• Now's your chance win big! Join our community and get entered to win a RTX 2060 GPU, plus more! Join here.

    Pi Cast Episode 3 streams live on Tuesday, August 4th at 2:30 pm ET (7:30 PM BST). Watch live right here!

    Catch Scharon on the Tom's Hardware Show live on Thursday, August 6th at 2:00 pm ET (7:00 PM BST). Click here!

Question CPU Usage Fluctuation, FPS Drops, and BSoD

Jun 27, 2020
3
0
10
0
tl;dr: my CPU usage is looking like a square wave and my performance is suffering big time.

For the past several weeks I have been experiencing strange frame drops in games and regular applications. I started monitoring my CPU on the Task Manager to see if there was something going on and I noticed semi-regular dips in CPU usage every so often. I left the monitor open on another screen and saw that the dips are directly connected to the drops in FPS. I ignored the issue for a bit until my PC started blue-screening. One would occur randomly, but would then open Pandora's Box and put me in a blue screen boot loop. At one point my PC booted into the BIOS screen telling me my overclock failed (I've never overclocked anything).

THINGS I'VE TRIED
-Updating my drivers. Got the newest AMD Radeon release and even flashed a BIOS update for my mobo.
-Checked all my temperatures. CPU and GPU and well within good temps and HWiNFO shows mobo is in good shape as well. I could get a screenshot if needed.
-Ran CrystalDisk to check state of drives. See results here
-Reset CMOS and disconnected a potentially problematic HDD. Did these two and one helped me get back on my computer after I got that overclock error & boot loop. The drive was only potentially problematic because its quite old and used to be my main drive before upgrading to an SSD. It saw basically no use but never caused trouble before and was in good health last I checked (~2 weeks ago).

Haven't blue screened since I reset the CMOS but not too much time has passed since I did this.

SPECS
CPU: AMD FX 8350 (no overclocks)
GPU: Stock Radeon RX 480 8gb from MSI
Mobo: ASUS M5A97 LE R2.0 (AMD 970 chipset with newest BIOS verison 2701)
RAM: Corsair XMS3 DDR3 16gb 4x4 1333MHz
Drives: Samsung 860 EVO 1TB SSD + SeaGate Barracuda 2TB HDD for bulk storage (disconnected drive not included here)
PSU: Corsair CX750M 80+ Bronze
OS: Windows 10 Home Premium Version 1909 OS Build 18363.900 (legit copy running latest update)

BSOD STOP CODES I GOT
WHEA_UNCORRECTABLE_ERROR
IRQL_GT_AT_SYSTEM_SERVICE
KERNEL_SECURITY_CHECK_FAILURE (x2)
REGISTRY_ERROR
REGISTRY_FILTER_DRIVER_EXCEPTION (what failed: farflt.sys)
CLOCK_WATCHDOG_TIMEOUT

(as a reminder I haven't Blue Screened since the CMOS reset and drive disconnect but not enough time has passed to rule anything out)

My main focus is to get the CPU usage back to normal and getting rid of these bad lag spikes.
Any help is appreciated and if any more info is needed please lmk. I hope the soluton isn't expensive.
 
Jan 9, 2020
53
4
45
2
Your motherboard VRM are surely throttling you down, do not force the system, can you put a 80+mm fan to blow air straight into the vrms? also undervolt a bit and disable all the turbo features. that will probably help you, also improving the case airflow will help a lot.
 
Jun 27, 2020
3
0
10
0
Your motherboard VRM are surely throttling you down, do not force the system, can you put a 80+mm fan to blow air straight into the vrms? also undervolt a bit and disable all the turbo features. that will probably help you, also improving the case airflow will help a lot.
Thanks so much for the reply. After your response I opened up HWiNFO and started monitoring the voltages. I've recorded a short video showing the relationship between the fluctuation and the Core VID voltage. Do you think your solution correlates with this? I'm not great with CPUs and Mobos so that's why I'm asking.
 
Jan 9, 2020
53
4
45
2
Yes, the problem there is straight related to the vrm temps when they are too hot, your motherboard drops the voltages down to prevent the MOSFETs from burning, so basically the motherboard forces a thermal throttle on the CPU to help the vrms to cool down, please try to put direct airflow to the vrms and report back if it helped by a bit (only a 980/990 chipset will probably fix this for good).

Also just for troubleshooting, you could try running the CPU with less cores, or at a lower speed like 3ghz, just to confirm that the vrms are the problem.

And I almost forgot, please check the CPU temps, that also could be causing that thermal throttle.
 
Jun 27, 2020
3
0
10
0
Yes, the problem there is straight related to the vrm temps when they are too hot, your motherboard drops the voltages down to prevent the MOSFETs from burning, so basically the motherboard forces a thermal throttle on the CPU to help the vrms to cool down, please try to put direct airflow to the vrms and report back if it helped by a bit (only a 980/990 chipset will probably fix this for good).

Also just for troubleshooting, you could try running the CPU with less cores, or at a lower speed like 3ghz, just to confirm that the vrms are the problem.

And I almost forgot, please check the CPU temps, that also could be causing that thermal throttle.
Yea CPU temps are fine at around 47°C average. It's just weird because the voltages are actually spiking and not dropping when the lag starts. I'll definitely see what I can do for airflow and underclock to make sure its the issue.
 
Last edited:

ASK THE COMMUNITY

TRENDING THREADS