Search Unity

  1. Megacity Metro Demo now available. Download now.
    Dismiss Notice
  2. Unity support for visionOS is now available. Learn more in our blog post.
    Dismiss Notice

[Level Design] Prefabs & Optimization?

Discussion in 'Getting Started' started by Shicksal, May 26, 2018.

  1. Shicksal

    Shicksal

    Joined:
    Jan 27, 2016
    Posts:
    1
    Hello everyone, new to the community and happy to be here. I'm fairly new to game development too, only ever getting as far as creating a single detailed level and functional character movement with animations. Early on I realized that programming is not my strong suit, but I have better potential in level design. I've recently started a project with an acquaintance who is fantastic with programming and I'm currently in the process of becoming more familiarized with good level design.

    As of right now I've been going through post processing early and just learned about prefabs, after tinkering with these and creating a few prefabs I worried about creating larger prefabs and the effect that might have on performance.

    Finally, my questions.
    1. What are some good level design practices to do going forward to keep things well optimized?
    2. Should my focus be more on good level streaming and proper culling than worrying about large prefabs?
    3. Is there a way to combine the models in the prefab like a collage of rocks into just the one rock?
    I was looking into a program call Mesh Combine Studio, and wondered if that would do the trick, however I feel I might be misunderstanding the usage of that tool.

    I hope I'm articulating my questions well, my main goal is to nip these issues in the bud and practice good level design from the ground up so I don't develop bad habits. If my third question didn't make sense, I can attempt to give an example. If you have any tutorials strictly on level design, paid for or free, I'd greatly appreciate to know about them! Thank you for your time!
     
  2. Deleted User

    Deleted User

    Guest

    1. Well it depends how big your scene is, what platform you're dev'ing for and the minimum specs of the game. If it's small lightbaking always helps with a very short cascade shadow map, the less the CSM the better. Then design your levels to occlude as much as possible, nothing worse than a long distance frustrum view of objects..

    In short the more you can see, the more it has to render and the worse performance will be.. If you're using a lot of foliage then make sure to look deep into LOD'ing. Use deferred rendering if you use a lot of lights and I could go on but:

    https://unity3d.com/learn/tutorials/topics/performance-optimization
    https://docs.unity3d.com/Manual/OptimizingGraphicsPerformance.html

    2. Don't optimise too early, read the above and decide from there.

    3. Yes, mesh combining obviously combines meshes but what you really need to do is limit draw calls. GPU's nowadays can handle a lot of tris / polys so that's less of a concern if you've properly baked your meshes.