War Thunder — best graphics settings (2026)

Typical impact 0-5% · 4% fps cost

In War Thunder, we recommend Texture Quality at Max (4% fps cost).

Controls the maximum mipmap resolution loaded for surface textures. Higher levels stream larger texture maps (2K/4K) from disk into VRAM via the texture streaming pool. The GPU samples these during fragment shading using the currently bound sampler state. The FPS cost is minimal when VRAM is sufficient because texture fetch latency is hidden by the cache hierarchy, but exceeding VRAM capacity triggers page-faulting and hitching as textures are swapped between system RAM and VRAM.

In War Thunder: Dagor Engine 6.5 manages a per-session texture streaming pool. At Low and Medium, 1K/2K diffuse and normal maps are loaded for vehicle hulls and terrain surfaces. High and Max stream 4K asset packs, notably for vehicle skins and hangar models. In a live match the VRAM delta between Medium and Max is roughly 600–900 MB; stay at High unless you have 6 GB+ VRAM free to avoid mid-match streaming stalls when many unique vehicle models are on screen simultaneously.

Typical impact 8-25% · 6% fps cost

In War Thunder, we recommend Shadow Quality at Medium (6% fps cost).

Controls shadow map resolution, filtering method, and cascade count for dynamic shadows. The engine renders the scene from each light source perspective into depth-only shadow map textures. Higher settings increase shadow map resolution (1024 to 4096 texels), add more cascaded shadow map splits for the directional light (improving near-field resolution), and enable softer PCF or PCSS filtering which requires more depth comparison samples per pixel during the lighting pass.

In War Thunder: Dagor Engine renders cascaded shadow maps for the directional sun light across all visible vehicles, terrain patches, and trees. Off eliminates all dynamic shadows entirely — a dramatic visual regression. Max increases cascade resolution to 4096 texels and adds an extra far-cascade split, requiring the engine to re-render visible shadow-casting geometry in additional passes. Ground forces maps with dozens of vehicles and dense foliage make this the heaviest single setting; Medium strikes the best balance at roughly half the shadow pass cost of Max.

Typical impact 2-15% · 3% fps cost

In War Thunder, we recommend Anti-Aliasing at TAA (3% fps cost).

Smooths jagged edges (aliasing) on geometric boundaries. FXAA is a single-pass edge-detection blur — cheap but softens the image. TAA accumulates multiple frames using motion vectors, sampling sub-pixel jitter offsets to reconstruct smoother edges — moderate cost with potential ghosting. SMAA uses pattern-matching edge detection with a more intelligent blend. MSAA runs the rasterizer at 2x/4x the sample count, evaluating coverage for each triangle edge — expensive because it multiplies ROP work and render target memory, but produces sharp geometry edges without blur.

In War Thunder: Dagor Engine offers four AA tiers. Off is raw aliased output. FXAA is a cheap single-pass edge blur — minimal cost but softens vehicle markings and reticle edges. TAA uses reprojection from the depth and motion vector buffers; it improves edge quality but introduces sub-pixel ghosting on fast-moving aircraft and shell tracers. 4x SSAA renders the scene at four times the pixel count before downsampling — the sharpest result but a 3–4x increase in GPU fragment work, essentially only viable on high-end GPUs at 1080p.

Typical impact 0-1% · 2% fps cost

In War Thunder, we recommend Anisotropic Filtering at 16x (2% fps cost).

A dedicated anisotropic texture filtering control that adjusts the maximum number of additional texture samples taken per pixel for surfaces at steep viewing angles. At 16x, up to 16 taps are taken along the anisotropy axis in the texture unit hardware. This computation is overlapped with ALU work in the shader pipeline, so even maximum settings cost virtually nothing on modern architectures. The visual payoff is significant — eliminates blurriness on ground planes and distant walls.

In War Thunder: War Thunder's terrain and road surfaces are viewed at extreme oblique angles, especially from aircraft. Dagor Engine handles AF in dedicated texture unit hardware, so the transition from Off (heavily blurred ground textures at range) to 16x (sharp runway markings and terrain detail at all angles) costs under 1% on any GPU made in the last decade. Set this to 16x permanently — the visual improvement to ground textures viewed from the cockpit is substantial and the performance cost is negligible.

Typical impact 3-15% · 3% fps cost

In War Thunder, we recommend Effect Quality at Medium (3% fps cost).

Controls the visual fidelity of gameplay effects including explosions, weapon impacts, ability VFX, and environmental interactions. Higher settings increase particle emitter counts per effect, use higher-resolution flipbook or mesh particles instead of simple sprites, enable GPU particle simulation via compute shaders, and add dynamic lighting from effects (each explosion spawning a temporary point light). The cost is highly variable — intense combat with multiple overlapping effects can produce 4-8x overdraw from layered transparent particles.

In War Thunder: Controls the resolution, particle count, and lighting interaction of combat VFX — shell impacts, engine fires, explosion fireballs, and smoke trails. At High, explosions spawn dense GPU-simulated particle systems with a temporary dynamic point light that re-evaluates nearby vehicle shading. In large Ground Forces battles with many simultaneous impacts, High can contribute noticeable frame-time spikes due to overdraw from layered transparent particles. Medium reduces emitter counts and disables per-explosion dynamic lights, recovering 4–8% in busy combat scenarios.

Typical impact 3-10% · 6% fps cost

In War Thunder, we recommend Terrain Quality at High (6% fps cost).

Controls terrain mesh tessellation density, heightmap resolution, and texture splatting layer count. Higher settings increase the number of terrain patches submitted to the tessellation pipeline (hull/domain shaders) and enable more material blend layers per patch — each layer requiring its own albedo, normal, and roughness texture fetches. Terrain rendering is particularly bandwidth-heavy because the large screen coverage means nearly every pixel on the ground plane runs the full multi-layer blend shader.

In War Thunder: Dagor Engine streams terrain at massive distances for aerial combat. Ultra loads highest-res heightmaps — important for ground forces maps.

Typical impact 3-12% · 2% fps cost

In War Thunder, we recommend Water Quality at Medium (2% fps cost).

Governs the fidelity of water surface rendering including wave simulation, tessellation, refraction, and reflection techniques. Higher settings enable GPU-computed FFT wave simulation in compute shaders, adaptive tessellation for displacement mapping on the water mesh, screen-space refraction via distorted depth buffer sampling, and planar or screen-space reflections. The reflection pass may render the scene a second time from a mirrored viewpoint, effectively doubling draw calls for visible water surfaces.

In War Thunder: War Thunder's naval and coastal maps use Dagor Engine's FFT-based water surface simulation. Low renders flat water with a simple specular highlight. Medium adds tiling normal-mapped wave detail and basic foam. High enables the full FFT wave displacement computed in a compute shader pass, plus screen-space refraction for underwater visibility and higher-frequency specular from dynamic wave normals. The cost is confined to pixels covering water surfaces — High is affordable on most maps, but on the large naval Pacific maps where water fills 70%+ of the screen, High can cost 6–10% versus Low.

Typical impact 3-12% · 6% fps cost

In War Thunder, we recommend Tree Quality at High (6% fps cost).

Controls tree model LOD tiers, branch geometry detail, and the distance at which trees transition from 3D meshes to billboard impostor sprites. Higher settings keep full geometry trees visible further, increasing triangle count and draw call overhead. Tree foliage is particularly expensive because alpha-tested or alpha-blended leaf cards cause significant overdraw — multiple semi-transparent layers are rasterized per pixel. Trees also cast complex shadow maps due to their irregular silhouettes.

In War Thunder: Dagor Engine uses a layered impostor system for trees — at Low, trees transition to flat billboard sprites at close range. High and Max keep full 3D branch-and-leaf geometry at greater distances and increase canopy density on forested European maps like Hurtgen Forest and Advance to the Rhine. Alpha-tested leaf cards cause significant overdraw in dense forest terrain; Max can increase foliage rendering cost by 8–12% in tree-heavy ground maps. For aerial players, Low is invisible from altitude. Ground forces players benefit most from High to correctly read cover and concealment.

Typical impact 2-8% · 4% fps cost

In War Thunder, we recommend Clouds at Medium (4% fps cost).

Controls cloud rendering method and complexity. "Fast" renders clouds as a flat 2D texture layer — a single textured quad at a fixed height. "Fancy" uses volumetric cloud rendering — ray-marching through a 3D noise field (Worley or Perlin-Worley noise) to produce realistic cloud formations with self-shadowing. The volumetric approach evaluates density and lighting at each ray march step, sampling the shadow map to determine which parts of the cloud are in sunlight. In games like War Thunder, volumetric clouds are gameplay-relevant — aircraft can hide in them.

In War Thunder: Volumetric clouds in War Thunder are used for aerial combat concealment. Disabling removes tactical cover — most players keep them on despite 5-8% cost.

Typical impact 3-12% · 4% fps cost

In War Thunder, we recommend Particle Quality at High (4% fps cost).

Controls particle system density, simulation complexity, and render quality. Higher settings increase maximum particle count per emitter, enable GPU-driven particle simulation in compute shaders (position, velocity, lifetime, collision), and use soft particle blending (sampling the depth buffer to fade particles near surface intersections). The overdraw cost from thousands of alpha-blended billboard quads is the primary performance concern — each particle that overlaps another requires a separate blending operation.

In War Thunder: Governs the maximum particle count for persistent environmental effects — vehicle engine exhaust, dust kicked up by ground movement, water splashes, and secondary debris. At High, Dagor Engine maintains a larger GPU particle buffer with per-particle physics-based collision against the terrain depth buffer. Low hard-caps emitter counts and disables terrain collision response for particles, which reduces both compute shader simulation cost and the overdraw from large lingering dust clouds behind fast-moving vehicles. The savings are most visible in combined-arms matches where 16 ground vehicles simultaneously generate exhaust and impact dust.

Typical impact 2-8% · 3% fps cost

In War Thunder, we recommend SSAO at Low (3% fps cost).

A dedicated Screen-Space Ambient Occlusion toggle that adds soft contact shadows in crevices and corners by sampling the depth buffer in a hemisphere around each pixel. The implementation dispatches a fullscreen compute shader pass that, for each pixel, takes 8-32 depth samples in a randomized hemisphere oriented along the surface normal (reconstructed from the depth buffer). Each sample tests whether nearby geometry occludes ambient light. A temporal filter and spatial blur pass clean up the noisy raw AO buffer. The result is multiplied into the ambient lighting term during the final composite. Cost scales linearly with sample count and screen resolution.

In War Thunder: Dagor Engine's SSAO runs as a post-process pass using depth buffer hemisphere sampling. Off removes the pass entirely — contact shadows between vehicle hull panels, wheel wells, and terrain crevices disappear, giving vehicles a flatter, less grounded appearance. Low uses a reduced sample count (approximately 8 taps) at half resolution with a spatial blur. High increases to a full-resolution pass with a higher sample count and tighter temporal filtering. The visual difference between Off and Low is significant for vehicle model fidelity; the difference between Low and High is subtle. Low provides the best cost-to-quality ratio.

Typical impact 15-60% · no measurable cost

In War Thunder, we recommend Render Resolution at 100% (no measurable cost).

Controls the internal 3D rendering resolution as a percentage of the display output resolution, independent of any AI upscaling (DLSS/FSR). The engine renders the scene at the specified fraction of native resolution — this scales the render target dimensions, directly reducing the number of fragment shader invocations, texture fetches, and ROP output pixels. A 50% render scale produces 25% of the total pixels of native resolution. The reduced-resolution image is then upscaled to display resolution using a simple bilinear or bicubic filter. This is the most direct FPS/quality tradeoff available.

In War Thunder: Without DLSS or FSR available in Dagor Engine, this is the only resolution scaling mechanism in War Thunder. At 75%, fragment shader invocations drop to 56% of native, producing a proportional FPS gain at the cost of image softness on vehicle markings and distant targets — a real readability concern in competitive play. 125% and 150% supersample for sharper edges on high-end GPUs. At 1080p on a mid-range GPU, dropping from 100% to 75% typically yields 25–35% more FPS. The bilinear upscale quality is inferior to DLSS/FSR, so the trade-off is sharper than in titles with AI upscaling support.

Typical impact 0% · no measurable cost

In War Thunder, we recommend V-Sync at On (no measurable cost).

Synchronizes the GPU's framebuffer swap with the monitor's vertical blanking interval to prevent screen tearing. When enabled, the GPU holds the completed frame until the monitor signals it is ready. If the GPU cannot maintain the refresh rate, VSync forces the frame to wait for the next blanking interval, causing framerate to drop to a fraction (e.g., 60fps to 30fps on a 60Hz display). This introduces up to one full frame of input latency. Triple buffering mitigates the fractional drop but adds more latency.

In War Thunder: Dagor Engine's VSync implementation locks swap to the monitor's refresh interval, eliminating tearing on the large sky and terrain surfaces where it is most visible in War Thunder. At framerates below the refresh rate the engine will drop to a synchronised half-rate (e.g., 30 fps on a 60 Hz display), adding up to one frame of input latency — problematic in air-to-air engagements where lead calculation depends on low-latency control response. Use the in-game frame cap instead of VSync if your hardware consistently exceeds the monitor refresh rate, or combine VSync with a slightly lower frame cap to prevent the half-rate drop.

Typical impact -30-80% · no measurable cost

In War Thunder, the recommended preset leaves NVIDIA DLSS off — little visual loss for the frames it returns.

Deep Learning Super Sampling — NVIDIA's AI-based temporal upscaling that runs on dedicated Tensor Core hardware. The engine renders at a lower internal resolution and feeds the reduced-resolution frame, motion vectors, and depth buffer to a neural network that reconstructs a high-resolution output. DLSS 3+ adds optical flow-based frame generation on Ada/Blackwell architectures. The FPS gain comes from rendering fewer pixels — Quality mode renders ~67% of native pixels, Performance ~50%, Ultra Performance ~33%.