Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Rules/Strategy (http://www.chiefdelphi.com/forums/forumdisplay.php?f=6)
-   -   Secret Sauce: Shot to Human Player (http://www.chiefdelphi.com/forums/showthread.php?t=127454)

IronicDeadBird 20-03-2014 19:23

Re: Secret Sauce: Shot to Human Player
 
Quote:

Originally Posted by marshall (Post 1361829)
How long does a 3 assist cycle take vs a 2 assist? And don't forget about a third robot dedicated to defense. It's a fun thought exercise to armchair on CD...

We ran the math on all of this right after kickoff and came up with a strategy that has paid off for us. I'm not going to claim it is 100% effective in all cases and I know it can be disrupted.

Fact is, it won us a regional and an engineering award and we have had a blast doing it. I'll take that any day.

Its hard to disagree with someone who won a regional with it but I still am going to (congrats to that by the way!) Winning is proof in itself that a strategy worked but the keyword here is worked. The thing that is blowing my mind is that in this game a quick 2 assist cycle strat could outweigh a quick 3 assist. I feel it would be poor game design to end up with a game that rewarded less work with more points. No offense to the GDC at all they are great guys and very innovative. This work:reward pay off could only be tested through live play and obviously the GDC doesn't have too much to work with when it comes to simulating a regional.

JohnFogarty 20-03-2014 19:47

Re: Secret Sauce: Shot to Human Player
 
Quote:

Originally Posted by IronicDeadBird (Post 1361921)
Its hard to disagree with someone who won a regional with it but I still am going to (congrats to that by the way!) Winning is proof in itself that a strategy worked but the keyword here is worked. The thing that is blowing my mind is that in this game a quick 2 assist cycle strat could outweigh a quick 3 assist. I feel it would be poor game design to end up with a game that rewarded less work with more points. No offense to the GDC at all they are great guys and very innovative. This work:reward pay off could only be tested through live play and obviously the GDC doesn't have too much to work with when it comes to simulating a regional.

Looking at the scores of the Greater Kansas City Regional and Orlando Regional Week 3, same as North Carolina, show that a 3 assist cycle would in-fact out score a 2 assist cycle.

marshall 20-03-2014 20:54

Re: Secret Sauce: Shot to Human Player
 
Quote:

Originally Posted by HannahWilson (Post 1361897)
Congrats to team 900 on winning the Raleigh regional! I really enjoyed watching your robot launch the ball so powerfully, and your team's strategy was amazing.

I have one question: when did your team have the idea of passing the ball over the truss to the human player? Did you plan to build your robot to shoot the ball that powerfully during the build season, or did you realize that strategy would work well during the competition? (I'm assuming the former.)

Hi Hannah,

We planned from the beginning to build our robot the way we did. After kickoff we did a couple of things. The first is that we read over the rules thoroughly to ensure we fully understood them. In addition, we also looked at statistical analysis of the game. From there, we began to think through possible strategies and the scores for them based on that analysis. (This lead to a 4D graph and some other stuff that I won't bore you with.)

I am proud to say that our strategy is based on that analysis and that it was arrived at on the day after kickoff. From there, it was a matter of designing a robot to accomplish the goals we set out to accomplish for our strategy.

Thank you for the compliments.


All times are GMT -5. The time now is 16:12.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi