PokerChipRace multiplayer challenge discussion


#1

Feel free to ask for help here. We’ll post additional information regularly about the contest on this topic, so stay tuned!


#2

Hi,

If a chip comes into contact with an oil droplet or another chip, the larger entity will absorb the smaller one and will grow in diameter (the areas will combine). The smaller entity is destroyed.
Once propelled, an entity will keep a constant speed by inertia, and can only be sped up or slowed by absorbing or expelling other entities. It will bounce off the edges of the table and entities of equal diameter.

What are the exact formulas for the new trajectory of the bigger chip/droplet that results of two chips/droplets merging ?
Also, how does an “acceleration” affect the speed vector ?

Regards,

Alex


#3

It seems we only may use one line to “Standard error stream:” using printErr( ) through JavaScript ?


#4

Also, I am totally new to this site but I find it unfortunate that we can’t split the project into several files. Or am I missing something ? Any OOP AI results in several classes in the same big file.


#5

Bonjour,
Au niveau de la manière dont s’ordonne le jeu est-ce du type tron-battle ou du type drone ?
dans le premier cas il y a un mouvement pour chaque joueur par tour ( chaque joueur obtient une position differente ).
Dans le deuxieme les mouvements sont tous effectué simultanément ( chaque joueur obtient la meme position ) ?
Yves Mocquard.


#6

Salut Yves,

Vu les traces quand tu déroules le jeu pas à pas, il semble que chacun joue à son tour et non tous en même temps.

Bon jeu !


#7

@LouJo @mocquard_yves: sur les drones, il y avait un pas par joueur dans les traces mais en réalité tout le monde jouait simultanément. Du peu que j’ai vu, ça a l’air d’être le cas ici aussi. D’autant plus qu’il n’y a aucun pas pour les mouvements des gouttes de pétrole alors qu’ils devraient se faire à part pour garder la logique d’un tour par tour.

@alex_servies: about the new trajectory (and acceleration), it’s probably based on real world physics, so depending on the mass (here the radius) and speed of both objects but I didn’t really look into it at the moment ^_^.
Also, if the single file editing is an issue, you can still code outside the editor with many files then merge them before sending. Depending on your language and/or IDE, there might be tools to do the merging automatically.


#8

Hello,

To precise the question:

  • when two entities collide, is the resultant speed proportionnal to weight (radius²)
  • when an entity hits a border, is there any loss of speed?
  • I observed once an entity which stopped losing 1/6 of its weight when accelerating, what is the limit size?
  • when accelerating toward a point, how much are we increasing?
  • when entities hit borders, do they react as light? (same angle each side) This question is particulary interesting when hitting a corner (what if we hit 2 borders at the same time when the trajectory does not intersect the corner? or what if we eat an entity which make us hit 2 borders at the same time?)

I might have other questions as I keep coding, Hope we can edit posts :slight_smile:

(sorry for the dumb questions, I’m not into testing anything for now)


#9

J’ai remarqué un comportement curieux dans des matchs contre une des IA.

Les messages suivants apparaissent dans la console :
melamrani pushes chip 1 toward NaN NaN (x y)
melamrani pushes chip 5 toward NaN NaN (x y)
melamrani pushes chip 9 toward NaN NaN (x y)

Et les jetons du joueur en question n’apparaissent pas dans l’animation, ils semblent “en dehors du terrain”. Sa taille (son score) décroit peu à peu jusqu’à se stabiliser, et au final survit jusqu’aux 300 tours avec un score minimum.

Bug d’affichage/de la simulation chez moi ou d’autres remarquent le même comportement ?


#10

That’s what I do actually :slight_smile:

After coding some basic things, I really need to know how accelerating affects the speed vector in order to make some trajectory predicting features.
It surely relies on real world physics, but I can’t figure out how EXACTLY absorbing and rejecting droplets affects the trajectory. Can someone share ideas ? I didn’t look at replays very closely though.

One fun thing, with an AI who only WAITs, I was top 10 for a while :stuck_out_tongue:

Alex


#11

Je confirme ! Ca m’arrive en arène et je perds…
Quand je regrde le replay ses jetons sont en dehors de la map, et donc je finis par perdre au bout de 300 tours…

Merci de régler ce bug :slight_smile:

ENGLISH:
I also noticed that. Even in arena matches, sometimes the ennemy just starts out of the map and wins after 300 turns. Please fix that :slight_smile:

Alex


#12

@Attractive: Sure would be nice to have several files, it is on our TODO list for the time being


#13

@mocquard_yves: this is a Game of Drones type of game: all movements are executed simultaneously (every player gets the same objects positions)


#14

@McKelian: You can output as many lines as you want. For example

printErr('Debug 1\n');
printErr('Debug 2\n'); 

#15

Oui, j’ai aussi remarqué cette curiosité


#16

@FredTreg Ok it’s nice for me. Did you fix something or my code contained a mistake :blush:


#17

J’ai remarqué un comportement curieux dans des matchs contre une des IA.

Les messages suivants apparaissent dans la console :
melamrani pushes chip 1 toward NaN NaN (x y)
melamrani pushes chip 5 toward NaN NaN (x y)
melamrani pushes chip 9 toward NaN NaN (x y)

On a corrigé ce problème, merci de nous l’avoir remonté :smiley:

ENGLISH: This bug has been fixed :smiley:


#18

For more details on the physics of the game :

http://www.codingame.com/forum/t/pokerchiprace-physics-explainations/268


#19

McKelian: Nope! we did not fix anything regarding printErr…


#20

Hello,

Quelqu’un peut m’expliquer ce que les différentes icones du leaderboard veulent dire ?
Autant il y en a qui sont évidentes, autant certaines (l’éclair ?) reste un mystère complet.

Et au niveau du classement, comment sont calculés les points ?

Sinon grand bravo pour ce jeu, je suis déjà accro :slight_smile: