high temperatures 8600k

[huti26]

Hello,

 

my temperature while being idle is always above 50° celsius according to HWMonitor.

This happens even when I am not overclocking my system.

 

I already double checked, that my amount of thermal paste is just right and that the cpu cooler is properly in place. My windows install is also fresh and I got all the proper drivers.

[LogicalDrm]

Pic of HWmonitor? Idle is rarely idle. With 15% of use and bit higher (25C+) room temps thats normal temps. You can get lower temps by increasing fan speeds, but 50C is far from danger levels.

[huti26]

Nevermind, I actually had the cooler upside down and therefore the fans the wrong way around.

 

Now I'm oced at 4,9 with 1.3volt and am at ~50° idle is that okay? I read that idle temps are not too relevant.

[JabroniBaloney]

You're fine.

[LogicalDrm]

On 5/2/2018 at 9:22 PM, huti26 said:

Nevermind, I actually had the cooler upside down and therefore the fans the wrong way around.

 

Now I'm oced at 4,9 with 1.3volt and am at ~50° idle is that okay? I read that idle temps are not too relevant.

Its still bit over what D15 is supposed to provide. But high room temps could explain that.

[huti26]

Reviving this one. Today I played csgo at the said settings and the game itself crashed a lot and I went up 90° cpu temp. When stresstesting with prime95 I go up to 90° too, but I neither exceed those 90° nor does any of my programs or the pc itself fully crash. Can I assume that something was faulty with csgo?

 

The airflow in my room is actually pretty lackluster, that could explain the high idle temps.

[LogicalDrm]

13 hours ago, huti26 said:

Reviving this one. Today I played csgo at the said settings and the game itself crashed a lot and I went up 90° cpu temp. When stresstesting with prime95 I go up to 90° too, but I neither exceed those 90° nor does any of my programs or the pc itself fully crash. Can I assume that something was faulty with csgo?

 

The airflow in my room is actually pretty lackluster, that could explain the high idle temps.

I would say that your OC is bit too much (too high voltages) or generally just unstable. Back it up a bit.

[huti26]

Alright, the crashes do not occur, when I go to stock voltages, so indeed my OC was the problem.

 

I guess I just lost the silicone lottery, thanks for the help.

[printf_cyber]

On 5/6/2018 at 3:59 PM, huti26 said:

Alright, the crashes do not occur, when I go to stock voltages, so indeed my OC was the problem.

 

I guess I just lost the silicone lottery, thanks for the help.

And whenever you bump up the voltages, make .0xxx change, and then check if you get the frequency you set, cause extra voltage will definitely decrease the life of your computer and increase the heat dissipation.

By the way the temps for an 'i5' are just as you are getting, you can check the Intel certification for the maximum safe temperature, and always start a topic on this forum before reporting them about the processor.

 

-Basil

[huti26]

Actually I got another crash today while being at stock speed and playing fortnite, so something is fishy here. I sent in the parts to get them exchanged. Is there any chance that my power supply is too weak? 

[JayRad]

29 minutes ago, huti26 said:

Actually I got another crash today while being at stock speed and playing fortnite, so something is fishy here. I sent in the parts to get them exchanged. Is there any chance that my power supply is too weak? 

It's definitely running way too hot, so something is for sure not adding up. I am also running the 8600k in the Gigabyte Ultra Gaming z370 mobo. My idle temps are easily floating in the mid 35C-40C with an overclock of 4.9GHz, and will spike to 70C+/- during heavy gaming. My ambient room temperature is on the cooler side, probably around 18-19C. 

 

If you are getting temperatures of 90C at stock clocks during gaming, you have got something wonky going on with either the CPU itself, or your heatsink/TIM placement. I think returning the CPU for another one was probably a good choice for good measure. Which thermal paste are you using by the way? (Not that it would itself attribute to these types of high temps solely)

 

And I believe that your PSU should suffice just fine. The RX480 looks to draw only about 170w peak out of your total.

[huti26]

I used the paste which came along the noctua.

 

I also used all those parts except for the motherboard and cpu & cooler on my "old" build, which had a ryzen 1600, an asus b350 instead and the stock cooler instead.

The old build never ran as hot as the new one, even when overclocking.

I guess if the next motherboard and cpu combination still runs hot, I might should check out, if the cooler is the problem. 

[JayRad]

17 hours ago, huti26 said:

I used the paste which came along the noctua.

 

I also used all those parts except for the motherboard and cpu & cooler on my "old" build, which had a ryzen 1600, an asus b350 instead and the stock cooler instead.

The old build never ran as hot as the new one, even when overclocking.

I guess if the next motherboard and cpu combination still runs hot, I might should check out, if the cooler is the problem. 

 

Yeah, see what the new replacements run like. This configuration should work quite well based on my experience with the same hardware. The Noctua you've got is a great cooler as well.

[huti26]

Alright, they didnt have 8600k in stock, so I got a 8700k instead & an Asrock z370M.

 

I didnt change anything so far, just booted and started prime95 and the temps you can see in the attached screenshot.

Arent these temps kind of too high for stock settings? The temps also keep very slowly but surely rising. I will upload a pic of how much thermal paste I used, just to make sure I didnt fail on that.

[huti26]

[WoodenMarker]

15 hours ago, huti26 said:

Alright, they didnt have 8600k in stock, so I got a 8700k instead & an Asrock z370M.

I didnt change anything so far, just booted and started prime95 and the temps you can see in the attached screenshot.

Arent these temps kind of too high for stock settings? The temps also keep very slowly but surely rising. I will upload a pic of how much thermal paste I used, just to make sure I didnt fail on that.

Yeah, that seems a bit high. Are you sure that the cooler is evenly mounted? Try reseating it and making sure to gradually tighten the screws a bit at a time as opposed to one side before the other.

How's your case airflow set up?

[huti26]

My Fan setup was the standard one of the case: 1 front fan for intake and 1 back fan for exhaust.

 

I was not happy with the noise of those fans, so I unplugged them for now and used the computer with an open case for today until my new fans arrive.

I let prime95 run again for 20mins with an open case and no case fans plugged in with the attached results.

Note that I now changed my cpu fan curve to the "silent" curve. On the previous one I was having the "standard" curve.

 

I will run the test again with the new fans later today.

 

[Dedayog]

I'd wager that with single fan intake and exhaust, coupled with that GPU... you've got too much heat in the case for the Noctua to handle.  It's getting heat soaked a bit.

 

That may be part of the issue, but probably not all.

 

With the case open, put a box fan or other house fan into it so the Noctua gets a whole lot of air and see how it cools the CPU.  That'll give you an idea of the Noctua is doing itsjob and it's the case that's the issue.

[huti26]

11 hours ago, jstudrawa said:

I'd wager that with single fan intake and exhaust, coupled with that GPU... you've got too much heat in the case for the Noctua to handle.  It's getting heat soaked a bit.

 

That may be part of the issue, but probably not all.

 

With the case open, put a box fan or other house fan into it so the Noctua gets a whole lot of air and see how it cools the CPU.  That'll give you an idea of the Noctua is doing itsjob and it's the case that's the issue.

I now reapplied thermal once more, reaseted the heatsink and made sure to do everything really as noctua advises you to. Attached are 2 Screenshots:

 

1. Closed Case, RPM limited to 1000.

2. Open Case, RPM permanently at 1500. Also I had a fan blowing at it, as you told me to.

 

The temperature difference is at only 4°. Is that how its supossed to be?

 

 

 

 

 

 

[Dedayog]

Others may be more versed in thermals, but it seems you're running at load at a decent temp.  If I am reading it right, you're pulling 130W from the CPU which is a lot?  So for that wattage draw, the temps are pretty good.

 

That said, I would have expected a little more disparity between closed and open+fan, to be honest.

 

Take this with a grain of salt, I'm no expert.

[huti26]

I undervolted the cpu now to 1,120V.

Prime 95 is at ~65° and fortnite as an example does hardly ever exceed 50° celsius. Im pretty happy with these results.

 

 

Sligthly off-topic, but I was running this system on stock settings with only xmp enabled and my computer completly crashed while playing fortnite.

Is it possible that the xmp was causing it?

[LogicalDrm]

6 hours ago, huti26 said:

Sligthly off-topic, but I was running this system on stock settings with only xmp enabled and my computer completly crashed while playing fortnite.

Is it possible that the xmp was causing it?

Not likely. You have some super-high voltages, probably PSU or something like that just wasn't up to task.

[huti26]

14 minutes ago, LoGiCalDrm said:

Not likely. You have some super-high voltages, probably PSU or something like that just wasn't up to task.

Which of my voltages are high? Only voltage I am knowledegeable about is the vcore.

[LogicalDrm]

13 minutes ago, huti26 said:

Which of my voltages are high? Only voltage I am knowledegeable about is the vcore.

The ones you set to 1.12V. I understood you were talking about something before you set manual volts (1.1V is actual stock volts). If it has crashed after you set new temps, it might not be stable at so low volts. XMP is automatic/semi-automatic process. If you think it might be issue, disable and see what happens.

[huti26]

Just now, LoGiCalDrm said:

The ones you set to 1.12V. I understood you were talking about something before you set manual volts (1.1V is actual stock volts). If it has crashed after you set new temps, it might not be stable at so low volts. XMP is automatic/semi-automatic process. If you think it might be issue, disable and see what happens.

I might have confused you there. Currently my vcore is at a fixed 1.12V and Im not overclocking my ram. With this setup I have been completly stable so far (played some games and ran prime95 for 45min).

 

At a different point of time, I had my vcore set on "auto" (which put it up to 1,28V as you can see in one of the previous screenshots) and I was trying to overclock my ram. When trying that, I crashed once while playing fortnite. After that I stopped overclocking my ram and I did not crash again so far.