mirror of
https://github.com/Kelsidavis/WoWee.git
synced 2026-05-09 10:33:51 +00:00
feat(rendering): add HiZ occlusion culling & fix WMO interior shadows
Implement GPU-driven Hierarchical-Z occlusion culling for M2 doodads using a depth pyramid built from the previous frame's depth buffer. The cull shader projects bounding spheres via prevViewProj (temporal reprojection) and samples the HiZ pyramid to reject hidden objects before the main render pass. Key implementation details: - Separate early compute submission (beginSingleTimeCommands + fence wait) eliminates 2-frame visibility staleness - Conservative safeguards prevent false culls: screen-edge guard, full VP row-vector AABB projection (Cauchy-Schwarz), 50% sphere inflation, depth bias, mip+1, min screen size threshold, camera motion dampening (auto-disable on fast rotations), and per-instance previouslyVisible flag tracking - Graceful fallback to frustum-only culling if HiZ init fails Fix dark WMO interiors by gating shadow map sampling on isInterior==0 in the WMO fragment shader. Interior groups (flag 0x2000) now rely solely on pre-baked MOCV vertex-color lighting + MOHD ambient color. Disable interiorDarken globally (was incorrectly darkening outdoor M2s when camera was inside a WMO). Use isInsideInteriorWMO() instead of isInsideWMO() for correct indoor detection. New files: - hiz_system.hpp/cpp: pyramid image management, compute pipeline, descriptors, mip-chain build dispatch, resize handling - hiz_build.comp.glsl: MAX-depth 2x2 reduction compute shader - m2_cull_hiz.comp.glsl: frustum + HiZ occlusion cull compute shader - test_indoor_shadows.cpp: 14 unit tests for shadow/interior contracts Modified: - CullUniformsGPU expanded 128->272 bytes (HiZ params, viewProj, prevViewProj) - Depth buffer images gain VK_IMAGE_USAGE_SAMPLED_BIT for HiZ reads - wmo.frag.glsl: interior branch before unlit, shadow skip for 0x2000 - Render graph: hiz_build + compute_cull disabled (run in early compute) - .gitignore: ignore compiled .spv binaries - MEGA_BONE_MAX_INSTANCES: 2048 -> 4096 Signed-off-by: Pavel Okhlopkov <pavel.okhlopkov@flant.com>
This commit is contained in:
Pavel Okhlopkov
parent
17c1e3ea3b
commit
4b9b3026f4
17 changed files with 1317 additions and 35 deletions
|
|
@ -23,6 +23,7 @@
|
|||
#include "rendering/character_preview.hpp"
|
||||
#include "rendering/wmo_renderer.hpp"
|
||||
#include "rendering/m2_renderer.hpp"
|
||||
#include "rendering/hiz_system.hpp"
|
||||
#include "rendering/minimap.hpp"
|
||||
#include "rendering/world_map.hpp"
|
||||
#include "rendering/quest_marker_renderer.hpp"
|
||||
|
|
@ -580,7 +581,6 @@ bool Renderer::initialize(core::Window* win) {
|
|||
overlaySystem_ = std::make_unique<OverlaySystem>(vkCtx);
|
||||
renderGraph_->registerResource("shadow_depth");
|
||||
renderGraph_->registerResource("reflection_texture");
|
||||
renderGraph_->registerResource("cull_visibility");
|
||||
renderGraph_->registerResource("scene_color");
|
||||
renderGraph_->registerResource("scene_depth");
|
||||
renderGraph_->registerResource("final_image");
|
||||
|
|
@ -672,6 +672,10 @@ void Renderer::shutdown() {
|
|||
}
|
||||
|
||||
LOG_DEBUG("Renderer::shutdown - m2Renderer...");
|
||||
if (hizSystem_) {
|
||||
hizSystem_->shutdown();
|
||||
hizSystem_.reset();
|
||||
}
|
||||
if (m2Renderer) {
|
||||
m2Renderer->shutdown();
|
||||
m2Renderer.reset();
|
||||
|
|
@ -798,6 +802,17 @@ void Renderer::applyMsaaChange() {
|
|||
|
||||
if (minimap) minimap->recreatePipelines();
|
||||
|
||||
// Resize HiZ pyramid (depth format/MSAA may have changed)
|
||||
if (hizSystem_) {
|
||||
auto ext = vkCtx->getSwapchainExtent();
|
||||
if (!hizSystem_->resize(ext.width, ext.height)) {
|
||||
LOG_WARNING("HiZ resize failed after MSAA change");
|
||||
if (m2Renderer) m2Renderer->setHiZSystem(nullptr);
|
||||
hizSystem_->shutdown();
|
||||
hizSystem_.reset();
|
||||
}
|
||||
}
|
||||
|
||||
// Selection circle + overlay + FSR use lazy init, just destroy them
|
||||
if (overlaySystem_) overlaySystem_->recreatePipelines();
|
||||
if (postProcessPipeline_) postProcessPipeline_->destroyAllResources(); // Will be lazily recreated in beginFrame()
|
||||
|
|
@ -846,6 +861,16 @@ void Renderer::beginFrame() {
|
|||
}
|
||||
// Recreate post-process resources for new swapchain dimensions
|
||||
if (postProcessPipeline_) postProcessPipeline_->handleSwapchainResize();
|
||||
// Resize HiZ depth pyramid for new swapchain dimensions
|
||||
if (hizSystem_) {
|
||||
auto ext = vkCtx->getSwapchainExtent();
|
||||
if (!hizSystem_->resize(ext.width, ext.height)) {
|
||||
LOG_WARNING("HiZ resize failed — disabling occlusion culling");
|
||||
if (m2Renderer) m2Renderer->setHiZSystem(nullptr);
|
||||
hizSystem_->shutdown();
|
||||
hizSystem_.reset();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// Acquire swapchain image and begin command buffer
|
||||
|
|
@ -864,6 +889,31 @@ void Renderer::beginFrame() {
|
|||
// Update per-frame UBO with current camera/lighting state
|
||||
updatePerFrameUBO();
|
||||
|
||||
// ── Early compute: HiZ pyramid build + M2 frustum/occlusion cull ──
|
||||
// These run in a SEPARATE command buffer submission so the GPU executes
|
||||
// them immediately. The CPU then reads the fresh visibility results
|
||||
// before recording the main render pass — eliminating the 2-frame
|
||||
// staleness that occurs when compute + render share one submission.
|
||||
if (m2Renderer && camera && vkCtx) {
|
||||
VkCommandBuffer computeCmd = vkCtx->beginSingleTimeCommands();
|
||||
uint32_t frame = vkCtx->getCurrentFrame();
|
||||
|
||||
// Build HiZ depth pyramid from previous frame's depth buffer
|
||||
if (hizSystem_ && hizSystem_->isReady()) {
|
||||
VkImage depthSrc = vkCtx->getDepthCopySourceImage();
|
||||
hizSystem_->buildPyramid(computeCmd, frame, depthSrc);
|
||||
}
|
||||
|
||||
// Dispatch GPU frustum + HiZ occlusion culling
|
||||
m2Renderer->dispatchCullCompute(computeCmd, frame, *camera);
|
||||
|
||||
vkCtx->endSingleTimeCommands(computeCmd);
|
||||
|
||||
// Ensure GPU→CPU buffer writes are visible to host (non-coherent memory).
|
||||
m2Renderer->invalidateCullOutput(frame);
|
||||
// Visibility results are now in cullOutputMapped_[frame], readable by CPU.
|
||||
}
|
||||
|
||||
// --- Off-screen pre-passes ---
|
||||
// Build frame graph: registers pre-passes as graph nodes with dependencies.
|
||||
// compile() topologically sorts; execute() runs them with auto barriers.
|
||||
|
|
@ -1489,7 +1539,9 @@ void Renderer::renderWorld(game::World* world, game::GameHandler* gameHandler) {
|
|||
if (parallelRecordingEnabled_) {
|
||||
// --- Pre-compute state + GPU allocations on main thread (not thread-safe) ---
|
||||
if (m2Renderer && cameraController) {
|
||||
m2Renderer->setInsideInterior(cameraController->isInsideWMO());
|
||||
// Use isInsideInteriorWMO (flag 0x2000) — not isInsideWMO which includes
|
||||
// outdoor WMO groups like archways/bridges that should receive shadows.
|
||||
m2Renderer->setInsideInterior(cameraController->isInsideInteriorWMO());
|
||||
m2Renderer->setOnTaxi(cameraController->isOnTaxi());
|
||||
}
|
||||
if (wmoRenderer) wmoRenderer->prepareRender();
|
||||
|
|
@ -1734,7 +1786,8 @@ void Renderer::renderWorld(game::World* world, game::GameHandler* gameHandler) {
|
|||
|
||||
if (m2Renderer && camera && !skipM2) {
|
||||
if (cameraController) {
|
||||
m2Renderer->setInsideInterior(cameraController->isInsideWMO());
|
||||
// Use isInsideInteriorWMO (flag 0x2000) for correct indoor detection
|
||||
m2Renderer->setInsideInterior(cameraController->isInsideInteriorWMO());
|
||||
m2Renderer->setOnTaxi(cameraController->isOnTaxi());
|
||||
}
|
||||
m2Renderer->prepareRender(frameIdx, *camera);
|
||||
|
|
@ -1887,6 +1940,23 @@ bool Renderer::initializeRenderers(pipeline::AssetManager* assetManager, const s
|
|||
spellVisualSystem_->initialize(m2Renderer.get());
|
||||
}
|
||||
}
|
||||
|
||||
// HiZ occlusion culling — temporal reprojection.
|
||||
// The HiZ pyramid is built from the previous frame's depth buffer. The cull
|
||||
// compute shader uses prevViewProj to project objects into the previous frame's
|
||||
// screen space so that depth samples match the pyramid, eliminating flicker
|
||||
// caused by camera movement between frames.
|
||||
if (!hizSystem_ && m2Renderer && vkCtx) {
|
||||
hizSystem_ = std::make_unique<HiZSystem>();
|
||||
auto extent = vkCtx->getSwapchainExtent();
|
||||
if (hizSystem_->initialize(vkCtx, extent.width, extent.height)) {
|
||||
m2Renderer->setHiZSystem(hizSystem_.get());
|
||||
LOG_INFO("HiZ occlusion culling initialized (", extent.width, "x", extent.height, ")");
|
||||
} else {
|
||||
LOG_WARNING("HiZ occlusion culling unavailable — falling back to frustum-only culling");
|
||||
hizSystem_.reset();
|
||||
}
|
||||
}
|
||||
if (!wmoRenderer) {
|
||||
wmoRenderer = std::make_unique<WMORenderer>();
|
||||
wmoRenderer->initialize(vkCtx, perFrameSetLayout, assetManager);
|
||||
|
|
@ -2627,7 +2697,6 @@ void Renderer::buildFrameGraph(game::GameHandler* gameHandler) {
|
|||
|
||||
auto shadowDepth = renderGraph_->findResource("shadow_depth");
|
||||
auto reflTex = renderGraph_->findResource("reflection_texture");
|
||||
auto cullVis = renderGraph_->findResource("cull_visibility");
|
||||
|
||||
// Minimap composites (no dependencies — standalone off-screen render target)
|
||||
renderGraph_->addPass("minimap_composite", {}, {},
|
||||
|
|
@ -2670,13 +2739,6 @@ void Renderer::buildFrameGraph(game::GameHandler* gameHandler) {
|
|||
renderReflectionPass();
|
||||
});
|
||||
|
||||
// GPU frustum cull compute → outputs cull_visibility
|
||||
renderGraph_->addPass("compute_cull", {}, {cullVis},
|
||||
[this](VkCommandBuffer cmd) {
|
||||
if (m2Renderer && camera)
|
||||
m2Renderer->dispatchCullCompute(cmd, vkCtx->getCurrentFrame(), *camera);
|
||||
});
|
||||
|
||||
renderGraph_->compile();
|
||||
}
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue