Jump to content
  • Advertisement
Sign in to follow this  
Burnt_Fyr

Sports Game Logistics

This topic is 2779 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I'm looking for some advice from those who have been down a similar road before. I'm currently working on a Curling game( Eh?) which will be my first start to finish game, assuming I arrive at some point of completion along the way. The game will involve the players team playing games throughout the year, acquiring skillpoints, money, and sponsorships along the way. First agaisnts teams in their local club, moving up to territorial championships, then on to the nationals, then onto the worlds. This has already become a logistics nightmare on paper; How to store and simulate play on all levels with such a large group of teams competing.

A simple example of an N-ary tree 4 levels deep, with n being between 16-32, x 5 players per team yields a plethora of players worth of stats to track. I realize that the majority of posts will be in favor of scaling down, an If something were to go it would be the top level of play(Worlds), and the game would become strict Canadiana.That yields a much more manageable amount of data, but limits the marketability of the game. That in itself is fine as I have no delusions of grandeur, but would like other ideas if possible

Appreciate any comments, hints, links, questions, etc.

Share this post


Link to post
Share on other sites
Advertisement
It depends to what level of detail you're looking for in order to simulate the non-player teams.

For a somewhat simplistic method, you'd get the skills and abilities of the players of each team and then combine that into a single (or a couple) number as the skill level of the team, then you'd throw in a bit of random chance and then determine via whatever method you use (some geometric thing, or just a simple 'if team1.skill > team2.skill') who beat who. You could also just assign 2 AI players for each team and simulate the game which depending on the complexity might not take as long as you think, but it would still be longer than simply doing a couple of equations to figure out who's likeliest to win and just assign that team as the winner.

Now, as to the stats, you need to decide whether you're just going to store the final stats for each team, or the stats per team per game, or the stats per player per game. Of course the final stats are easier and less costly to store, and as you go down to per team and per player, it takes up more and more space. But the only complexity you should get, storage-wise is the sheer amount of stats that you're storing.

General off-the-top-of-my-head requirements
  • Final stats per team: just a bunch of stats for the team.
  • Stats per match per team: Stats for the team (for ease of access), and a list of match ids and the stats (match id just an arbitrary identifier for each match between two teams) for that particular match.
  • Stats per player per match: Same as the "per match per team", except you have a <match id, stats> list for each player and tally those to get the teams stats for that game, and then for all games as the final for the team.
    So, not unworkable, just confusing and daunting in the sheer amount of data you have to work with. So, unless you're planning on going full Electronic Arts-scale commerical with the game, you'll probably wanna dumb it down a bit.

Share this post


Link to post
Share on other sites

It depends to what level of detail you're looking for in order to simulate the non-player teams.

For a somewhat simplistic method, you'd get the skills and abilities of the players of each team and then combine that into a single (or a couple) number as the skill level of the team, then you'd throw in a bit of random chance and then determine via whatever method you use (some geometric thing, or just a simple 'if team1.skill > team2.skill') who beat who. You could also just assign 2 AI players for each team and simulate the game which depending on the complexity might not take as long as you think, but it would still be longer than simply doing a couple of equations to figure out who's likeliest to win and just assign that team as the winner.

Now, as to the stats, you need to decide whether you're just going to store the final stats for each team, or the stats per team per game, or the stats per player per game. Of course the final stats are easier and less costly to store, and as you go down to per team and per player, it takes up more and more space. But the only complexity you should get, storage-wise is the sheer amount of stats that you're storing.

General off-the-top-of-my-head requirements
  • Final stats per team: just a bunch of stats for the team.
  • Stats per match per team: Stats for the team (for ease of access), and a list of match ids and the stats (match id just an arbitrary identifier for each match between two teams) for that particular match.
  • Stats per player per match: Same as the "per match per team", except you have a <match id, stats> list for each player and tally those to get the teams stats for that game, and then for all games as the final for the team.
    So, not unworkable, just confusing and daunting in the sheer amount of data you have to work with. So, unless you're planning on going full Electronic Arts-scale commerical with the game, you'll probably wanna dumb it down a bit.


That was my main concern... The sheer amount of players to represent such a large scale. The thought of creating all that data, and the storage required was mind boggling. Even just in Canada, I'm around 64k players, worth of names, statistics, and skills to create. I've decided to drop down to just Canada for now, which drops down to about 1/60th or so of the data I would need. I'm still thinking it's a lot, but with some type of procedural name/stat/skill creation I think I'll be OK. It's funny you mention EA, as I'm using Tiger woods 2010 as sort of a template for the feel of the game, just swapping curling for golf.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

GameDev.net is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!