As you are likely aware, our 2018 Global competition ends at the end of the year. We were throwing ideas around for exciting competitions in 2019, and wanted to hear from everyone here about what they would like to see. We want to hear what you guys are most interested in, and if you have ideas you want to share with the team for competitions large and small. Here are some ideas we have been throwing around, as examples.
Use a more fair format to select the top 8 for a single elimination broadcasts
Full support for teams and team competitions
Ability to apply to create sponsored competitions for just a group of friends or on campus
Ability to to create community or invite-only competitions
King of the hill competitions, where winning a competition lets you face the reigning champion, and remaining a champion for a while earns you prizes over time
Frequent but minimally or unprized ‘practice’ or ‘for fun’ competitions
I feel like tiered competitions could be a good idea, for smaller prizes of course:
A player selects their algo of choice
All algos in the contest battle it out as they do now, receiving a representative ELO by the end of the week or after X amounts of battles
They are then split into groups of twenty or so, where every group does a round robin tournament, the winner of each group could receive a small prize
I think this is good because it gives everybody the chance to win something, even if it’s very little - otherwise, there’s only a small selection of the many who have entered who are in with a shot of winning anything, so it’d be the same people winning prizes.
All those ideas seem pretty fun. I don’t know how strenuous on your systems it would be, but it would be cool to have a competition where you play every other algo, and whoever has the most wins, wins the tournament. Also, I think it would be interesting to have classifications for algos, like static/dynamic, maze/ping rush/alternating sides/other, and then have small competitions to face other algos with the same classifications, thus similar to your own, so you know how effective yours is in comparison.
This behavior will disappear once the adaptiveness increases. The basis behind all those static strategies are very effective ways of attacking the opponent. See it as a set of screwdrivers to open the box on the other side. The most effective adaptive algorithms will probably figure out a way to use the best attack method for a given situation, making it very difficult to classify them.
Haha, ok you’re totally right that there would have to be a way to limit the entries for any kind of round-robin system like that.
I think the local events so far have all required in-person participation, so maybe there could be some online “local” events.
Edit: what got me thinking of it was the C1Ryan challenge, which surprisingly has just 118 entrants. But I think we’d all enjoy seeing how our algo matched up again Ryan. Maybe he just does an exhibition match against everyone in the group - not saying he needs to pay every one who beats him $100
Maybe have every algo matched against his as a goalpost to enter the rest of the tournament? And then ryan hosts a twitch stream laughing uncontrollably while destroying all the contestants?
That might actually be more entertaining though. (Not because of the money, we can just change it to: the winner gets 200,-)
I think this would be the essense of “challenge C1Ryan”
[edit] it looks like there are only 118 participants in the challenge, so this is actually very possible. You guys could handle the code bullet challenge, so you guys should be able to easily handle a @RegularRyan boss battle . If there would be no algo that can beat @RegularRyan I will have to pay my respects to a superior programmer! Guys back me up on this one xD, I think everybody wants a chance to face off @RegularRyan , can we have poll?
not gonna lie, thinking of different types of competitions is pretty difficult. i mean, due to the way this game works, algo a will always beat algo b in any match, so that cuts off several potential options, although the ease of simulation and use of algos opens up several others. here are some ideas i had:
algo vs. human: by the end of the global competition, there will without a doubt be algos strong enough to beat decently experienced humans, and without knowing the code behind them, they will prove to be challenging opponents. my idea was to have a “team algo” and a “team human”, but of course there are other permutations, like having each player play alongside their algo.
team tournaments: these have a bit of an entry barrier because not only do you need to have a time commitment, but you also need to be pretty good at the game. that being said, i could see an official league of sorts being super cool. the top players would be collaborating in different teams to build the best algos, while the replays would be publicly posted after every match for the rest of us to check out.
stronger bosses: this would be hard to implement, but if you could have bosses much stronger than any strategy that has been thought of yet, with the first person to beat the final boss getting a prize, i feel like the community would have a fun time figuring out how different strategies fare against it. you could also ban the use of by hand.
This would be awesome! This could be interesting to host internal competitions in our company or to sponsor a local competition where other devs can join in.
It’s been neat to hear from people still in High School working on algos. Maybe there could be a series of competitions based on age or even school status? Would make for neat bragging rights
. If there would be no algo that can beat