No Rest for the Wicked: Best Steam Deck and ROG Ally X Settings

Table of Contents (14)

No Rest for the Wicked handheld answer: Steam Deck at 30 FPS, Ally X with compromises

No Rest for the Wicked works on ROG Ally X and Steam Deck, with a big asterisk: both handhelds need deliberate settings rather than a default-install-and-play approach. Steam Deck has the clearer route, centred on a 30 FPS cap, the Performance preset, and 50% render scale. ROG Ally X has more room to chase higher responsiveness, but published Ally-family results rely on 25-30 W-class power profiles, dynamic resolution scaling, and a visibly reduced internal render resolution.

This guide applies to the PC Steam Early Access build, which launched on April 18, 2024. Hotfixes, Proton releases, SteamOS builds, and the game’s own rendering options can materially change the result. Treat these as sensible baselines to test after every major patch, not eternal magic numbers. There is also no official Steam Deck Verified label established in the available material, so the practical answer here comes from patch notes and handheld-tested settings rather than a tidy green badge.

Handheld settings at a glance

Handheld	Recommended starting point	Frame-rate target	Important caveat
ROG Ally X	Performance preset; 50% render scale; dynamic resolution on; use FSR when the current game menu provides it; 30 W-class Turbo or manual profile	30 FPS for steadier play; 40 FPS dynamic-resolution target if image quality is secondary	Published Ally-family reports place performance around 30-40 FPS, with dips into the 20s under load.
Steam Deck LCD	Performance preset; 50% render scale; motion blur 0%; dynamic resolution off; 30 FPS cap	30 FPS	Use GE-Proton only if the default compatibility layer creates launch or performance trouble.
Steam Deck OLED	Start with the LCD preset	30 FPS	No OLED-specific settings test in the available handheld material establishes a different preset.

If you only want the blunt answer, that table is it. Steam Deck should be treated as a 30 FPS machine for this game. ROG Ally X can push harder, but the cleanest way to think about it is as a fork in the road: either lock it to 30 for steadier frame-times, or run a high-power profile and let dynamic resolution and upscaling chase something closer to 40 with noticeably softer image quality. The mistake is trying to pretend you can have all three at once: low power draw, sharp image quality, and stable higher frame rates.

Patch history matters more here than in most handheld guides

No Rest for the Wicked is one of those Early Access games where the patch number matters almost as much as the hardware. Early launch-window advice was built around harsher compromises because the game was rougher on handhelds and the graphics menu offered fewer escape hatches. Later builds changed that picture in ways that directly affect what you should turn on.

Patch 1 added significant CPU and GPU performance improvements, including traversal and heavy-area optimizations. It also improved Steam Deck controller icon legibility, which sounds minor until you are playing on a seven-inch screen and trying not to misread prompts in the middle of a fight. Patch 1 also brought broader control improvements such as custom key rebinding, which matters for handheld players using back buttons or a non-default controller layout.

Patch 2 is the real dividing line for graphics advice because it added DLSS and FSR upscaling support and included explicit Steam Deck texture-resolution and memory-budget optimizations. In plain English: pre-Patch-2 advice often leans on dynamic resolution scaling and a heavily reduced render scale because that was the practical survival kit at the time. Post-Patch-2 builds are more likely to let you use a proper upscaler in the graphics menu, which is usually the cleaner path on both the Ally and the Deck if it is available in your current build.

The hotfixes matter too. Hotfix 2 improved rendering performance, the Balanced and Performance quality modes, visual effects and fog, and the Ship Prologue. Hotfix 3 introduced camera zoom adjustment and automatic camera zoom for handheld devices and Performance Mode. Later notes continued to mention performance work around CPU spikes, memory usage, culling, and content streaming. That does not magically make every old preset obsolete, but it does explain why a 2024 handheld post and a 2025 handheld video can both be useful while still recommending slightly different mixes of render scale, dynamic resolution, and FSR.

That is also why any settings guide that pretends to be timeless is overselling it. Moon Studios has already shown a willingness to keep tuning performance during Early Access, and the studio has delayed version 1.0 to March 2027 for more polish, performance work, class-system refinement, and additional testing. For handheld players, that is not filler-roadmap talk. It means the correct preset is a moving target by design.

ROG Ally X: use the hardware headroom carefully

The Ally X is the handheld to choose if responsiveness matters more than battery endurance and pristine image quality. A documented Ally configuration uses 1920×1080 in Fullscreen Exclusive mode, V-Sync off, a 120 FPS in-game cap, the Performance preset, 50% render scale, motion blur disabled, dynamic resolution enabled, and a 40 FPS dynamic-resolution target. Set VRAM to 6 GB or higher on the Ally-family baseline.

The 120 FPS cap looks odd beside a 40 FPS target, so here is the sane way to read it. The cap and the target are not doing the same job. A high cap stops the game from hitting an artificially low ceiling, while the dynamic resolution target tells the game what performance level it should try to preserve by reducing internal image quality on the fly. That setup only makes sense if your goal is to chase a variable result above 30 and accept blurrier reconstruction when scenes get ugly.

If your goal is steadier play, do not overthink it: lock to 30 FPS and build everything else around making that 30 feel clean. That is the real decision tree on Ally X.

ROG Ally X decision tree: pick one path and stop fighting yourself

Path 1: Steady handheld play. Use a high-power mode, keep the Performance preset, keep 50% render scale as a starting point, and use FSR if the current build exposes it in the graphics menu. Then cap the game to 30 FPS and check frame-time consistency rather than the highest number you glimpse in a quiet room. This is the preset for people who want the Ally X to feel stable instead of merely fast-looking.

Path 2: Chase a more responsive feel. Use a 30 W-class Turbo or manual high-power profile, keep Fullscreen Exclusive, leave V-Sync off, use the Performance preset, enable dynamic resolution scaling, and set the target to 40 if the current build allows it. If FSR is available, pair it with the lower render scale rather than expecting native-looking 1080p from a handheld APU. This is the preset for people who are willing to trade image clarity for a snappier result.

Path 3: The fallback when busy areas wreck both plans. Keep the high-power profile, drop expectations back to 30 FPS, and treat the graphics menu as a stability tool. No one wins an argument with Sacrament, Nameless Pass, or another demanding area by insisting the same settings that looked fine in the prologue must also hold under heavier load.

ROG Ally X menu walkthrough: the safest baseline to build from

Start outside the game. In Armoury Crate, go to Settings > Operating Mode > Manual Mode if you want a custom high-power setup, or use the device’s Turbo-class mode if you prefer the simpler route. The exact wording can vary by Armoury Crate version, but the practical point does not: if you are trying to push No Rest for the Wicked beyond a conservative 30 FPS experience, this is not the game for a low-power profile.

Then go into the game’s graphics options and build the profile in this order:

• Display mode: Fullscreen Exclusive on the documented Ally setup.
• Resolution: 1920×1080 in that same documented profile, with the understanding that internal render resolution is being cut hard underneath it.
• V-Sync: Off.
• Quality preset: Performance as the baseline.
• Render scale: 50% as the starting point.
• Motion blur: Off, including framerate-independent motion blur if that separate toggle exists in your build.
• Dynamic resolution scaling: On if you are chasing the 40-target route; optional if you decide 30 FPS stability matters more.
• Upscaling: Use FSR only when the live build actually provides it in the menu. Patch 2 is the major cutover here.

That last point is worth repeating because it is where a lot of old advice gets messy. One early Ally settings listing included FSR Performance while also reflecting a moment when broad upscaling support was still limited. Patch 2 later added FSR and improved stability, which means the clean rule for 2026 is simple: if the upscaler is present in the game menu, use it; if it is not there, do not try to fake it with system-level hacks.

There is no documented Ally X battery-runtime figure in the current handheld material. High-power 30 W-class operation is therefore best treated as a performance setting, not a portable-play promise. The Ally X can make this game feel more responsive than Steam Deck, but it earns that advantage by feeding the APU far more power and by accepting a much softer image under load.

Later ROG Ally X testing examples reinforce the same broad point rather than replacing it. One August 2024 test used FSR 2.2 with 8 GB VRAM shown on-device and moved between different quality modes depending on area, including Balanced Medium in the prologue, Quality Low and High on the beach, Balanced Low in Sacrament and Nameless Pass, and Performance Low in Orban Glades. That is useful not because it gives you one sacred preset, but because it shows the real personality of this game on the device: area-to-area tuning is normal, and demanding zones can force uglier settings than the opening hours.

Steam Deck LCD: a conservative 30 FPS setup is the right call

Steam Deck’s practical baseline is less glamorous and more dependable: Performance preset, 50% render scale, motion blur at 0%, Framerate Independent Motion Blur disabled, dynamic resolution disabled, and a 30 FPS cap. A May 2024 completion run used that combination with 4 GB VRAM allocated and no CryoUtilities. That is exactly the kind of low-risk configuration worth preserving on a handheld where stability matters more than a fleeting higher counter.

If that sounds conservative, good. It is supposed to. The Steam Deck guidance in the available material keeps circling back to the same answer because the machine does not have enough spare performance for fake optimism. The win condition is not seeing 40 once. The win condition is making combat, traversal, and camera movement feel coherent for an entire session.

Steam Deck menu walkthrough: build the baseline before you troubleshoot

Start in the game’s graphics options and set these first:

• Quality preset: Performance.
• Render scale: 50%.
• Motion blur: 0%.
• Framerate Independent Motion Blur: Off.
• Dynamic Resolution Scaling: Off for the baseline Deck setup.
• Target FPS: 30 if the current build exposes that option.

Then use the Steam Deck performance menu to finish the job. Open the Quick Access menu, go to the performance controls, enable a per-game profile if you use them, and apply a 30 FPS frame limit. Whether you prefer the system-side cap, the in-game 30 target, or both, the principle is the same: keep the frame-time graph boring. On Steam Deck, boring is a compliment.

Set the game’s audio output to Handheld. Disabling controller vibration is also reasonable if the constant haptics become distracting during combat. Neither adjustment replaces graphics tuning, but both make extended portable sessions less irritating.

The 4 GB VRAM note from that early completion run is useful as a documented example, not a universal commandment. If you already have a working Deck setup, do not rebuild your entire system around one old anecdote. The stronger through-line is simpler: 30 FPS, Performance preset, and 50% render scale have remained the repeatable core of Steam Deck advice even as patches and SteamOS versions changed around them.

When Proton matters, use it as a tool, not a superstition

Proton compatibility needs a measured approach. GE-Proton9-4 was a useful point-in-time recommendation during the game’s launch period, installed through ProtonUp-Qt and selected in the game’s Steam compatibility settings. It should be a fallback today, used when the default Proton version fails to launch the game or produces clearly worse behaviour. Chasing an old compatibility layer purely because a 2024 guide named it is how handheld troubleshooting turns into a weekend project.

The better order of operations is this:

1. Update SteamOS.
2. Update the game.
3. Test the current default Proton first.
4. If launch behaviour or performance is obviously bad, try GE-Proton through ProtonUp-Qt.
5. Retest the same area after the change instead of swapping three things at once and guessing which one helped.

SteamOS version matters here. SteamOS 3.6.24 Beta included a fix specifically addressing a No Rest for the Wicked performance regression in March 2025. Keep SteamOS and the game updated before rebuilding your graphics preset around an older complaint; a patch-level regression can make otherwise sensible settings look broken.

Later point-in-time Deck tests still landed in the same neighborhood. A June 2025 Steam Deck test for the Breach Refined update used SteamOS 3.7.8 and still recommended a 30 FPS cap. Another 2025 clip used a Performance preset with FSR 3.1.5 Quality and a 30 FPS lock on SteamOS 3.7.20. Those are not universal prescriptions, but they are a useful sanity check: even after later updates, the Deck story still revolves around a steady 30, not a triumphant leap to a broadly stable 40.

Steam Deck OLED: begin with LCD settings, then check the frame-time graph

No OLED-specific No Rest for the Wicked settings test in the available material establishes a separate graphics profile. So do not invent one. Start with the LCD configuration, keep the 30 FPS cap in place, and treat any change from there as your own local experiment rather than a sourced recommendation.

That also means avoiding a common trap: assuming the OLED model should automatically get a more ambitious preset because the hardware feels nicer elsewhere. The only defensible advice from the material here is to use the same baseline and judge it by frame-time consistency during combat, traversal, and effects-heavy scenes. If the graph stays calm and input feels predictable, the preset is doing its job. If it spikes whenever the world gets busy, that is your answer.

There are also no sourced battery-life figures for No Rest for the Wicked on either Steam Deck variant. Avoid setting expectations around a specific runtime until a test records the game version, power limit, brightness, wireless state, and actual minutes played. Handheld battery claims without those basics are usually just optimism wearing a percentage sign.

What the later patches changed for handheld players

If you are returning after a long break, the cleanest way to think about the update history is this:

• Launch / early hotfix era: harsher reliance on low render scale, Performance mode, and raw compromise.
• Patch 1: meaningful CPU and GPU optimization work, especially in traversal and heavier areas, plus better Steam Deck prompt legibility.
• Patch 2: the major handheld-friendly change because it added DLSS/FSR and Steam Deck-specific texture and memory tuning.
• Hotfix 2 and Hotfix 3: continued work on rendering performance, fog and effects, quality modes, the prologue, and handheld camera behaviour.
• Later SteamOS and game updates: more reason to revisit old presets instead of assuming a 2024 guide still maps perfectly to a 2026 build.

That patch arc is also why older Ally advice and newer Deck advice can sound like they disagree when they are really talking about different builds. Before Patch 2, dynamic resolution and lower render scale carried more of the load. After Patch 2, built-in upscaling became part of the conversation. The right response is not to pick a favorite old guide and declare the others wrong. It is to ask a much more useful question: which version of the game was that advice written for?

When the game stutters, change the right thing

No Rest for the Wicked can punish random tweaking because several of its rough edges overlap. Stutter may come from simple GPU overload, but it can also come from memory pressure, content streaming, or one of the CPU-heavy situations that later patches tried to smooth out. That is why turning every slider down at once often feels dramatic but teaches you nothing.

A cleaner troubleshooting approach looks like this:

• If performance collapses in dense areas or while moving between zones: suspect traversal, streaming, or heavy-area load before assuming your entire preset is wrong. Patch 1 and later hotfixes specifically targeted this class of problem.
• If the image looks stable in still scenes but falls apart during action: decide whether dynamic resolution is the culprit or the rescue tool. On Ally X, it can be worth keeping on for the higher-target profile. On Deck, the baseline advice keeps it off and simply accepts the 30 FPS target.
• If the game feels worse after an update: check your SteamOS version and Proton behaviour before rebuilding every graphics setting from scratch. SteamOS 3.6.24 Beta fixing a regression is the exact kind of reminder that platform software can be the problem.
• If frame pacing is ugly despite acceptable average performance: use a strong 30 FPS cap and recheck the same test area. A lower but even result usually feels better on handheld than a higher number with constant swings.
• If controls or camera readability feel off: remember that some updates improved controller prompts and handheld camera behaviour. Not every problem is a raw FPS problem wearing a disguise.

The temptation, especially on Ally X, is to think there must be one heroic preset that makes the game sharp, smooth, and efficient all at once. The available testing does not support that fantasy. What it supports is a more boring but more useful truth: area-specific pressure still matters, and the best preset is the one that survives the game’s difficult zones without turning every sword swing into a frame-time weather report.

Early Access makes every preset temporary

No Rest for the Wicked has received performance-focused hotfixes throughout Early Access, including Hotfix 3, Hotfix 5, and Hotfix 6 in April and May 2024. Moon Studios also plans further performance work as it moves toward version 1.0, now scheduled for March 2027 on PC via Steam and PlayStation 5 after moving from the previous October 2026 window.

The added development time covers general performance improvements, class-system refinement, and additional testing. A closed beta is planned ahead of an open beta in December 2026. Xbox Series X|S and Nintendo Switch 2 versions are planned for later dates after the March 2027 PC and PlayStation 5 launch window. Further development updates are set to continue through the Wicked Inside developer diaries.

That roadmap is good news for handheld owners, but the useful takeaway today is narrower: revisit the graphics menu after major updates, because an old 50% render-scale preset may no longer be the best trade. Patch 2 already proved that a single update can change the upscaling conversation entirely. There is no reason to assume the current state is the final state.

What to avoid

• Do not use unverified registry edits, undervolts, or aggressive third-party system tweaks to rescue performance. The game’s patches and SteamOS updates already move the needle enough to make those fixes fragile.
• Do not expect the Ally X to hold a clean 40 FPS everywhere at 1080p. Its higher-wattage profile helps, but busy scenes can still fall below that target.
• Do not treat a 120 FPS cap, a 40 FPS dynamic target, and a manual 30 FPS lock as one combined magic preset. They belong to different goals.
• Do not remove the Steam Deck 30 FPS cap simply because a quiet area renders faster. A stable cap produces more predictable input and camera motion than a frame rate that continually rises and collapses.
• Do not apply an old ProtonGE version as a permanent fix without checking the current default Proton behaviour first.
• Do not assume Steam Deck OLED has a separate sourced preset. Start with the LCD baseline unless you are doing your own testing.
• Do not make battery-life promises from thin air. No sourced runtime figure in the available material establishes one for either handheld.

The short version has not changed: Steam Deck owners should aim for a locked 30 FPS and build around it. ROG Ally X owners can push harder, but only by spending power and sacrificing image quality. In an Early Access game this patch-sensitive, that is not a compromise to be embarrassed about. It is the honest answer.

Visible source link referenced in current demand: https://www.reddit.com/r/NoRestForTheWicked/comments/1w3zscj/how_does_it_run_on_a_rog_ally_x/