Strategies for yutnori - waiting for the tool

      Strategies for yutnori - waiting for the tool

      Strategies for yutnori

      Since we must try to win the game against the computer, but also to accumulate as many points as possible, it is advisable, in order of relevance:

      • Overlap your pawns, in order to decrease the throws (since each roll causes you to lose 10 points).
      • Select the throws in order to increase the chances of eating the opponent's chits, so as to slow down the computer and gain points.
      • Select the jumps so as to increase the chances of ending up on special boxes (vertices or center) to take shortcuts.
      • Whenever it is potentially profitable, increase the probability of throwing Yut and Mo (to repeat the throw).
      • Tactically manage the increased probability of stepping back to eat opposing checkers or take shortcuts.


      I thought it might be useful : VIDEO YOUTUBE YUTNORI

      Post was edited 27 times, last by “Stark73” ().

      We also want to try to put these strategies into the bot so it plays good :)
      But for this, could you please test how well these work and then kinda rank them like which ones should the bot prefer, when is Yut and Mo profitable etc.


      If you want to support the work of us, please renew/activate your uploaded.net premium account with the following link:
      >>> ul.to/ref/13987238 <<<
      Goodmorning everyone.
      These "strategies" are from the italian wiki page of the yutnori minigames, and they are pretty good ones.
      I've played the minigame quite a lot of times already and I have to say this is one of the most frustrating sh*t I've ever experienced :D
      Anyway, I think this game is ~70/80% luck based and what we can do to change the outcome is limited, and it's pretty much all summed up in the few guidelines above.

      Overlapping the 2 pawns is, most of the times, crucial to win and also to get the silver chest.
      If you lose or if you win with less than 150 points you get the bronze chest, so whenever the computer gets a few good rolls it's pretty much all about luck.

      I think that eating the enemy pawn should be priority only when the computer overlaps its 2 pawns before we do OR if his 2 pawns are already overlapped and it's ahead of us.
      Of course, if the computer goes for a shortcut and we don't, we can't try to eat him.
      Also, the computer can select the throws just like we can, so in certain situations trying to not get eaten by the computer is more important than ending up in the shortcuts.

      The yut/mo throws are very situational... but you can always go for them if the 2 shortcuts have already been surpassed and if the computer is not ahead of you by 1 to 3 spaces.

      I'd like to see someone else's thought on this because I feel like I am missing up on something, as I got the bronze chest most of the times. (or maybe it was VERY bad luck :D)
      Just one thing you have to keep in mind, things like "in certain situations trying to not get eaten by the computer is more important than ending up in the shortcuts" are okay for humans
      but in order for the bot you have to specify this exactly ;)
      the bot is very stupid and can not "see" which strategy is the best, you have to give him crystal clear criteria - the simpler these are, the better


      If you want to support the work of us, please renew/activate your uploaded.net premium account with the following link:
      >>> ul.to/ref/13987238 <<<
      Show Spoiler
      <iframe width="560" height="315" src="https://www.youtube.com/embed/ob0ojcPKVNQ" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

      This is a video should be easy to guess how to make gold or at least a nice score, I will update it in the post.
      Instead of dealing with this unnecessary event, it made more sense to build an alchemical purification bot.
      Because this mini game is not as easy as a monster card game.
      It made more sense to develop puzzle bot, grab the king, or okey card bot.
      This is a very strange mini game, you can win if your PC wants ...

      We thought about some strategies already!
      There are two tasks, #1 deciding which token to move, #2 deciding which field to select for the next turn (Yut, Gae, etc.)

      For task #1 so far we thought about the following ranking the bot decides on:
      1. Merge tokens
      2. Kick enemy back to start
      3. Move to field in the upper right/left corner or field in the middle of the board
      4. If none of the 3 applies, then move token that is further behind

      For task #2 we didnt think about anything yet, this is still to be done...but here what we want to do is to consider for each token the possibilities of moving -1, 1, 2, 3, 4 and 5 spaces and then also ranking the possible outcomes according to the list from task #1

      Please give us feedback for these strategies! :)
      Important: You have to formulate your rules for game playing as clear as possible there must be no "room for interpretations"



      @Duvetlue Dont worry alchemy is next, but we first focus on this one because its just temporary ;)


      If you want to support the work of us, please renew/activate your uploaded.net premium account with the following link:
      >>> ul.to/ref/13987238 <<<
      Yesterday with 100+ tries I only got 1 gold chest and many bronzes, today in 10 minutes I got 3 gold chests already. It's insane how luck-based this game is.

      Anyway I think that task #1 is the most important one and those rules you listed in order of priority are good enough.

      Task #2 is, to be honest, not THAT important: I don't really feel like selecting the field does much difference and the odds are only slightly incresed.
      This is how the fields should be selected:
      1. If none of our tokens are on the ground, select the farthest field (Mo)
      2. If only one of our tokens is on the ground, select the field where the token is
      3. If both of our tokens are on the ground and are not yet merged, the one that is more behind should go for the field where the other one is. If the one that is more behind is on one of the 2 corners AND the other token is only 1 field far, then try to go for a -1 (back-do)
      4. If both of our tokens are on the ground and are already merged, select the field that corrisponds to one of the 2 corners; again, if the corner is only 1 field behind then try to go for -1 (back-do)
      5. Go for the farthest one (Mo) if the tokens are already merged and on the second corner (the one on the left) OR if the second corner is 2+ fields behind
      6. Go for 3 (Geol) if the tokens are already merged and on the first corner (the one on the right)
      7. At the very first roll (the one that decides who starts) go for -1 (back-do)

      There would be some exceptions to these rules but you are not building an AI so of course we have to keep it as simple as possible. ;)
      Tell me if these that I've listed look good to you or if I should try to explain in a more simple way.


      By the way each round can take up to 1 to 3 minutes so I'd consider adding an option to just end the game if the score is below ~100/110/120 , this can save a lot of time considering how many rounds the bot has to play. What do you guys think?

      Post was edited 3 times, last by “Hypnox” ().

      You should never merge your tokens,NEVER.The PC will always eat you.This is just how this game works...
      We need to find a way to win silver everytime.



      I found that if u dont eat the PC tokens,it will not eat you next round.Just eat it's tokens when u are very far from him,like 5spaces or more.