You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Top 10 Correlated Spike Culprits (Strictly Gated):
Expanded the spike culprit list to display up to the Top 10 highest-impact mods for any recorded hitch.
Implemented the strict \Test-IsSpikeWorthy\ gatekeeper: candidate mods must exceed tangible execution thresholds (\FrameTaxRaw > 0.02, permanent loops, in-hook queries, throttled hooks, or massive world meshes/VRAM load).
Zero Padding: Completely eliminated artificial padding loops. Harmless cosmetic packs (< 1 ms [Imperceptible]\ / passive static UI) are never falsely blamed for frame spikes. If only 3 worthy mods exist, only 3 are displayed.
World / Chunk Geometry vs. Character Skinned Mesh Differentiation:
Accurately distinguishes static world tile/chunk geometry and vehicles (\�oxel, props, world tiles) from character attachments (\clothing, \hair, \skinned).
Character hair and clothing mods will never be blamed for chunk cache rebuild stalls (\chunk cache XX ms: YY builds) or given unfair risk penalties.
Stutter Roster Sorting Overhaul:
The master active mod roster now sorts primarily by Stutter Impact Severity (\CRITICAL\ -> \HIGH\ -> \MODERATE\ -> \LOW\ -> \NEGLIGIBLE) and predicted latency burst rather than raw asset count.
Passive cosmetic mods (< 1 ms) always settle cleanly at the bottom.
Color coded per severity rank: Red for CRITICAL, Yellow for HIGH, DarkYellow for MODERATE, Cyan for LOW, and DarkGray for NEGLIGIBLE.
Precision Frame Tax Weighting:
Queries executed inside throttled / periodic hooks (\PermanentHooks == 0) are scaled down from full continuous per-frame weight to reflect real-world interval execution.
Bundled & Verified:
Synchronized standalone polyglot wrapper \Scan-PZModPerformance.bat\ and documentation.