ua4192

Members
  • Content count

    32
  • Joined

  • Last visited

  1. ahh. ok. no problema. Many thanks for your quick answer.
  2. no answers? sorry if the question seemed silly.
  3. Hi. Would it be possible to add a special case for AutoLOD method when quality will be set to zero? As addLODLevel is not working properly for imported meshes (complex meshes extracted from GLTF files) and instances. For this reason, I am thinking to return to autoLOD method. This was working properly. In fact my goal is by now just hide very small meshes the problem is that if you specify zero quality in autoLOD method the decimation process takes a lot. Maybe would be a good idea for this case return a null mesh, this would be immediate. COuld anyone explain me how to modify the code to add this special case? Maybe the async decimation process should have an extra distance to be specified to return null mesh. This would be very useful to hide small meshes or very far away. Of course the first priority should be to return add this LOD null mesh and after the others. In fact in addLODLevel method you have this option, but when I try to use it it doesn't work properly. Does it make sense for you? Best regards
  4. It seems that the method addLODLevel when used with meshes extracted from method ImportMesh doesn't work properly or maybe the method works in async way like ImportMesh. This is what I am doing and each time I load the scene, the total number of meshes is different. First I import all GLTF into my scene but make them not active. I wait until all meshes have been loaded using a whendone() function on ImportMeshCall. This ImportMeshCall has a countdown and when it reach to zero it means that I have all meshes loaded and saved in an array of meshes. When I have all meshes loaded and saved in the array, I invoke a function (make_lod_assets) to addLODLevel, by now only one (null when distance is greater than x) After having finished to add all LODs, I invoke to another function (loadscene) which starts to build the scene. As I want to keep the same structure than CATIA, I build invisible unitary boxes for all assemblies containing the CATParts (my meshes) and I apply the transformation matrix (rotation+position) in a relative and inherited way. I am using instances because I want to use HW acceleration. When this function (loadscene) ends it returns the scene and it is rendered. My meshes (extracted from GLTF) don't have textures My problem are: - the addLODLevel seems to work in a async way -Sometimes the instances are not properly positioned My code is quite big and not possible to be integrated in PG, but I can attach it.
  5. Dear all. Is it possible to use clones and LOD with imported meshes? With instances I have lots of problems: 1.- When I set a distance and null mesh to be loaded, then the instance is not properly positioned and rotated. 2.- Some instances are not loaded, with the same code the number of intances vary and should not. 3.- The only mesh that seems to work is the null LOD the other meshes with different qualities not. I can attach the code I am using. Is there any way to use LOD with different quality meshes with clones?
  6. Do you mean to use LOD? I am already using 4 levels of detail. My problem with this approach is that LOD cannot be used with closes and with instances now I have the same problem than before about inherited transformations. Anothe problem is that I am using an arcrotatecamera and the LOD distance seems not to be working properly. Sometimes I have to be very near from the object to make it visible. Best regards
  7. It seems very interesting, despites it seems not to be compatible with all browsers. I have tried to load an example: https://www.babylonjs.com/demos/simd/ But it doesn't work. It gets hang while loading, no progress. Do you have any other example? I have not found too much documentation. Many thanks in advanced. You are the master
  8. Afrter many tries The scene is still too big to be managed. Could I do anything else? Best regards
  9. Many thanks. : ) I will try and give you feedback
  10. Hi all. I am trying to apply different optimizations on my big complex scene with 4000 meshes and 40000 instances. The scene is very complex (12.5M vertices ) My idea is to combine diferent optimization methods probided by babylon. AUTOLOD+MESHSUBDIVIDE+OCTREES+SCENEOPTIMIZATION. My main problem is that when I try to apply the scene optimization method SceneOptimizerOptions.HighDegradationAllowed I always get the message that the optimization has been successfully but many times the scene is cleaned and the number of scene total vertices decrease until 0. I think there is something wrong. Here the code: BABYLON.SceneOptimizer.OptimizeAsync(scene, BABYLON.SceneOptimizerOptions.HighDegradationAllowed(15, 5000), function(){ if ( mydebug ) { console.log("Optimization successful"); } }, function(){ console.log("Optimization failed"); }); I have also realize that the function provides the mesage only one time. I thought that the function was executed each 5 seconds (5000ms) if the target of FPS of 15 was not possible. I am thing in serializing big scenes, but I have never tried and don't know if this will help. Many many thanks in advanced.
  11. Hi. Many thanks for your support. Finally at the end I found a workaround. I create a symbolic box (a-dimensional) and make parent of meshes included in GLTF. IN this way I have not to apply any transformation on the GLTF meshes, I do it on the box instead. Many many thanks : )
  12. Hi all. I have a problem. In my scene I am using instances. The problem I have is that the first time I use the mesh (imported GLTF) I am positioning and rotating the GLTF according to the product structure inherited transformation matrix. My problem is that the instances seem to take this transformation and are not properly transformed. Is there any way to reset the already applied transformations(relative rotation + position to its parent)from the imported meshes (GLTF) and create the instances with no transformations? Many thanks in advanced.
  13. Hi again. It seems that the problem is in the final box render. here you can see https://www.babylonjs-playground.com/#10FF4M#6 The progressive bounding boxes ofthe scene are fine, but the one I render out of the loop is not properly rendered. And the coordinates are the same!!! Many thanks again in advanced for your support.
  14. here the PG https://www.babylonjs-playground.com/#10FF4M#5
  15. Hi. The attachment can be copied and pasted in PG. Due to proxy constraints in my company I cannot doi it directly. I will try to do it from my mobile.