Why you SHOULD use Seeded Randomness!

Поделиться
HTML-код

Комментарии • 26

  • @davidcodeandart
    @davidcodeandart 19 дней назад +3

    I also use seeds to reduce save sizes in my game, it's very handy! :D though saves can get messed up lightly if I change the generation process

    • @alchemysheep
      @alchemysheep  18 дней назад +1

      Oh yeah I also do that, but you have to be careful just because of the changes yeah xD

  • @Efra_EMX
    @Efra_EMX 2 месяца назад +12

    This video appeared in such an eerie timing, because I just started a new game project that revolves around seeded randomness as its main mechanic lol

    • @alchemysheep
      @alchemysheep  2 месяца назад +2

      Hahah, nice :D what kind of a game are you making, I am intrigued :)

    • @Efra_EMX
      @Efra_EMX 2 месяца назад +2

      @@alchemysheep Since it's for a game competition, I think I need to keep it a secret for the time being xD

    • @alchemysheep
      @alchemysheep  20 дней назад +1

      Oh fair enough, you can still let me know about it when the time comes if you’ll remember :)

  • @glenfoxh
    @glenfoxh 20 дней назад +2

    I discovered all that you said, years ago when working with Dark Basic Pro. I was making several types of maze games. The one that was the most finished, used a seed value to save each level. Yeah, instead of saving the layout of the level, all I had to save, was one number. That makes for one very small save file.
    Helps that the game never saved any progress made in the level, for each level was short and meant to play through in one go. It helped the player a bit, for if they had to leave the game, they could just go back in, and everything would be the same as their last attempt at that level. Making each run of that level, more easy as you got to know it. But, as said, each level was short, very short, to the point, one really never needed to replay any level they passed.
    And yes, grabbing or setting a seed value, can help a lot for in debugging. Many times I had a maze made, and it looked perfect. On occasion as I ran it over and over with different seed values, the game would break, and I would not only have to find out why, but how to fix it so it does not happen again. That is the hard part.
    In Dark Basic Pro, I had also found, if you set the seed yourself, at the start of the program, and you setup a physics sim, so long as you use the same seed, the all physics events will play out in the same exact way, before any user input interrupts the cycle. Meaning a physics sim could be used for level generation, and still only the seed would need to be saved to regenerate the same level again, with physics pushing things or doping into place before the start of each level regeneration.
    Yeah, seeded random generators, are awesome for games.
    If used right, it's also a good way to add a bit of replayability to a game. So no two runs are ever quite the same, if you want to see something new, as far as game layout.

  • @SeahorseGames
    @SeahorseGames 28 дней назад +4

    Excellent videos man, keep it up, perfect RUclips game dev content

  • @forthebeans5011
    @forthebeans5011 Месяц назад +2

    Love the concise explanations!

  • @immortalsun
    @immortalsun Месяц назад +1

    The RUclips algorithm has blessed me again! Surprised this doesn’t have more views. I love concise explanations of interesting topics like this.
    As a rookie game dev interested in making roguelikes, this is actually super helpful!

    • @alchemysheep
      @alchemysheep  Месяц назад +1

      Thank you, I am really glad it helped :D

  • @bc4198
    @bc4198 6 дней назад +1

    Oh, interesting - I like it! I thought this was going to be "bounded randomness" or whatever the term is, for "random but doesn't suck"... For example, a guarantee that all the options will happen sometimes - especially before any of them repeat a bunch of times. I guess you could seed them all once, like weather in Animal Crossing, but ultimately I don't want a 1/1000 chance of getting a Prismatic Shard - I want a goddamn guarantee that I will not have to break more than 1000 rocks without getting one. Or for MMO tag games like Among Us, Lockdown Protocol, etc., nobody should be "it" five times in a row, while other people never are. I mean, I'm open to a "don't make me the killer" setting - but, in practice, people care less about "Wow, what are the odds?!" than they do about the game feeling even and fair.
    It turns out pure randomness isn't actually fun. After 10k hours on 6 islands, with never catching a trevali; or 20+ hours grinding rocks in Stardew, we just stopped playing.

    • @alchemysheep
      @alchemysheep  2 дня назад

      Thank you!
      Yeah I think what you are talking about might be “pseudo randomness”, where you control the randomness so it doesn’t destroy your experience if you are really really unlucky (and the other way around as well) and that is also a really important component of game development, I would imagine almost every major game is using it these days :)

  • @anon_y_mousse
    @anon_y_mousse Месяц назад +1

    Completely agree.

  • @youtubeuniversity3638
    @youtubeuniversity3638 Месяц назад +2

    2:59 Any pros or cons to doing some modifications over others?

    • @alchemysheep
      @alchemysheep  28 дней назад +1

      This is a great question, it is one part of the seeded randomness that is not exactly clear to me yet. It is definitely something that I still have to look into more, if I get some good info I will try to remember and comment it here!

    • @youtubeuniversity3638
      @youtubeuniversity3638 28 дней назад +2

      @@alchemysheep Thank you.

  • @galactic8608
    @galactic8608 2 месяца назад +2

    Nice

  • @Dany-rx7rs
    @Dany-rx7rs 2 месяца назад +2

    😮😮

  • @rzrx1337
    @rzrx1337 Месяц назад +1

    Deckbuilders are not roguelike games.

    • @alchemysheep
      @alchemysheep  Месяц назад +2

      That is true, but a lot of them that have appeared lately are roguelikes as far as I have seen.