Fix toLowerInPlace() which was accidentally self-recursive (would stack
overflow on any Lua string lowering). Remove 30 redundant
if(addonEventCallback_) wrappers around pure fireAddonEvent blocks.
Extract color constants in performance_hud.cpp (24 inline literals).
Replace ~37 remaining C-style casts with static_cast across 16 files.
Extract named color constants (kColorRed/Green/Yellow/Gray) and dialog
window flags (kDialogFlags) in game_screen.cpp, replacing 72 inline
literals. Normalize keybinding_manager.hpp to #pragma once.
- Extract guidToUnitId(), getQuestTitle(), findQuestLogEntry() helpers
to replace 14 duplicated GUID-to-unitId patterns and 7 quest log
search patterns in game_handler.cpp
- Remove duplicate #include in renderer.cpp
- Remove commented-out model cleanup code in terrain_manager.cpp
- Replace C-style casts with static_cast in auth and transport code
Request 2 queues from the graphics family when available (NVIDIA
exposes 16, AMD 2+). Upload batches now submit to queue[1] while
rendering uses queue[0], enabling parallel GPU transfers without
queue-family ownership transfer barriers (same family).
Falls back to single-queue path on GPUs with only 1 queue in the
graphics family. Transfer command pool is separate to avoid contention.
processReadyTiles was calling advanceFinalization with a step limit of 1
but a single step (texture upload or M2 model load) could take 1060ms.
Replace the step counter with an 8ms wall-clock time budget (16ms during
taxi) so finalization yields to the render loop before causing a visible
stall. Heavy tiles spread across multiple frames instead of blocking.
When parallel recording is active, the scene pass uses
VK_SUBPASS_CONTENTS_SECONDARY_COMMAND_BUFFERS. Post-processing paths
(FSR/FXAA) end the scene pass and begin a new INLINE render pass for
the swapchain output. ImGui rendering must use the correct mode —
secondary buffers for SECONDARY passes, direct calls for INLINE.
Previously the check used a static condition based on enabled features
(!fsr && !fsr2 && !fxaa && parallel), which could mismatch if a
feature was enabled but initialization failed. Replace with
endFrameInlineMode_ flag that tracks the actual current render pass
mode at runtime, eliminating the validation error
VUID-vkCmdDrawIndexed-commandBuffer-recording that caused intermittent
NVIDIA driver crashes.
Validation layers revealed 9965 VkSamplers allocated against a device
limit of 4000 — every VkTexture created its own sampler even when
configurations were identical. This exhausted NVIDIA's sampler pool
and caused intermittent SIGSEGV in vkCmdBeginRenderPass.
Add a thread-safe sampler cache in VkContext that deduplicates samplers
by FNV-1a hash of all 14 VkSamplerCreateInfo fields. All texture,
render target, renderer, water, and loading screen sampler creation
now goes through getOrCreateSampler(). Textures set ownsSampler_=false
so shared samplers aren't double-freed.
Also auto-disable anisotropy in the cache when the physical device
doesn't support the samplerAnisotropy feature, fixing the validation
error VUID-VkSamplerCreateInfo-anisotropyEnable-01070.
VkPipelineCache causes vkCmdBeginRenderPass to SIGSEGV inside
libnvidia-glcore.so on NVIDIA 590.x drivers. Skip pipeline cache
creation on NVIDIA GPUs — NVIDIA drivers already provide built-in
shader disk caching, so the Vulkan-level cache is redundant.
Pipeline cache still works on AMD and other vendors.
The AMD FidelityFX FSR3 runtime corrupts Vulkan driver state when
context creation fails on NVIDIA GPUs, causing vkCmdBeginRenderPass
to SIGSEGV inside libnvidia-glcore. Gate FSR3 frame gen initialization
behind isAmdGpu() check — FSR2 upscaling still works on all GPUs.
When ffxCreateContext for the upscaler fails (e.g. on NVIDIA with the
AMD FidelityFX runtime), the shutdown() path called dlclose() on the
runtime library which could hang — the library's global destructors may
block waiting for GPU operations that never completed.
Skip dlclose() on context creation failure: just clean up function
pointers and mark as failed. The library stays loaded (harmless) and
the game continues with FSR2 fallback instead of hanging.
Log GPU name and vendor ID during VkContext initialization for easier
debugging of GPU-specific issues (FSR3, driver compat, etc.). Add
isAmdGpu()/isNvidiaGpu() accessors.
Temporarily log SMSG_PLAY_SOUND and SMSG_PLAY_OBJECT_SOUND at WARN
level (sound ID, name, file path) to diagnose unidentified ambient
NPC sounds reported by the user.
Create a VkPipelineCache at device init, loaded from disk if available.
All 65 pipeline creation calls across 19 renderer files now use the
shared cache. On shutdown, the cache is serialized to disk so subsequent
launches skip redundant shader compilation.
Cache path: ~/.local/share/wowee/pipeline_cache.bin (Linux),
~/Library/Caches/wowee/ (macOS), %APPDATA%\wowee\ (Windows).
Stale/corrupt caches are handled gracefully (fallback to empty cache).
Inventory sort: clicking "Sort Bags" now generates CMSG_SWAP_ITEM packets
to move items server-side (one swap per frame to avoid race conditions).
Client-side sort runs immediately for visual preview; server swaps follow.
New Inventory::computeSortSwaps() computes minimal swap sequence using
selection-sort permutation on quality→itemId→stackCount comparator.
World map: fix continent bounds derivation that used intersection (max/min)
instead of union (min/max) of child zone bounds, causing continent views
to display zoomed-in/clipped.
Update README.md and docs/status.md with current features, release info,
and known gaps (v1.8.2-preview, 664 opcode handlers, NPC voices, bag
independence, CharSections auto-detect, quest GO server limitation).
The wall/floor classification threshold was lowered from 0.65 to 0.35
in a prior optimization commit, causing surfaces at 35-65° from
horizontal (steep walls, angled building geometry) to be classified as
floors and skipped during wall collision. This allowed the player to
clip through angled WMO walls.
Restore the threshold to 0.65 (cos 50°) in both the collision grid
builder and the runtime checkWallCollision skip, matching the
MAX_WALK_SLOPE limit used for slope-slide physics.
CharSections.dbc has different field layouts between stock WotLK (textures
at field 4-6) and Classic/TBC/Turtle/HD-textured WotLK (VariationIndex at
field 4). Add detectCharSectionsFields() that probes field-4 values at
runtime to determine the correct layout, so both stock and modded clients
work without JSON changes.
Also add BLOODELF_MALE/FEMALE and DRAENEI_MALE/FEMALE voice types to the
NPC voice system — previously all Blood Elf and Draenei NPCs fell through
to GENERIC (random dwarf/gnome/night elf/orc mix).
Lava M2 models used independent static-local start times in pass 1
and pass 2 for UV scroll animation. Since static locals initialize
on first call, the two timers started at slightly different times
(microseconds to frames apart), causing a permanent UV offset mismatch
between passes — visible as texture flicker/jumping on lava surfaces.
Replace both function-scoped statics with a single file-scoped
kLavaAnimStart constant, ensuring both passes compute identical UV
offsets from the same epoch.
PR #19 (572bb4ef) swapped CharSections.dbc field indices, placing
Texture1-3 at fields 4-6 and VariationIndex/ColorIndex at 8-9. Binary
analysis of the actual DBC files (Classic, TBC, Turtle — all identical
layout, no WotLK-specific override) confirms the correct order is:
Field 4 = VariationIndex
Field 5 = ColorIndex
Field 6 = Texture1 (string)
Field 7 = Texture2 (string)
Field 8 = Texture3 (string)
Field 9 = Flags
With the wrong indices, VariationIndex/ColorIndex reads returned string
offsets (garbage values that never matched), so all CharSections lookups
failed silently — producing white untextured character models at the
login screen and in-world.
Fixes all 4 expansion JSON layouts, hardcoded fallbacks in
character_preview.cpp, application.cpp, and character_create_screen.cpp.
Also handles the single-layer edge case (body skin only, no face/underwear)
by loading the texture directly instead of skipping compositing.
Apply at-rest values from M2 color alpha and transparency animation
tracks to batch rendering opacity. This fixes models that should render
as semi-transparent (ghosts, ethereal effects, fading doodads) but were
previously rendering at full opacity.
The fix multiplies colorAlphas[batch.colorIndex] and
textureWeights[batch.transparencyIndex] into batchOpacity during model
setup. Zero values are skipped to avoid the edge case where animated
tracks start at 0 (invisible) and animate up — baking that first
keyframe would make the entire batch permanently invisible.
Add setZoneName() to LoadingScreen and display the map name from Map.dbc
as large gold text with drop shadow above the progress bar. Shown in both
render() and renderOverlay() paths. Zone name is resolved from gameHandler's
getMapName(mapId) during world load. Improves feedback during zone transitions.
Cloud renderer's density was hardcoded at 0.35 and never updated from the
DBC-driven cloudDensity parameter. Now setDensity() is called each frame
with the lighting manager's cloud coverage value. Active weather (rain/
snow/storm) additionally boosts cloud density by up to 0.4 so clouds
visibly thicken during storms.
Reduce flare intensity when sun is near the horizon via smoothstep on
sunDir.z (0→0.25 range). Apply amber/orange color shift to flare elements
at sunrise/sunset for a warm golden glow. Prevents overly bright flares
at low sun angles while enhancing atmospheric mood.
Add Weather::Type::STORM enum value and wire it from SMSG_WEATHER type 3.
Storm particles are faster (70 units/s vs rain's 50), wind-angled at 15+
units lateral velocity with gusty turbulence, darker blue-grey tint, and
shorter lifetime. Previously storms rendered identically to rain.
Add QuestPoi struct and setQuestPois() to WorldMap, render quest objective
markers as cyan circles with golden outlines and quest title labels. Wire
gossipPois_ (from SMSG_QUEST_POI_QUERY_RESPONSE) through GameScreen to the
world map so quest objectives are visible alongside party dots and taxi nodes.
The spell visual failed-model cache was never cleared across world
changes, so models that failed to load during initial asset loading
(before MPQ/CASC data was fully indexed) would never retry. Now clears
spellVisualFailedModels_ in resetCombatVisualState() alongside the
active spell visual cleanup, giving failed models a fresh attempt on
each world entry.
Active spell visual M2 instances were never cleaned up when the player
teleported to a different map or re-entered the world. Orphaned effects
could linger visually from the previous combat session. Now properly
removes all active spell visual instances in resetCombatVisualState(),
which is called on every world entry.
Add indoor/outdoor state macro conditionals using the renderer's WMO
interior detection. Essential for mount macros that need to select
ground mounts indoors vs flying mounts outdoors. The Renderer now
caches the insideWmo state in playerIndoors_ and exposes it via
isPlayerIndoors(). Updated /macrohelp to list the new conditionals.
Spell visual M2 models that fail to load (missing file, empty model,
or GPU upload failure) were re-attempted on every subsequent spell cast,
causing repeated file I/O during combat. Now caches failed model IDs in
spellVisualFailedModels_ so they are skipped on subsequent attempts.
Spell visual effects previously used a fixed 3.5s duration for all
effects, causing some to linger too long and overlap during combat.
Now queries the M2 model's default animation duration via the new
getInstanceAnimDuration() method and clamps it to 0.5-5s. Effects
without animations fall back to a 2s default. This makes spell impacts
feel more responsive and reduces visual clutter.
Captures the Vulkan swapchain image to PNG via stb_image_write.
Screenshots saved to ~/.wowee/screenshots/ with timestamped filenames.
Cross-platform: BGRA→RGBA swizzle, localtime_r/localtime_s.
Models that lose all instances are no longer immediately evicted from
GPU memory. Instead they get a 60-second grace period, preventing the
thrash cycle where GO models (barrels, chests, herbs) were evicted
every 5 seconds and re-loaded when the same object type respawned.
M2Renderer::removeInstance() was calling rebuildSpatialIndex() for every
single removal, causing 25-90ms frame hitches during entity despawns.
Now uses O(1) lookup via instanceIndexById, incremental spatial grid
cell removal, and swap-remove from the instance vector. The auxiliary
index vectors are rebuilt cheaply since they're small.
The captureSceneHistory barrier was using srcAccessMask=0 with
VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT when transitioning the swapchain
image from PRESENT_SRC_KHR to TRANSFER_SRC_OPTIMAL. This does not
flush the GPU's color attachment write caches, causing VK_ERROR_DEVICE_LOST
on strict drivers (AMD, Mali) that require explicit cache invalidation
before transfer reads.
Fix: use VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT + COLOR_ATTACHMENT_OUTPUT
as the source mask so color writes are properly made visible to the
transfer unit before the image copy begins.
Also remove the now-unnecessary "requires FSR" restriction in the
settings UI — water refraction can be enabled independently of FSR.
- CameraController::resetAngles(): new method that only resets yaw/pitch
without teleporting the player. R key now calls resetAngles() instead
of reset() so pressing R no longer moves the character to spawn.
The full reset() (position + angles) is still used on world-entry and
respawn via application.cpp.
- packet_parsers_classic: parseSpellStart now calls
skipClassicSpellCastTargets() to consume all target payload bytes
(UNIT, ITEM, SOURCE_LOCATION, DEST_LOCATION, etc.) instead of only
handling UNIT/OBJECT. Prevents packet-read corruption for ground-
targeted AoE spells.
- packet_parsers_tbc: added skipTbcSpellCastTargets() static helper
(uint32 targetFlags, full payload coverage including TRADE_ITEM and
STRING targets). parseSpellStart now uses it.
When the player dies and releases spirit, the world map now renders a
bone-white X cross at the corpse's location (matching the existing
minimap skull marker). The marker appears only when the player is a
ghost with an unclaimed corpse on the same map, and shows a "Your
corpse" tooltip on hover. Implemented via setCorpsePos() on WorldMap,
called from renderWorldMap() using getCorpseCanonicalPos().
Add gold diamond markers for every flight master the player has already
discovered (knownTaxiMask_), read from TaxiNodes.dbc and filtered to the
current continent/map being displayed:
- WorldMapTaxiNode struct carries canonical WoW coords + known flag
- WorldMap::setTaxiNodes() accepts the per-frame list from game_screen
- renderImGuiOverlay() projects each known node to UV, draws a gold
diamond (AddQuadFilled) with a dark outline, and shows the node name
as a tooltip on hover
- GameHandler::isKnownTaxiNode(id) checks knownTaxiMask_[] efficiently
- Markers update live — newly discovered nodes appear without reopening
the map
loadSpellVisualDbc() now builds two distinct maps:
spellVisualCastPath_ — visualId → M2 via SpellVisual.CastKit chain
spellVisualImpactPath_ — visualId → M2 via SpellVisual.ImpactKit chain
playSpellVisual() accepts useImpactKit=false (default, cast) / true (impact).
SMSG_PLAY_SPELL_IMPACT passes useImpactKit=true so impact effects (explosions,
debuff indicators) use the ImpactKit model instead of the CastKit model.
Added ImpactKit field to all four dbc_layouts.json files.
Parse SMSG_PLAY_SPELL_VISUAL (casterGuid + visualId) and spawn a
transient M2 spell effect at the caster's world position.
DBC chain: SpellVisual.dbc → SpellVisualKit.dbc → SpellVisualEffectName.dbc
Lookup priority: CastKit.SpecialEffect0, fallback to MissileModel.
Models are lazy-loaded and cached by path; instances auto-expire after 3.5s.
DBC layouts added to all four expansion layout files (Classic/TBC/WotLK/Turtle).
Replace the static filled circle with a red triangle arrow that
rotates to match the character's current facing direction.
Uses the same render-space yaw convention as the 3D scene so
the arrow matches in-world orientation.
Prevents unbounded GPU memory growth in long play sessions where the player
visits many zones. Tiles are inserted into a FIFO deque; when the count of
successfully-loaded tiles exceeds MAX_TILE_CACHE (128), the oldest entry is
destroyed and removed from both the cache map and the deque.
At 256×256×4 bytes per tile this caps minimap GPU usage at ~32 MB.
Apply a cold blue-grey fullscreen overlay when the player is in ghost
form, creating a desaturated, muted appearance that clearly signals the
death state. Uses the existing overlay pipeline infrastructure. Applied
in both parallel and non-parallel rendering paths, after underwater tint
but before brightness adjustment so UI elements remain unaffected.
Remove stray X-flip in minimap display shader that mirrored the map
horizontally (West on right instead of East). Fix arrow rotation
fallback path (missing negation) and add character-facing-relative
arrow in rotateWithCamera mode.
Compact key ring: 24px slots in 8-column grid, only show rows with
items, hide when empty. Add Show Key Ring toggle in Settings with
persistence.
Move 5-second brute-force HMAC-SHA1 code pattern searches to a
background thread via std::async. The main loop now detects PAGE_A/B
checks, launches the response builder async, and drains the result
in update() — encrypting and sending on the main thread to keep
wardenCrypto_ RC4 state thread-safe.
Also adds Turtle WoW PE binary support (isTurtle flag, dedicated exe
search, runtime patches), searchCodePattern with result caching,
writeLE32 public API, and Warden scan entry verification.
