Last thing, I notice that everytime I open an application such as chrome, my cpu will skyrocket from like 35-65, then back down to the 30's again. Is that also supposed to be normal? That was happening even before the OC.

I would really appreciate it if someone could confirm this information, Thank you :)
 
Solution
The voltage should be running at 1.26 volts. To check and make sure, run AIDA64 with only Stress CPU and check the CPU voltage. Should be at 1.26 volts.

Most likely you will run into Thermal limits, before you are able to get much higher. But really it all depends on how high of a temperature you are willing to get up to. Most people recommend keeping the Core temps below 90C. I myself have a lower clock set to keep the temperature lower (65-70C).

Keep me posted on the stability of your system. If you are able to achieve a stable overclock of 4.7Ghz at 1.26 volts, that would be awesome =)

SkylerJacobs

Reputable
Jan 26, 2015
685
0
5,360
*Can you please post your computers full specs?

*Some stress tests, such as Prime95 and AIDA64 (When stressing FPU), will increase the voltage past what is set in the BIOS. This seems to only occur when AVX instructions are used.

*I am guessing you have the Power options in windows set to "high performance"?

*The temperature should not be jumping that high just from opening an application.

*What is the temperature of the room which the computer is in? (Ambient Temp)
 

ali13245

Reputable
Oct 14, 2014
108
0
4,690
Full PC Specs:
- Intel Core i7 4790k
- MSI Z97 Gaming 7 motherboard
- Crucial Ballistix Tactical Tracer 8GB Ram 2x4GB @1866MHz (OC with XMP on MOBO)
- 2x GTX 980's in SLI @Stock
- 1TB WD Caviar Blue
- 256GB Crucial MX100 SSD
- Corsair H100i liquid cooler
- Corsair RM750 Watt PSU
- Fractal Design Define R4 Mid Tower Case
- Cheap ASUS CD/DVD Optical Drive

Power options are set to balanced mode

Ambient temp: 11 degrees celcius

Right now as I am typing this the temp of the cpu is hovering between 45-50 degrees :S I have nothing opened but chrome, steam, and corsair link
 

SkylerJacobs

Reputable
Jan 26, 2015
685
0
5,360
11 celcius? I hope you mean 21 celcius

Your temperatures should not be that high when just at idle/small load. First thing that comes to mind from your description of things is incorrect installation of the CPU cooler. But, there could also be other reasons.

Looking at Corsair link, is the pump running?

Any problems while installing the CPU cooler?

Did you use the included thermal paste that was on the CPU cooler? Or did you use another Thermal paste? If so which one?

Is the 1.26 volts set in bios Adaptive or Manual?
 

ali13245

Reputable
Oct 14, 2014
108
0
4,690
Hmm... It prob is 21 Celsius, my bad.

Yes I removed and installed the cooler twice now. Here is the weird thing before I removed the cooler, I was using the stock h100i thermal paste that is already pre applied, and everything was going fine. I am now using AS5, which I am kind of regretting now... Seeing that My temps seemed to have gotten worst. I heard other poeple were having problems with AS5, and they said that there is a 200 hour cure time. Is that why my temps are all over the place? I even cleaned both the cpu and the cpu cooler with 91% isopropyl alcohol, and then wiped both down with a microfiber cloth.

The 1.26 Volts is set to adaptive, I don't think there is a manual option. There is auto,adaptive, and I think offset (which I have no idea what it does), and a couple of others.

* Just so you know, I replaced the thermal compound about two days ago.
 

SkylerJacobs

Reputable
Jan 26, 2015
685
0
5,360
What method did you use when applying the thermal paste?

I had not seen the picture you posted in your original post.
As for the temps you were seeing for idle 34-36, was this after stress testing?
If so, then i would say your temps are fine.
Load temps seem to be fine for how how of an overclock it is, even when running with "Stress FPU" ticked.

Im going to assume your ambient is a bit higher than you think. Do you know the degrees Fahrenheit?
 

ali13245

Reputable
Oct 14, 2014
108
0
4,690
I think the degrees fahrenheit is 52, not too sure :/

First method I used for applying thermal paste was the "grain of rice" method such as the one linus suggests. That gave me very poor temps under load. I noticed that bf4 was running around 65 Celsius. So I removed the paste and discovered the paste did not cover the bottom left corner of the heat spreader, So I reapplied using the 'Straight line" method that Arctic silver suggests doing, and that is what is giving the current temps right now. My idle temps when not doing anyhting is from 34-36, and when I'm browsing, youtube, etc.. it idles from 38-40 Celsius. The reason the temps are slightly high is also because of the two cards?

What about when I open chrome, Is the cpu fluctuation normal?

Right now I have steam opened pre loading gta 5, and my temp as of this moment is 48-50 Celsius, Is this supposed to be normal? When I closed steam It drops back down to the low-mid 30's

 

SkylerJacobs

Reputable
Jan 26, 2015
685
0
5,360
You were right about the ambient being 11C if its 52 fahrenheit ^^ With an ambient that low i am suprised your CPU is not reflecting and showing a lower temp at idle, but as you said, the reason the temps are slightly high is because of the two cards.

I myself usually do a pea sized drop in the middle of the CPU. the line method works good too.

What are your GPU temps? Stock cooler, or what version are they?

Upon opening chrome on my own computer, i saw a jump of about 3-5 degrees Celsius. This correlates with the information you were providing (about a 4-5 degree increase). I think everything is fine, you are just seeing slightly higher temps because of the SLI setup.

 

ali13245

Reputable
Oct 14, 2014
108
0
4,690
I am two gtx 980's from evga which has the ACX 2.0 coolers on them. The top card usually idles as 47 Celsius, and the bottom at 40 Celsius, but this is due to them running at low rpm set by me.

So what about the CPU fluctuation when I open chrome, Should I be worried about that?

What about my current temp with steam opened?(48-50- Celsius)

*When I close steam, temps drop back to mid-high 30's*

Thank You for all the information you have shared with me, I really appreciate it :)
 

SkylerJacobs

Reputable
Jan 26, 2015
685
0
5,360
Not a problem, just here to help. I am learning at the same time as well. After doing a ton of searching about SLI temperatures, your system seems to be running as it should.

I am interested though, what are your GPU temperatures when playing games?


My single GPU system runs on the cold side.

EVGA GTX780 SC 6GB runs at 30C idle, 73-75 when playing games such as Middle Earth:Shadow of Mordor, or The Crew. (The higher temperatures are due to being overclocked. At stock the card rarely, if at all, hits 70C.)

Currently i am seeing CPU core temps of 40-45C with steam, chrome, skype, curse, and Shadow of Mordor paused and minimized.

With your overclock (4.7Ghz) and the voltage required (1.26 volts) i am going to say that the temps are normal, and shouldn't be a problem.
 

ali13245

Reputable
Oct 14, 2014
108
0
4,690
I believe my top card runs between 70-75 Celsius, and the bottom card is usually 10 Celsius lower (so in this case ~65 Celsius)

For the overclock, So far nothing has crashed which is a good sign. I am going to leave the system on overnight stress testing just make some final confirmations of stability. If I succeed, I will try bumping it up at 4.8GHz with the same Voltage (hopefully) if it can handle it.

One more question. So is my 4.7GHz overclock @1.260 volts as I set in the bios, or is it @1.312 like cpu-z shows. I would like to know which of the two values is the true voltage it is running at, because if it is stable 1.260 volts, that is great, because it leaves me alot of headroom to bump it to 4.8GHz :)
 

SkylerJacobs

Reputable
Jan 26, 2015
685
0
5,360
The voltage should be running at 1.26 volts. To check and make sure, run AIDA64 with only Stress CPU and check the CPU voltage. Should be at 1.26 volts.

Most likely you will run into Thermal limits, before you are able to get much higher. But really it all depends on how high of a temperature you are willing to get up to. Most people recommend keeping the Core temps below 90C. I myself have a lower clock set to keep the temperature lower (65-70C).

Keep me posted on the stability of your system. If you are able to achieve a stable overclock of 4.7Ghz at 1.26 volts, that would be awesome =)
 
Solution

ali13245

Reputable
Oct 14, 2014
108
0
4,690
Alright thx for all the help :).

I sure will, I'll probably have an update by tomorrow if the system is stable at 4.7GHz

Is it okay to leave Aida64 running overnight, Lets say... around 15-20 hours of stress testing, is that safe?
 

SkylerJacobs

Reputable
Jan 26, 2015
685
0
5,360
I tested my system for right around 12 hours to ensure stability. Should be safe, as long as temperatures are in check, and no thermal throttling is happening.

Also, if you want to see what kind of temperatures you will get in real world situations, i recommend ASUS Realbench. Realbench benchmark tests the system with the following: Image Editing, H.264 Video Encoding, OpenCL (Graphics Cards), and Heavy Multitasking.This gives a realistic representation of the temperatures you will see while performing said tasks.
I have provided the link to where you can download Realbench if you wish to do so.
http://rog.asus.com/rog-pro/realbench-v2-leaderboard/

Realbench also has a Stress Test built into it. It is what i used for stress testing my CPU when i was trying to find my lowest overclock voltage. This stress test only gives options up to 8 hours.

**Remember to run RealTemp to monitor temperatures while running Realbench**
 

ali13245

Reputable
Oct 14, 2014
108
0
4,690
So I have good news (so far)

The system has been stress testing for 4 hours and 10 min now, and everything seems to be running stable at 4.7GHz @1.312V (CPU-Z reading). The avg cpu temp on aida64 is 72.4 so far. I will leave the system on for another 12 hours, making a total of ~18 hours. I am assuming by that time if the system is still stable, the overclock is good?
 

SkylerJacobs

Reputable
Jan 26, 2015
685
0
5,360
if it makes it that far, then yes i assume it is good to go. From what you are saying, i am assuming you are running the stress test with "Stress FPU". This is where the voltage increase is at. Increasing from the 1.26 volts set in bios up to 1.312V. I suggest running without Stress FPU in order to assure stability @ 1.26V.

Either that, or run the stress test within Realbench (Which wont increase the voltage) to ensure stability at the BIOS set 1.26V.
 

ali13245

Reputable
Oct 14, 2014
108
0
4,690
Okay So I have the final results of the stress test. The test ran for 17 hours straight, and there were no crashes :). I can assume now that my 4790k is stable at 4.7GHz.

http://postimg.org/image/c41mbeku7/full/

And you were right about the 1.26V, as you can see from the pic above, I was stress testing with "Stress FPU" checked. I tested without it and CPU-Z showed it was running at 1.26V. So that means my CPU is stable at 4.7GHz @1.26V?

I also ran GTA 5 with the overclock, and my cpu temp was averaging between 60-65 Celsius, is that temperature safe? Also remember that I am running two 980's in this system.

Thank You.
 

ali13245

Reputable
Oct 14, 2014
108
0
4,690
Okay Then I'm all set, however there is still one thing that is worrying me, and that is the fluctuation of cpu temps. I am worried that this will cause issues later on down the line. Eveytime I launch something, for example chrome, my temps skyrocket then immediatly come back down. I still don't know what to do about this, or whether it is supposed to be normal. I notice that the highest fluctuations occur when running chrome, so I am not sure if chrome is the problem.