Although a lot is going on behind the scenes with some of these action button presses.ĭuring the beta phase we'll also see some official Unity terrain improvements which may help a lot in some instances, both in creation and run-time performance, so that's all happening at a good time.Ī longer term goal is also to explore the Entity Component System and other optimizations that could perhaps speed things up considerably via parallel processing (eg for multi-terrain generation). Earlier in development when more of the code was experimental it was even slower, but I'm sure there are more improvements to be had from code optimization. So improving processing speed is definitely a high priority and will be given more focus once the system is fully functional. Still got a way to go, but I think CityGen3D could be really useful in that regard, especially if it can be sped up for a more iterative workflow, like you suggest. I had an idea for a game, but realised even with all the amazing terrain and road tools available, it was still quite difficult to get a sandbox scene setup just to try some gameplay out in an urban environment that looked good in first person. This is actually how CityGen3D came to be. So something that could take days or weeks of manual work, just to try out some game logic in a suitable environment, can be done almost automatically. What I did in the video was fairly simple, but you can achieve far more unique looking and varied environments with some creativity in the setup. In the sense that you can build fairly detailed and realistic scenes to test your game idea on relatively very quickly.

Even in it's current state, I think it works well as a prototyping tool, just because of the level of automation it provides.

Pixelactive cityscape 1.8