Taking Spring AI making to the next level?
Moderators: hoijui, Moderators
Taking Spring AI making to the next level?
man this is SO not fair, im currently stuck with my AI, thinking out a way to make it choose its buildings, while there seems to be 1000's of AIs coming out! I think maybe it would be time to start taking proactive steps towards this whole AI thing!
Spring is the only game i know of that is both great and has a proper AI interface (buggy but nevertheless extremely powerful). The TA setting is perfect for developing AIs too since simple behaviour is very easy to make but complex and smart behaviour is also possible!
I was thinking maybe we should grow this side of the community a bit more, make a few changes on the interface to make it more AI-friendly (eg allow communication between AI's) and start publicising this whole thing more!
Sure its a far-fetched dream, but wouldnt it be cool as hell if we got loads more people into making AI's? The forums would be more lively, we would get more experienced programmers to help spring's development...
Hell we could have weekly AI tournaments in both single and team settings! Or have the wiki more complete and have AI templates posted so beginners could have a go too!
I'm not asking for the second coming of jesus here! If each of us does their part in telling as many people as we can about spring and the AI scene, it will help a lot already!
dont you guys agree it would be just too cool to have a bigger AI community here?
Spring is the only game i know of that is both great and has a proper AI interface (buggy but nevertheless extremely powerful). The TA setting is perfect for developing AIs too since simple behaviour is very easy to make but complex and smart behaviour is also possible!
I was thinking maybe we should grow this side of the community a bit more, make a few changes on the interface to make it more AI-friendly (eg allow communication between AI's) and start publicising this whole thing more!
Sure its a far-fetched dream, but wouldnt it be cool as hell if we got loads more people into making AI's? The forums would be more lively, we would get more experienced programmers to help spring's development...
Hell we could have weekly AI tournaments in both single and team settings! Or have the wiki more complete and have AI templates posted so beginners could have a go too!
I'm not asking for the second coming of jesus here! If each of us does their part in telling as many people as we can about spring and the AI scene, it will help a lot already!
dont you guys agree it would be just too cool to have a bigger AI community here?
- Guessmyname
- Posts: 3301
- Joined: 28 Apr 2005, 21:07
It wouldnt confuse them since there could be a ladder of the best AI's.
At the moment people would be confused, since there are 4 AI's out and about and theres no evidence of which plays better besides heated discussions on fairness of some south-american maps...
as to your note:
Probably, but:
1. Why would you need/want to? c++ is more powerful than LUA AFAIK
2. This would be off-topic
At the moment people would be confused, since there are 4 AI's out and about and theres no evidence of which plays better besides heated discussions on fairness of some south-american maps...
as to your note:
Probably, but:
1. Why would you need/want to? c++ is more powerful than LUA AFAIK
2. This would be off-topic

I think we should first have a fair AI tournament or something. An someone who's not involved in any of the AI should select a date, announce it, and download all the AI's and let them play against eachother. We could even have a script generator for it, to generate startup scripts for every combination of AI's in a list 

One flaw. What about AAI? A virgin install of AAI would get pwned since it doesn't have as vetran knowledge as it would need.Zaphod wrote:I think we should first have a fair AI tournament or something. An someone who's not involved in any of the AI should select a date, announce it, and download all the AI's and let them play against eachother. We could even have a script generator for it, to generate startup scripts for every combination of AI's in a list
Re: Taking Spring AI making to the next level?
i think the community already grows fast, doesnt it? Taking MyAI developed by Berion and KAI developed by Krogothe into account there are at least two further ais that will add to spring's ai scene.krogothe wrote:!
dont you guys agree it would be just too cool to have a bigger AI community here?
in the near future there'll be about 5-7 different ais
1 TAI - alantai*
2 NTAI - alantai
3 ZCAIN - cain*
4 JCAI -zaphod*
5 AAI - submarine
6 OTAI - veylon
7 KAI - krogothe
8 MyAI - beryon
9 5MatrixAI - Triaxx
* AI is discontinued or on pause.
We have 9 AI projects and counting.
If you want the ball rolling, release source code, dont even wait for people to ask for it, just bundle the source with every release and update.
2 NTAI - alantai
3 ZCAIN - cain*
4 JCAI -zaphod*
5 AAI - submarine
6 OTAI - veylon
7 KAI - krogothe
8 MyAI - beryon
9 5MatrixAI - Triaxx
* AI is discontinued or on pause.
We have 9 AI projects and counting.
If you want the ball rolling, release source code, dont even wait for people to ask for it, just bundle the source with every release and update.
- Lindir The Green
- Posts: 815
- Joined: 04 May 2005, 15:09
as far as a tourney goes, why not have each AI face off against each other in a series of 6 games. This way, one AI is truely shown to be the better one, and any adaptive or learning AIs have a chance to learn about their new foes in the first few games.
I would really enjoy seeing that, replays especially. Any suggestions for a map? B. Battlefields apparently is uneven, so a different map is in order.
I would really enjoy seeing that, replays especially. Any suggestions for a map? B. Battlefields apparently is uneven, so a different map is in order.
That might eb a problem.
OTAI for example, has a chance of beign either 2 things.
A brilliant expansionist AI capable of causign NTAI major bother..
Or a pansy that builds mexes around the startpos tyrign to get metal and palces them everywhere but the mex spots.
AAI has flukes where it goes straight to lvl 2 then stalls halfway.
NTAI has maps it doesn't like e.g small divide.
Only NTAI 0.28.10 final (see AI forum in next 20 mins) is even capable of building a somewhat decent sea base.
OTAI for example, has a chance of beign either 2 things.
A brilliant expansionist AI capable of causign NTAI major bother..
Or a pansy that builds mexes around the startpos tyrign to get metal and palces them everywhere but the mex spots.
AAI has flukes where it goes straight to lvl 2 then stalls halfway.
NTAI has maps it doesn't like e.g small divide.
Only NTAI 0.28.10 final (see AI forum in next 20 mins) is even capable of building a somewhat decent sea base.