Hi everyone,
I recently upgraded my GPU from an RTX 3060 12GB to an MSI RTX 5060 Ti 8GB Gaming Trio OC, and since then I’m experiencing noticeable microstutter / hitching, especially in Valorant and EA FC/FIFA.
My specs:
CPU: Ryzen 7 5800X
Motherboard: MSI MPG X570 Gaming Plus / MS-7C37
RAM: 32GB G.Skill DDR4 3600 MHz, XMP enabled
GPU: MSI RTX 5060 Ti 8GB Gaming Trio OC
Old GPU: RTX 3060 12GB
PSU: Gigabyte P850GM 850W Gold, fully modular
SSD: Samsung 970 EVO Plus 500GB NVMe
HDD: 2TB SATA HDD
Monitor: AOC 1440p 165Hz
OS: Windows 11 Home
BIOS: 7C37vAR1 / A.R1, 09/11/2025
The problem:
In Valorant, FPS is high, usually around 300+ FPS, and GPU render time is very low, around 0.5ms, but I see visible spikes in the Valorant performance graph. The spikes appear mainly in:
CPU Game Time
Total Frame Time
Game Latency CPU
GPU time stays low, so it does not look GPU-bound.
Things I already checked/tried:
- DDU clean NVIDIA driver install
- Tried clean driver setup without extra audio/overlay components
- Disabled NVIDIA HD Audio
- Disabled overlays
- Tested with Chrome/Discord closed
- Tested different FPS caps: 165 / 240 / 300
- OCCT CPU/GPU/RAM/Power tests: no errors
- HWiNFO PSU rails: +12V stable at 12.000V, +5V and +3.3V normal
- SSD health looks good, Samsung 970 EVO Plus at 96% life, no warnings
- Disk activity during stutter looks low
- RAM is running correctly at 3600 MHz
- ReBAR is enabled
- Secure Boot enabled
- TPM enabled
- HVCI / Memory Integrity enabled because Vanguard required it
- Page file is set to automatic
- BIOS updated to latest MSI beta BIOS with “TPM firmware for better game compatibility”
- Cleared TPM once
- SFC found and repaired some Windows files, but later scan/log looked clean
LatencyMon earlier showed some driver latency with:
nvlddmkm.sys
HDAudBus.sys
dxgkrnl.sys
Wdf01000.sys
But later LatencyMon looked green and said the system appears suitable for real-time audio.
One suspicious thing:
Event Viewer repeatedly shows TPM Event ID 14 even after clearing TPM:
The device driver for the Trusted Platform Module (TPM) encountered a non-recoverable error in the TPM hardware...
I also got TPM-WMI attestation errors. The measured boot JSON said TPM is present/responsive, but EkCertIsAvailable = false.
TPM PowerShell check shows:
TpmReady: True
ClearRequired: False
RestartRequired: False
So TPM is technically ready, but Event ID 14 still appears.
In HWiNFO / PresentMon, the stutter seems to match CPU-side frametime spikes, not GPU usage or VRAM. VRAM usage in Valorant is only around 3GB out of 8GB.
My question:
What would you check next for CPU Game Time spikes in Valorant/EA FC when the GPU is clearly not maxed out?
Could this be related to:
- AMD fTPM / TPM Event ID 14
- HVCI / VBS / Vanguard security stack
- BIOS CPU power-state behavior
- Windows scheduler / driver latency
- motherboard issue
- CPU issue
I’m trying to avoid randomly replacing parts because the PSU voltages, SSD health, RAM, temps, and OCCT stability all look fine.
Any advice from people who had Ryzen fTPM stutter, Valorant CPU Game Time spikes, or TPM Event ID 14 would be appreciated.