# ThinInstances lighting behaviour seems off **URL:** \ **Category:** Questions\ **Created:** [October 24, 2025, 1:11pm UTC](https://forum.babylonjs.com/t/thininstances-lighting-behaviour-seems-off/61121 "2025-10-24T13:11:07Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![Tarks](https://sea2.discourse-cdn.com/flex024/user_avatar/forum.babylonjs.com/tarks/32/48474_2.png) [@Tarks](https://forum.babylonjs.com/u/Tarks)\ **Post date:** [October 24, 2025, 1:11pm UTC](https://forum.babylonjs.com/t/thininstances-lighting-behaviour-seems-off/61121/1 "2025-10-24T13:11:07Z")
Not sure if it’s the lighting, or the materials or what, but everything’s darker **Without ThinInstances** ![image](https://us1.discourse-cdn.com/flex024/uploads/babylonjs/original/3X/b/9/b95bb8a869c2c23b801d9913d8a214d2353061d4.jpeg) **With ThinInstances** ![image](https://us1.discourse-cdn.com/flex024/uploads/babylonjs/original/3X/2/9/29e9dca38904c77d2bc1fad84ce5766e49819493.png) I’ve tried exporting a serialized version of the scene to help **Playgrounds** **Without ThinInstances** > **[Babylon.js Playground](https://playground.babylonjs.com/#P3PB9N%232)** > > Babylon.js playground is a live editor for Babylon.js WebGL and WebGPU 3D scenes **With ThinInstances** > **[Babylon.js Playground](https://playground.babylonjs.com/#S2FTTA)** > > Babylon.js playground is a live editor for Babylon.js WebGL and WebGPU 3D scenes ---
**Author:** ![Deltakosh](https://sea2.discourse-cdn.com/flex024/user_avatar/forum.babylonjs.com/deltakosh/32/26635_2.png) [@Deltakosh](https://forum.babylonjs.com/u/Deltakosh)\ **Post date:** [October 24, 2025, 3:25pm UTC](https://forum.babylonjs.com/t/thininstances-lighting-behaviour-seems-off/61121/2 "2025-10-24T15:25:26Z")
cc @Evgeni_Popov ---
**Author:** ![kzhsw](https://avatars.discourse-cdn.com/v4/letter/k/848f3c/32.png) [@kzhsw](https://forum.babylonjs.com/u/kzhsw)\ **Post date:** [October 25, 2025, 12:50am UTC](https://forum.babylonjs.com/t/thininstances-lighting-behaviour-seems-off/61121/3 "2025-10-25T00:50:30Z")
Is this better now? ``` const mesh = nS.getMeshById('Clone of Hill_merged') as BABYLON.Mesh; mesh.sideOrientation = 1; ``` > **[Babylon.js Playground](https://playground.babylonjs.com/#LEDJEG)** > > Babylon.js playground is a live editor for Babylon.js WebGL and WebGPU 3D scenes ---
**Author:** ![Tarks](https://sea2.discourse-cdn.com/flex024/user_avatar/forum.babylonjs.com/tarks/32/48474_2.png) [@Tarks](https://forum.babylonjs.com/u/Tarks)\ **Post date:** [October 25, 2025, 11:21am UTC](https://forum.babylonjs.com/t/thininstances-lighting-behaviour-seems-off/61121/4 "2025-10-25T11:21:26Z")
Yes, thankyou ! What have I not understood? This is the part of my code that loads meshes, I’m calling `mergeMeshes` on all objects set to be thinInstances, but not sure how causing winding issues. ```auto const meshContainer = meshAssetContainer.instantiateModelsToScene(); let rootMesh = meshContainer.rootNodes[0] as Mesh; rootMesh.rotation = babylon.Vector3.Zero(); // Babylon applies a scaling of x:1,y:1, **z:-1** to all meshes loaded. // I assume that's a hack to get meshes to align with its rendering orientation rootMesh.bakeCurrentTransformIntoVertices(); rootMesh.metadata = {}; if (shouldCreateThinInstance) { if (!this.isInsideNullEngine()) { const allmeshes = rootMesh.getChildMeshes() as Mesh[]; const mergedMesh = Mesh.MergeMeshes(allmeshes, true, true, undefined, false, true); rootMesh = mergedMesh!; rootMesh.alwaysSelectAsActiveMesh = true; rootMesh.thinInstanceEnablePicking = true; const instanceId = rootMesh.thinInstanceAddSelf(); rootMesh.thinInstanceSetMatrixAt( instanceId, babylon.Matrix.Translation(transformComponent.position.x, transformComponent.position.y, transformComponent.position.z) ); // Todo : this would be replicated if we did this on the server... meshComponent.thinInstanceId = instanceId; this.assetLoader.addAssetCacheItem(thinInstanceCacheId, rootMesh); return { createType: "thinInstance", mesh: rootMesh, thinInstanceId: instanceId, thinInstanceCacheIdForVisual: thinInstanceCacheId }; } else { return { createType: "nullEngineResult" }; // On the server we get // MeshHandler.ts:108 BJS - [16:22:45]: Thin Instances are not supported on this device as Instanced Array extension not supported // Becuase it's using the null engine } } return { createType: "mesh", mesh: rootMesh }; } ``` If I apply the sideOrientation = 1 to everything I get the ‘problem’ meshes working and the meshes that were fine misbehaving. ![image](https://us1.discourse-cdn.com/flex024/uploads/babylonjs/original/3X/d/9/d9203b7935d6220f4b685c64837c304c69d0d32a.jpeg) Checking the normals in blender implies they’re the ‘right’ way round too ![image](https://us1.discourse-cdn.com/flex024/uploads/babylonjs/original/3X/6/3/63f1d32138506fb2e9fdaf148f632bf7a141e284.jpeg) ---
**Author:** ![Evgeni\_Popov](https://sea2.discourse-cdn.com/flex024/user_avatar/forum.babylonjs.com/evgeni_popov/32/4432_2.png) [@Evgeni\_Popov](https://forum.babylonjs.com/u/Evgeni_Popov)\ **Post date:** [October 25, 2025, 11:54am UTC](https://forum.babylonjs.com/t/thininstances-lighting-behaviour-seems-off/61121/5 "2025-10-25T11:54:28Z")
I think this post should help you (but the entire thread is very informative): > [@Automatically Convert Instances to Thin Instances?](https://forum.babylonjs.com/t/automatically-convert-instances-to-thin-instances/44236/13): > > Here’s a working PG: The main problem was that the glTF loader sets the mesh.overrideMaterialSideOrientation property, which takes precedence over the material.sideOrientation property, meaning the new material side orientation was not taken into account. ---
**Author:** ![Sucher](https://avatars.discourse-cdn.com/v4/letter/s/50afbb/32.png) [@Sucher](https://forum.babylonjs.com/u/Sucher)\ **Post date:** [November 7, 2025, 6:57am UTC](https://forum.babylonjs.com/t/thininstances-lighting-behaviour-seems-off/61121/6 "2025-11-07T06:57:19Z")
Our objective is to develop a dataset and approach addressing the autonomous or assisted harvesting[,](https://seminolecountyclerkofcourt.click/) loading and measuring of logs, ie cut stems and branches.