Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

It's true. Speaking as a gamedev, more people should really study how Notch was able to make procedural gen so interesting to look at. It's rather involved, and took a lot of testing to get right. And even that starts to get boring after awhile.

Every environment you see in this video will probably look roughly the same: http://www.youtube.com/watch?v=RNVgVl6v6YU

The reason is because most programmers simply do 'procedural' gen like "if desert, then X. If temperate, than Y. If..."

But you can achieve much more interesting results by using weighting instead of special casing.



Procedural generation along similar lines has been used successfully for most of the vegetation in Pixar's "Brave", see http://www.cgw.com/Publications/CGW/2012/Volume-35-Issue-4-J... for some background.


> more people should really study how Notch was able to make procedural gen so interesting to look at.

Could you elaborate or link an article on this?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: