Taking this blog in a slightly different direction, I'll write a bit about my thoughts on DotA AI and the retirement of BuffMePlz.
I recently received an email from BMP informing me of his decision, as well as his offer of general guidance to lead my own AI project. My first reaction was "Why didn't I get an email like this sooner?"
I believe that if an AI project had been worked together by a team of people, then it wouldn't have come down to this. It's really too much to take on by one person.
The idea of posting my Pre-Alpha builds of RGB's AI for the newest version at the time was to spark some interest in development in an otherwise dead scene. I knew BMP wasn't interested in wasting his time with people that thought they "could help" but couldn't do anything other than offer another opinion. This was a way of proving that there were certainly people interested in helping, myself and Cipd included
That was almost a year ago when my interest for developing AI was at it's peak, and when I received no response regarding those maps, I moved on to the next idea. This was to create an Open AI for DotA that could be ported to the latest version almost instantly, and Open in the sense that it would be very easy for others to contribute (especially open source).
Even though I completed the code-base for this project, it still needed some real AI to go along with it. This was back in January, and once again with no response or input from the ones I contacted, and the continuation of BMPs' ports of RGBs' maps, I decided that it was a wasted effort.
What I'm getting at here is that it's basically too little too late. I believe DotA for Warcraft 3 is near it's death. With Heroes of Newerth released just days ago, Starcraft 2 coming out this year, and the possibility of IceFrog moving on to Valve, by the time I created a decent AI, DotA would be on a different platform.
I don't see myself working on an AI for DotA now or in the future, and the only way I could see this changing is if some very ambitious developers came forward and teamed up as it should be done.
If you have a comment or question you can leave it in this post and I will reply there as well.
A little side note: I always thought it would be a fun game mode to combine 6v6 DotA with DotA AI. You could even play 12v12 in this scenario! (just for fun of course)
If you're confused reading all those words, let me tell you straight, he is declined to continue the Dota AI Map Project. He feels that it's already too late for that and by the time he finish the map, Dota is already dead beaten by HoN or Starcraft 2. However, he's also saying that if there's another person who can teamed up with him on developing the map, he will reconsider to continue the map.
Well, i can't make Dota map, so let's just support GoD-Tony and maybe he could change his mind. But regardless, i respect his decision and hope we can find another solution for this. Dota AI map is already become an important aspect of Dota. It helps people learn the game and encourage newbie to play Dota. And without it, Dota will never be the same..
No comments:
Post a Comment