iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
There is no universal best-settings preset for every game and PC. The reliable way to improve FPS without making the scene unreadable is to measure a repeatable gameplay moment, determine whether the GPU or CPU is limiting performance, change one setting at a time, and judge both frame-time behavior and moving-image clarity. Because the game name is still written as [Game], exact menu labels, recommended values, visibility-critical effects, patch behavior, and upscaler support cannot be stated responsibly. Use the method below, then apply the game’s own labels and documentation.
Start with a baseline you can repeat
Settings changes are useful only when you can tell what changed. Do not begin by lowering every option or selecting the lowest preset. First create a baseline in the same game mode, location, camera direction, and display resolution that you can reproduce.
Choose a demanding gameplay scene
Use a real section of play rather than a static menu or an empty room. A crowded fight, busy town, large outdoor view, weather effect, or area with many visible objects is more useful because it exposes the limitation that matters during play. Repeat the same route or encounter for at least 30 seconds. If the game has a built-in benchmark, use it consistently, but also check an actual gameplay scene because a benchmark may not contain the visibility problems you care about.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteRecord more than the average FPS
Write down the display resolution, render scale, graphics preset, upscaler mode, frame-generation state, and whether the game is running in exclusive full-screen, borderless, or windowed mode. Record average FPS and the behavior of frame times if your overlay provides it. A high average can hide uneven delivery: brief stalls and large frame-time spikes are often more noticeable than a small change in the average.
#1 Best Overall
- Powered by Radeon RX 9070 XT
- WINDFORCE Cooling System
- Hawk Fan
- Server-grade Thermal Conductive Gel
- RGB Lighting
- Keep the test scene, camera movement, and run length constant.
- Change only one setting or one tightly related feature group per pass.
- Capture a short clip or screenshot while moving the camera so you can inspect shimmering, ghosting, blur, and lost detail.
- Return to the baseline if a change is not clearly beneficial.
Find the bottleneck before lowering settings
A graphics setting can help only when it reduces work on the component that is limiting the frame rate. NVIDIA’s guidance distinguishes a GPU-bound case, where a faster graphics card can raise frame rate, from a CPU-bound case, where a faster GPU may not increase FPS. The same distinction determines which settings are worth testing.
Signs of a GPU-bound run
If GPU utilization remains high while one or more CPU cores have spare capacity, reducing internal resolution, render scale, ray-traced effects, shadows, reflections, volumetric effects, or other expensive rendering features is a sensible test. The useful signal is a repeatable reduction in GPU load accompanied by better frame times, not a particular utilization percentage.
Signs of a CPU-bound run
If the GPU is not fully occupied while the frame rate stalls in crowded scenes, during heavy simulation, or near a fixed ceiling, lowering image-quality options may produce little improvement. View distance, crowd density, physics, world simulation, and other game-specific controls can affect CPU work, but their names and impact differ by title. A graphics-card upgrade is not a reliable remedy for a CPU-bound result.
Free tools Windows power users keep installed
One-click scans. No signup required.
Use a controlled A/B test
| Observation | What to test next | What not to assume |
|---|---|---|
| GPU load is high and lowering a visual effect improves frame time | Continue testing GPU-heavy options and render resolution | That every visual option will scale the same way |
| GPU load falls but FPS barely changes | Check CPU utilization, per-core activity, simulation settings, and frame caps | That the graphics card is necessarily defective |
| Average FPS rises but frame-time spikes remain | Compare the 95th-percentile or worst visible frame-time behavior and investigate CPU or streaming stalls | That a higher average automatically feels smoother |
| The image becomes harder to read while moving | Restore the visibility-critical option and test a different setting | That a measurable FPS gain is worth any clarity loss |
Use this tuning loop for every setting
- Save the baseline. Note all relevant settings and capture the same moving scene.
- Choose one candidate. Start with a setting that plausibly addresses the diagnosed bottleneck, not the first option in the menu.
- Make a moderate change. Move one quality level, reduce render scale by a clearly measurable step, or switch between the game’s documented upscaler modes. Avoid changing several controls at once.
- Repeat the exact scene. Use the same route, camera motion, and run time.
- Check frame rate and frame times. Keep the change only if it improves the behavior you care about, not merely the headline average.
- Inspect motion at native output. Look for disappearing small objects, unstable edges, ghost trails, shimmering, excessive blur, and changes in target readability.
- Keep a setting only when both tests pass. If performance improves but visibility suffers, restore it and test another candidate.
- Recheck after major updates. A game patch, driver change, or new rendering mode can alter the cost and image quality of a feature.
Which settings are worth testing first?
The order below is a decision framework, not a universal value table. The game represented by [Game] has not been identified, so its exact setting names and relative costs are unknown.
Rank #2
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Powered by GeForce RTX 5070 Ti
- Integrated with 16GB GDDR7 256bit memory interface
- PCIe 5.0
- WINDFORCE cooling system
| Setting group | Why test it | Visibility risk to inspect |
|---|---|---|
| Display resolution and render scale | Reducing the number of pixels or the internal render resolution can be a high-impact GPU test. | Fine geometry, text, distant targets, and edge definition may become less clear. |
| Upscaler or reconstruction mode | It may render internally at a lower resolution and reconstruct the output, offering a performance/clarity trade-off. | Inspect ghosting, shimmering, motion blur, and detail stability rather than judging a still image. |
| Ray tracing and advanced lighting | These effects can be costly in a GPU-bound scene. | Dark areas, reflections, contact shadows, or object separation may change. |
| Shadows, reflections, ambient occlusion, and volumetrics | They are common GPU-heavy candidates when the game exposes separate controls. | Check whether lighting cues, silhouettes, and scene depth become harder to read. |
| Anti-aliasing and post-processing | Different methods can trade edge stability for GPU cost. | Watch moving edges, foliage, wires, and distant geometry for shimmer or blur. |
| View distance, crowd, foliage, and particle density | These may affect either rendering or CPU simulation, depending on the implementation. | Lower density can remove useful landmarks, objects, or environmental cues. |
| Texture quality | Test it independently rather than assuming it is an easy FPS gain. | Lower textures can make surfaces, signs, and object markings less legible. |
Do not lower textures automatically
NVIDIA’s 2019 Control guide measured essentially similar performance across several texture levels in its test, while Ultra carried a small measured cost. That example shows why each option needs its own evidence. Texture quality may consume video memory without being the frame-rate limiter; lowering it can make surfaces less legible while delivering little improvement. Test it only after measuring your own run.
Prioritize settings that preserve useful cues
Visibility is not the same as maximum visual quality. Identify what you must read while moving: player silhouettes, enemy outlines, muzzle flashes, item highlights, cover edges, subtitles, objective markers, or distant landmarks. Keep the control that governs those cues at the highest level your bottleneck allows, and look for savings in effects that add atmosphere without carrying information you need. The correct control is game-specific, so verify it in the game’s own settings descriptions.
Resolution, render scale, and upscaling
Resolution and internal render scale are high-impact test candidates in a GPU-bound run, but they are not guaranteed wins. A lower internal resolution can improve performance while leaving menus or interface elements at native resolution in some implementations. That behavior is implementation-specific, so check the result in [Game] rather than assuming the interface will remain sharp.
Recommended Free Tools
What DLSS Super Resolution does
NVIDIA defines DLSS Super Resolution as reconstruction of a higher-resolution output from a lower-resolution render. The available modes, minimum hardware, image quality, and performance vary by game and by implementation. Compare the modes in motion at the resolution you actually play; a still screenshot can conceal temporal artifacts.
Rank #3
- Powered by the NVIDIA Blackwell architecture and DLSS 4. System Requirements: Minimum 850W PSU with 16-pin 12V-2x6 (12VHPWR) connector required. Verify before purchasing.
- Military-grade components deliver rock-solid power and longer lifespan for ultimate durability. Compatibility: 348mm (13.7") length, 3.6 slots, 4.3 lbs. Confirm case clearance and slot spacing. GPU bracket included.
- Protective PCB coating helps protect against short circuits caused by moisture, dust, or debris
- 3.6-slot design with massive fin array optimized for airflow from three Axial-tech fans
- Phase-change GPU thermal pad helps ensure optimal thermal performance and longevity, outlasting traditional thermal paste for graphics cards under heavy loads
Should you use an upscaler or lower the display resolution?
Test both when the game exposes them. An upscaler may preserve a native-resolution output while reducing internal rendering work, whereas lowering the display resolution changes the final output itself. Which looks better depends on the title’s reconstruction, anti-aliasing, sharpening, HUD behavior, and scene content. Keep the option that gives stable frame times without making moving targets or important scene cues harder to distinguish.
Do not generalize the Control numbers
NVIDIA’s official 2019 Control guide reported DLSS frame-rate increases of as much as 89% in a higher-fidelity mode and up to 197% in lower-resolution performance modes under its stated test conditions. Those are vendor-published, game-specific maximum results, not typical gains promised for another title. The same guide reported that its 1920×1080 test became CPU-bottlenecked at around 80 FPS. That ceiling belongs to the tested Control configuration and should not be treated as a universal CPU limit.
Frame generation and responsiveness
Frame generation creates additional displayed frames; it does not simply make the game’s simulation run faster. A displayed frame-rate increase can therefore coexist with input latency or a CPU/GPU bottleneck that still limits responsiveness. Evaluate camera control, aiming, and frame-time behavior in addition to the counter in the corner.
Use the game’s implementation, not a generic promise
NVIDIA’s driver documentation describes DLSS Frame Generation, Ray Reconstruction, DLAA, and related features, while noting that game-specific libraries and options vary by title and age. If [Game] does not expose a feature, a driver toggle cannot reliably add the same behavior. Read the game’s own support notes and check whether the option is available for your GPU and version.
Rank #4
- AI Performance: 767 AI TOPS
- OC mode: 2632 MHz (OC mode)/ 2602 MHz (Default mode)
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Axial-tech fan design features a smaller fan hub that facilitates longer blades and a barrier ring that increases downward air pressure
- A 2.5-slot design maximizes compatibility and cooling efficiency for superior performance in small chassis
Do not combine competing features blindly
NVIDIA states that native DLSS Frame Generation and Smooth Motion are competing technologies and should not be used together; its documentation warns that doing so causes lower performance and visual artifacts. Apply that warning only when both options are actually present, but follow it exactly if the game and driver expose them.
Consider latency controls separately
NVIDIA Reflex is designed to manage CPU/GPU work and the render queue. A queue can affect responsiveness even when an FPS counter looks high. If the game offers Reflex, test it as its own setting and judge input response in the same repeatable scene. Do not claim that a latency option increases FPS; its purpose is render-queue and responsiveness management.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Using an automated advisor carefully
AMD’s Radeon Game Advisor illustrates a useful iterative process: monitor gameplay, review average FPS and 95th-percentile frame time, apply a recommended in-game change, and run the report again. It can suggest changes such as lowering a quality preset, advanced rendering options, render scale, or display resolution.
The documented tool has important qualifications. It requires exclusive full-screen operation, documents support for Radeon RX 400-series and newer graphics, and excludes APU graphics. Software availability and compatibility can change, so verify the current AMD documentation before relying on installation or support details. Treat any recommendation as a test proposal, not an automatic answer: confirm the result in your own scene and inspect moving-image clarity.
Best Value
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Powered by GeForce RTX 5060
- Integrated with 8GB GDDR7 128bit memory interface
- PCIe 5.0
- WINDFORCE cooling system
When a hardware upgrade is justified
Do not buy a graphics card because a preset feels slow. First establish that the game is GPU-bound, that reducing GPU-heavy settings produces a meaningful improvement, and that the remaining visual quality is acceptable. Only then is a graphics-card upgrade a plausible FPS remedy. If the run is CPU-bound, a faster GPU may leave the limiting behavior unchanged. The game, resolution, target FPS, processor, memory, and current graphics card are all needed before selecting a specific upgrade; none are known from the placeholder title.
Troubleshoot disappointing results
Lowering every setting barely changes FPS
Return to the baseline and check whether the GPU was the limiter. Inspect CPU activity by core, frame-time spikes, simulation-heavy locations, frame caps, and background processes. If the game is CPU-bound, continue lowering graphics quality only if it improves image stability or eliminates a separate rendering problem; do not expect a large FPS increase.
The average rises but the game still feels uneven
Compare frame-time behavior, especially the 95th percentile if your tool reports it. Repeat the run to rule out a one-off streaming event. Keep the setting that reduces visible spikes, even if its average FPS is slightly lower than a setting with unstable delivery.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →The image looks acceptable when paused but poor in motion
Inspect the same scene while panning and while targets move. Temporal reconstruction, anti-aliasing, motion blur, and reduced internal resolution can produce trails, shimmer, or lost detail that a still capture hides. Restore the last visibility-preserving setting and test a different performance candidate.
An option is missing
Feature availability depends on the game, rendering API, GPU, driver, edition, and patch. Check the game’s current documentation and hardware requirements. Do not substitute a similarly named driver feature or assume that a setting documented for another title exists here.
The game refuses to run the advisor
Check whether the tool requires exclusive full-screen, whether the GPU is in the documented Radeon RX 400-series-or-newer range, and whether the system uses excluded APU graphics. If any requirement fails, use the manual baseline-and-A/B workflow instead.
A practical worksheet for [Game]
- Write the exact game version, display resolution, refresh target, GPU, CPU, and memory configuration.
- Record the baseline preset, render scale, anti-aliasing, upscaler, frame generation, latency option, and frame cap.
- Choose one demanding, repeatable scene and record average FPS plus frame-time behavior.
- Classify the run as probably GPU-bound, probably CPU-bound, or uncertain.
- Test one candidate setting that matches that diagnosis.
- Check performance in the same scene and inspect the scene while moving.
- Keep the change only if it improves the target behavior without removing necessary visual cues.
- After all changes, replay a different demanding scene to ensure the result was not specific to one location.
Exact values should be added only after [Game] is replaced with a named title and its current settings, patch, hardware support, and visibility-critical controls are verified. Until then, a measured workflow is safer than a copied preset.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
