PC shutting down while playing higher demanding games.

In the last few days I have noticed my PC would shut down in a instant when I'm playing higher demanding games (Crysis 3, Witcher 3 etc) from anywhere ranged 10 - 40 minutes of gameplay.

This doesn't occur if I'm playing SIMS 3, counter strike GO, Bioshock, Skyrim etc or any older game that isn't graphically intensive.

Could the issue be my current CPU or GPU? The AMD processor does have turbo boost and I did noticed the clock speed would go higher, I can hear the fans get louder as well. I know that my video card is drawing power from the motherboard itself so I'm not sure what else could cause the random shutting down.

Any suggestions would be appreciated.

check to see if the temps on cpu, system etc become too high over time with intensive tasks , or gpu, cpu oc'd too unstable, or check total power of components requirements vs power supply rating.

Power off means overheating or power issue.

Make sure temps are not exceeding 65C on the CPU (ideally it should be under 60C) and not exceeding 90C on the GPU (ideally it should be 80C or under, upwards of 85C is okay, but lower is better).

You can use HWMonitor to help check temps if you don't already have programs to do so for your GPU and CPU.

Also, have you cleaned out the fans/vents on the computer case any time soon? Build up of dirt/dust/gunk can hinder air flow and reduce the cooling on the CPU and GPU.

After some testing the only number that climbed too high was my CPU, it would push close to 4.4Ghz. I disabled the turbo boost via AMD software and inside the system settings > processor power managements I lowered the maximum and minimum to 99% processor state each and it lowered my CPU down to 3.00- 3.2GHz steady doesn't rise or fall. That seems to solved the issue none of my games caused the shutdown.

However I effectively lowered my CPU power for stability what do I need in order to push it back to its normal 100%? Larger power supply with better connectors and cooling?

@neatfeatguy: That will be my next agenda when I get a moment to do it. It's been about 2-3 months since the last time I cleaned it from inside the fans. I haven't done the HWMonitor as of yet I will post the results once I get some time do it. Thank you for the advice.

FX 6300 should only turbo to 4.1GHz. If it's hitting 4.4 then it's possible the turbo limit was increased in the BIOS. I would suggest resetting the BIOS. You can do this from the BIOS or when you're computer is off, you can remove the CMOS battery (leave it out for about 30 seconds) on the motherboard or there may be a CMOS reset button on the motherboard you can use.

What you need to do is pinpoint where the problem is - heat or power.

AMD Phenom II and FX series processors have a thermal limit of 62C (give or take a couple degree). Once they hit that 62C area, the CPU will start to throttle and as the temps get higher the CPU will slow down and eventually the system will power off to try and prevent damage if the temps get too high.

If the issue is power it could be happening during the turbo boost when your CPU is pulling max power draw. Bad/failing PSU or outlet the computer is plugged into isn't providing clean power, this could be causing the system to power off as well.

HWMonitor will also list your power on the +12V rail - this should be right around 12V. It will also list your 5V and 3.3V. If you see a large discrepancy in the voltage, it could very well be the PSU that's at fault.

set your power settings in windows to max.

enter your bios when your pc starts and select the following load optimal/performance defaults, save and exit. this should set the amd cpu cool and quiet function, c-state function and cpu and system fan speed to adjust automatically as necessary. this will allow for auto system oc, while preserving system heat management function. your can also load power saving default if you wish

other than that make sure their is enough power for the components and system ventilation is adequate.

@Sokol4ever: Actually from what I read in your specs, you have no aftermarket cooler.

With FX6300 going to 4,4GHz, it would be better to have at least a non-aftermarket cooler.

It could be a heating issue. I figured I won't need additional cooler since it only uses 95watts of power but you could very well be right. Once I get home from work I will pinpoint where the issue is with HWMonitor. I did managed to manually in the power settings adjust my CPU to 99% for max and minimum processor state and that had been keeping the Ghz at 3.2 steady. I have a feeling I will need a larger case with better connectors and cooling. Especially since I will have either a new Pascal card or GTX 970 down the road once money permits.

I will post the results later today.

Running the HWMonitor at the moment. This is what comes up.

If my eyes are reading your screenshot correctly, the temp listed under your FX 6300 is very high. 80C is hot for an AMD chip.

I just ran my work PC through Intel Burn Test and the i3 on her never went over 55C according to hwmonitor.

You have 1 of 2 things going on.

1. you have a bad temp sensor and the system isn't really running the CPU at 80C as of your screenshot (most likely this isn't the case)
2. you have very poor cooling for the CPU

If you're running the stock heatsink for your CPU, I would suggest you look into a better one. Constant high temps and shutdowns can very likely damage your system.

You should be reading the temps off AMD overdrive. Recommended by the people who actually make the processors, and 125W can be in the 60's Celsius and 95W can be in the 70's Celsius. And again all from AMD, they also go by core temp.

@Sokol4ever: As I thought, you should have problem with the cooling.

You should try using an aftermarket cooler. You will not believe in your eyes the temp difference a just $20-30 cooler can do...

Like @neatfeatguy said, with so many shutdowns you might damage your system.

Also I noticed one thing more.

On the left its your system and on the right its my system.

So, what's wrong with your 3,3V, 5V and 12V? The numbers are way off.

Look at my system's numbers. They should be like this.

I hope this is just a faulty sensor and nothing more or it might be the source of the problem.

Also, I see two sensors on the motherboard. TMPIN0 is your CPU (probably next to socket sensor) temp and TMPIN1 is your motherboard sensor. They seem normal comparing to how much heat your CPU emits...

ps: Although I disabled energy savings from Windows, I should have 3,5GHz min on 4 cores not, 3,7GHz. Time to visit BIOS again... xD

@Coseniath said:

@Sokol4ever: As I thought, you should have problem with the cooling.

You should try using an aftermarket cooler. You will not believe in your eyes the temp difference a just $20-30 cooler can do...

Like @neatfeatguy said, with so many shutdowns you might damage your system.

Also I noticed one thing more.

On the left its your system and on the right its my system.

So, what's wrong with your 3,3V, 5V and 12V? The numbers are way off.

Look at my system's numbers. They should be like this.

I hope this is just a faulty sensor and nothing more or it might be the source of the problem.

Also, I see two sensors on the motherboard. TMPIN0 is your CPU (probably next to socket sensor) temp and TMPIN1 is your motherboard sensor. They seem normal comparing to how much heat your CPU emits...

ps: Although I disabled energy savings from Windows, I should have 3,5GHz min on 4 cores not, 3,7GHz. Time to visit BIOS again... xD

I also noticed the low readings on the voltage, but not sure what to think of it. It could be a mix of things - failing PSU and overheating CPU.

I'd strongly suggest to look in your BIOS at the voltage readouts to see what it shows in there for the 12V and so on. On my PC at home when I was in the BIOS the other day I checked what the computer was showing for voltage on my trip through settings and 12V showed it was right around 12.09 (like the screen shot of Coseniath's voltage readings).

I'd first start with fixing your heat issue - clearly 80C is running hot on your CPU and is the most likely reason for shut downs, especially when you get into heavy use (gaming/benching). Once your heat issue is resolved, if the problem persists then your next step is to get a new PSU.

Or get both - aftermarket heatsink and PSU so you tackle both things at once. If your system has a failing PSU it can cause damage or if you're constantly overheating it can cause damage - each one will result in catastrophic failure on the CPU and/or motherboard or even more.

Quick story about a failed PSU: My younger brother had my old hardware except he got his own PSU about 7 years back. The PSU was rather cheap and I told him he shouldn't have gotten it, but he insisted it was good enough since our step-dad (knows some stuff about computers) said it would work for him. My brother tucked his tower into the shelf-opening area that the tower just fit into perfectly so the air flow was greatly restricted. After about 2 years his PSU overheated and failed - it took the CPU, MB, RAM and his HDD with it.

I re-tested after blowing some compressed air and it got little better but not much at all. CPU will still push to 80C after 30 minutes workload on heavy gaming.

At this point I will be getting aftermarket heatsink and PSU.

Thank you for all the help folks, as always extremely helpful. :)