I released a development update for my game which includes footage of two projects I’ve been working; both using Unity. Whilst both are driving/simulation related, they are very different in their backend. The driving simulator interfaces to a real car via CANBUS and uses 6 PC’s all running the client to provide 360 degree views, whilst outputting all the cues for motion simulation. It is controlled remotely via an Instructor Operator Station (IOS) which connects to the server automatically via Lidgren/UDP.
KartSim is a game I’ve been working on for quite a while, having originally come up with the idea for it back in 2005. Apart from the web services and sculpting the driver, I’ve developed everything else. It uses a custom physics engine, client/server architecture using Lidgren for low level networking and NGUI for all the GUI elements.
There are more screenshots on the Facebook Page. I’m more than happy to answer any questions about the development of either project and how I went about implementing any features you’re interested in.
What were the reasons you decided to go with your own vehicle physics rather than Unity’s physics? Did you go with entirely different SDK (Bullet, ODE) or build your own solution from scratch?
Did you evaluate other networking solutions (i.e. Photon, Tasharen, Unity built-in networking) before deciding to go with Lidgren? What were some of the issues you encountered with high velocity networked physics and if you’re able to share, what solutions did you come to?
Do you plan to upgrade the project to use Unity 5?
Fan Questions:
Does the game feature underpasses, bridges, ramps, and other stunt track elements?
Will it be possible to import one’s own decals and locals for karts, signs, uniforms?
Also, for the vehicle simulator, was that also architected with Unity? In either event, how were you able to simulate a very large (100km x 100km) environment without memory and floating point precision limitations?
Looking very forward to KartSim’s eventual release.