Page 1 of 2 12 LastLast
Results 1 to 10 of 12
  1. #1
    Junior Member
    Cybernetik is offline

    Posts
    4

    Announcing: YPP Bot Suite; Help Needed

    Hello mates!

    I am announcing here the beginning of a new large project I am commencing work on; a complete suite of bots for YPP. I will be going puzzle by puzzle, implementing solutions, features, and generally improving on existing bots, or creating entirely new ones in the case that there are no good bots around.

    You may say, "Why would you make new bots when there are already perfectly good ones around?"

    Well, most of the decent bots around cost money. I intend to release my entire suite, bot by bot, entirely free, though I may keep source hidden to prevent OOO from reverse engineering it; but i'll worry about that later.

    While each puzzle game is different from other puzzles, I will, however, be implementing some of the same features in every member.

    For one, when it comes to my undetectability scheme, I plan to analyze packet data, human mouse behavior, etc., and make the bots not only practically undetectable, but entirely impossible to detect. That is, indistinguishable from a human player. I will also be researching, and even testing on my own YPP's own detection methods, but I bet you guys have already discovered some things regarding them, and how to circumvent them. Just know that each and every one of my bots will have 0% chance for detection.

    Maybe I need not worry about detection so much if the current detection scheme is crap, but i'd like to do it both for fun, and to stop the bots from ever being detected if they were to improve their detection mechanisms, which seems unlikely because it'll be a dead game soon enough.

    In addition to the regular duty puzzles, I am also thinking of creating several bots that interact on different levels with the game. For example, a bot that automatically calculates prices across the ocean, and is programmed to run shops in the most efficient way.

    Also, when it comes to the parlor games, the level of bot interaction will be highly variable. I do, however, intend to create the most useful bot that can be practically made for each parlor game.

    Or even a bot web of Blockade bnaving bots ensuring efficient play, or even learning algorithms about the behavior of ships.

    Is there a good Bnav bot out there? I really want to make one... this may be my first contribution, though maybe not since it seems like it'd be pretty difficult.

    I'm an inexperienced but decent C# programmer with a shitload of time on his hands for the next 20 months, and I aim to complete the entire suite by September 2016, if YPP is even around. I bet there will be some stragglers.

    You may not have reason to pay any attention to me, but when I begin releasing a bot every so often, I hope you'll see I am capable and willing to take on this giant project to essentially master YPP.

    I'm searching for sources and methods on my own, but something I would really appreciate would be if an experienced botter would give me the breakdown of methods and for anyone who has elusive sources, to give them to me.

    I would also enjoy working with others, if anyone else would care to embark on this massive project with me. Programming is fun.

    For either of these things, PM me or reply to this thread, thank you.

    Also, suggestions are priceless, thank you.

    - Cybernetik

  2. #2
    Senior Member
    Money is offline

    Posts
    282
    Good luck.

  3. #3
    Member
    EmperorPiece is offline

    Posts
    39
    If you make an Alchemy bot that would be amazing!

  4. #4
    Junior Member
    Cybernetik is offline

    Posts
    4
    Sure, i'll get working on an Alchemy bot!

  5. #5
    Senior Member
    Xalyx is offline

    Posts
    326
    The most fundamentally needed bot is forage, which would open up 2 new areas of content and income for people that purely bot. Other than that, bnav. Other than those two puzzles, anything else is not really helping all that much. We've had 3 forage bots started on this forum that amounted to nothing, not to mention dozens of other "projects" that amount to nothing. Hopefully you don't end up on that list but... yea.

  6. #6
    #Return
    Scarecrow is offline

    Posts
    3,443
    I can offer you advice and guide you along the way. As for actually helping you program I'll probably wait until I see results. Plenty have people have done this, then ended up doing nothing.

    Skype: thescarecrow-botsupply

  7. #7
    Junior Member
    Cybernetik is offline

    Posts
    4
    I see.

    Perhaps this was a bit too ambitious, I was just feeling really motivated, haha, but I haven't made anything as of yet.

    I think i'll work on what is needed, instead of what is not, though I am still a bit disturbed by pricetags, and enjoy problem solving and creating so much i'd be willing to work on it for free, just for the sake of it.

  8. #8
    Junior Member
    Talisvaer is offline

    Posts
    23
    Quote Originally Posted by Cybernetik View Post
    I see.

    Perhaps this was a bit too ambitious, I was just feeling really motivated, haha, but I haven't made anything as of yet.

    I think i'll work on what is needed, instead of what is not, though I am still a bit disturbed by pricetags, and enjoy problem solving and creating so much i'd be willing to work on it for free, just for the sake of it.
    Do you have a skype? I'd love to help make the GUI's for your bots

  9. #9
    Senior Member
    trickshot17 is offline

    Posts
    110
    A few things.

    1.) YPP (at least when I was starting out bot-building) doesn't check your mouse-movement all the time... My first attempt at an HMM was a linear constant velocity mouse movement, and did not get banned with that. My guess is that they either audit accounts from time to time or just check suspicious accounts.
    2.) I've open sourced some nice stuff for image recognition and mouse movement as well as a few complete bots (see http://bot-supply.com/forum/puzzle-p...ree-tools.html).
    3.) Go from easy to hard. Everyone wants an alchemy bot and a good rumble bot, but if you can't write a bilge bot, you have no business touching the other puzzles. (for some perspective, an experienced coder can probably whip out a Legendary-attainable bilge bot in a day if the HMM and image recognition libraries are available)
    4.) When you approach a puzzle, start by figuring out how to find the best move given a game-state, then AFTERWARDS figure out how to read the screen into a game state. Screen reading is generally easy and tedious/boring, but takes a while, but the algorithms are the hardest part, and a background in AI is really helpful for some of them. If you can't come up with reasonable moves, the scaffolding is wasted effort.

  10. #10
    Junior Member
    Talisvaer is offline

    Posts
    23
    Quote Originally Posted by trickshot17 View Post
    A few things.

    1.) YPP (at least when I was starting out bot-building) doesn't check your mouse-movement all the time... My first attempt at an HMM was a linear constant velocity mouse movement, and did not get banned with that. My guess is that they either audit accounts from time to time or just check suspicious accounts.
    2.) I've open sourced some nice stuff for image recognition and mouse movement as well as a few complete bots (see http://bot-supply.com/forum/puzzle-p...ree-tools.html).
    3.) Go from easy to hard. Everyone wants an alchemy bot and a good rumble bot, but if you can't write a bilge bot, you have no business touching the other puzzles. (for some perspective, an experienced coder can probably whip out a Legendary-attainable bilge bot in a day if the HMM and image recognition libraries are available)
    4.) When you approach a puzzle, start by figuring out how to find the best move given a game-state, then AFTERWARDS figure out how to read the screen into a game state. Screen reading is generally easy and tedious/boring, but takes a while, but the algorithms are the hardest part, and a background in AI is really helpful for some of them. If you can't come up with reasonable moves, the scaffolding is wasted effort.
    Do you have a skype? I'm trying to find an experienced programmer.

Page 1 of 2 12 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •