Possibility of having a battle consist of more than 3 matches

Entelect Challenge 2017 Tournament Format
1

Hey guys its been a while and have I been waiting…

When I first saw the challenge I thought. This is too easy. Then I started thinking about the problem and I actually found quite a number of “problems” to deal with. That and I have never used Java before. But managed to get the sample bot working and started building my bot. The Golden ticket system is genius.

For the best part I take things as they come and are not one to complain etc. I was wondering about the following.

In bomberman, We had about 400 rounds max. This made for a rather long match. This year on a match that is 14*14 tiles the max number of rounds before a win is 196. And thats and that is extreme Chances of 2 bots both going this long is very close to impossible if your building a serious bot. So my question is. Would there be a possibility to make a match best out of 3? My main reason for asking is. A average bot would be able to beat a near perfect bot if the one bot gets a good “Start” where best out of three would surely produce better results. And on a 10 V 10 match, I see an average of about 50 matches.

With regards to Round Robin matches. This wouldn’t be a problem. Since the number of matches should reflect bot strength.

When it comes to the Double Elimination. I would consider a best out of three system. Just to avoid those scenarios where a average bot beats an “Elite” bot. If the bots could face on 77, 1010 and 14*14 I believe it would be very cool. But I understand that time is a factor. Regardless. Nicely done on the competition. I know things will be a lot of fun as usual. And with as many events as we have. Theres much less waiting involved.

My main concern is. Due to the nature of Battleship. I dont feel a bot’s real strength could not be measured in a single match. (Specifically looking at the double Elimination format). I do intend to get my golden ticket earlier. But thats hopeful more than confident. Giving myself 1 and a half months to build he perfect bot. And then tweaking as needed through the year. But if the Double elimination does come. I believe we need to find a way to “Stablize” randomness a little. I would agree that the larger the map, the smaller the chance of a bot winning randomly. So if the 14*14 map is used for Double Eliminations. It could

2 Likes
2

Hi Cyberdevs,

I like your idea but I will have to discuss it with my colleagues. We will get back to you with a more conclusive answer once we have thought of all the possible impacts this have.

3

Hi Cyberdevs,

After some discussion with my colleague, we have decided to stick with double elimination as our main tournament mode as well sticking to a single match between the bots, the reason for this is that we still have a time frame that we have to run all the matches in, however we will be using 14 x 14 sized maps for our tournament mode. We believe with the upcoming changes and rules to the game engine, that elite bots will be favoured and stand a better chance of winning. Thanks for your input, really appreciated,

4

This is cool thank you.

I noticed that there were some content that is currently irrelevant and figured that you would make things much more interesting as time progresses. Not that its not interesting. I just remember the amount of problems I had with Bomberman last year. It took me almost a month just to get the thing to move. My bot just needs a few tweaks and it will be complete.

I only started noticing that there is a few data in the reference bot that hints on more complexity after my post. And reading through the challenge write-up again. I am also looking forward to the rule changes as they come. My battleship bot was harder than I thought initially. I just got very lucky that most of my plans worked out of the box. So I didn’t need to rewrite everything over and over. So now its those super hard tweaks that will give you that 1% increase in win chance.