Find a stable graphics setup and work out which change actually helps your PC. Start with your symptom, then compare the same scene with one setting changed.
Reviewed 2026-10-08. Official patch references through Buddy System Hotfix 2. The comparisons below are troubleshooting advice, not a hardware benchmark.
Compare the same scene at a lower resolution or a lower render-scale preset, using only options your game exposes.
What the result tells you: A clear improvement points toward rendering load. Little change means you should also check CPU load, power limits and background work; it does not identify a single cause.
Repeat the same short route, then compare a separate session. Note whether the hitch repeats at the same location.
What the result tells you: An improved second pass is useful diagnostic information, but it cannot by itself distinguish shader work, asset streaming or another transient load.
Compare with Frame Generation off, keeping resolution and upscaling unchanged.
What the result tells you: Generated frames can make motion look smoother without the same increase in game simulation or input updates. Record both responsiveness and the counter.
Install the current update, then change one of these toggles at a time and repeat the same scene.
What the result tells you: EA 1.2 specifically fixes Reflex / Frame Generation conflicts and particle flicker with VSync plus Reflex. A remaining issue needs the exact toggle combination.
Compare another location in that save, then a separate new save on the same branch. Keep your original save.
What the result tells you: A difference narrows the report to the location or save state. Hotfix 1 removes leftover base data that contributed to save bloat, but does not promise a particular FPS gain.
Record the game build and driver version, then compare the same scene and graphics settings in a separate session.
What the result tells you: Unknown Worlds notes that outdated third-party modifications can cause crashes or unexpected behavior after a hotfix; that warning does not establish the cause of every performance change.
If the game crashes or never reaches a playable scene, use the crash and black-screen checklist first. Co-op connection failures belong in the co-op guide.
Choose a short route you can repeat in the same save. Record the starting settings, change one variable, then compare. Available options depend on the hardware and game build.
Keep the output resolution fixed when comparing upscalers. Change one preset, then replay the same route.
Compare: FPS, image clarity in motion and tool or foliage artifacts.
Start with a lower preset to establish a stable baseline. Raise it one level at a time once that baseline is usable.
Compare: Frame consistency as well as average FPS; record the actual preset name.
After the preset comparison, test one available option at a time. Keep screenshots of your original settings.
Compare: Whether the visual improvement is worth its measured cost on your hardware. This page does not assign unmeasured FPS costs to individual settings.
Establish a baseline with it off, then test a supported mode. RTX 5000-series Dynamic / 5x / 6x modes are covered below.
Compare: Responsiveness, image artifacts and stability alongside displayed FPS.
Record both states and whether Frame Generation uses a fixed multiplier or Dynamic mode. If a combination misbehaves, change one toggle and restart if requested.
Compare: Tearing, latency and repeatable flicker. Hotfix 2 disables VSync when DLSS Dynamic FrameGen is selected.
Use a consistent power mode and close optional recording tools or overlays for one comparison.
Compare: Whether the problem follows a background application or a power-state change. This is a diagnostic comparison, not a confirmed game-specific fix.
When testing Frame Generation, write down the multiplier and whether the FPS overlay includes generated frames. A larger displayed number alone does not prove lower input latency or a faster native render rate.
The shared requirements below match our PC requirements page. Minimum and recommended labels are hardware baselines; we have not measured a guaranteed FPS result for either tier.
For drivers, use the GPU manufacturer’s official release and record its version. If a problem began with a driver change, that timing belongs in the report; it does not establish which version is best for every player.
Copy this checklist into your notes before changing settings. Two results are only comparable when you know what changed between them.
Use the bug report guide if a problem repeats. EA 1.2 also lets the in-game feedback tool attach a save file, which can help the developer reproduce a save-specific issue.
Start with a lower graphics preset and compare the same route at your chosen resolution. Establish performance with Frame Generation off, then test available upscaling and Frame Generation options one at a time. There is no verified universal best-settings preset or GPU benchmark table on this page.
The published minimum and recommended requirements do not establish a 60 FPS result for your resolution, save or build. Compare your PC with the requirements, then measure a repeatable route. Compare frame rate on the same route and graphics settings.
Yes. Unknown Worlds’ Hotfix 3 notes say DLSS was upgraded to 4.5, with fixes for crashes, settings not saving and Frame Generation being unavailable in some game versions.
The Buddy System notes identify Dynamic Frame Generation and the 5x / 6x modes as RTX 5000-series features. They are not promised for every RTX GPU. Check the GPU, driver, game build and the options actually exposed in your settings menu.
Buddy System Hotfix 2, released October 6, explicitly fixes VSync so it is disabled when DLSS Dynamic FrameGen is selected. EA 1.2 had added VSync compatibility with DLSS Frame Generation generally; Dynamic mode now has this specific behavior. Record the Frame Generation mode when comparing settings.
No. Record when it happens and whether it repeats on a second pass or after restarting. Shader compilation is one possible explanation; recurring stutter still warrants troubleshooting. Persistent, reproducible stutter is useful feedback.
Diagnose the bottleneck first. A lower-resolution comparison, storage activity, memory use and the same route on a separate save provide more useful information than the symptom alone. Use those comparisons to identify which component is limiting the game before choosing an upgrade.
The method of comparing one setting at a time still helps, but Windows GPU drivers and Nvidia-only Frame Generation features are not Steam Deck instructions. Check the separate Steam Deck page and the current store compatibility information for handheld-specific context.
Compare your actual CPU, GPU, RAM and DirectX support with the published baseline.
Follow the startup, save, branch and storefront diagnosis if you cannot reach normal play.
Keep handheld compatibility separate from Windows and Nvidia settings.
Turn a repeatable hitch or graphics failure into a useful report.