• 2d shooting game library
• game AI
• game observer mode
• server/client works like mmorpg
• server run game logic and decision
• client get environment info from server and send AI action to serve
• separated NPC(AI) server, game logic server, client reception server

• (x)ubuntu 13.10
• python 2.7.x
• wxpython 2.8.x

• twisted version game server and npc server
• run game server with s arg
• run npc server with int(number of ai npc) arg

• twisted version wxpython game client
• see wxgameclient.py option

• game server
• no need wxpython, and can be run by pypy
• fork 4 process
  • main : main process
  • game logic : GameLogicServer class : game logic decision
  • npc : NPCServer class : server side NPC AI
  • tcp : TCPServer class : client connection reception
• see help by -h option

• ui client / observer
• need wxpython
• client option -s serverip : server ip/name to connect -t teamname : AI team name to use (in sever) : if -t is ommitted client is work as observer mode (no ai team , just visualize game state)

• non ui AI client for server load test

• common library code

• old standalone wxgame2

http://kasw.blogspot.kr/2014/01/github.html

http://kasw.blogspot.kr/2014/03/wxgame2.html

http://kasw.blogspot.kr/2014/03/wxgame2_15.html

http://kasw.blogspot.kr/2014/03/wxgame2_22.html

wxgame2.py

• some code refine, bug fix

wxgame2server.py, wxgame2client.py

• full refactoring for client/server version

• C/S communication using file/pickle

wxgame2server.py, wxgame2client.py

• code refactoring
• full tcp networked client/server
• data packet changed to json
• AI client base work ( in progress )
• team score print

• performance tunning
• observer mode
• remove resource not used
• resource load code rework
• fix cpu usage issue

server loop changed to select

inter server communication use pipe

profile and time-run add for performance test

standalone npc server(for load test)

server process added

• server/client works like mmorpg
• server run game logic and decision
• client get environment info from server and send AI action to serve
• separated NPC(AI) server, game logic server, client reception server

twisted version added

• twservers.py
• twclient.py

zlib compressed json : Vector2 => (x, y)

client send
{
	cmd : makeTeam,
	teamname : teamname
}
Server send
{
	cmd : teamInfo
	teamname : teamname,
	teamid : teamid
}

client send
{
	cmd='reqState',
}
server send state
{
	'cmd': 'gameState',
	'frameinfo': {k: v for k, v in self.frameinfo.iteritems() if k not in ['stat']},
	'objplayers': [og.serialize() for og in self.dispgroup['objplayers']],
	'effectObjs': self.dispgroup['effectObjs'].serialize()
}

client send
{
	cmd='act',
	team=self.myteam,
	actions=actionjson,
}
server send
{
	cmd='actACK',
}

server send to server
{
	cmd: del
}