Computer slows down after running for about 15-20 minutes

Nibbows

Commendable
Jun 6, 2016
5
0
1,520
I've tried everything I can think of, including driver updates, uninstalling the last OS update, cleaning out the dust from my computer, and nothing has fixed the problem. It only started a few days ago. If I restart it, it works perfectly fine for a little while, and then it just becomes sluggish all at once. If I'm in a game, I just see the FPS go from 60 straight down to 10-15 (depends on the game) and it becomes unplayable. It even causes regular web browsing and simple computer operations to become sluggish. I'll post what information I can, and if any else is needed, I will do so upon request. I have no idea what fix to try next, or how to find out if I have some hardware going bad. Any help would be greatly appreciated.

OS: Windows 10 Pro 64-bit
Processor: Intel(R) Core(TM) i7-4790k CPU @ 4.00GHz
RAM: 16 GB
GPU: NVIDIA GeForce GTX 760
Motherboard: MSI Z97 GAMING 5 (MS-7917)
Storage: 512GB Samsung SSD 850 PRO
3TB Seagate SATA
 
Solution
Sooo... It ended up being the Windows 10 update after all... I had uninstalled the update which was supposed to fix the problem according to a google search I did, if that was what was causing it, but I just happened to come across a "restore previous windows build" option in the settings, and that completely reverted it back to the previous update. The problem is now gone. I should have just went with my gut and kept pursuing that as the problem. Thank you so much for the help anyways everybody. I was getting pretty worried I was going to have to reformat or buy a new piece of hardware.
First though is overheating somewhere. What are your temperature readings? Free download of HWMonitor will tell you if you don't have good software.

If turning the computer off and leaving it for half an hour or so restores the 'goodie', it strongly suggests some sort of thermal issue.
 

Nibbows

Commendable
Jun 6, 2016
5
0
1,520


I've watched the temps, and nothing really changes once it goes into "slow mode". I've asked some people about my temps, and everybody says they seem fine, but I'm not exactly that knowledgeable when it comes to temps. Restarting immediately fixes the problem temporarily, not after waiting a bit for anything to cool down.

It is currently running sluggish, and These are the temps.
CPU 40c
Motherboard 41c
GPU 64c
Storage 35c on SATA
37c on SSD

 
When you restart, does it take another 20 minutes to slow, or does it slow much faster (5 minutes or so)?

For a start, using the stock cooler, 40C is far too low for intense use or high-performance gaming on that CPU. (I have one)

What are your CPU frequencies and voltages? HWMonitor and/or CPU-Z can tell you.

There are other possibilities, but thermal issues is where we should start and confirm or eliminate.
 

Nibbows

Commendable
Jun 6, 2016
5
0
1,520


When I restart, it takes about the same amount of time for it to slow down, regardless of how long I wait between turning it back on.

Also, I apologize for not mentioning this before, but I have a Corsair water cooling unit, I believe it's the H100, although I'm not sure how to check exactly which one it is. It's been a couple years since I built this.

As for the voltages, it says VID (hoping that's the right one cause all the others say 0) 1.032 min and 1.037 max. I'm not sure where to find the frequencies though.
 
OK, those symptoms could still be thermal (or not) but they suggest cooer issues rather than CPU issues. To track this down, we are going to need a systematic approach and a lot of data. HWMonitor can tell you most of what we need to know at the moment. We need to know the processor speed, voltages, usage, and temperature when it runs fast, and we need to know how this changes, and in what order, when it slows down.

I'm assuming you have no big background tasks that kick in and start doing something when you are gaming and that you are not getting important parts of your work from an internet connection that might slow down.
 

Nibbows

Commendable
Jun 6, 2016
5
0
1,520


I just double checked, and it slows down even if I leave it alone doing nothing, without even starting a game, yet restarting immediately fixes the problem, for about the same amount of time every time.

I'll try to post what information I thought was relevant, but I'm not too savvy on all of this. I could post the whole thing of each one here, or wherever if it'd be more convenient. Just not sure the best way to go about sharing the data.

I noticed the Voltage 7 on the CPU VCORE is higher after the restart. Could it be a power supply issue or something? Just random guessing at this point, since I honestly have no idea anymore.

After restart:

Hardware monitor Nuvoton NCT6792
Voltage 2 3.39 Volts [0xD4] (AVCC)
Voltage 3 3.39 Volts [0xD4] (3VCC)
Voltage 7 1.04 Volts [0x82] (CPU VCORE)
Voltage 8 1.79 Volts [0x70] (VCCIN)
Voltage 9 1.50 Volts [0x5E] (DRAM)
Voltage 10 11.26 Volts [0x80] (+12V)
Voltage 11 5.16 Volts [0x81] (+5V)
Temperature 0 35°C (95°F) [0x23] (SYSTIN)
Temperature 3 41°C (105°F) [0x29] (TMPIN3)
Fan 1 2156 RPM [0x86C] (CPUFANIN)
Fan 3 1036 RPM [0x40C] (AUXFANIN1)

Hardware monitor ACPI
Temperature 0 28°C (82°F) [0xBC2] (TZ00)
Temperature 1 30°C (85°F) [0xBD6] (TZ01)

Hardware monitor Intel I/O
Clock Speed 0 399.91 MHz [0x18F] (Graphics)

Hardware monitor NVIDIA NVAPI
Voltage 0 0.98 Volts [0x3CF] (VIN0)
Power 0 8.00 pc (GPU)
Temperature 0 40°C (104°F) [0x28] (TMPIN0)
Fan 0 1410 RPM [0x582] (FANIN0)
Fan PWM 0 42 pc [0x2A] (FANPWMIN0)
Clock Speed 0 135.00 MHz [0x87] (Graphics)
Clock Speed 1 324.00 MHz [0x87] (Memory)

Temperature 0 42°C (107°F) [0x3A] (Core #0)
Temperature 1 42°C (107°F) [0x3A] (Package)
Power 0 31.98 W (Package)
Power 1 24.78 W (IA Cores)
Power 2 0.00 W (GT)
Power 3 7.20 W (Uncore)
Power 4 2.48 W (DRAM)
Voltage 0 1.04 Volts (VID)
Voltage 1 +0.00 Volts (IA Offset)
Voltage 2 +0.00 Volts (GT Offset)
Voltage 3 +0.00 Volts (LLC/Ring Offset)
Voltage 4 +0.00 Volts (System Agent Offset)

Slow idle:

Hardware monitor Nuvoton NCT6792
Voltage 2 3.36 Volts [0xD2] (AVCC)
Voltage 3 3.38 Volts [0xD3] (3VCC)
Voltage 7 0.72 Volts [0x5A] (CPU VCORE)
Voltage 8 1.79 Volts [0x70] (VCCIN)
Voltage 9 1.50 Volts [0x5E] (DRAM)
Voltage 10 11.18 Volts [0x7F] (+12V)
Voltage 11 5.12 Volts [0x80] (+5V)
Temperature 0 36°C (96°F) [0x24] (SYSTIN)
Temperature 3 42°C (107°F) [0x2A] (TMPIN3)
Fan 1 2160 RPM [0x870] (CPUFANIN)
Fan 3 999 RPM [0x3E7] (AUXFANIN1)

Hardware monitor ACPI
Temperature 0 28°C (82°F) [0xBC2] (TZ00)
Temperature 1 30°C (85°F) [0xBD6] (TZ01)

Hardware monitor Intel I/O
Clock Speed 0 600.07 MHz [0x258] (Graphics)

Hardware monitor NVIDIA NVAPI
Voltage 0 0.98 Volts [0x3CF] (VIN0)
Power 0 58.61 pc (GPU)
Temperature 0 59°C (138°F) [0x3B] (TMPIN0)
Fan 0 1770 RPM [0x6EA] (FANIN0)
Fan PWM 0 53 pc [0x35] (FANPWMIN0)
Clock Speed 0 1201.94 MHz [0x4B1] (Graphics)
Clock Speed 1 3004.68 MHz [0x4B1] (Memory)

Temperature 0 34°C (93°F) [0x42] (Core #0)
Temperature 1 38°C (100°F) [0x3E] (Package)
Power 0 24.12 W (Package)
Power 1 16.57 W (IA Cores)
Power 2 0.00 W (GT)
Power 3 7.54 W (Uncore)
Power 4 2.39 W (DRAM)
Voltage 0 1.04 Volts (VID)
Voltage 1 +0.00 Volts (IA Offset)
Voltage 2 +0.00 Volts (GT Offset)
Voltage 3 +0.00 Volts (LLC/Ring Offset)
Voltage 4 +0.00 Volts (System Agent Offset)

Slow in game:

Hardware monitor Nuvoton NCT6792
Voltage 2 3.36 Volts [0xD2] (AVCC)
Voltage 3 3.38 Volts [0xD3] (3VCC)
Voltage 7 0.76 Volts [0x5F] (CPU VCORE)
Voltage 8 1.79 Volts [0x70] (VCCIN)
Voltage 9 1.50 Volts [0x5E] (DRAM)
Voltage 10 11.18 Volts [0x7F] (+12V)
Voltage 11 5.12 Volts [0x80] (+5V)
Temperature 0 35°C (95°F) [0x23] (SYSTIN)
Temperature 3 39°C (102°F) [0x27] (TMPIN3)
Fan 1 2160 RPM [0x870] (CPUFANIN)
Fan 3 996 RPM [0x3E4] (AUXFANIN1)

Hardware monitor ACPI
Temperature 0 28°C (82°F) [0xBC2] (TZ00)
Temperature 1 30°C (85°F) [0xBD6] (TZ01)

Hardware monitor Intel I/O
Clock Speed 0 599.86 MHz [0x257] (Graphics)

Hardware monitor NVIDIA NVAPI
Voltage 0 0.98 Volts [0x3CF] (VIN0)
Power 0 61.32 pc (GPU)
Temperature 0 61°C (141°F) [0x3D] (TMPIN0)
Fan 0 1740 RPM [0x6CC] (FANIN0)
Fan PWM 0 53 pc [0x35] (FANPWMIN0)
Clock Speed 0 1201.94 MHz [0x4B1] (Graphics)
Clock Speed 1 3004.68 MHz [0x4B1] (Memory)

Temperature 0 35°C (95°F) [0x41] (Core #0)
Temperature 1 40°C (104°F) [0x3C] (Package)
Power 0 29.02 W (Package)
Power 1 21.52 W (IA Cores)
Power 2 0.01 W (GT)
Power 3 7.50 W (Uncore)
Power 4 2.49 W (DRAM)
Voltage 0 1.04 Volts (VID)
Voltage 1 +0.00 Volts (IA Offset)
Voltage 2 +0.00 Volts (GT Offset)
Voltage 3 +0.00 Volts (LLC/Ring Offset)
Voltage 4 +0.00 Volts (System Agent Offset)
 

PreevBR

Commendable
Feb 22, 2016
163
0
1,710


Not sure of the clocks, you said in 'slow idle' but the GPU usage is high (Or atleast doing something in 3d, see its clocks) and at the same time the CPU vCore drops and the temperatures stay the same (Apart from GPU which gets to 60°C).
Could you try downloading another hardware monitor program like Open Hardware Monitor and manually checking CPU vCore, CPU Clocks and CPU usages while on idle and while its getting slow.
 
I too agree that you should check the information with another piece of software.

Failure at the same time, independent of load, still suggests a thermal issue, but one in an unusual place, like memory or the PSU.

There is also the possibility of a PSU issue, but we need data to zero in on the correct target.
 
Have you ruled out software issues? I bring it up because I just had a similar issue due to my antivirus software getting stuck. About 10 minutes after every reboot it would start background scanning the winsxs (win update) folder. It wasn't a lot of bandwidth, but thousands of seek requests hammered the drive and brought everything to a crawl. While that shouldn't faze a SSD at all, enough hardware interrupts would still slow the system.

Just to rule it out, can you open Task Manager and Resource Monitor and check the Disk tab when it's going slow? See if anything unusual is running, or creating a lot of activity. Or if any process loads right as it slows down, which could indicate some form of malware.
 

Nibbows

Commendable
Jun 6, 2016
5
0
1,520
Sooo... It ended up being the Windows 10 update after all... I had uninstalled the update which was supposed to fix the problem according to a google search I did, if that was what was causing it, but I just happened to come across a "restore previous windows build" option in the settings, and that completely reverted it back to the previous update. The problem is now gone. I should have just went with my gut and kept pursuing that as the problem. Thank you so much for the help anyways everybody. I was getting pretty worried I was going to have to reformat or buy a new piece of hardware.
 
Solution