5
\$\begingroup\$

Core War is a game about two or more programs written in an abstract assembly language (REDCODE) dueling/jousting/fighting. The goal of it is to get the enemy program to have no more running threads.

Brainfuck Jousting is a Core Wars like game using a Brainfuck derivative instead of REDCODE. It has a bit of activity on the #esoteric IRC channel, but it's very thin. It has an abstraction syntax to make writing programs in it somewhat easier. Its goal is to zero out the enemy 'flag'

Do things like these and their relatives belong on PPCG? If so, how would they be handled and executed? What would be the scoring criteria? How would matchmaking be done?

| |
\$\endgroup\$
  • 1
    \$\begingroup\$ If you are looking to write your controller, I've developed KoTHComm which makes writing these challenges way easier. I love helping others out (especially because it shows me the pain points of my framework), so feel free to ping me in TNB. \$\endgroup\$ – Nathan Merrill Feb 5 '18 at 20:41
3
\$\begingroup\$

This would be perfectly appropriate for PPCG as a challenge! To quote the excerpt:

King-of-the-hill indicates a game where the submissions interact with and compete against each other in some form of game.

In fact, a challenge very similar to this has been done before!

A lot of your questions (scoring, execution, matchmaking, etc.) come down to the individual challenge, but you can find helpful information on the tag wiki. An example of one scoring and matchmaking scheme you could use is to have every bot fight every other bot once, and each game count as 1 point for a win and 0 points for a win. Or you could do it in a March Madness style bracket tournament. However, these are the kind of things that could be solved in the sandbox. Sine KOTH challenges are notoriously hard to balance and implement, I would recommend posting your challenge there to get valuable feedback before going live.

| |
\$\endgroup\$
  • 2
    \$\begingroup\$ Its worth noting that the setup between how the game arena is constructed can lead to a given challenge being more (or less!) interesting. Formic Functions I think became amazing precisely due to the 16-ant free-for-all arena style, whereas Petri Dilemma requires paired-off matches, but the whole "how pairs are chosen" (randomly, from a shifting population based on winners) made that challenge work well. A "March Madness" bracket style would have ruined both of them. \$\endgroup\$ – Draco18s no longer trusts SE Feb 5 '18 at 19:52
  • \$\begingroup\$ @Draco18s True, which is exactly why IMO every KOTH should be sandboxed \$\endgroup\$ – James Feb 5 '18 at 19:59
  • \$\begingroup\$ A variant of Core War has also been done here. \$\endgroup\$ – Zgarb Feb 5 '18 at 20:00
  • \$\begingroup\$ My comment was more born out of the realization that Formic Function's own author made in the chat that a lot of the decisions about the playing field were arbitrary and ended up just working out very well. Sandboxing may not have known how well it was going to work prior to there being several of the entries. It had been sandboxed and several things were changed, but my point is, the sandbox can't see everything. e.g. this comment and trichoplax's reply a few minutes later. \$\endgroup\$ – Draco18s no longer trusts SE Feb 5 '18 at 20:19

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .