Age of Thieves - v 1.00 with A.I. !!!

Date 17-Sep-2013 17:07:02
Topic: software OS4

I'm proud to present my game Age of Thieves v 1.0. Age of Thieves is family strategy computer board game based on real a game. Version 1.0 has everything what you need to fun, up to 4 players included A.I. opponent.

Click READ MORE ===>

download link: Age of Thieves

Some questions and answers about my production:
Q: From last published version 0.8 the most new option is A.I...
A: A.I. is currently set to be equal opponent for intermediate player. In my first plan was write best A.I. on the world but I must little decreased my ambition during coding. Most users not interested how work A.I., but little info I must add to explain some things. Typical A.I. can be based on statistical search methods of position in game, my A.I. is based on decision tree. During coding I found some disadvantage of selected method. Most important is hard and slow adding any improvements. Only with big effort I finish my engine on current version to able identify and deal with the most typical problems in this game. Any next improvements will be possible and must be done but not so fast.

But even if some advanced players can easy win with my engine it is not bad itself. It know some advanced tricks, It can deal with limited meeple (simple A.I. is able set all his meeple and do nothing to end of game). In most situations A.I. is not bad in positional game. For me is stronger then A.I. from similar games existed on java or Android systems, but I found that it depend on human strategy. A.I. has some weaknesses - can not be changed depending on the opponent of game style. Also some branches of my decision tree are thick at this moment and causing not best A.I. moves. To balance it using pure own calculations many code need be added...

At last I can promising that A.I. is coded without any cheating. His only advantage over human is knowledge how tiles can be hidden on stack - it legal knowledge - human also can note for self how pieces are used or not but typically human think differently then computers.

Q: What about other features?
A: I want not delay release time so I concentrate on stable game engine. I resolve many users requests but some things like fullscreen or picking piece on start move was little above my coding knowledge or I can't easy deal with some coding language limitations (game is coded in PortablE). I want add more options in future because 1.0 is not equal final version.

Q: Is your game exist on other operating systems?
A: No! Is Amiga OS4 native game coded from start no using any part of code similar games. It has unique user interface on my own idea.

Q: Is game can be ported to Amiga family operating systems like Aros Morphos?
A: Not now. Source code should be highly compatibile but I not have proper machine and other operating systems experience to do port. Maybe in future...

Q: Is this full game?
A: No, game is donationware but there is limit of donations what I want. So I hope that game soon will be freeware! In undonated game you can play single game to end with prerandomized game tiles stack. Every donation is only for my satisfaction, not for earn money, because coding game for 5 months for my donation limit is not any salary.

Q: Some final words?
A: 1.0. version is good number to hear some players comments. Every comment can help develop this game. How do you like this game? How strong is A.I. for you? Is game fast enought (game is very fast on my sam460 with radeon 4650)? Develop this game more or try something new (I'm little tired to get game to current stage)? At last usually I want say thanks to Chris Handley who help me more solving some coding problems and usually improved his PortablE coding language to be able deal with my code.

This article comes from AmigaWorld - Amiga Community Portal

The URL for this story is: