[ Browse Entries ]

Thank you for participating!

Hello. So, having the usual LD craze ended, we’ll have yet another MiniLD coming this month!

Theme: “Determinism” (i.e. no randomness)

In this MiniLD, we’re going to make a game without using any randomness. A game that feels deterministic. Random number generators (RNGs) are often indispensable parts of video game mechanics, but here’s the challenge: Can we make a fun experience without relying on them? The idea has been long in my mind, but now it’s time for trying out this idea!


  • A game with no random element.
  • Team works, using assets are allowed.
  • Choose whatever genre/topic you may please.
    • Here are some ideas: Racing game, Rhythm game, Clairvoyance Gambling (e.g. You know the results in advance), A “detective” game (You find the cause of an event from its consequences), Unconventional strategy game (e.g. a turn-based platformer), Space golf, etc. etc.
  • Submission Date: Sep. 19th ~ Oct. 2nd, 2016 (Submission closed)


  • Can’t we use any random number at all? – Well, I don’t mind if you use random numbers for particle effects, etc. But the core game mechanics shouldn’t rely on randomness.
  • How do you define (pseudo-)randomness? – If a player feels something is random, I’d say it’s random. Whether if it’s actually predictable or not doesn’t matter. What matters is the user’s perception.

    Edit: a little clarification: I think the game should encourage a player to exploit its non-randomness. Just using a fixed seed doesn’t qualify as a good deterministic game.

  • This sucks! I hate a mechanics theme! – I’m sorry, but hey, I’m just offering an idea for experiments. Because, after all, we all love experimenting with radical new ideas, right?

That’s it. Have fun~~

Edit, Oct. 2nd:

We had more than 30 awesome entries! Thank you everyone who participated!
Also I want to personally thank @ratking, who immensely helped me for this.

56 Responses to “Mini LD #70: Determinism – Sep. 19th ~ 30th, 2016 (Ended)”

  1. Ghen says:

    awesome!, never participated on miniLD before, seems interesting theme πŸ˜€

  2. Geckoo1337 says:

    However, can we use a fake random number generation – as a value in an array which has been selected by a constant? This challenge seems interesting for developers. I am in ++

  3. McMutton says:

    Sounds interessting. Just wondering should you always know the result of your action before you do it, or should it just not be based on random numbers?

    As example I’ll take a golf game as you’ve mentioned:

    Deterministic, but player still doesn’t know the outcome:
    If the power of the shot is a timing based action, I don’t know the exact result of my action, but no random numbers are involved. (well… the player is kind of a random component)

    Deterministic, and player knows the outcome:
    Player can input the values (power,angle, etc) himself and is shown how the shot will line up, before he actually commits his action.

  4. euske says:

    @Geckoo1337 and @McMutton

    Hm, a good question. I think if a certain pattern is repeatable (i.e. a player can try the same pattern again and again) it’s not random. But this is a slippery slope, because any random game with a fixed seed can be claimed as non random. I think the game should encourage a player to take advantage of its non-randomness.

  5. ReactorScram says:

    I love making my racing games deterministic! Finally I am vindicated!

    I might iterate on my LD36 game instead of starting anew though

  6. Tijn says:

    What an awesome keynote video!

  7. klianc09 says:

    Theme is really inspiring, but why that specific restriction to not just use a fixed seed? That’s not less deterministic than anything else a computer already does.
    Determinism is nothing special in games. Can’t get any more deterministic than any puzzle game, story-driven game, action game with pre-made levels, strategy game, or …. board game like chess? Ok, multiplayer games are ruled out, because any person appears to be a Random Number Generator to another person.

    Trying to view it from an alternate viewpoint, my first thoughts were:

    Create a rouge-lite and

    EITHER have a fixed seed and only one PRNG, so if you do exactly the same actions again, everything will be generated exactly the same, and if you do just one thing differently, anything could be different. Sorta like a butterfly effect. That would allow for people to share 100% repeatable walkthroughs, and people could try to explore new paths to find “alternate dimensions”. Heck, you could make sharable savegames by simply exporting (and importing) the series of inputs to a textfile.

    OR have a fixed seed when the program starts the first time, so that everybody will go through the same generated content/levels when they play. You could potentially watch someone play it on Youtube for the first time, and they will play through the exact same levels you did, so you can compare your performance to theirs. There could be like a separate leaderboard for each Play Session, so you could compare your score of your 37th try to other people’s 37th try, who have gone through completely the same experiences/levels as you have. Of course this could be easily cheated by deleting the locally saved seed to start anew.

    Both approaches don’t bring anything new to the table, what couldn’t be done with pre-made levels. Sharable savegames can be created by other means and leaderboards unique to a seed where you can only try once is something e.g. Spelunky already did with the daily challenge.

    PS: Just writing this down because I wanted to focus on something else this month, so maybe someone else can use this. But I’m actually getting hyped to just drop everything and prototype the first idea… dammit, self-discipline!

  8. caryoscelus says:

    Hmm, i suppose this only makes sense as a challenge if you make a kind of game that does normally rely on randomness.

  9. Oller125 says:

    This’ll be interesting. I’m thinking of maybe one of those games that exploits the fact that the level will behave the same every time. It’ll probably be one of those games that has the “ghost” or time travel mechanic. Basically, you might have your first iteration die on spikes so that your next iteration can clear them and complete the level.

  10. I’ve had an idea for a while now for a Tokyo train system simulator, and this sounds like a good opportunity to create it.

  11. kyyninen says:

    So here’s what I’m thinking…

    Turn-based tactics game, somewhat similar to FFT(A). Always wanted to make a clone of those games. Combat is skill-based: attack check would be character skill + d20 + other modifiers vs. armor + other modifiers. HOWEVER, every combat encounter would have fixed “random” seed, that is visible to player and AI. To make things interesting, player has limited-per-encounter resources that can manipulate the seed; insert new numbers, swap rolls around, destroy an entry etc.

    AI is difficult. Maybe enemies would have very simple routines: choose the shortest route to nearest player party member, attack. Ranged enemies try to always be on max range away from chosen target. If there’s several different tiles that are equally viable, start from top and go clockwise until viable tile is found. Some enemies would choose the strongest possible target, some the weakest.

    Boss enemies could manipulate the seed. If next turn is player’s and next number from seed is too high, burn it. If attacking player, use limited resource to insert highest possible roll.

    Since random generation is not allowed, biggest problem I see is content creation. There should be lots of combat challenges of varying difficulty, preferably a short campaign. Character creation and level ups, loot. Hmm. Maybe combat is not tile-based, but plain old JRPG-style.

  12. scriptorum says:

    I think what’s throwing me off here is “If a player feels something is random, I’d say it’s random.” Couldn’t that arguably knock out games of dexterity and memory where arbitrary failures of reflexes or memory create an unpredictable outcome? Similarly, as @klianc09 says, a board game feels random since opponents can make unpredictable choices.

    What that leave are puzzles. I love puzzles. I’ve made a few. But as @caryoscelus suggests, the challenge only makes sense if you take a different genre with a random element, and translate it to a deterministic platform. That could be fun, but I’m concerned that all that’s making is just a puzzle in a different genre’s garb. AKA a puzzle with a complicated backstory. ;^)

    • iwiws says:

      Guys, I don’t see why Board Games would be out.

      Most AIs on board/strategy games use a bit of random, heavily weighted to have they do something that make sense but isn’t predictable. But it could be fun to have, for example, a game of 7 Wonders where the card order is fixed, and in which you have to find the good way to beat the AIs, influencing them to take some cards and making sure you get enough points.

  13. CyberStarLight says:

    As McMutton said, even if you make a game 100% deterministic and predictable, the player IS the random component. so as long as the player can in any way affect the level, randomness will be injected in. like the butterfly effect.

    In that case the player will perceive randomness as a result of his changing actions.
    As long as the cascading effect is small and slow the player can follow, and play a “deterministic” game. like in the old game “The Incredible Machine”. where the player abuses the non-randomness.
    What about games that are deterministic but too complex to follow?

    what about instead, exploring the butterfly effect? placing the player in a seemingly random world which is actually just a complex predictable system, when the player learn enough about it?
    I think this is an interesting experiment with randomness too, so maybe we should not shy away from perceived randomness as long as the mechanics revolve around non-randomness.

    Just leaving my thoughts about experimenting with non-randomness here.

    * If we are talking about just a game that is not random, then I would argue there is more of that then random games.
    non-random: Platformers, first person-shooters, turn-based strategy, real time strategy, a lot of competitive multiplayer games, puzzles, point and click, shmups, bullet hell, fighting games, sports games, and probably more.

    • McMutton says:

      Well you can make nearly any game “non-random”. Randomness in a game is just a easy way to increase the replay value of a game.
      Just a few examples:
      – remove procedural generated dungeons and use designed levels
      – remove damage ranges from weapons/monsters and use always the same dmg output
      – remove random elements from your AI and have a pure finite state machine
      – don’t shuffle a card deck, but use them in a specific order (actually done in most card game tutorials)

      So it’s not really hard to make a “non-random” game… not even hard to make an interessting non random game. But in most cases it’s not easy to make a “non-random” game with high replay value. (what I personally interpreted as the “objective” of this mini-LD, since Euske mentioned that the player should be encouraged to abuse the non-randomness of the game)

      Of course you can have a really complex system, where the player can’t anticipate the outcome…. but wouldn’t that “feel random to the player” again?

      Personally I (probably) won’t participate in this MiniLD, since I currently favor playing around with at least “some” random mechanics. But I still like the idea of the theme, but depending on the interpretation it could be quite difficult as well.

  14. Ieyfo says:

    When does this start, if its when submissions start its very unclear.


  15. euske says:

    The submission is open from Sep. 19th to 30th, but you can start working before this date.

  16. FLDude64 says:

    Welp time for my 2nd chance

  17. QualityCatGames says:

    well… i have an idea!
    going to try testing the limits here


  18. Syrapt0r says:

    I am SO in πŸ˜€
    I have a pretty nice idea for this theme which I think could turn out amazing if I manage to implement it the way I picture it in my head. As always, good luck to everyone participating!

  19. yangamedev says:

    Terrific theme! Perfectly timed for me as I’ve been playing PC versions of Blood Bowl & WHQ and am surprised at just how fun dice rolls aren’t. Been musing about how I’d make these tabletop games without dice for the past couple of weeks and then this theme pops up!

    An AI you can predict 100% (win or lose) is no fun. Making the AI unpredictable without randomness is the biggest challenge here, it’ll have to change strategies when things aren’t going well (or to let the player win of course).

    Or I’ll just be lazy and make it multiplayer. Either way works :)

    • sorceress says:

      Turn based strategy games could feature a deterministic AI. The main problem is that once you learn how to win, you’ll be able to win every time. Possible approaches are:

      1) Leave it like that, as a small game with no replay value.
      2) Make the AI very strong, so there is an expectation of losing. All joy must come from exploring the game space, and making failure fun.
      3) Make the AI deterministic but chaotic. The same situation leads to a different AI response because it doesn’t like doing the same thing twice. Maybe the AI expresses regret in the game, and communicates this to the player. “I regret making that move. I won’t do it next time you play against me!”
      4) Upon winning the game, the game level (and AI?) could mutate, so the same strategy may not succeed in subsequent plays. The game basically reinvents itself each time it is conquered.

      • yangamedev says:

        Thanks for the feedback — can’t get excited about zero replay but I’ll see how it ends up.

        Since any pattern you can’t discern appears random, obfuscation is the name of the game. I’d rather it be a pattern based on inputs visible to the player (# of units alive => behaviour) rather than some black box math function (option = name.GetHashCode() % options.count). I’m worried chaos/entropy is itself predictable, damn humans.

        To ensure I finish something, I’m going in this order:

        [1] Zombie — gathers all options and randomly selects [baseline for comparison]
        [2] Active — evaluates the board for risk/reward and implements a single strategy
        [3] Reactive — evaluates the board and switches strategy based on risk/reward
        [4] Predictive — evaluates future boards and switches strategy based on risk/reward
        [5] Genetic — mutate tunables based on win/loss, maintain balance to stay challenging but winnable
        [6] Adaptive — cram a neural net in there and claim wonky behaviour is a “feature”

  20. Agecaf says:

    Interesting! I started working on something that could work very nicely with determinism… Just a “simple” bullet hell where you can mess with time : D

    I’ll try to get it to a demoable state by then!

  21. resty says:

    I’m going to take some time at weekend to make a little game. Got a nice idea after some thinking.
    Just to test whether it will be fun.

  22. g_o says:

    Lol when you had trouble with that machine … it’s like you failed as a japanese XD

  23. HawkWhale says:

    Would it be acceptable to have the game mechanics be determined/rhythmic, but the aspects and concepts of the game seem “random”?

  24. anegri says:

    This is a great theme! and as someone else said awesome keynote video… I will try my hardest to complete this miniLD and submit a game on theme

  25. Arowx says:

    What about using fractals they are deterministic, what about noise functions aren’t they deterministic?

    Also if we are using a computed random number generator in essence they are not random, just random enough over ‘smaller’ sample sizes to provide an approximation of randomness.

    The harder challenge would have been make a game that uses true randomness. Probably involving 3d printing a wireless dice or building a radio receiver that listens to the background noise of the universe to generate random numbers.

  26. pvtroswold says:

    I once had an argument with my brother about randomness years ago, in which I made the claim that there isn’t any ‘true’ randomness. Anything that is governed by the laws of physics or otherwise limited/inhibited by any restraint or confine must be in a sense deterministic. For example, the role of a typical die will always land on a number 1-6 exclusively, depending on how you drop it, the wind, gravity, weight of the die, and the surface you drop it on.

    But ‘random’ is still a very useful term inasmuch as we can’t predict a given outcome. If something is so complicated, or the mechanisms are hidden from us, that we can’t reliably calculate quickly or precisely enough with the tool of our brain, we can call it ‘random.’ To me at least, the word random is 100% equal to the idea of ‘unpredictability.’

    Also games. Hopefully games. Good grief, two weeks to make games in!

    • I once argued to a friend that there is no free will because of the same things. Everything in the cosmos is deterministic because of the laws of physics, so even the things you think have been determined at the lowest levels by these laws… He didn’t like that. πŸ˜›

      • pvtroswold says:

        Haha that could actually be a great game idea for the theme. Incidentally, I don’t think I like it much myself. But that opens a whole new can of worms with the idea of free will. But I think you just might be exactly right! However, interestingly, if you think about it, just because what we do may be pre-known or pre-determined (whatever you want to call it), doesn’t mean we don’t really experience the ‘now’ and actually make meaningful real decisions.

      • sorceress says:

        what about quantum uncertainty? the outcome of two particles colliding is dependent on both their positions and velocities, neither of which have exact values.

        • Agecaf says:

          And let us not forget about differential equations without continuous dependency. Continuous dependency means “small enough changes in initial data lead to small changes in the solution”; The heat equation, the equation for a spring, all the “easy ones” are continuously dependent. But not all. For example, the equations of the weather are not continuously dependent, so we’ll never be able to predict the weather (this is the so-called butterfly effect). (That’s when probability studies kick in, and why the weather is forecasted with probability percentages)

          That being said, I actually prefer thinking that determinism exists retroactively; I’ll be someday in the future, and that future me would know what happened, and that was what was determined to happen. In other worlds, I like to think “the book of fate is already written, but no one will ever be able to read it”.

          THAT BEING SAID… wtf is up with quantum erasure.

  27. Stuntddude says:

    Given any sufficiently complicated physical system, small changes can become amplified at every step and turn the end result into something unpredictable and chaotic. But what if you could control those changes on an arbitrarily fine level? With enough super-high-precision tweaking, you could eventually coax a chaotic system to give you the result you want. In that vein, I have what I think could be a really cool idea for this MiniLD.

  28. josiest says:

    I have a somewhat vague concept. I’m new to LD and game jams in general though, how early can we start designing. Like if I wanted to do compo style, will I need to wait until the 19th?

    • NoahCagle says:

      You can definitely start thinking about what kind of game you’re going to create. But you can’t begin any development until the 19th.

    • euske says:

      Uh, actually it is not that strict. You can start working before 19th. The reason that I chose the submission dates was mostly technical (we don’t want to overlap with the previous LD). But we have to close submission at 30th, because MiniLD is supposed to be a monthly event.

  29. Avant-Marde says:

    Aw, first time ever I manage to catch a mini LD when it starts and not in the middle or afterwards, and the theme doesn’t interest me. :p

    I feel like I make a lot of deterministic stuff anyway, so it’s not something I need to have as a theme to do (much like I feel it’s really silly that people would need “no violence” as a theme before they make a game that doesn’t have it, which is why I’m always baffled when it ends up in the final voting rounds)… My last entry was almost completely deterministic.

    Probably holding out till the next one, to see if the next theme suits me better. But good luck to everyone who’s in on this!

  30. AntReek Games says:

    I’m quite new to Ludum Dare. I was impressed by the theme and thought i should make a game. My doubt is where should I submit my game for this jam?

    Thank You

  31. NoahCagle says:

    I love this theme. This is very interesting, yet very difficult. But over the years, I’ve seen great things come from Ludum Dare’s competitors. I can’t wait to see what everybody comes up with!

  32. dollarone says:

    I think the place to submit is http://ludumdare.com/compo/minild-70/?action=preview once it’s been set up (I assume that happens once LD is complete)

  33. I made a simulation of Tokyo’s Yamanote line in two nights worth of work. It was a good excuse for me to put the train station jingles into a game, which I will always remember from my trip to Tokyo in 2014. All you can do is toggle the map and names, so I guess it’s debatable whether or not it’s actually a game. I did learn how to use two cameras/viewports in a game at the same time, and it’s actually really easy to do in Unity. I’m surprised that it’s not used more often in games, but I guess it does take more graphical processing power.

    Here’s the link to my submission on Itch.io (hopefully the Ludum Dare submission page opens soon). https://gatechgrad.itch.io/toukyoutochikatetsu

  34. Wan says:

    Excellent keynote, worth watching until the end πŸ˜‰

    It’s also a good idea for a MiniLD theme! I’ve always wanted to try making a racing game anyway so I might enter if I get enough time.

  35. euske says:

    The submission is open now!

  36. huminaboz says:

    Starting on a late submission this weekend, hope it’s not TOO late :)

  37. huminaboz says:

    Welp, after 6 days of working on this game nonstop, it’s ready. A metaphor for the machine of reality and our daily lives… but maybe in a fun way, heh. It’s called ‘Negative’ if you check the submissions.

  38. 7heSAMA says:

    What time do submissions close tomorrow night?

  39. euske says:

    Technically, the submissions are open for the most part of weekend. It will close at Sunday 8PM, UTC.

  40. yangamedev says:

    Thanks for the extension, end of month is brutal for milestones at work

  41. euske says:

    Thank you everyone who participated!

Leave a Reply

You must be logged in to post a comment.

[cache: storing page]