@babylonjs/loaders 9.16.2 → 9.17.1

package/SPLAT/gaussianSplattingStream.d.ts

@@ -472,7 +472,14 @@ export declare class GaussianSplattingStream extends GaussianSplattingMesh {
      * cheap per-node frustum test every frame so the off-screen LOD bias tracks camera rotation. The LOD
      * re-evaluation is throttled to at most every {@link _lodUpdateInterval} frames once the camera has
      * translated far enough, but also runs immediately whenever a node enters/leaves the frustum (so its
-     * detail upgrades/downgrades promptly) or a cap change forces it. Active ranges rebuild on any LOD change.
+     * detail upgrades/downgrades promptly), a node whose cooldown just expired still needs to switch LOD,
+     * or a cap change forces it. Active ranges rebuild on any LOD change.
+     *
+     * The cooldown-expiry trigger lets a node reach its already-computed target level as soon as its
+     * cooldown clears, rather than waiting for the camera to move. This matters right from load: a
+     * node's base-layer decode is itself applied as a switch (from no active level to the base one), so
+     * it starts the same cooldown a later switch would — this trigger is what lets the node progress past
+     * that base level promptly once it expires, even at a fixed camera pose.
      */
     private _onLodFrame;
     /**

package/SPLAT/gaussianSplattingStream.js

@@ -1340,15 +1340,26 @@ export class GaussianSplattingStream extends GaussianSplattingMesh {
      * cheap per-node frustum test every frame so the off-screen LOD bias tracks camera rotation. The LOD
      * re-evaluation is throttled to at most every {@link _lodUpdateInterval} frames once the camera has
      * translated far enough, but also runs immediately whenever a node enters/leaves the frustum (so its
-     * detail upgrades/downgrades promptly) or a cap change forces it. Active ranges rebuild on any LOD change.
+     * detail upgrades/downgrades promptly), a node whose cooldown just expired still needs to switch LOD,
+     * or a cap change forces it. Active ranges rebuild on any LOD change.
+     *
+     * The cooldown-expiry trigger lets a node reach its already-computed target level as soon as its
+     * cooldown clears, rather than waiting for the camera to move. This matters right from load: a
+     * node's base-layer decode is itself applied as a switch (from no active level to the base one), so
+     * it starts the same cooldown a later switch would — this trigger is what lets the node progress past
+     * that base level promptly once it expires, even at a fixed camera pose.
      */
     _onLodFrame() {
         if (this._disposed || !this._baseLayerReady) {
             return;
         }
+        let cooldownExpiredWithPendingSwitch = false;
         for (const node of this._leafNodes) {
             if (node.lodCooldown && node.lodCooldown > 0) {
                 node.lodCooldown--;
+                if (node.lodCooldown === 0 && node.targetLevel !== undefined && node.targetLevel !== node.activeLod) {
+                    cooldownExpiredWithPendingSwitch = true;
+                }
             }
         }
         // Tick eviction cooldowns: unreferenced files are freed once their cooldown elapses (budgeted streaming).
@@ -1360,7 +1371,7 @@ export class GaussianSplattingStream extends GaussianSplattingMesh {
         // Per-node frustum test runs every frame (cheap) so the off-screen LOD bias tracks camera rotation,
         // not just the translation that gates the throttled LOD re-evaluation below.
         const frustumChanged = this._updateNodeFrustum();
-        let runLodEval = this._forceLodUpdate || frustumChanged;
+        let runLodEval = this._forceLodUpdate || frustumChanged || cooldownExpiredWithPendingSwitch;
         if (!runLodEval && ++this._framesSinceLodUpdate >= this._lodUpdateInterval) {
             const camera = this._scene.activeCamera;
             const threshold = this._lodUpdateDistance;