Unity additive scenes collision and lighting help

Chubbs

Hi,

I'm looking for some advice/thoughts/ experiences on how best to structure collision handling in unity between the Level Designers and Art team. 

The project I'm working on has split remote teams with the graphics side of things handled by one team and the Code/ level design handled another. We've split the levels down to Graphics and Logic. When playing/testing the game we additively loading these scenes. This allows us to work in tandem and removes the risk of scene merge errors. Having tested UnityYAMLMerge we weren't happy with the result on complex scenes and so opted for this workflow.

At the moment the LD's are blocking out levels in the Logic scene with basic geo and collisions. The Art team then does a pass in the Graphics scene adding the static meshes and collision across the level. The LD's then remove there block out meshes and the majority of the collisions.

This has generally work well apart from the occasional collision issue from some static meshes with colliders that are unsuitable. With this workflow we are also splitting collisions across the graphics scene and the logic scene with approximately 85% of them being the graphics scene and the rest in Logic.  

What do people think? is there a better way to handle this? how have teams of around 12- 20 people handled this issue in Unity?

Regarding lighting with additive scenes we assigning the Graphics Scene as the Lighting Master so it handles baking of all the light map data.

Thanks for your time.

RyanB

If you bake your lighting in 3DS Max/Maya/Blender and make a lightmap for each object then you won't have to worry about baking in Unity.  Using V-Ray will massively speed up your render times at the cost of some pipeline setup in whatever program you choose.

How are the Art sides static meshes with collides "unsuitable?"

Chubbs

The baking needs to be done inside Unity for a myriad of reasons, its relatively fast a the moment so its not a huge issue. Regarding the colliders occasionally the angles are a little too sharp and cause the player to get stuck. 

Near Fantastica

Chubbs said:

The baking needs to be done inside Unity for a myriad of reasons, its relatively fast a the moment so its not a huge issue. Regarding the colliders occasionally the angles are a little too sharp and cause the player to get stuck. 

When that happens I just put box colliders on empty Game objects and use them to block off acute angles created by other colliders. Or you can create a collision model for the whole scene.

Chubbs said:

Hi,

I'm looking for some advice/thoughts/ experiences on how best to structure collision handling in unity between the Level Designers and Art team. 

The project I'm working on has split remote teams with the graphics side of things handled by one team and the Code/ level design handled another. We've split the levels down to Graphics and Logic. When playing/testing the game we additively loading these scenes. This allows us to work in tandem and removes the risk of scene merge errors. Having tested UnityYAMLMerge we weren't happy with the result on complex scenes and so opted for this workflow.

At the moment the LD's are blocking out levels in the Logic scene with basic geo and collisions. The Art team then does a pass in the Graphics scene adding the static meshes and collision across the level. The LD's then remove there block out meshes and the majority of the collisions.

This has generally work well apart from the occasional collision issue from some static meshes with colliders that are unsuitable. With this workflow we are also splitting collisions across the graphics scene and the logic scene with approximately 85% of them being the graphics scene and the rest in Logic.  

What do people think? is there a better way to handle this? how have teams of around 12- 20 people handled this issue in Unity?

Regarding lighting with additive scenes we assigning the Graphics Scene as the Lighting Master so it handles baking of all the light map data.

Thanks for your time.

My team works in a similar manner. Art creates the scene, colliders, lighting, occlusion, pathing, etc. DEVs create their own scene with all of the logical and loads in the art scene and lighting data at runtime.