let's do this contest again some time...

...but add a few things!

With just one "person" on each team and the very limited arsenal of things to use I believe this game is based too much on the teleporter (and thus a random event). I prefer having the better strategist win when playing a "strategic game" like capture the flag...at least the majority of the time.

A little bit of randomness can make things interesting but if it's a determining factor in the "game" then it kinda' spoils the fun...

Next time around let's have:
- 3-5 "person" teams
- give each "person" a death ray weapon...or maybe a couple/"person"
- let a team member "self-destruct" to kill an opposing member in the same square

now that would be fun!!

*starts planning the next office paint-ball "field-trip"*

would each "player" be controlled by a different person's script? - or would one script control the moves of every player in a team?

i'd say have one script control an entire team...

it would be interesting if you had a standard form for the data.txt file (one file between a whole team)... how different scripts would work with each other. I.e you could have defensive & attacking scripts... etc... that all talk to each other and share one data-source.

this already exists, it's called Unreal Tournament ;)

The strategy in this game is to take into account the random factor and adjust your code to take advantage of this while preventing your opponent from doing the same. I would explain all the various strategies that you can apply to this game and how they are impacted by the random factor but that would kind of give away the logic I am applying to my entry. Without the teleport this game would degenerate into the racer strategy and the defender strategy which wouldn't be very fun.

Bottom line for this game is this:
The crux of your design needs to focus on when you use the 5 locates and how you respond to what you have learned.

I have to agree with Taffy and disagree with Baker.

My program rushes with the teleporter and then beelines from where it lands to the enemy flag.

I sincerely doubt anything is more efficient than this. The sticky traps are esp. useless unless you're trying to deathray them. It takes you a turn to lay them, but it only slows them a turn too!

I guess you could try to make an algorithm using the locates to approximate where they're going. Still, that's a lot more logic to do for a questionable benefit over just rushing with the teleporter.

I guess we'll see at the final results, but I wouldn't be surprised if the winners all just rushed w/ the teleporter.

lol... my thought exactly ;)

hmm... taffy... I've heard that nick somewhere before... you wouldn't happen to play STV:EF would you?

I'll bet you I can write a script that defeates the rusher's 9 times out of 10 and I'm pretty sure I won't be alone. I still haven't had time to look at the command line thing so I'll probably not get an entry in because we are running out of time and I don't want to enter and illegal entry.

B

If you simply rush you are likely to be shot, and if you rush but use some random element to make your moves less predictable (i.e. slower) you are increasingly likely to lose to others who are rushing.

Isn't that from 'Clayfighter'? An old SNES game...

Baker, I think you're describing a script that tries to hunt the rusher, which wouldn't be too hard. But then you would lose to a "moderate" script that doesn't purely rush. But then the moderate script loses to the rushing script!

Ahaha! Is it down to a rock-paper-scissors deal?
Rusher-hunter-moderate

I see your point, but that's why they call it logic. You need to make a best guess as to what the opponent is doing and try and counter it accordingly. Any script that doesn't account for the what the opposition is doing is destined to fail.

B

Er, nope, if you spend turns locating, then you're precluding yourself from being a rusher! And I estimate it would take at least 3 locates to determine what the opponent's doing...

So what if you are precluded from being a rusher? Rushers are like lemmings and even a modest amount of logic will kill them 9 times out of 10.

but what if a runner goes random...

it wan't be slower, and u just cant predict where he is going (other than the flag, ;) )

consider this two paths:



both robots take same number of steps to get from '1' to 'l', but i bet u can't predict second's robot path...