Quick answer: Increase r.Streaming.PoolSize in DefaultEngine.ini to something like 2000 or 3000 for modern PCs. For lasting fixes, reduce texture resolution on LOD0, set LOD bias on large textures, and disable streaming on always-loaded textures.

Here is how to fix Unreal “Texture streaming pool over budget” warnings. Your game runs, textures appear blurry at random, and the log or screen shows a streaming pool warning. Cranking r.Streaming.PoolSize makes it go away but hides the real problem. The actual issue is that your level is asking for more texture data than the GPU pool can hold at once, and the streamer is evicting or demoting mips to fit.

The Symptom

On-screen log: Texture streaming pool over 1024.00 MiB budget, usage 1289.12 MiB. Textures appear blurry or low-res intermittently. Distant objects’ textures pop in as the camera approaches. In packaged builds the warning is usually suppressed but blurry textures remain.

Variant: one camera angle is fine, another pans over a blurry wall. Or the game starts clean but gets worse the longer you play a map.

What Causes This

Pool size too small for the scene. The streaming pool is a GPU memory budget (default 1000 MiB on PC). Every texture visible on screen contributes to usage. Big scenes with high-res textures easily exceed this.

Non-streaming textures eating the budget. UI textures, render targets, and any texture with Never Stream = true are loaded at full size all the time. They do not count against the streaming pool but they count against total GPU memory, and many teams leave large textures as non-streaming by mistake.

High texture resolutions on small objects. A 4K texture on a pebble wastes 16 MB of budget. Matching texture resolution to the screen-space size of the object is the cheapest fix.

LOD bias wrong. Texture assets have LODBias and MaxTextureSize. Default is to stream full resolution when near, which is fine, but shipping packages can strip mips if MaxTextureSize is set wrong.

Cooked data mismatch. The editor may show 2048 textures but packaging cooks them at a different size due to TextureGroup settings. Production builds differ from editor.

The Fix

Step 1: Measure actual pool usage. In the console:

r.Streaming.PoolSize 2000
stat streaming

The stat streaming HUD shows Current Used, Max Required, and pool budget. If Required is 1800 MiB and pool is 1000 MiB, you have an 800 MiB deficit. Either raise the pool or reduce required.

Step 2: Raise pool size in DefaultEngine.ini.

; Config/DefaultEngine.ini
[/Script/Engine.RendererSettings]
r.Streaming.PoolSize=2000
r.Streaming.LimitPoolSizeToVRAM=1

2000 MiB is a reasonable value for modern PC GPUs with 6 GB+ VRAM. On consoles, match the platform’s memory budget — PS5 and Xbox Series X have less headroom than 2026-era PCs.

Step 3: Reduce high-res textures on small objects. In the Content Browser, filter by Texture. Sort by Resource Size. Anything over 4 MB and used only on small props should probably be 1024 instead of 4096:

// In the Texture editor, set:
// Compression Settings: Default (DXT1/5)
// Max Texture Size: 1024
// LOD Group: World or WorldSpecular

Or set it via C++ for batch processing:

for (UTexture2D* Tex : Textures)
{
    Tex->MaxTextureSize = 1024;
    Tex->PostEditChange();
}

Step 4: Use Texture Groups. Texture Groups let you apply LOD bias across whole categories. For UI, set NeverStream. For Characters, apply moderate bias in shipping. Edit in Project Settings > Engine > Texture Streaming > TextureLODGroups, or via config:

[/Script/Engine.TextureLODSettings]
@TextureGroup=Group=World,LODBias=0,MaxLODSize=2048
@TextureGroup=Group=Character,LODBias=0,MaxLODSize=2048
@TextureGroup=Group=UI,LODBias=0,NeverStream=true

Step 5: Disable streaming on always-visible assets. For textures that are in every frame (sky, UI, main character), NeverStream = true keeps them resident full-size and avoids pool churn. Count them against overall GPU memory, not streaming budget.

Debug with Texture Streaming Stats

The most useful tools:

Green = ideal resolution. Red = under-streamed (pool exhausted). Blue = over-streamed (more resolution than needed).

Per-Actor Streaming Control

For specific actors that should prefetch textures regardless of camera distance (cinematic actor, hero NPC):

MeshComp->PrestreamTextures(5.0f, true);

This tells the streamer to bring those textures to full mip for the next 5 seconds regardless of screen size. Useful right before a cutscene triggers so cutscene actors do not show blurry.

Understanding the issue

This bug class falls into a pattern that's worth understanding beyond the specific case. In Unreal Engine, the underlying behavior is shaped by how the engine layers its abstractions - the public API you call, the runtime systems that respond, and the platform-specific implementations underneath. A bug at any layer can produce symptoms that look like they originate at a different layer. Triaging effectively means recognizing which layer the symptom belongs to, even when the gameplay code is what's visible.

The specific bug described above is the kind that surfaces during integration rather than unit testing. It depends on a combination of factors: the asset configuration, the runtime state, the platform's specific behavior. In isolation, each piece looks correct; in combination, the bug emerges. This is why thorough integration testing - playing the actual game in realistic conditions - catches things that automated tests miss.

Why this happens

Bugs of this class are particularly easy to ship past internal QA because they often depend on specific runtime conditions - hardware combinations, network states, or asset configurations that QA didn't reproduce. Players hit them in the wild, file reports that are hard to repro, and the bug accumulates negative reviews while engineering tries to recreate the failure mode.

At the engine level, the behavior comes from a deliberate design decision in Unreal. The engine team chose a particular trade-off - usually performance versus convenience, or generality versus specificity - and that trade-off has consequences when you push against it. Understanding the trade-off is what turns 'this bug is mysterious' into 'this bug is the expected consequence of this design'.

Verifying the fix

For shipping games, the safest verification is a staged rollout. Apply the fix to 1% of players for 24 hours; watch the affected metric; expand if green. Skipping the staged rollout means the verification is the entire player base, which is too high a stakes for most fixes.

Reproducibility is the prerequisite for verification. If you can't reliably reproduce the bug pre-fix, you can't reliably verify it post-fix. Spend time getting a clean reproduction before you write any fix code. The fix is fast once you understand the reproduction; the reproduction is the slow part.

Variations to watch for

There's almost always a less obvious case where the same problem applies. The reported case is the one a player hit; the related cases hide because they're rarer or affect fewer players. After fixing the reported case, search the codebase for the pattern - one fix often unlocks several.

Adjacent bugs often share a root cause. After fixing the case you've found, spend an hour searching the codebase for similar patterns. What's the same call with different arguments? The same data flow with a different entity type? The same lifecycle issue in a sibling system? Each match is a candidate for the same fix, or a related fix that prevents future bugs of the same class.

In production

Live games surface this bug class at scale. What's a rare edge case in development becomes a daily occurrence once you have a few thousand concurrent players. The class isn't 'this player has a unique setup'; it's 'one in N thousand sessions will trigger this exact combination'.

When triaging a similar issue in production, prioritize gathering data over hypothesizing causes. A player report describes a symptom; what you need is a build SHA, a session timestamp, and ideally a screen recording or session replay. With those, the bug becomes tractable. Without them, you're guessing at hypothetical reproductions that may not match what the player actually hit.

Performance considerations

Performance implications matter when this bug class scales with player count or asset count. A bug that fires once per session is annoying; a bug that fires once per frame compounds. After fixing, profile the affected code path under realistic load. The fix that's correct for one entity may be too slow for ten thousand.

Diagnostic approach

Before applying any fix, gather enough context to be confident you're addressing the actual cause and not a similar-looking symptom. The cheapest diagnostic step is reproducing the bug deterministically - if you can't get the same failure twice in a row, your fix attempts will be hard to evaluate. Lock down the reproduction first.

For Unreal-specific diagnostics, the editor's profiler is the canonical starting point. Capture a representative frame with the symptom present; compare against a frame without the symptom; the diff often points directly at the cause. If the symptom is non-deterministic, capture multiple frames and look for the pattern - the cause is usually a state transition or a specific input value rather than a continuous effect.

Tooling and ecosystem

The tooling around this bug class matters as much as the fix itself. Good logging, accessible profilers, and clear error messages turn 30-minute investigations into 5-minute ones. If your project doesn't have visibility into this code path, the first fix should add the visibility - the second fix uses it.

Within Unreal, the relevant diagnostic surfaces include the standard frame debugger, memory profiler, and engine-specific debug overlays. Each one shows a different facet of what's happening. The frame debugger reveals draw call ordering and state transitions; the memory profiler shows allocation patterns; the debug overlay reveals per-system state. Bugs that resist one tool usually surrender to another - the trick is knowing which tool to reach for first.

Edge cases and pitfalls

Platform-specific edge cases are worth enumerating explicitly. iOS handles backgrounding differently than Android; Windows handles focus changes differently than macOS. A fix that works on the development platform may not work on every target. Test on each shipping platform deliberately.

When writing a regression test for this fix, focus on the boundary conditions that surfaced the original bug. Tests that exercise the happy path catch obvious regressions; tests that exercise the boundary catch the subtler regressions that look like new bugs but are really the original returning. The latter are the tests that earn their keep over the long life of the project.

Team communication

Document the fix and its rationale in the commit message or attached engineering doc. Future engineers will encounter related issues; the rationale tells them whether your fix is reusable or specific to the case at hand. Without rationale, the fix gets reverted or copied incorrectly.

If this fix touches a system several engineers work in, a short writeup in the team's engineering channel helps. Not a full design doc - a paragraph explaining what was wrong, what's fixed, and what to watch for. Future engineers encountering similar symptoms will search for the fix; making it findable is a small investment that pays back later.

“Pool over budget means your scene wants more mips than you reserved. Raise the pool or lower the ask.”

Related Issues

For performance profiling more broadly, see Physics Constraint Drive Not Working. For content pipeline questions, SplineComponent Tangent Wrong Direction covers related asset issues.

Measure first, then raise pool or reduce ask. Small objects do not need 4K textures.