Search Unity

  1. Unity 6 Preview is now available. To find out what's new, have a look at our Unity 6 Preview blog post.
    Dismiss Notice
  2. Unity is excited to announce that we will be collaborating with TheXPlace for a summer game jam from June 13 - June 19. Learn more.
    Dismiss Notice
  3. Dismiss Notice

Linear vs. nonlinear complexity and puzzles

Discussion in 'Game Design' started by RockoDyne, Mar 12, 2015.

  1. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    So something that's been bugging me for a few days is how to make puzzles for nonlinear games.

    What started this whole escapade was an innocent comment about how Ocarina of Time might have been better if its dungeons could be done in any order. I immediately struck this down because it means there is no longer any capacity to build puzzles that are expanding on each other. You end up with a bunch of dungeons that are mostly of uniform difficulty (for the puzzles). Zelda 1 has this issue as far as I remember.

    This naturally got me into thinking about how to actually pull this off. The main way to do this that stuck out was to focus on dungeon specific mechanics. Any specific items needed could either be procured on-site, or, if any items are required, the player is tested at the entrance whether they already posses the item (ideally to avoid hours of rubbing you junk against a pedestal of plot progression mid dungeon). Instead of puzzles based on items, the puzzles become more about manipulating the dungeon itself.

    The potential third option in all of this might just be nonlinear puzzles, or puzzles with more than one way to solve it. Instead of thinking you need a fire arrow, you end up needing fire and a point of force, with the difficulty of pulling off any given method being the gating mechanism. This way any new item found opens up new properties that can be combined. This seems doable, but also like a logistical nightmare. It's pros are basically that it fits the spirit of nonlinear exploration, but the number of design considerations are astronomical. Comparatively, nonlinear level design is easier than this since it allows players the option to completely ignore a puzzle to look for a route that fits their aptitude better.

    So any clues how this can be done?
     
  2. Kiwasi

    Kiwasi

    Joined:
    Dec 5, 2013
    Posts:
    16,860
    Cheat. Present the dungeons in any order. But present the puzzles in a set difficulty order.

    :)
     
  3. Gigiwoo

    Gigiwoo

    Joined:
    Mar 16, 2011
    Posts:
    2,981
    How do you solve progression? Non-linearity may take away the player's sense of progress - both in pixel-skills and human-skills. Skyrim (et al) solves this by presenting an open-world and making the monsters get more difficult with you. What they did is VERY hard to get right - thus the awards for game of the year.

    Another possible solution is randomness, as in Rogue-likes.

    Gigi
     
  4. wccrawford

    wccrawford

    Joined:
    Sep 30, 2011
    Posts:
    2,040
    Ittle Dew did it. Each dungeon could be passed by a couple different mechanics, so you could pick your rewards in any order and still complete the game. In fact, it added to replayability, because if you picked a different order the second time, it forced you to play the dungeons differently.

    And it basically gated the dungeon by having very early puzzles that obvious required one or the special powers, preventing the player from advancing until they had it.
     
  5. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    Hadn't actually thought much about progression past inventory size and puzzle difficulty. I haven't even thought much about combat. Might actually be a curious twist on the zelda formula to make dungeon items be combat focused, with fewer puzzle applications. So instead of dungeon items being needed as keys to progress, the real impediment becomes beef gates. I have to admit, I kind of like that idea.

    We'll see. At this point this is just a thought experiment on player randomness. In essence, what would zelda 1 look like if it was made today?

    DM's would approve, but my inner level designer(*plinkett voice* who I keep in the basement */plinkett*) won't have it.

    It's been on my list, and I'll definitely get around to it one of these days.
     
  6. Gigiwoo

    Gigiwoo

    Joined:
    Mar 16, 2011
    Posts:
    2,981
    Hasn't this been addressed? Pick one of 100 rogue-likes, Indie Steam RPG's, or your favorite AAA. Legend of Zelda (Wii), Diablo 3, and Skyrim for example. All of them demonstrate 20 years of RPG evolution.

    Gigi
     
    Kiwasi likes this.
  7. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    The problem with them is that they don't handle exploration particularly well. Sure, it's an activity you will do in those games, but it's not the single most important aspect of play (or even in their top five). Exploration is just another avenue towards grinding. Maybe once in a blue moon Skyrim will toss something atypical like http://www.uesp.net/wiki/Skyrim:Kagrenzel at you, but the majority of the time exploring without a quest isn't needed or interesting (god forbid you explore a dungeon to later find a quest for that dungeon). Exploration as a goal in and of itself isn't something those games actually encourage, much less need. Hell, in roguelikes you can extrapolate exploration as a function of the probability to attain better/necessary loot (in effect, you explore to fill a meter of probability).

    If anything, I'm looking to draw more from metroidvania games then RPG's. The idea isn't to give the player a better stick, but to give them something with different attributes. You don't give the player a better pawn, you give them a rook instead.
     
    GarBenjamin likes this.
  8. Gigiwoo

    Gigiwoo

    Joined:
    Mar 16, 2011
    Posts:
    2,981
    I suppose there are different types of exploration and also that players' desire to explore will differ.

    Skyrim felt wide open to me - offering me nearly infinite possibilities. After I completed the tutorial, I accidentally took a wrong turn. So that I ended up enjoying dozens (hundreds?) of side quests, activities, and tangents before my son pointed out that there was a main quest line! I eventually finished the game, and along the way, I explored whatever I felt like doing - look over there! That looks fun. Magic? Sure. Killing guards? Hmmm... Thieves guild? Nah... Of course my favorite example of exploration is Minecraft, where there are no explicit tasks, only one implicit task, and an infinity of player-driven tasks. You might enjoy, "Design Better Games! Flow, Motivation, and Fun'.

    Gigi
     
  9. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    That's kind of ironic considering it's actually the most directed elder scrolls game. For comparison play Morrowind. After creating your character, you're given a letter and told to deliver it to a dude in Balmora, which you might be given a hand drawn map with directions to. That's it, and you're on your own. Even if you run straight to him (or rather walk, you will actually walk to places), he'll say to go join guilds or monkey around. The game practically begs you to just go out and soak it in, and most of what you'll be soaking up is the culture and lore of the world. The game just oozes history out of it's orifices that you'll actually be able to connect to the present.
     
    Kellyrayj, wccrawford and Gigiwoo like this.
  10. AndrewGrayGames

    AndrewGrayGames

    Joined:
    Nov 19, 2009
    Posts:
    3,821
    Seconded.

    In my project, Sara the Shieldmage [/obligatory plug], I started off going for an open-world approach. I found out quickly that that's hard...like, really hard, especially given my choice of genre (Eastern RPGs tell stories, Western RPGs let players find their own.) So, I took @BoredMormon's advice - In my design, I'm cheating like mad.

    There is a main quest thread that, if followed, will eventually take you around the world in pursuit of defeating the Big Bad, like any self-respecting JRPG would. However, the main quest is just one rather 'vague' thread that in terms of the larger game is really not much more than breadcrumbs. The real action comes from gaining party members that let you solve particular in-game problems (in this fight I need two tanky characters, because this boss dispels temporary HP buffs, but Sara's the only one with the ability to put temporary HP on other members in the first place! Hmmm...)

    Additionally, there are specific quests for things like opening the docks at Solumeria (the town and sub-region of my first continent that I'm working on) so you can get between the first two continents, that you will have to do as part of the main quest, but the player can opt to attempt earlier, if they so choose. The player has no real choice in the matter - if you want to beat the game, you'll need a ship at some point - but I do leave the player a choice on when, and even how they'll go about achieving that implied but required objective.
     
    wccrawford and BrandyStarbrite like this.
  11. wccrawford

    wccrawford

    Joined:
    Sep 30, 2011
    Posts:
    2,040
    You can call it 'cheating', but I just see it as interesting world design. It's perfectly fine that something is too hard (or impossible) to progress through without something else. Half the fun for me is figuring things out. It tends to be better when it's obvious that it's impossible than when it's not, but that's design work.

    Zelda games tend to strike me as that. They've got a lot of things that are walled off by various mechanics, and you're expected to remember that there were upright logs blocking your way in 1 area, and when you find the hammer you learn you can smash logs into the ground... And you're supposed to put those 2 facts together. The game designer clearly laid that out, but the act of figuring it out for yourself is still fun. They're still basically linear games, but the exploration adds a lot to them.

    Asvarduil, your game sounds like it's even more open than that, which is even better IMO. I look forward to seeing it in the future.
     
    AndrewGrayGames likes this.
  12. AndrewGrayGames

    AndrewGrayGames

    Joined:
    Nov 19, 2009
    Posts:
    3,821
    Well, there is still a degree of cheating involved.

    In Zelda, each dungeon and every 'challenge' between dungeons awards an item that 'solves' some mechanic (e.g. in A Link to the Past, the first Dark World dungeon gives you the Hammer. However, the hammer allows you to get to one of the smithy brothers who got trapped in the Dark World; getting him back to the Light World lets you temper the Master Sword into the 'Tempered Master Sword', which greatly improves your attack power.

    My game is fundamentally still a Eastern RPG. Gear doesn't change a character's abilities; it just makes numbers bigger. Even with my current design of having to explore to find things that let you 'build' your own abilities - another instance of cheating, right now I'm defining on paper what abilities I want to be possible, and figuring out what I'm going to make the player do to unlock a set of those - those don't necessarily solve a varied number of in-game puzzles, either; they just let you do different things in battles (though, I guess that's similar. After all, the shield-dispelling boss example above is an example of such a puzzle. The Woden fight will be, too. I've already designed that one, and I'm considering implementing it in a WIP thread iteration soon, to 'check' the idea.

    The other thing left on my mechanical plate is the politics mechanic. Before, I'd intended to just implement it on the Senat Continent, where the big bad there has burrowed herself in, as a politician (see what I did there? There's a pun in that. Clever, isn't it?) and is causing all sorts of trouble. Going up and challenging her to a duel will lead to a non-standard game over; she's just too politically powerful. She won't get any cool Emperor Palpatine lines, but suffice to say that the Ancile piece she's trying to keep out of your hands will be handed to you on a silver platter when you outfox her.

    Now, I've been thinking it would be interesting to implement that around the game world, because it would lead to some different quest progressions; if you're not careful, you can lock yourself out of specific sidequests. You actually have to read NPC text, for once (though, I'll make it short. Something I learned from my writing in The Hero's Journey: a load of tripe doesn't smell any better than a single strip.)

    It would be even more powerful in a way, because I can use it to subtly imply to the player that they are affecting the game world in a notable way. Even more powerful - getting in a certain NPCs good graces can pretty much cause other 'unexpected' effects. In good with a local Lord who has a troop of crack guards? Why should you go into that cave hunting bandits? Best of all, you still get the reward for quest completion. Players love it when previous hard work pays off in cool, and unexpected ways. As long as the NPC text reflects that, such that it's clear that their old work is the reason they're getting quest rewards on a silver platter, it could be pretty awesome indeed.
     
  13. Gigiwoo

    Gigiwoo

    Joined:
    Mar 16, 2011
    Posts:
    2,981
    When I was younger, I used to find that kind of game fun. The older I got, the more I prefered less sophisticated games. I'm not investing my life into a game, I'm simply looking for an engaging experience with goals that I can continue to work toward. If I want to be frustrated by an excellent game, I'll go play LoL. :)

    Gigi
     
  14. AndrewGrayGames

    AndrewGrayGames

    Joined:
    Nov 19, 2009
    Posts:
    3,821
    I think this is a very very very important remark.

    The reason I speedrun Dragon Warrior I for the GBC isn't because it's a JRPG. It isn't because it's the first JRPG. It sure isn't because the game is complicated. It's because, not exploiting bugs, if you know what you're doing you can play through the game in roughly 2:30:00-3:00:00. Seriously, I've got the videos to prove it. If you do exploit bugs, that time drops to roughly 1:30:00, I believe (hint: every time you die in the Game Boy Color version, the Dragon+ guarding the princess respawns. He's worth 950 XP. Have fun with that.)

    I speedrun that game, because I can enjoy it in my limited free time. I can't throw 60 hours at a $60 game anymore; I have to go lower, because bills suck, and it takes hours to get money to pay them.

    I think where gaming has gone wrong these days, is that people making games have conflated the presence of both complexity and length with quality, when in truth, quality is best shown by the absence of unnecessary complexity, or unnecessary length.

    A good game lasts long enough to be entertaining. That's hard to quantify, but it means your game has no filler or pointless 'errands.'

    A good game is also easy to grasp, and as stated other places, 'just makes sense.' Having stats solely to influence other stats that themselves influence stats is superfluous; World of Warcraft's mechanical changes in Mists of Pandaria got a lot of flak, but one thing I think they got correct, was removing 'Bonus Damage/Bonus Healing', in favor of Intellect influencing that, and dropping Mana/5 sec in favor of Spirit influencing that. Reducing cognitive load is never a bad thing, because the most compelling decisions aren't decided by how complicated they are - it's like the moral decisions in Mass Effect/Dragon Age/KotOR.

    When you have two options that work equally well, have benefits as well as downsides, but are ultimately an expression of you, that's when decision-making gets interesting. More filler won't make that the case. More stats and complexity won't make that the case. And, for us small developers, it just makes it harder for us to make the game in the first place.
     
    Gigiwoo likes this.
  15. Gigiwoo

    Gigiwoo

    Joined:
    Mar 16, 2011
    Posts:
    2,981
    Can even this be called into question? I've been playing 'Heroes Charge' for almost 3 months. It's not particularly great gameplay, it's not remotely unique, and the UI is too complicated in some places. And at the same time, I enjoy logging in to engage in a few battles, tick off some daily quests, and level up my heroes.

    Heroes Charge also has one of the MOST clever monetizations I've yet seen. For $30, I can purchase 2820 gems. OR, for $2.99 I get 120 free gems every day for 30 days, for a total of 3600 gems. It's enough gems to purchase most of the things I want, it encourages me to log in every day (engagement), and it's a bloody fantastic value. It's monetization for the non-whales and it's perfectly done.

    Like most RPG's, Heroes Charge is full of 'pointless errands' which are primarily about grinding xp, beating mobs, and collecting gear, so I can level up my characters, in order to do more of the same. And, yet I enjoy it - 3 months worth = $2.99 * 3. A happy customer is always right.

    PS - It's also linear.

    Gigi
     
    AndrewGrayGames likes this.
  16. wccrawford

    wccrawford

    Joined:
    Sep 30, 2011
    Posts:
    2,040
    What I find interesting is that I once enjoyed complexity, I once enjoyed mindless fun and clear goals... But now they've both palled. I've moved on to what I call "adventure".

    I enjoy learning new things, discovering new things, solving new problems. Puzzle games are great for this, so long as they keep it fresh. Games with a good, decent-paced story tend to be great. They typically introduce new places, people, and ideas at a good pace. Scifi seems to be the best at introducing new concepts, but even fantasy does a fair bit of it.

    I will happily playing clicker/incremental/idle games if it means they feed me something interesting while I do it. If they don't, I generally find a way to cheat or give it up as a bad investment of my time.

    The Clicker Heroes monetization that you love so much? It ruins a game for me. Daily things that I have to do or risk losing rewards? They kill a game for me. It feels like a job that I'm getting inadequately paid for. I log in to do the job, and then I no longer care about the actual game behind it. Clearly the monetization works for a lot of people, but I'm definitely over it already.

    Likewise, games that don't let you *think* also fail for me. Presenting a puzzle and then placing a time limit on it destroys the fun. I don't care if it doesn't make sense that I can take forever solving the puzzle. I like to think, and anything that forces me into repetitive restarts just interrupts that thinking without any benefit for me.

    Finally, I hate random loot. It's bad enough when it's random on the enemies you kill, but when they sell you keys or tickets (with real money!) to a chance at random loot? Ugh. Absolutely disastrous on my fun-level. I have absolutely no confidence that they've designed the game without taking that monetization into account from the start, which means that if you don't do it, you'll be purposefully slowed down just so they can make more money.

    I don't want my fun downgraded because they think they can make more money if they do so.

    This all ties back into the topic because I feel that a good, structured experience if far superior to a random experience, and an open world is superior to a linear one because there's more to discover and figure out. Things have meaning for you to discover, and you might have to travel far and wide to do it.
     
    Gigiwoo and AndrewGrayGames like this.
  17. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    And here is pretty much the entire reason for this thread. The hope was that the player will look at a puzzle, look at their inventory, and be able to MacGyver their way through it. Ideally they should only give up when it's clear they do not possess anything with a needed property, or their skills aren't enough to make their current gear work. A player randomly exploring things shouldn't feel like they need to acquire every available item. They don't need to find the lone fork in the world to fight the flying spaghetti monsters.
     
    AndrewGrayGames and wccrawford like this.
  18. AndrewGrayGames

    AndrewGrayGames

    Joined:
    Nov 19, 2009
    Posts:
    3,821
    And that in turn is a powerful statement. The player should be given the major tools through 'just doing what's asked of them.' If you want those extra, juicy bits because you're a completionist, then you'll need to go looking. That does not mean shell out $20 for the strategy guide.*

    *: I jest. I want them to buy the strategy guide from me. Muahaha! Oh wait, that's evil.
     
  19. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    In theory, there wouldn't even need to be explicitly important items if there are several items that have similar functions. Just finding a single item in a class could solve most/many of the puzzles that simply require an effect from that class. An example would be needing a fire started, or maybe not even started, just having fire moved. Then add a traversal element, like it's otherwise a spot out of reach, and now you need the capability to start the fire at range or move yourself into a close enough position.
     
    AndrewGrayGames likes this.