Hi,
We have an old engine that was used to render our very large scene (30M polys) with dynamic loading.
Recently we decided to try Unity features, and our goal is to render our scene on Unity.
I’m new to Unity, and I would like to know how can I handle this problem?
Is there a prefer way to organize the scene to support this kind of scene? ( Any reference would be gr8 ).
How the occlusion calling handles dynamic loading?
Tnx
30 million polygons is quite a lot, what type of rendering engine did you use in the past?
You are not giving a lot of information for people to suggest how this could be achieved, since we practically only know you have something that consists of 30 million polygons.
The rendering engine is 12 years old DirectX engine. Our scene is a large city which has many buildings. We have offline process that
builds a visibility graph for the occlusion culling algorithm. So every frame there are ~3M polys in view. I have tried unity with that amount of polys and it preform much better.
How can I load our scene and manage it in unity? I want to use unity built in tools and not reinventing the wheel.
The goal is to be able to load a scene which has ~30M polys and have a decent frame rate. Any help would be gr8.
Anyone?
3 million seems more manageable, and should be fine on recent hardware. You need occlusion culling like Umbra (Unity Pro) to improve performance, though. A city scene is exactly what they use to showcase Umbra. Unity lets you try Pro for a month, so I’d say it’s worth a try.
Build the scene, then read up on how to use Umbra (there were a few articles in the blog section about performance tuning recently - check under Company links).
Thank you for your answer. I have tested unity occ which works well. I saw that in order to use it,
You need to turn on some flags to the objects and bake the scene. But our scene is huge and it won’t be
Possible to load 30m polys and manage them. So any suggestions here?