My friend Mark and I created a game in 4 weeks as a final research project entry into an AI competition. We Won 3 of the 4 awards including best game and best graphics etc… It’s mainly just an AI simulation of evolution. I’ve posted a final build on my website if you’d like to give it a try. It’s a Simple 4 turret tower defense using NEAT. We plan to make some real sellable games in the future. This one was just to show off the AI. Read the article on my site then download the demo we showed off in the competition at the bottom of the post! Enjoy! (PS. NEAT learns as you play so it gets increasingly harder, you can’t win).
I’m thinking of looking through your source materials and, in Unity JavaScript, creating a “cut-down” version of the NEAT system for a simplified game project.
this IS the page from the founders of NEAT, it’s pretty complicated, Real-Time NEAt is about 10k lines of code. This is currently the best resource for it
I’m working on converting the C# we used for neat to a real-time java script library and maybe even a plug in for Unity so people don’t have to attend seminars and college classes to understand it enough to use it in their projects.
We’re also working on 2 projects we’re considering making for a sellable title, one kinda like the initial stages of spore except with REAL evolution and not that scripted garbage, and the other is muh secret I’d love to hear any ideas or thoughts you have tho. I’m always down to collab or help.
Hi thanks, unfortunately that page is somewhat evasive on the actual topic (and seems full of dead and circular links), you might almost imagine it’s being purposefully abstruse. I was hoping to find a generalized explanation.
Thank you. I’m going to wade through that right now.
Since you’re working with the algorithm, you should give a “lite” breakdown of it on this title. From what I’ve read thus far, it seems a cross-over between a Neural Net AI solution, and a Genetic Algorithm AI solution, with some “mutations” thrown in that add the “complexification” and increased resolution of the problem-solving. But I’m sure that I don’t understand the full connotations of this algorithm.