Add SpellCastAnimCallback to GameHandler, triggered on SMSG_SPELL_START
(start=true) and cleared on SMSG_SPELL_GO / SMSG_SPELL_FAILURE
(start=false) for both the player and other units.
Connect the callback in Application to play animation 3 (SpellCast) on
the player character, NPCs, and other players when they begin a cast.
The cast animation is one-shot (loop=false) so it auto-returns to Stand
when complete via the existing return-to-idle logic.
Also fire stop-cast on spell failure to cancel any stuck cast pose.
SMSG_EMOTE packets for NPCs and other players were received but the
emoteAnimCallback_ was never wired to the rendering layer. Register
the callback in application.cpp so that when the server sends an emote
animation ID, the corresponding CharacterRenderer instance plays it as
a one-shot animation (loop=false), falling back to idle on completion.
Lookups check creatureInstances_ first, then playerInstances_ so both
NPCs and other online players respond to server emote packets.
After reconnect, `creaturePermanentFailureGuids_` and `deadCreatureGuids_`
could retain stale entries for GUIDs not tracked in `creatureInstances_`
(creatures that failed to load or died before being spawned). These stale
entries would silently block re-spawning or cause wrong death state on the
fresh CREATE_OBJECTs the server sends after reconnect.
Clear both caches in the reconnect-to-same-map path so server state is
authoritative after every reconnect.
The previous reconnect fix caused loadOnlineWorldTerrain to run, which
cleared and reloaded all terrain tiles — unnecessarily heavy for a
reconnect where the map hasn't changed.
New path: when isInitialEntry=true and mapId==loadedMapId_, despawn all
tracked creature/player/GO instances from the renderer (proper cleanup),
clear all pending spawn queues, update player position, and return — the
terrain stays loaded and the server's fresh CREATE_OBJECTs repopulate
entities normally.
On disconnect/reconnect to the same map, entityManager was not cleared
and creatureInstances_ still held old entries from the previous session.
When the server re-sent CREATE_OBJECT for the same GUIDs, the spawn
callback's early-return guard (creatureInstances_.count(guid)) silently
dropped every NPC re-spawn, leaving the world empty.
Fixes:
- disconnect() now calls entityManager.clear() to purge stale entities
- WorldEntryCallback gains a bool isInitialEntry parameter (true on first
login or reconnect, false on in-world teleport/flight landing)
- Same-map optimization path skipped when isInitialEntry=true, so
loadOnlineWorldTerrain runs its full cleanup and properly despawns old
creature/player instances before the server refreshes them
- Nameplates: player names always rendered regardless of V-key toggle;
separate cull distance 40u (players/target) vs 20u (NPCs); cyan name
color for other players; fade alpha scales with cull distance
- Level-up: add expanding golden ring burst (3 staggered waves, 420u
max radius) + full-screen flash to renderDingEffect(); M2 LevelUp.m2
is still attempted as a bonus on top
- Vanilla tile loading: add AssetManager::setBaseFallbackPath() so that
when the primary manifest is an expansion-specific DBC-only subset
(e.g. Data/expansions/vanilla/), world terrain files fall back to
the base Data/ extraction; wired in Application::initialize()
- Warden: map a null guard page at address 0x0 in the Unicorn emulator
so NULL-pointer reads in the module don't crash with UC_ERR_MAP;
execution continues past the NULL read for better diagnostics
- game: clear pendingNameQueries on player out-of-range and DESTROY_OBJECT so
re-entering players get a fresh name query instead of being silently skipped
- game: add 5s periodic name resync scan that re-queries players with empty names
and no pending query, recovering from dropped CMSG_NAME_QUERY responses
- warden: fix UC_ERR_MAP by moving HEAP_BASE from 0x200000 to 0x20000000; the old
heap [0x200000, 0x1200000) overlapped the module at 0x400000, causing Unicorn to
reject the heap mapping and abort emulator initialisation
- warden: add early overlap check between module and heap regions to catch future
layout bugs at init time
- assets: add loadDBCOptional() which logs at DEBUG level when a DBC is absent,
for files that are not distributed on all expansions
- assets: use loadDBCOptional for Item.dbc (absent on Vanilla 1.12 clients) and
fall back to server-sent itemInfoCache displayInfoId for NPC weapon resolution
processAsyncNpcCompositeResults() had no per-frame budget cap, so when
many NPCs finished async skin compositing simultaneously (e.g. right
after world load), all results were finalized in a single frame causing
up to 284ms frame stalls. Apply the same 2ms budget pattern used by
processAsyncCreatureResults. Load screen still processes all pending
composites without the cap (unlimited=true).
The selection circle was positioned using the entity's game-logic
interpolator (entity->getX/Y/Z), while the actual M2 model is
positioned by CharacterRenderer's independent interpolator (moveInstanceTo).
These two systems can drift apart during movement, causing the circle
to appear under the wrong position relative to the visible model.
Fix: add CharacterRenderer::getInstancePosition / Application::getRenderPositionForGuid
and use the renderer's inst.position for XY (with footZ override for Z)
so the circle always tracks the rendered model exactly. Falls back to
the entity game-logic position when no CharacterRenderer instance exists.
Hearthstone post-teleport fix:
- Expand same-map hearthstone precache from 5x5 to 9x9 tiles so workers
have more tiles parsed before the player arrives at the bind point
- After same-map teleport arrival, enqueue the full load-radius tile grid
(17x17 = 289 tiles) at the new position so background workers immediately
start loading all WMOs/M2s visible from the new location
Quest tracker improvements:
- Clicking a quest in the tracker now opens the Quest Log (L)
- Remove NoInputs flag so the tracker window receives mouse events
- Show only tracked quests in tracker; fall back to all quests if none tracked
- Add Track/Untrack button in Quest Log details panel
- Abandoning a quest automatically untracks it
- Track state stored in GameHandler::trackedQuestIds_ (per-session)
Model batch submesh IDs and NPC geoset lists fire on every NPC spawn and
produce excessive log noise in normal gameplay. Downgrade to LOG_DEBUG.
Also downgrade per-equipment-slot DBC lookups from LOG_INFO to LOG_DEBUG.
When the player starts casting Hearthstone (spell IDs 6948/8690),
trigger background terrain loading at the bind point so tiles are
ready when the teleport fires.
- Add HearthstonePreloadCallback to GameHandler, called from
handleSpellStart when a Hearthstone cast begins.
- Application callback enqueues a 5×5 tile grid around the bind
point via precacheTiles() (same-map) or starts a file-cache warm
via startWorldPreload() (cross-map) during the ~10 s cast time.
- On same-map teleport arrival, call processAllReadyTiles() to
GPU-upload any tiles that finished parsing during the cast before
the first frame at the new position.
Fixes: player landing in unloaded terrain and falling after Hearthstone.
Hang/GPU device lost fix:
- M2_INSTANCES and WMO_INSTANCES finalization phases now create instances
incrementally (32 per step / 4 per step) instead of all at once, eliminating
the >1s main-thread stalls that caused GPU fence timeouts and device loss
M2 two-pass transparent rendering:
- Opaque/alpha-test batches render in pass 1, transparent/additive in pass 2
(back-to-front sorted) to fix wing transparency showing terrain instead of
trees — adds hasTransparentBatches flag to skip models with no transparency
Tile streaming improvements:
- Sort new load queue entries nearest-first so critical tiles load before
distant ones during fast taxi flight
- Increase taxi load radius 6→8 tiles, unload 9→12 for better coverage
Water refraction gated on FSR:
- Disable water refraction when FSR is not active (bugged without upscaling)
- Auto-disable refraction if FSR is turned off while refraction was on
Game object M2 models share the same default facing (+renderX) as
character models, so apply the same π/2 offset instead of π when
computing renderYawM2go from canonical yaw. This corrects street signs
and hanging shop signs that were 90° off after the server-yaw formula
fix.
Nameplates (health bar + name label) are now only rendered for the
currently targeted entity, matching WoW's default UI behaviour and
reducing visual noise.
Add PlayPositionalSoundCallback that carries both soundId and sourceGuid.
In Application, look up the source entity position and play via
AudioEngine::playSound3D(); fall back to playSound2D() when the entity
is unknown. Also read the 8-byte sourceGuid field from the packet
(previously the full 12-byte payload was ignored).
Add PlaySoundCallback to GameHandler (same pattern as PlayMusicCallback).
When SMSG_PLAY_SOUND arrives, resolve the soundId through SoundEntries.dbc
(fields 3-12 = files, field 23 = DirectoryBase) and play the first found
file as a 2-D sound effect via AudioEngine::playSound2D(). Previously the
opcode was parsed and dropped.
Correct SoundEntries.dbc field access in the PlayMusic callback: file names are at
fields 3-12 (not 2-11) and DirectoryBase is at field 23 (not 22). Field 2 is the
Name label string, not a file path.
Remove dead detectVoiceType(creatureEntry) from NpcVoiceManager — it was never
called; actual voice detection uses detectVoiceTypeFromDisplayId() in Application.
- Add Escape Menu → Instance Lockouts button opening a new panel
that lists active lockouts with instance name (from Map.dbc),
difficulty, time-until-reset countdown, and locked/extended status.
map name lookup is cached on first open.
- Fix uninitialized ChatType in sendChatMessage (default to SAY)
- Remove unused startWorld variable in handleMonsterMoveTransport
- Remove unused modelCached variable in spawnOnlineCreature
Eliminates all -Wunused-but-set-variable and -Wmaybe-uninitialized
warnings in the main translation units.
Add PlayMusicCallback to GameHandler so SMSG_PLAY_MUSIC (and the
vanilla 0x0103 alias) dispatch a soundId to the registered handler
instead of being silently consumed. Application.cpp registers the
callback, loads SoundEntries.dbc, resolves the first non-empty
Name+DirectoryBase into an MPQ path, and passes it to MusicManager
for non-looping playback. Resolves the TODO in the SMSG_PLAY_MUSIC
handler.
- Parse SMSG_ACHIEVEMENT_EARNED (guid + achievementId + PackedTime date)
and fire AchievementEarnedCallback for self, chat notify for others
- Add renderAchievementToast() to GameScreen: slides in from right,
gold-bordered panel with "Achievement Earned!" title + ID, 5s duration
with 0.4s slide-in/out animation and fade at end
- Add triggerAchievementToast(uint32_t) public method on GameScreen
- Wire AchievementEarnedCallback in application.cpp
- Add playAchievementAlert() to UiSoundManager, loads
Sound\Interface\AchievementSound.wav with level-up fallback
- SMSG_ALL_ACHIEVEMENT_DATA silently consumed (no tracker UI yet)
- Selective neighborhood clamp: only modify history when there's actual
motion or disocclusion — static pixels pass history through untouched,
preventing jitter-chasing from the shifting variance box
- Tonemapped accumulation: Reinhard tonemap before blend compresses bright
edges so they don't disproportionately cause jitter
- Jitter-aware sample weighting: blend 3-20% based on sample proximity
- Soft MV dead zone: smoothstep instead of step avoids spatial discontinuity
- Aggressive velocity response: 30%/px motion, 50% cap, 80% disocclusion
- Terrain loading: radius 3 (49 tiles) to prevent spawn hitches,
processOneReadyTile for smooth progress bar updates
FSR2 temporal upscaling:
- De-jitter scene color sampling (outUV - jitterUV) for frame-to-frame
consistency, eliminating the primary source of temporal jitter
- Remove luminance instability dampening (was causing excessive blur)
- Simplify to uniform 8% blend (de-jittered values are consistent)
- Gamma 2.0 for moderate neighborhood clamping
- Motion vector dead zone: zero sub-0.01px motion from float precision noise
Loading screen:
- Reduce tile load radius from 3 to 2 (25 tiles) for faster loading
- Process one tile per iteration for smooth progress bar updates
- Normal map CPU work (luminance→blur→Sobel) moved to background threads,
main thread only does GPU upload (~1-2ms vs 15-22ms per texture)
- Load screen warmup now waits until ALL spawn/equipment/gameobject queues
are drained before transitioning (prevents naked character, NPC pop-in)
- Exit condition: min 2s + 5 consecutive empty iterations, hard cap 15s
- Equipment queue processes 8 items per warmup iteration instead of 1
- Added LoadingScreen::renderOverlay() for future world-behind-loading use
Loading screen now calls processCreatureSpawnQueue(unlimited=true) which
removes the 1-upload-per-frame cap and 2ms time budget, allowing all pending
creature models to upload to GPU in bulk. Also increases concurrent async
background loads from 4 to 16 during load screen. Replaces 40-line inline
duplicate of processAsyncCreatureResults with the shared function.
Each loadTexture call was generating a normal/height map inline (3 full-image
passes: luminance + blur + Sobel). For models with 15-20 textures this added
30-40ms to the 70ms model upload. Now deferred to a per-frame budget (2/frame
in-game, 10/frame during load screen). Models render without POM until their
normal maps are ready.
Move all DBC lookups (CharSections, ItemDisplayInfo), texture path resolution,
and BLP decoding for humanoid NPCs to background threads. Only GPU texture
uploads remain on the main thread via pre-decoded BLP cache.
Move CPU-heavy BLP texture decoding from main thread to background worker
threads for all hot paths: terrain M2 models, WMO doodad M2s, WMO textures,
creature models, and gameobject WMOs. Each renderer (M2, WMO, Character) now
accepts a pre-decoded BLP cache that loadTexture() checks before falling back
to synchronous decode.
Defer WMO normal/height map generation (3 per-pixel passes: luminance, box
blur, Sobel) during terrain streaming finalization — this was the dominant
remaining bottleneck after BLP pre-decoding.
Terrain streaming stalls: 1576ms → 124ms worst case.
- Replace per-frame VMA alloc/free of material UBOs with a ring buffer in
CharacterRenderer (~500 allocations/frame eliminated)
- Batch all ready terrain tiles into a single GPU upload during load screen
(processAllReadyTiles instead of one-at-a-time with individual fence waits)
- Lift per-frame creature/GO spawn budgets during load screen warmup phase
- Add background world preloader: saves last world position to disk, pre-warms
AssetManager file cache with ADT files starting at app init (login screen)
so terrain workers get instant cache hits when Enter World is clicked
- Distance-filter expensive collision guard to 8-unit melee range
- Merge 3 CharacterRenderer update loops into single pass
- Time-budget instrumentation for slow update stages (>3ms threshold)
- Count-based async creature model upload budget (max 3/frame in-game)
- 1-per-frame game object spawn + per-doodad time budget for transport loading
- Use deque for creature spawn queue to avoid O(n) front-erase
Was waiting for all ~50 tiles (radius 4) to fully prepare + finalize
before entering the game. Now loads only the immediate surrounding tiles
during the loading screen, then restores the full radius for in-game
streaming. setLoadRadius just sets an int — actual loading happens lazily
via background workers during the game loop.
- Async creature model loading: M2 file I/O and parsing on background threads
via std::async, GPU upload on main thread when ready (MAX_ASYNC_CREATURE_LOADS=4)
- CharSections.dbc lookup cache: O(1) hash lookup instead of O(N) full DBC scan
per humanoid NPC spawn (was scanning thousands of records twice per spawn)
- Frame time budget: 4ms cap on creature spawn processing per frame
- Wolf/worg model name check cached per modelId (was doing tolower+find per
hostile creature per frame)
- Weapon attach throttle: max 2 per 1s tick (was attempting all unweaponized NPCs)
- Separate texture application tracking (displayIdTexturesApplied_) so async-loaded
models still get skin/equipment textures applied correctly
- Add case-insensitive "glass" detection for WMO window materials
- Make instance (WMO-only) glass highly transparent (12-35% alpha)
so underwater scenes are visible through Deeprun Tram windows
- Keep normal world windows at existing opacity (40-95% alpha)
- Disable shadow mapping for interior WMO groups to fix dark
indoor areas like Ironforge
- Fix NULL renderer pointers by moving TransportManager connection after
initializeRenderers for WMO-only maps
- Fix tram direction by negating DBC TransportAnimation X/Y local offsets
before serverToCanonical conversion
- Implement client-side M2 transport boarding via proximity detection
(server doesn't send transport attachment for trams)
- Use position-delta approach: player keeps normal movement while
transport's frame-to-frame motion is applied on top
- Prevent server movement packets from clearing client-side M2 transport
state (isClientM2Transport guard)
- Fix getPlayerWorldPosition for M2 transports: simple canonical addition
instead of render-space matrix multiplication
- Add movementSuppressTimer to camera controller that forces all movement
keys to read as false, preventing held W key from carrying through
loading screens (fixes always-running-forward after instance portals)
- Increase shadow frustum default from 60 to 72 units (+20%)
- Make shadow distance configurable via setShadowDistance() (40-200 range)
- Add shadow distance slider in Video settings tab (persisted to config)
- Stronger wall collision push (0.35/0.15) and swept push (0.45/0.25)
for interior/exterior WMOs to reduce clipping through tunnel walls
- Use all triangles (not just pre-classified walls) for collision checks
- Allow invisible collidable triangles (MOPY 0x01 without 0x20) to block
- Pass insideWMO flag to all collision callers, match swim sweep to ground
- Widen swept hit detection radius from 0.15 to 0.25
- Restrict camera zoom to 12 units inside WMO interiors
- Fix /unstuck launching player above WMOs: remove +20 fallback, use
gravity when no floor found
- Slash and Enter keys always focus chat unless already typing
The glm::quat(w,x,y,z) constructor was receiving swapped X/Y components,
causing doodads like the Deeprun Tram gears to be oriented horizontally
instead of vertically. Also use createInstanceWithMatrix for instance WMO
doodads to preserve full rotation from the quaternion.
Classify light shafts, portals, spotlights, bubbles, and similar M2
doodads as spell effects so they render with additive blending instead
of as solid opaque objects.
Set camera yaw from server orientation on world load so teleports face
the correct direction.
Reduce area trigger minimum radius (3.0 sphere, 4.0 box) to prevent
premature portal firing near tram entrances.
WMO interior doodads (gears, decorations) were blocking player movement
via M2 collision. Skip collision for all WMO doodad M2 instances since
the WMO itself handles wall collision.
Also filter WMO wall collision using MOPY per-triangle flags: only
rendered+collidable triangles block the player, skipping invisible
collision hulls.
Revert tram portal extended range (no longer needed with collision fix).
- Equipment-driven geoset selection: read GeosetGroup1 from ItemDisplayInfo
for legs/feet/chest to pick covered mesh variants (1302+ pants, 402+ boots,
802+ sleeves) instead of always defaulting to bare geosets
- Prevent per-instance skin override from replacing baked/composited armor
textures on equipped NPCs
- Set bald NPC hair texture slot to skin texture so scalp isn't white
- Skip white fallback textures in per-instance hair overrides
- Remove debug texture dump, reduce NPC logging to DEBUG level
- Remove bogus 2-byte skip after materialId in MLIQ parser that shifted
all vertex heights and tile flags by 2 bytes (garbage data)
- Skip liquid loading for interior WMO groups (flag 0x2000) to prevent
indoor water from rendering as outdoor canal water
- Clear movement inputs on teleport/portal to prevent auto-running after
zone transfer (held keys persist through loading screen)
- Fix Stormwind barracks floor: interior WMO groups named "facade" were
incorrectly marked as LOD shells and hidden when close. Add !isIndoor
guard to all LOD detection conditions so interior groups always render.
- Fix water exit stair clipping: anchor lastGroundZ to current position
on swim exit, set grounded=true for full step-up budget, add upward
velocity boost to clear stair lip geometry.
- Re-enable NPC humanoid equipment geosets (kEnableNpcHumanoidOverrides)
so guards render with proper armor instead of underwear.
- Keep instance portal GameObjects animated (spinning/glowing) instead
of freezing all GO animations indiscriminately.
- Fix equipment disappearing after instance round-trip by resetting
dirty tracking on world reload.
- Fix multi-doodad-set loading: load both set 0 (global) and placement-
specific doodad set, with dedup to avoid double-loading.
- Clear placedWmoIds in softReset/unloadAll to prevent stale dedup.
- Apply MODF rotation to instance WMOs, snap player to WMO floor.
- Re-enable rebuildSpatialIndex in setInstanceTransform.
- Store precomputeFloorCache results in precomputed grid.
- Add F8 debug key for WMO floor diagnostics at player position.
- Expand mapIdToName with all Classic/TBC/WotLK instance map IDs.
- NPC hair/skin textures now use per-instance overrides instead of shared
model-level textures, so each NPC shows its own hair color/style
- Hair/skin DBC lookup runs for every NPC instance (including cached models)
rather than only on first load
- Fix keyframe binary search to use float comparison matching original
linear scan semantics
- Split M2 instances into fast-path index lists (animated, particle-only,
particle-all, smoke) to avoid iterating all 46K instances per frame
- Cache model flags (hasAnimation, disableAnimation, isSmoke, etc.) on
M2Instance struct to eliminate per-frame hash lookups
- Replace full rebuildSpatialIndex on position/transform updates with
incremental grid cell remove+add, preventing 8.5ms/frame rebuild cost
- Advance animTime for all instances (texture UV animation) but only
compute bones and particles for the ~3K that need it
M2_UPDATE: 10.7ms → 2.0ms, FPS: 35 → 55-59
When the GPU device is lost (unrecoverable Vulkan error), the app now
closes cleanly instead of looping with a black screen. Also adds
vk_context.hpp include for the isDeviceLost() check.
