My RTX 5080 and Ryzen 7 7800X3D get high FPS, but every game feels choppy. Capping at my monitor's 165 Hz refresh rate helps a little, and NVIDIA Smooth Motion helps more, but neither fixes it.
I've already reinstalled Windows, updated the BIOS and done a clean graphics driver installation. What should I check next?
Start with a frametime graph rather than the FPS counter. A high average can hide spikes, and an FPS number jumping around doesn't necessarily mean individual frames are arriving unevenly. Check whether the visible hitch matches a spike in frame delivery.
Try a repeatable scene with Smooth Motion off, then compare it with Smooth Motion on. The improvement is useful information, but it doesn't by itself identify a scheduling problem or any other cause.
Check that Windows is actually set to 165 Hz and that the monitor is plugged into the graphics card. If the display supports variable refresh and it's enabled, try a cap around 160 or 162 FPS rather than right at 165. You can also test V-Sync in the NVIDIA control panel and compare fullscreen with borderless where the game offers both.
Very high uncapped FPS can put more pressure on the CPU, but a CPU limit doesn't automatically explain bad pacing. Try a lower cap too and see whether the frametime spikes change.
Strip back the background stuff for a test. Close game trackers, overlays and other monitoring tools, then add them back one at a time if the stutter changes. A clean Windows install doesn't rule out something that was installed again afterward.
LatencyMon may also give you a lead on driver latency while the issue is happening. Treat its results as clues, not proof that a particular driver is the cause.
For BIOS troubleshooting, keep it simple. Leave CPU tuning on auto and compare memory with EXPO enabled and disabled. Change one setting at a time so you can tell what actually made a difference. I wouldn't start manually adjusting voltages just to chase an unexplained stutter.
Does it hitch only when moving the mouse, or also during steady movement without mouse input? If mouse movement makes it worse, try a lower polling rate as a quick test. That won't diagnose every kind of stutter, but it's an easy variable to check.
Answers are reviewed before publication.
9g__sprocketfile+31:
It's worth testing variable refresh on and off as well. If camera movement looks different between the two, that helps separate display behavior from actual frametime spikes.