The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →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
Open-world horror RPG stutter usually has a specific trigger: a shader is compiled as you enter an area, the engine is streaming assets during traversal, or one system resource has reached its limit. Crashes commonly point to GPU memory pressure, unstable driver or overclock settings, overlays, or a Windows timeout-and-recovery reset. Long loading screens improve only when storage is the bottleneck or the game actually implements a fast-loading API such as DirectStorage.
Use the workflow below to identify the limiting resource before changing settings or buying hardware. Shader preparation, driver-cache rebuilding, GPU-crash checks, targeted graphics reductions, and NVMe compatibility checks address different problems; none is a universal FPS or loading-time fix.
Start by identifying what is actually stalling
A hitch is a frame-time event, not simply a low average frame rate. An open-world horror game can report a high average FPS while pausing for a fraction of a second whenever it enters a new street, room, forest section, or other streaming cell. Record what happens at the exact moment of the hitch instead of changing several settings at once.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errors- Reproduce the same route. Load the same save, walk or drive through the same transition, and note whether the pause happens once, repeatedly, or only after a restart.
- Capture frame-time behavior. Use the game’s frame-time graph if it has one, or a trusted performance overlay. A tall, isolated spike suggests a one-time operation; repeated spikes at every traversal boundary suggest streaming, shader preparation, or a CPU traversal limit.
- Record utilization at the spike. Note GPU utilization, dedicated-video-memory (VRAM) use, CPU utilization, system-memory use, and storage activity. A resource that reaches its limit while the others remain comparatively idle is the first place to investigate.
- Classify the trigger. Mark whether the hitch occurs during first launch, after a graphics-driver update, while entering a previously unseen area, during a loading screen, or under sustained heavy rendering. Each pattern has a different fix.
Do not use an average FPS number to choose a fix. For example, lowering resolution can help a GPU-limited scene but will not remove a traversal hitch caused by CPU work or storage reads.
#1 Best Overall
- This product has been replaced by our latest generation. Please search for the SANDISK Optimus GX 7100 NVMe SSD
- HIGH-OCTANE GAMING. Experience speeds up to 7,250MB/s read and 6,900MB/s write (1-2TB models), with up to 35% faster performance than previous generation.
- PURPOSE-BUILT. Designed for serious on-the-go gamers, with a PCIe Gen4 interface and SANDISK’s next generation TLC 3D NAND.
- MORE TIME TO CLEAR THAT CHECKPOINT. Built with laptops and handheld gaming devices in mind, with up to 100% more power efficiency over the previous generation.
- DO MORE WITH DASHBOARD. Ensure your drive is optimized for prime performance with the downloadable WD_BLACK Dashboard (Windows only).
Why shader compilation causes a sudden pause
Modern engines cannot always ship every hardware-specific shader in a finished form. When a material, lighting path, post-processing effect, or ray-tracing permutation is needed, the engine may ask the graphics driver to compile it. If that request arrives immediately before the draw call, the render thread waits.
“Shader compilation stuttering happens when a render engine discovers that it needs to compile a new shader right before it uses it for drawing, so everything stops while waiting for the driver to finish compilation.”
That explanation comes from Unreal Engine engineers Kenzo ter Elst, Daniele Vettorel, Allan Bentham, and Mihnea Balta. In an open-world horror RPG, the effect is easy to notice because entering a new area can expose a different combination of fog, decals, foliage, reflections, shadows, and character materials. The game may have been smooth in the previous area and still hitch when the new combination is encountered.
Launch-time preparation versus traversal stutter
Some games show a shader-compilation or shader-preparation step when they start. Let it finish before judging traversal performance. A game that is still preparing shaders can hitch repeatedly during the first minutes, then become smoother once the relevant permutations are compiled and cached.
Microsoft describes the trade-off this way: “Compiling those shaders on the player’s device at launch, or ‘just-in-time’ during gameplay, leads to longer load times, in-game shader compilation stutter, and higher power consumption.” A longer first launch can therefore be the mechanism that prevents later just-in-time pauses.
What changes after a graphics-driver update
Installing a new graphics driver clears the driver’s shader cache. The game and driver must rebuild entries that were created under the previous driver, so a first run after the update can look worse even when the new driver is functioning correctly.
Rank #2
- MEET THE NEXT GEN: Consider this a cheat code; Our Samsung 990 PRO Gen4 SSD helps you reach near max performance with lightning-fast speeds; Whether you’re a hardcore gamer or a tech guru, you’ll get power efficiency built for the final boss
- REACH THE NEXT LEVEL: Gen4 steps up with faster transfer speeds and high-performance bandwidth; With a more than 55% improvement in random performance compared to 980 PRO, it’s here for heavy computing and faster loading
- THE FASTEST SSD FROM THE WORLD'S FLASH MEMORY BRAND: The speed you need for any occasion; With read and write speeds up to 7450/6900 MB/s you’ll reach near max performance of PCIe 4.0 powering through for any use
- PLAY WITHOUT LIMITS: Give yourself some space with storage capacities from 1TB to 4TB; Sync all your saves and reign supreme in gaming, video editing, data analysis and more
- IT’S A POWER MOVE: Save the power for your performance; Get power efficiency all while experiencing up to 50% improved performance per watt over the 980 PRO; It makes every move more effective with less consumption
Epic gives a concrete engine example: Fortnite takes around 20–30 seconds longer to load into a Battle Royale match when its driver cache is empty. That is an example measured for that game and engine behavior, not a loading-time promise for every horror RPG.
What to do after the cache is cleared
- Start the game once and allow its shader-preparation or compilation stage to complete. Keep the game open long enough to reach the main menu and, if appropriate, a short practice or low-risk session.
- Repeat a representative route once the preparation step has finished. Compare frame-time spikes with the first run rather than comparing only average FPS.
- Do not repeatedly delete the shader cache to “fix” stutter. Clearing it forces the same compilation work again and can make every first traversal look like a fresh installation.
- If the game never completes preparation, crashes during it, or becomes worse on every run, treat that as a separate stability problem. Check driver installation, overlays, memory pressure, and the game’s patch notes before clearing anything again.
Secondary coverage of Microsoft’s Advanced Shader Delivery demonstration has cited a 95% first-load reduction, from 1.5 minutes to 4 seconds on a specified test system. That result is tied to the demonstration system and workflow; it is not a universal expectation for a retail game, so use the game’s own observed preparation time as the baseline.
Separate shader hitches from traversal and streaming hitches
A traversal hitch occurs as the player crosses an open-world boundary and the engine builds or streams the next area. It can involve CPU-side visibility and simulation work, storage reads, decompression, and GPU resource creation. Shader compilation may be one part of the event, but the remedy depends on the resource that saturates.
| Observed pattern | Likely limiting resource | First action |
|---|---|---|
| Large spike on first launch or first visit to an effect-heavy area; later visits are smoother | Shader preparation or an empty driver cache | Allow preparation to finish; avoid repeatedly clearing the cache |
| Hitch at the same world boundary on every pass, with CPU activity or traversal work peaking | CPU-side streaming, visibility, crowd, or simulation work | Lower view distance, crowd, and other CPU-heavy settings one at a time |
| Hitch coincides with high disk activity and an area-load request | Storage throughput or asset decompression | Confirm the game drive and controller; consider NVMe only after verifying compatibility |
| Hitch or crash while dedicated VRAM is near its limit | VRAM pressure | Lower texture quality and ray-tracing features before unrelated settings |
| Consistently low GPU utilization while CPU usage or one busy thread limits frame delivery | CPU or engine traversal limit | Reduce view distance, crowd, and simulation settings rather than resolution first |
| Frame rate drops with GPU utilization near full | GPU rendering load | Lower resolution or upscaling quality first |
These categories can overlap. A new area may request shaders, read assets from storage, and increase CPU visibility work in the same frame. Use the utilization record to decide which change is most likely to remove the largest part of the spike.
A practical diagnostic workflow
1. Establish a repeatable baseline
Choose one save and one route that reliably produces the problem. Record the game version, graphics-driver version, display resolution, upscaler mode, ray-tracing state, and whether the game is on a hard disk, SATA SSD, or NVMe SSD. Keep this note so a later patch or driver change can be compared with the same conditions.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
2. Watch VRAM and system memory
Open Task Manager with Ctrl + Shift + Esc and use the Performance view while reproducing the issue. Watch dedicated GPU memory and system memory, then check whether another application is consuming GPU resources. A near-full VRAM budget can force eviction and reloading of textures or other resources; system-memory pressure can add paging and worsen loading behavior.
Rank #3
- HUGE SPEED BOOST: Get random read/write speeds that are 40%/55% faster than 980 PRO; Experience up to 1400K/1550K IOPS, while sequential read/write speeds up to 7,450/6,900 MB/s reach near the max performance of PCIe 4.0*
- BREAKTHROUGH POWER EFFICIENCY: Use less power and get more performance; Enjoy up to 50% improved performance per watt over 980 PRO, plus optimal power efficiency with max PCIe 4.0 performance**
- SMART THERMAL CONTROL: Samsung's own nickel-coated controller delivers effective thermal control; With its slim size, 990 PRO is a perfect fit for desktops and laptops that meet the PCI-SIG D8 standard***
- THE CHAMPION MAKER: Up to 65% improvement in random performance enables faster loads for an ultimate gaming experience on PS5 and DirectStorage PC games****
- SAMSUNG MAGICIAN SOFTWARE: Get the most out of your SSD with Samsung Magician's advanced yet intuitive optimization tools; Monitor drive health, protect valuable data, and receive important updates for your 990 PRO
3. Check for competing 3D applications and overlays
Close recording tools, browser windows using hardware acceleration, monitoring overlays, game launchers with in-game overlays, and other 3D applications for one test run. Epic’s GPU-crash guidance specifically recommends removing competing 3D applications and overlays when investigating GPU failures. If the hitch or crash disappears, re-enable tools one at a time to identify the conflict rather than permanently changing every setting.
4. Test the post-driver-update case separately
If the problem began immediately after a driver installation, run the game once with shader preparation enabled and allow the cache to rebuild. If crashes remain, follow the game publisher’s support procedure for a clean graphics-driver installation. A clean installation is a recovery step for a suspected driver-state problem, not a routine stutter cure.
5. Change one variable and keep a rollback
Change one graphics option, replay the same route, and record the result. Save the original value or use a screenshot before each change. Patches and driver updates can change shader-cache and streaming behavior, so a setting that helped one build may be unnecessary or counterproductive later.
Fix first-run and UE5-style shader stutter
Allow the complete preparation pass
If the title presents a visible shader-compilation or preparation screen, let it reach completion before entering a demanding area. Interrupting that process, repeatedly restarting, or deleting the cache prevents the game from retaining the work it has already done.
Distinguish compilation from ordinary low performance
A compilation hitch is often an isolated, severe frame-time spike tied to a new effect. If the frame rate is low continuously, the limiting factor is more likely GPU load, CPU simulation, VRAM, or another resource. Lowering every setting will not repair a cache that is still being rebuilt.
Use a clean driver installation only when stability requires it
For persistent crashes that began with a driver change, use the graphics driver’s clean-install option or the game publisher’s documented clean-install procedure. Reboot, launch the game without overlays, and repeat the baseline route. Do not combine a clean install, a cache deletion, and multiple graphics changes in the same test; otherwise you cannot tell which action mattered.
Rank #4
- BREAKTHROUGH PCIe 5.0 PERFORMANCE: Supercharge your workflow and gaming with PCIe 5.0, boasting up to 14,700/13,400 MB/s¹ sequential read/write speeds. Tackle massive files and power up your gaming with Gen5—twice as fast as the 990 PRO SSD.
- EVERY TASK, TURBOCHARGED: Speed past productivity limits. With random read/write speeds up to 1,850K/2,600K IOPS¹, enjoy fast game loads, seamless AI apps, and efficient multitasking. Virtually no lag, no limits—just nonstop performance.
- THINK FAST, CREATE FASTER: With random read/write speeds of up to 1,850K/2,600K IOPS¹, the 9100 PRO SSD fuels seamless AI content creation, swift loads, and smooth gameplay. Work, play, and create at lightning speed.
- SPEED, WHENEVER YOU NEED: From laptops, desktop PCs, to PS5 with added heatsink devices², experience blazing PCIe 5.0 speeds and up to 8TB of storage. Perfect for video editing, gaming, and creative tasks, with the compatibility to match your device.
- STAY COOL, RUN FAST: Push limits, not temperatures. A 5nm controller boosts power efficiency up to 49% over the 990 PRO SSD³, while advanced thermal control keeps performance smooth and reliable.
Stop crashes related to GPU timeouts and memory pressure
Crashes mentioning DXGI_ERROR_DEVICE_REMOVED, device reset, or a lost graphics device often occur when the GPU cannot complete work within Windows’ timeout-and-detection-and-recovery (TDR) window. Epic’s GPU-crash guidance connects long GPU operations with Windows TDR resets. The message does not prove that the game is defective; it indicates that the graphics device stopped responding quickly enough for Windows to recover it.
Use this order of checks
- Check dedicated VRAM and system memory. Reproduce the crash while watching Task Manager. If VRAM is at or near its budget, lower texture quality and ray-tracing features first. If system memory is constrained, close other applications and remove overlays before retesting.
- Remove competing 3D work. Exit other games, GPU-accelerated capture or rendering programs, and overlays. Test the game by itself.
- Return overclocks to stable settings. Restore GPU, CPU, and memory tuning to stock values. A setting that is stable in a lightweight game can fail during a long, shader-heavy open-world scene.
- Install the graphics driver cleanly when support guidance calls for it. This addresses damaged or conflicting driver state. Restart before the next test and allow shader preparation to rebuild.
- Reduce the specific load that triggers the crash. For a GPU-limited scene, lower resolution or upscaling quality. For VRAM pressure, reduce textures and ray tracing. Do not begin by reducing settings unrelated to the measured limit.
Do not treat a longer TDR timeout as the first fix. It can hide a stability problem and does not reduce the GPU work that caused the reset. Use the game publisher’s supported troubleshooting path instead.
Choose graphics settings by the bottleneck
| Measured bottleneck | Settings to reduce first | Why this order matters |
|---|---|---|
| GPU rendering load | Output resolution, then upscaling quality | These directly reduce the number or quality of pixels the GPU must render |
| CPU or traversal work | View distance, crowd density, simulation, and other CPU-heavy options | These reduce visibility checks, world updates, and simulation rather than pixel shading |
| VRAM pressure | Texture quality and ray-tracing features | These use large texture and acceleration-structure allocations; reducing unrelated options may leave the memory peak unchanged |
| System-memory pressure | Close competing applications and overlays; then reassess the game’s memory settings | Free memory reduces paging and removes external GPU or CPU work |
Apply the smallest change that addresses the observed limit. A horror game’s atmosphere often depends more on lighting, fog, shadows, and texture detail than on a maximum view-distance setting, so a targeted reduction can preserve the visual character better than dropping every option to low.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Will moving the game to an NVMe SSD reduce loading times?
It can, but only when storage activity is the bottleneck or when the game uses a storage API designed to exploit NVMe throughput. An NVMe drive is not a universal stutter or FPS fix, and it cannot repair shader compilation, a CPU traversal limit, VRAM exhaustion, or a GPU crash.
Microsoft describes DirectStorage as an API that lets developers unlock the potential of high-speed NVMe drives for loading game assets. The game must implement DirectStorage; simply installing the game on an NVMe device does not add that implementation.
Windows 11 requirements to verify
Microsoft’s Windows 11 requirements state that DirectStorage needs an NVMe SSD used with the Standard NVM Express Controller driver and a DirectX 12 GPU with Shader Model 6.0 support. Verify all of the following before buying hardware for this purpose:
Best Value
- PCIe 4.0 Performance: Delivers up to 7,100 MB/s read and 6,000 MB/s write speeds for quicker game load times, bootups, and smooth multitasking
- Spacious 1TB SSD: Provides space for AAA games, apps, and media with standard Gen4 NVMe performance for casual gamers and home users
- Broad Compatibility: Works seamlessly with laptops, desktops, and select gaming consoles including ROG Ally X, Lenovo Legion Go, and AYANEO Kun. Also backward compatible with PCIe Gen3 systems for flexible upgrades
- Better Productivity: Up to 2x faster than previous Gen3 generation. Improve performance for real world tasks like booting Windows, starting applications like Adobe Photoshop and Illustrator, and working in applications like Microsoft Excel and PowerPoint
- Trusted Micron Quality: Built with advanced G8 NAND and thermal control for reliable Gen4 performance trusted by gamers and home users
- The game lists DirectStorage support or otherwise documents an NVMe-aware asset-loading path.
- The target drive is NVMe, not merely a different SATA device.
- Windows is using the Standard NVM Express Controller driver for that drive.
- The GPU supports DirectX 12 and Shader Model 6.0.
- Storage activity is actually high during the measured loading delay.
If storage activity is low while the CPU is decompressing assets or the game is compiling shaders, an NVMe purchase is unlikely to remove the observed pause. Measure first, then decide whether the hardware intervention matches the bottleneck.
Compare the main fixes before spending money
| Fix | Best symptom match | Primary limit addressed | Expected benefit | Cost and reversibility | Compatibility requirements |
|---|---|---|---|---|---|
| Let shader preparation finish | First-run or post-driver-update hitch | Shader compilation and empty driver cache | Fewer just-in-time shader spikes after preparation | No purchase; naturally reversible | The game must provide a preparation or compilation path |
| Close overlays and competing 3D applications | GPU crash, device reset, or unexplained instability | Competing GPU, CPU, and memory use | Removes external contention and a common troubleshooting variable | No purchase; reversible | Applications must be identifiable and closable |
| Clean graphics-driver installation | Crashes that began after a driver change | Driver state and cache rebuilding | Restores a supported driver state when the publisher recommends it | No purchase; reversible by installing another supported driver | Follow the game or driver maker’s supported procedure |
| Lower resolution or upscaling quality | GPU-limited frame rate or GPU-heavy crash | GPU rendering workload | More GPU headroom; image quality changes | No purchase; immediately reversible | Game must expose resolution or upscaling controls |
| Lower view distance, crowd, or simulation settings | Traversal hitch with CPU activity peaking | CPU-side world and visibility work | Less traversal work; may reduce world detail or simulation density | No purchase; immediately reversible | Game must expose the relevant options |
| Lower textures and ray tracing | VRAM pressure or memory-related GPU crash | Dedicated GPU memory | Reduces allocation pressure; visual detail changes | No purchase; immediately reversible | Game must expose texture and ray-tracing controls |
| Move the game to NVMe | Loading delay with storage activity high | Storage throughput and supported asset loading | Potentially shorter asset-loading time when the game and system use the drive effectively | Hardware purchase; location can be changed back | NVMe, Standard NVM Express Controller, DirectX 12/Shader Model 6.0 GPU, and game-side DirectStorage support where applicable |
Common mistakes that make stutter worse
Clearing caches after every bad run
An empty cache is precisely the state that requires compilation. Delete or rebuild it only when a supported troubleshooting procedure calls for it, not as a routine response to a single hitch.
Changing ten settings before measuring
Mass changes can hide the real limit and make it impossible to restore a stable configuration. Use the same route, adjust one option, and keep a rollback note.
Free tools Windows power users keep installed
One-click scans. No signup required.
Expecting an NVMe drive to increase FPS
Storage can shorten an asset-loading wait, but it does not increase the GPU’s shader throughput or the CPU’s simulation capacity. A storage upgrade is justified by measured storage activity or documented DirectStorage support, not by the word “open world” alone.
Assuming every post-update hitch is a broken driver
A driver update clears the shader cache, so the first run can be slower without indicating a fault. Escalate to a clean installation when crashes persist, preparation never completes, or the publisher’s support instructions call for it.
Ignoring overlays during crash diagnosis
Overlay and capture software can consume GPU memory or submit additional graphics work. Test with them closed before replacing hardware or changing system-level timeout settings.
A repeatable recovery plan
- Write down the game and driver versions, display settings, storage location, and the exact route that hitches.
- Reproduce it while recording frame time, GPU utilization, VRAM, CPU, system memory, and storage activity.
- If it is a first-run or post-driver shader event, complete the preparation pass and retest without clearing the cache.
- If it is a crash, check VRAM and system memory, close overlays and other 3D applications, restore stock clocks, and perform a clean driver installation if the publisher recommends it.
- If the GPU is the limit, lower resolution or upscaling quality. If the CPU or traversal path is the limit, lower view distance, crowd, or simulation. If VRAM is the limit, lower textures and ray tracing.
- If storage is the limit, verify the drive and controller, then check whether the game supports DirectStorage before purchasing an NVMe drive.
- Test one change at a time, keep the previous value available, and repeat the same route after every patch or driver update.
This process separates a temporary cache rebuild from a persistent engine or hardware limit and prevents an expensive storage upgrade from being used to solve the wrong problem.
Recommended Free Tools
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.
