CPU over heating.

[tipsy]

I'm thinking of taking my PC  cabinet to somebody, who can look into my problem. But it's around 48 degrees outside. So I thought to share my problem here first. I used to have 16GBX2 RAM installed. But currently I'm using 16GBX1, but that ain't helping either. 

 

Issue:-

 

1) If I use two RAM 16GBX2, I sometimes get no display problem.  Yes, I don't put both RAM one after another,  like 1a then 2b channel . 

 

2) After using 16GBX1, everything runs ok but if I play "Horizon Forbidden West", I get lag and PC shuts down or...... no display with the fan and CPU still running.

 

3)  I also get a warning "CPU overheating. Check if cpu is attached properly. (or something like that, will write it down next time) .

 

My PC specs:-

 

Processor = i9 12900k

MB= z690-A pro

RAM = DDR5 16GB ( I have spare too)

GPU =  RTX3060 .   When I opened cpu-z to check for sys info, it showed UHD Graphics 770  Perf level: "Current".  Then I have to select RTX 3060 from the dropdown. Is this normal? 

PSU= 750watss corsair 

 

I will check CPU cooler later and edit it here. It's from MSI though with 3 additional fans attached.

 

All parts are new. I can take pic of the cabinet from inside for more clarification if needed. 

 

 

 

 

@EVIL BUTTERS

 

 

 

 

 

 

 

Edited May 21 by tipsy

[Snuffs99]

I'd turn the UHD integrated gfx off if possible, just not needed if you have a 3060, if it is being used by your system as your primary display instead of your 3060 then there is a good chance its the integrated gfx causing the overheating??.

 

Check in device manager and then "disable device" or do it via bios, should have an option regarding integrated/primary gfx etc.

 

If all is as it should be and your still having shutdowns would be nice to know the CPu temp at shutdown so something like hwinfo etc can give a good real time temp of CPU, GFx etc etc.

 

 

 

 

 

Edited May 21 by Snuffs99

[Element']

Like Snuffs said I’d monitor the CPU temp while you’re gaming or even at idle using CoreTemp or something similar: https://www.alcpu.com/CoreTemp/

 

How old do you think the thermal paste is? 

 

PC semi clean on the inside? 

 

Some pics wouldn’t hurt.

[Aft3rmath]

If you're using an AIO CPU cooler, you may need to adjust the fan and pump settings in the bios to enable higher speeds.  By default, bios had mine set up for normal fans in "quiet mode" instead of a water pump.  My 7900X runs at about 65°C and the output air from my AIO is pretty hot.  If your output air isnt noticably hotter than the ambient air, I would guess that there is either an air flow or water flow issue.

[Lt Steiner]

1. Going by your description I do not think that your RAM is at fault (as you get the system crashes regardless of RAM configuration). But to make sure you can test your RAM easily with Memtest 86. Just put it on a USB flash drive and boot from said flash drive.
2. Does the screen just turn dark or do you get a Blue Screen of Death or any other message? Do you find anything about the crashes in the Eventviewer? To open the Eventviewer, press the Windows-Key+R, type in eventvwr, and press Enter. Windows usually puts there reports about any critical system error that has happened.
3. CPU-Z shows you at what frequency your card runs. "Current mode" is the current frequency in idle. If you hover over it with your mouse, it should afaik also tell you at what frequency the card would run in boost mode.

It sounds a lot like an overheating issue to me. You should check both the CPU's as well as the GPU's temperatures in idle and under load. For the former, I can recommend using CoreTemp for reading the temperatures and Prime95 for creating the load. For the latter I can recommend using GPU-Z for reading the temperatures and Furmark for creating the stress.

 

Because of the error message you have received and because the parts are all new, my guess is that the CPU cooler is not properly attached to the CPU socket. This can happen when the person building the PC uses the wrong set of screws for the type of socket. Badly applied thermal paste is probably not the culprit as it would have to be applied really badly to cause such grave issues (not in the sense of too little, but either way, way too much or no thermal paste at all) and is unlikely as coolers usually come with thermal paste already pre-applied nowadays. This would then explain why removing one of the DIMMs (RAM), and thus creating better air flow, helps alleviate the issue and why crashes are more likely when you tax the CPU with a modern game.

[tipsy]

@Snuffs99  I have turned UHD off.

 

@Element'  Horizon forbidden runs in full screen mode, I'm not sure how to check for temp. while in game.   ET or any other app running fine. though. Yeah, I need something to have log of  CPU temp. 

 

@Aft3rmath  When my system got ready, it used to give out hot air. But yea.. currently there isn't much hot air that I can feel. No water cooling in my system.

 

@Lt Steiner  Last time I tried that.. my USB got f***ed up and had to erase all data. I think RAM is fine.  I will go through ur post and post the result tomorrow.

 

Thanks everyone for help. 

 

 

[EVIL BUTTERS]

4 hours ago, tipsy said:

I'm thinking of taking my PC  cabinet to somebody, who can look into my problem. But it's around 48 degrees outside. So I thought to share my problem here first. I used to have 16GBX2 RAM installed. But currently I'm using 16GBX1, but that ain't helping either. 

 

Issue:-

 

1) If I use two RAM 16GBX2, I sometimes get no display problem.  Yes, I don't put both RAM one after another,  like 1a then 2b channel . 

 

2) After using 16GBX1, everything runs ok but if I play "Horizon Forbidden West", I get lag and PC shuts down or...... no display with the fan and CPU still running.

 

3)  I also get a warning "CPU overheating. Check if cpu is attached properly. (or something like that, will write it down next time) .

 

My PC specs:-

 

Processor = i9 12900k

MB= z690-A pro

RAM = DDR5 16GB ( I have spare too)

GPU =  RTX3060 .   When I opened cpu-z to check for sys info, it showed UHD Graphics 770  Perf level: "Current".  Then I have to select RTX 3060 from the dropdown. Is this normal? 

PSU= 750watss corsair 

 

I will check CPU cooler later and edit it here. It's from MSI though with 3 additional fans attached.

 

All parts are new. I can take pic of the cabinet from inside for more clarification if needed. 

 

 

 

 

@EVIL BUTTERS

 

 

 

 

 

 

 

look in task manager and check temps run a stress program and see real time temps.  Core temp 1.18.1 is a good tracker for your temps .  Did you put Thermal paste down?  LOL  Did you take off the plastic film on the Aio??  SUCK ITTTTT

[Betinho]

There are many ways to see where your CPU temperature problem is coming from:

GPU:
1) Check the NVIDIA control panel settings. You should make sure that you use the best performance, more than the graphics, you will play better and your CPU will not heat up.

The game itself:
2) Check the ET resolution (EJ I play with 1024 x 768), why do you need more?

Processor:
3) Every 6 months you can change your thermal paste, in case you see repeated heating in your processor.

 

Here I leave the NVIDIA configuration, which helps me play with more performance, taking away the quality of the graphics of course. Try this along with also reducing the resolution of the game. This helps a lot to lower the temperature of the game.

 

 

[Dimo4ka]

Hello @tipsy

1) first of all, open the BIOS before starting the PC, any BIOS has the voltage for the PC components and the temperature in idle mode. Also check the voltage, there may be a problem with the power supply. You need to find the characteristics of your components and look at the permissible temperature values at idle and under load.

2) if everything is in order in the BIOS and the temperatures are within the permissible idle mode values, download any stress test program that loads your components. There can be a lot of problems with overheating, poor contact of the component with the cooler, the cooler could move away from the mount due to vibrations

3) the computer may reboot due to the video card, if it has problems, if there was overheating, it begins to slowly fall off, with the processor, in addition to the fact that the CPU itself may overheat, the processor power zone may overheat, your PC is very powerful, so pay attention and to this. I advise you to replace the thermal paste and check the ventilation of the case and the correct placement of coolers

4) ideally you need to look at the error code why it turns off, when the BSOD screen comes up, then an error is written there, the error code indicates the reason why it went into reboot, if it’s overheating, then which component is overheating

[*Kiba*]

6 hours ago, Dimo4ka said:

Hello @tipsy

1) first of all, open the BIOS before starting the PC, any BIOS has the voltage for the PC components and the temperature in idle mode. Also check the voltage, there may be a problem with the power supply. You need to find the characteristics of your components and look at the permissible temperature values at idle and under load.

2) if everything is in order in the BIOS and the temperatures are within the permissible idle mode values, download any stress test program that loads your components. There can be a lot of problems with overheating, poor contact of the component with the cooler, the cooler could move away from the mount due to vibrations

3) the computer may reboot due to the video card, if it has problems, if there was overheating, it begins to slowly fall off, with the processor, in addition to the fact that the CPU itself may overheat, the processor power zone may overheat, your PC is very powerful, so pay attention and to this. I advise you to replace the thermal paste and check the ventilation of the case and the correct placement of coolers

4) ideally you need to look at the error code why it turns off, when the BSOD screen comes up, then an error is written there, the error code indicates the reason why it went into reboot, if it’s overheating, then which component is overheating

it's good thing i read walls of text he saying pretty much what i was going to say it could just be bad  thermal paste it would be good to do some stress test to  any good  program out there for that will work just fine like  benchmark i myself don't have money to replace parts that break so i don't really mess around with them but there is a number of youtube videos on how it works and what you need to do  

[tipsy]

For those who have missed it, I like to clarify...   My PC runs fine anytime, except when I play the game "Horizon West Zero".  ET  is running smoothly with 125 fps and so do other applications. The problem shows up after a few minutes. of playing "HFW"  and I believe if I play any other game that uses heavy resources, my CPU will heat up. 

 

Again thanks for the help, I will go with solutions step by step. Currently, I have just disabled the default GPU on the CPU, that didn't help though. 

[Dimo4ka]

1 hour ago, tipsy said:

For those who have missed it, I like to clarify...   My PC runs fine anytime, except when I play the game "Horizon West Zero".  ET  is running smoothly with 125 fps and so do other applications. The problem shows up after a few minutes. of playing "HFW"  and I believe if I play any other game that uses heavy resources, my CPU will heat up. 

 

Again thanks for the help, I will go with solutions step by step. Currently, I have just disabled the default GPU on the CPU, that didn't help though. 

HWF consumes much more computer resources, you need to find the component that is overheating and simply fix this problem

[Vice86]

I had the same problem with my stock Laptop. But the mobo throttled the CPU speed, to a value where it could basically run windows without failing.

turns out, that the heatsink wasn't properly attached with proper heat paste by default, from Acer manufacturing.

I then renewed the heat paste, and it was just fine eversince.

 

maybe try to renew the heatpaste, and maybe get a better fancooler, especially if you consider your outside temperatures. 

when distributing the heat paste, most people do it wrong, and the CPU develops hotspots which inturn develops a chain reaction when under load. 

I don't even use the X-mode when distributing, I do more like a thin layer, like buttering a toast. 

this will require more effort, but will be rewarding. been doing it like that since my first PC in 1998. 

I ve never had unevenly distributed heatpaste. but gotta be extra careful to not use too much, otherwise it will be counterproductive.

 

 

best heatpaste is CTG8 in my personal experience, but I didn't test around much since the MX-4. 

Default CPU-Stock coolers must always be replaced. there is no option. 

 

[tipsy]

44 minutes ago, Dimo4ka said:

HWF consumes much more computer resources, you need to find the component that is overheating and simply fix this problem

 

Yes, for now I have tried AIDA64 to do stress testing and it shows CPU overheating. I need to try thermal pasting again and re-installation of cooling fan and see how ventilation goes. 

 

P.S.  I'm trying to insert image, I can't see option. the fk is wrong with me lol. 

 

oh N.M , I figured it out ..  "You have used 189.73 MB of your 97.66 MB attachment limit."     @daredevil

 

Edited May 22 by tipsy

[-=HipKat=-]

1 hour ago, tipsy said:

 

Yes, for now I have tried AIDA64 to do stress testing and it shows CPU overheating. I need to try thermal pasting again and re-installation of cooling fan and see how ventilation goes. 

 

P.S.  I'm trying to insert image, I can't see option. the fk is wrong with me lol. 

 

oh N.M , I figured it out ..  "You have used 189.73 MB of your 97.66 MB attachment limit."     @daredevil

 

Other than Thermal paste, which you said a few weeks ago is newly replaced, have you checked on the airflow of your fans through the case?? I saw on JayzTwoCents on YT last week that a lot of people get the fans reversed on their CPU Cooler, particularly

Edited May 22 by -=HipKat=-