Search Unity

  1. Megacity Metro Demo now available. Download now.
    Dismiss Notice
  2. Unity support for visionOS is now available. Learn more in our blog post.
    Dismiss Notice

Procedural vs Handcrafted

Discussion in 'Game Design' started by tiggus, Apr 1, 2016.

Thread Status:
Not open for further replies.
  1. tiggus

    tiggus

    Joined:
    Sep 2, 2010
    Posts:
    1,240
    I'm not sure if anyone else has posted a topic like this here already, if so, my apologies I don't usually read this forum. With that out of the way, here is my question.

    I find myself often debating whether it is better to proceduralize as much level design/content/items/etc. as I can or go the hand crafted route where every item is meticulously placed/named/designed by hand. What do you think?

    Here is what I think after gnawing on this beast for a good long while and several failed roguelikes(aborted due to non-funness(new word?)). Handcrafted is the way to go. This may sound like heresy if you are of my former mindset, because everywhere you read about this topic you find the strong opinion that going procedural when possible allows you to compete by giving you replayability and more content. Ie. play this game and you'll never get bored because it's different every time!

    I think that statement is often uttered when thinking of some very early procedural game experience we had when they first were gaining in popularity. After that first one, I have very rarely played one through(as in try to actually finish) many times after a couple play sessions.

    Static but well thought out content on the other hand is like a story. You start reading it, and if done well you want to find out what's at the end. You want to figure out the riddle in that insanely cool looking room that only appears once in the game, like many other unique rooms with their own riddles and unique loot. After that, you put it away, and maybe sometime 7-10 years down the road you pick it up again and read/play it again.

    I think the idea that your game is highly replayable is a bit conceited. Maybe for a few people, yes, it will be. But why short the players with fairly generic randomly generated content for that small chance when you could read them a really good story instead that conveys your vision perfectly and leaves them satisfied at the end.

    Anyways! Curious if anyone has thoughts on this I haven't read elsewhere on the internets.

    edit: obviously this does not apply to certain genres of games, such as simulations. You can't do a Dwarf Fortress that is not procedural.
     
    Last edited: Apr 1, 2016
  2. JoeStrout

    JoeStrout

    Joined:
    Jan 14, 2011
    Posts:
    9,859
    I think it hinges largely on another big design decision: is your game meant to be played through basically once, or many times?

    Back in the Good Old Days(tm), all games were meant to be short and played many times, so as to ingest more quarters. Nowadays, most games have gone the other way: game "content" is measured in hours, because you're expected to play through it just once and then move on. Die-hard fans may give it another play-through or two, but that's not really the intent.

    Single-play games depend on a story, dramatic revelations, etc. to keep the game interesting, just like a book or movie. Multi-play games depend on other things: exploration, skill-building, steep achievements, multiple play styles, etc.

    If you're making a single-play game, then I think it's obvious that hand-crafted content would be superior to procedural (assuming you can afford to hand-craft as much content as your game really needs). And if you're making a multi-play game where exploration is an important part of the fun, then (just as obviously) procedural generation is da bomb. This is not just an old-timey phenomenon; all the gamers in my family (from age 11 to, let's say, more than 25) get out Cavern (an iOS roguelike) for weeks at a time, every few months, and have probably clocked 20 to 40 playthroughs each, largely because of the procedural levels. Similarly, I still fire up Minecraft now and then, and the chief draw in doing so is exploring the procedural terrain I happen to find myself in. If it were always the same, I would seriously not bother.

    EDIT: And I also think it's a bit of a conceit to think that gamers will care about your story. Cut scene or dialog? Just show me which button I have to mash to get through it and get back to the game!
     
  3. JohnnyA

    JohnnyA

    Joined:
    Apr 9, 2010
    Posts:
    5,041
    Liked just for this :)
     
  4. tiggus

    tiggus

    Joined:
    Sep 2, 2010
    Posts:
    1,240
    Haha fair enough, but I wasn't really referring to a narrated story like in a cutscene. More the fact that every area has character and a purpose with a bit of a backstory that is told by the environment(and items/enemies/whathaveyou) itself. I find that very very difficult to auto generate and can usually instantly tell when a level has been designed by hand versus algorithm.
     
    Martin_H likes this.
  5. Martin_H

    Martin_H

    Joined:
    Jul 11, 2015
    Posts:
    4,436

    I think it largely depends on a combination of factors what makes the better choice for a game. Minecraft is perfect with procedural generation. It has emphasis on exploration and user generated experiences. Narrative isn't being fed to the player, he creates it through his decisions. Probably most avid Minecraft players will have a wealth of "stories" to tell you that stuck with them because they stood out in their experience. I doubt that you could come even close to that kind of impact if Minecraft had one big handcrafted world. Here procedural generation is a perfect fit for the mechanics of the game.

    But take Call of Duty singleplayer for example, a game that relies on high intensity pacing with handcrafted setpieces and heavily scripted progression through linear levels. I don't see how that could be enhanced with procedural generation. Likewise all competitive multiplayer games, where map-knowledge is a key skill to be better than other players. If Call of Duty or Counterstrike multiplayer levels were completely different every time, it would lower the skill ceiling imho.

    There are sensible compromises that can be made though. Afaik there is one multiplayer shooter that I can't remember the name of, that basically assembles maps from a combination of 3 random level slices, that get stitched together to form one big rectangular level area. I think this approach sounds promising because you can "learn" the individual parts, but the shuffling of the combinations keeps the experience fresh for longer I'd imagine. I never played it myself though, so maybe it's a crap idea and I just don't know it.

    For me personally the most important factor in solo dev projects is "what are you good at and what do you like to do?". I hate making level content manually. That's literally what killed my first game project. I hadn't anticipated this, but for future projects I'll not try anything that doesn't at least have a heavy procedural aspect to the level generation. Combining modular setpieces might work, but I just fail when I need to go 100% manual. I see myself having a lot more patience for tweaking procedural forms of content generation. Though it hasn't been put to test yet, so maybe I just don't like making games and my current assessment of the situation is incorrect ^^.

    Maybe drabness?
     
    syscrusher and tiggus like this.
  6. Tomnnn

    Tomnnn

    Joined:
    May 23, 2013
    Posts:
    4,148
    Handcrafted pieces, procedural generation & assembly. :p

    Focused on narrative? Make larger pieces.
    Focused on mechanics? Make smaller pieces.
     
    syscrusher, Kiwasi and Martin_H like this.
  7. JoeStrout

    JoeStrout

    Joined:
    Jan 14, 2011
    Posts:
    9,859
    Sure, I understand, but I think many players just won't care about the backstory. They're in it for the skill-building, or the sense of accomplishment they get from building a highly upgraded character/base/mech/whatever.

    Take Warcraft III, one of my all-time favorite games. It had a solo campaign I quite enjoyed, with hand-crafted maps and a story. And that story was... um... something about orcs and humans hating each other. That is literally all I remember. But I do remember the fun of building up bases, exploring maps, unlocking cool little flying machines, etc.

    So for me, that game would probably have been better with procedural generation. The things I actually enjoyed about it — mainly exploration & development, and occasional combat with terrain constraints — would have lasted longer with random levels, and I really didn't care at all about any story.
     
    AndrewGrayGames, Martin_H and tiggus like this.
  8. beige

    beige

    Joined:
    Mar 1, 2013
    Posts:
    18
    I would say it really depends on the game and what you plan to get out of it, as well as the intention

    the above example of minecraft (and your own mentioned dwarf fortress) are games that would only really work procedurally generated; their experiences made by discovery, and reducing that personal discovery down to a preset might cheapen the experience. I also think your missing a lot of the other strengths of procedural generation- one of which is personalization. A procedural game experience is personal too the player, each has their own stories of how it happened, their own unique memorable moments.

    another reason for use of proc-gen is speed and team size - it can't be argued a lot of games are judged on how long it takes to reach the 'end', and a proc gen game can expand that with much less time and investment - and a lower development cost ; something very attractive to the indie market where cash is a lot shorter.

    another thing I've seen proc-gen used well is in challenge; like in rougelikes, it can give a game without the cushion of experience - where you have to rely on your own reaction and skill rather than being used to the challenge. if the game

    Handcrafting gives you that extra element of control, to fully influence everything that the player experiences - it gives you control to ensure you can hit directly with the intentional tone and pace. It also helps to create more unique moments - especially in the like of set-pieces, and defintley in the case of storytelling, both direct and indirect. Another main advantage of hand crafting is ramping difficulty, to keep the game flowing smoothley.
    conversely to the personal experience of a proc-gen game, a hand crafted game breeds a shared experience - one that people can talk about and emphasis together.

    I'll grant I do feel some developers make use of proc-gen as a short cut or easy way out, as well as something to pass the buck on. I'd say, in a way, even proc-gen should be hand crafted, as the engine should be robust enough to ensure it works and that its not just cycling the same thing or making something night impssible because 'its proc gen so we can;t help it' is lazy. Cloudberry kingdom is my example of what a proc-gen system should be like - its designed to ensure a variety of challenges on each

    I'd actually say that many of the games I've enjoyed have had elements of both; shadow of mordor and aliens isolation are good examples of games that are generally hand-crafted but have just enough elements in them that are personal to the player it sticks with them; the named orc warbands and their stories, the hiding places and items

    soooo, hand crafted or proc-gen? it depends on everything else as well to me. one cog in the machine :D

    I'll be honest, maybe just the environment I've grown up in, but myself and most of the people I've talked to about games are of the mind a game you can actually replay and want to replay is more fun - dismissing the idea of replayability and those who aim for it as conciet is just being concieted yourself; sorry to say, but if no-one wants to experience what you made more than once, then that says to me that what you made just wasn't that interesting, procedural or hand crafted
     
    syscrusher and tiggus like this.
  9. tiggus

    tiggus

    Joined:
    Sep 2, 2010
    Posts:
    1,240
    Good points and thoughtful responses. To be clear I am not trying to say anyone else is wrong, just a personal reflection. I do think a hybrid approach can work well. I should have included MineCraft as the type of game I think is meant to be done procedurally much like a simulation as that is a core tenet of the game.

    Perhaps this is more of a observation as a gamer than a game developer for me as I have traditionally played tons of roguelike games but I am just not feeling the magic anymore with some of the recent ones I have downloaded. I got one that looked like a really cool space shooter and it just left me with this feeling of "drabness" once I realized pretty much all the dungeons were going to be the same with varying levels of souped up enemies. But for some reason when I play a game and I find named items that actually interact with the environment or a npc in a specific way it triggers that little "aha cool" moment for me still.
     
  10. Martin_H

    Martin_H

    Joined:
    Jul 11, 2015
    Posts:
    4,436
    I agree. I've played multiple times through some of my all time favorite games like Jagged Alliance 2 and System Shock 2. It's quite possible I'll play through them again in the future.

    My prediction is that you'll have less and less of the "aha cool" moments as you progress on your journey to become a bitter old man. I don't think this necessarily has much to do with the games. I notice this myself as it gets harder and harder for me to get pulled in by a game and forget the time while playing. This war of mine was the most recent one that managed this.
     
    Kiwasi and JoeStrout like this.
  11. frosted

    frosted

    Joined:
    Jan 17, 2014
    Posts:
    4,044
    It's worth noting that there's also a lot of different kinds of 'procedural generation'.

    The one I think that gives the most bang for the buck, as well as the best gameplay experience is procedural interaction, that is, simulating many players or agents and giving the player a fun and interesting system to work within.

    This kind of thing is usually seen in strategy titles, but also has a respectable showing in RPG/Strategy hybrids:
    - Jagged Alliance 2
    - Mount & Blade
    - XCOM (original)
    - Crusader Kings 2

    In terms of random maps and the like, often the rule is based on your loss conditions, the more important the loss condition is or the more frequently the player may reach a loss condition, the more value procedural provides. For games without loss conditions it's frequently not so important, since they're intended for single play through.
     
  12. JoeStrout

    JoeStrout

    Joined:
    Jan 14, 2011
    Posts:
    9,859
    Yes, I think you've hit a key point here. The choice of single (usually long) play-through vs. multiple (usually short) play-throughs is very fundamental, and impacts nearly every aspect of the design, including loss conditions, permadeath, and the current topic of proc-gen vs. hand-crafted levels.

    @tiggus, it sounds like you're burned out on roguelikes and into more story-driven games at this point. I can respect that, even if I don't feel it myself. You would naturally gravitate to single play-through designs then, and in that case, I agree that your proc-gen tools can gravitate to the bottom of your toolbox.
     
    frosted, tiggus and Martin_H like this.
  13. tiggus

    tiggus

    Joined:
    Sep 2, 2010
    Posts:
    1,240
    Oh trust me I am already well along that path :p It is that increasingly difficult search for "aha cool" moments that makes me pontificate about things like this.
     
    Martin_H likes this.
  14. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    As much as this is probably the prevailing notion, it seems horribly misguided. For one, it's intrinsically tied to the notion that levels are the content, ergo more levels means more content. With just about any roguelike though, it should become apparent that the real content is the mechanics. Learning and mastering the systems becomes the real goal of the game, and not getting that stupid amulet of yendor, which is the equivalent of beating all the levels.

    To really go down the rabbit hole, the next matter to consider is how well proc gen provides for expert level mastery... Short answer is it doesn't. Okay, that's not entirely truthful. Breadth of mastery is usually the most important factor, know to respond to scenario X with operation A. Depth of mastery is rarely present; where scenarios X, Y, and Z don't synergize to require the player to use operation D and E, but are only additive and only rely on the usual operations A, B, and C; or there is never any time where expert level mastery is demonstratable; where scenarios X, Y, and Z occurring together is almost unfathomable. And this is just me complaining about the skill ceiling.

    Skill floors either never change (it happens more in ARPGs) or are intrinsically tied to RNG and the breadth of mastery. In the majority of cases, learning anything is purely an exercise in trial and error. How mechanics work is never explained, much less in detail, nor are there ever times where the player can freely or safely experiment. You are left with players with the vaguest sense of what things do and little to no concrete knowledge since any experimenting they attempt is always in an environment with variables they likely can't account for.

    So you're left with a game that is fundamentally about system mastery, made impenetrable entirely through the obscurity of it's mechanics, with little ability to show deep mastery. It's completely antithetical to "easy to learn, difficult to master," but at least you get your money's worth in terms of playtime...
     
    LMan likes this.
  15. JoeStrout

    JoeStrout

    Joined:
    Jan 14, 2011
    Posts:
    9,859
    TL;DR: @RockoDyne doesn't care for Roguelikes. :)
     
    TeagansDad, Kiwasi and Martin_H like this.
  16. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    Back to the main topic...

    I actually bounce back and forth on this, in large part depending on whether I feel like being a level designer at the moment. Proc gen always sounds promising, but I know I can always produce more interesting results myself. The current project being shelved was sort of in the middle, a player created/arranged overworld with an underworld essentially seeded by what the player put on top. I sort of stalled trying to figure out how the art would work.

    I'm going to classify it as "I'm critical because I love." Well, that and Eighties sensibility can die in a fire.
     
    tiggus likes this.
  17. tiggus

    tiggus

    Joined:
    Sep 2, 2010
    Posts:
    1,240
    This is pretty much where I am at the moment. My last attempt at a roguelike was mechanically functional, but boring. Entirely possible I am just not capable of designing a fun procedural game and it is my own limitation.
     
  18. Martin_H

    Martin_H

    Joined:
    Jul 11, 2015
    Posts:
    4,436
    Could it be that the process of setting up procedural generation is so technical that one easily loses track of the "making it fun to play" part, because you are too much preoccupied with the how instead of the why?
     
  19. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    Making level generation any good is almost always about throwing more features and complexity into it. There are a thousand techniques to make a roguelike dungeon, but any of them on their own are just mediocre. God forbid if you want it to make any kind of coherent sense, too.
     
    Kiwasi, Martin_H and JoeStrout like this.
  20. Deleted User

    Deleted User

    Guest

    Well that's not strictly true is it, you don't play games like Skyrim / Witcher or Mass effect for their exceptional strategy and combat systems do you? Last time I checked they did pretty well, as much as I'd like to see more engaging systems in RPG's as the format seems mainly copy and paste.

    It of course depends what type of game you're into, but some seem to mistake a plot as an obstacle. What they really mean is the plot sucks and they want to engage in something actually fun.

    As for procedural content, yeah it can work in your favour if you're not the one doing it. Like Houdini for example which contains a lot of procedural art creation tools and / or something like the parametric modelling solutions I've seen hanging around this forum.

    You're generally better off doing it by hand in a modular / re-use fashion. I do use it in a fashion for e.g. if I'm randomising windows on buildings etc. but for the most part I try to avoid.
     
  21. JoeStrout

    JoeStrout

    Joined:
    Jan 14, 2011
    Posts:
    9,859
    The only one of those I've plaid was Skyrim, and wait, did it have a story? I do recall some cut scenes at the beginning, notable mainly because Patrick Stewart was the voice of the king, but I don't remember what he was on about. I do remember lots and lots of time leveling up my character, gathering potion ingredients, enchanting items in clever combos, and eventually getting my character up to the point where she could attack while completely invisible (and the stupid but oddly amusing reactions of my victims when I did that).

    So, yeah, it really is true. Everything I enjoyed and remember about that game would have worked just as well with procedural content; the baked-in linear story, I think I finished, but I honestly couldn't tell you the plot. Something about demon gates = bad, and if you close them all you get to be emperor, or something. Who cares? It was a big world to explore and there was lots to accomplish.

    Yeah, I shouldn't overstate the case. A good story can be fun, once, and certainly some players do care about it. But others will just click through. They won't read text on the screen, and if you read it out loud to them, they may or may not listen (depending largely on whether there's any way to click past it).

    But if you're designing a play-once game, there is really no reason not to use hand-crafted content, and to put as much story and depth into it as you can afford. Players may or may not care, but I don't think it can hurt. (Unless you use one of these amazingly stupid designs where, after failing at some frustratingly difficult task, you send the player back to the last save point before the long-winded unskippable cut scene and make them watch it again for 30 seconds before they can again spend 10 seconds failing at the task. If you do that, you should be banned by law from getting within 10 miles of any game studio.)
     
    Kiwasi, Martin_H and GarBenjamin like this.
  22. Deleted User

    Deleted User

    Guest

    There are people who don't care for stories and that's fine, but you have to remember just because YOU don't like them. It doesn't mean many don't care..

    You should of heard the up-roar over Mass Effect 3's ending, a metric boat load of people must of been paying attention to complain about it in the first place.

    I'm a massive RPG fan and they don't half lay it thick in places (for no real reason), even I skip through the majority of it even though I find plots very interesting.

    I find it interesting that quite a few dev's eliminate factors because THEY believe it's the right thing to do without reaching out and questioning a player base first.

    P.S I think you're talking about Oblivion not Skyrim :D..
     
    AndrewGrayGames, frosted and Teila like this.
  23. JoeStrout

    JoeStrout

    Joined:
    Jan 14, 2011
    Posts:
    9,859
    You're absolutely right, but it goes both ways... just because you love a good story, doesn't mean most players do.

    So, at times like this, we can survey our market and try to build something they'll like... or we can build a game we would like. I've heard sensible arguments on both sides of that decision.

    And yeah, you're right again, I can never keep Oblivion and Skyrim apart in my head. It was indeed Oblivion I was thinking of.
     
  24. Martin_H

    Martin_H

    Joined:
    Jul 11, 2015
    Posts:
    4,436
    I didn't play Oblivion but I did play through Skyrim. I remember so little of its main quest, that I didn't really notice you were talking about an entirely different game. I thought you might be remembering some things wrong or mixing in DLC storylines maybe and the invisibility might have been a character build I hadn't tried. So I still think your argument mostly holds up for Skyrim too, with one exception. There was an aspect to Skyrim that I didn't really partake in, but based on how the internet reacted at the time I think many people were sharing stories and experiences and saying things like "You need to go there, you'll find something awesome, trust me". That's one kind of thing that doesn't work with procedural worlds unless several people share the same world-seed. I've contemplated how it could be a good thing to have the first (few) playthrough(s) of a procedural game be a predetermined seed, so that the dev can properly playtest for a "first impression" and be sure that no too hard scenario happens before the player can find his way around the game. Also it would give the possibility to better help noobs stuck on their first playthrough (assuming it's not a roguelike). And for anyone that doesn't like all this, it would be very easy to skip ahead to a random seed.

    I've tried to play through Oblivion, Witcher 1, Witcher 2 and Mass Effect 1, and all were too boring for me gameplay-wise. A good story alone usually doesn't keep me playing. I do appreciate good narrative and lore a great deal, when it is fused with a great game, like Deus Ex Human Revolution or Dishonored. I think I've read every piece of lore I found there, except the newspapers.
     
  25. tiggus

    tiggus

    Joined:
    Sep 2, 2010
    Posts:
    1,240
    I'm definitely a fan of building something I would like, but I'm not a pro either so that alleviates many of the real world constraints around game making for money. What Martin said about "go there and find X awesome thing" is the kind of stuff that I remember and like so it makes sense that I am gravitating towards non procedural environments.
     
    Martin_H likes this.
  26. Teila

    Teila

    Joined:
    Jan 13, 2013
    Posts:
    6,932
    That depends on your game and your targeted audience, which you should know before you start your game. :) My intended audience would care very much. Yours, maybe not.
     
  27. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    Can we at least admit that edge cases don't represent the majority? Neither extreme of people who must have a good story nor those who give absolutely no S***s about story are not most players even combined. The majority like a game for its strengths, with the biggest non-technical hurdle being its pacing. What players actually care about is whether there is something engrossing and whether it maintains their interest. No one should choose between gameplay or story since they both serve the same function.
     
    Kiwasi, frosted, JoeStrout and 2 others like this.
  28. frosted

    frosted

    Joined:
    Jan 17, 2014
    Posts:
    4,044
    For me, personally, as a gamer. I think there is good procedural generation and then there is bad proc gen.

    One of my pet peeves as a gamer is when I feel like I'm not really playing a game as much as trying to figure out what the level designer wants me to do. You see this most in strategy games where a map or scenario has some very specific kind of challenge or specific sets of obstacles or assets that you clearly need to make specific use of. But you also see this kind of thing in action games where the way that cover is distributed through a level means that you're really being led through a very specific path.

    Really good proc gen removes all of that. And the experience I'm having is my own. It's unique. I'm not trying to figure out what the level designer intended, I'm responding to the state or configuration of the system on the fly.

    JA2 is a great example, as is Mount and Blade and Crusader Kings - but these use AI to produce context and it's a very rare and unique kind of procedural generation. King's Bounty does proc gen in a very simple fashion by just randomizing which units you can hire, items you can buy and enemies you can face. The map, quests, etc are all the same each play through, but the stuff that really matters: the game play choices all vary significantly from one play through to the next.

    Other great examples are 4x games like Civ, since these are so dependent on the map conditions, each of my civ play throughs really feel unique. Each game feels like it's mine.

    Diablo's levels are not good proc gen. I probably really wouldn't care if Diablo just had 4 versions of each dungeon and picked one at random or if it was generated randomly. I don't think the randomized levels in diablo matter at all. In diablo, the randomization that matters is the equipment. When that crazy sword drops, I feel like it's unique, it's my sword. The entire reason diablo worked as a game is because that sword feels unique, that loot drop feels rare and special because it was random.

    XCOM:EU did not have proc gen maps, but it didn't matter. You could still play it 100 times because it had enough variation of each map to keep things moving along (most of the time). XCOM2 has proc gen maps and to be honest, I barely notice the difference (I have many hundreds of hours of XCOM, this is not a casual observation). FTL has excellent proc gen, the shops and the items are all different, so you constantly need to re-evaluate your choices and options between playthroughs.

    I think that often we talk about proc gen, we're almost always talking about the 'bad kind' of procedural generation, the diablo map kind of proc gen.
     
    JoeStrout likes this.
  29. Teravisor

    Teravisor

    Joined:
    Dec 29, 2014
    Posts:
    654
    Actually this is easily explainable: 90%+ of quests in TES series are side quests. If you rush main quests, you pass them within an hour or two while doing all side quests would take you several tens of hours.
    Yet only in main quests I've seen things that I actually remembered: hand-crafted castles, non-procedural large dungeons (side quests usually had quite small kill-through with little story) which really felt differently from 90%+ of other content...
     
  30. Billy4184

    Billy4184

    Joined:
    Jul 7, 2014
    Posts:
    6,008
    @JoeStrout I think it's a bit presumptuous to think that gamers don't care about story. A great story is something I savor and it can really make me feel as if I'm experiencing something meaningful and not just some button-mashing contest. All the great games I've played had great stories, and in fact that's probably the main reason why they were great. Homeworld, Age of Empires, these games that I played quite some time ago I remember because of that feeling of being something, of taking part in something, which is all due to the story.

    And a bad story can destroy a game. To be fair, it wasn't really bad, but I've always been a fan of Splinter Cell games and the lack of story substance in the recent games basically ruined that for me, Conviction was crap story-wise although I did replay missions for the slick gameplay. But it's not a memorable game for me. Also, I stopped playing Nexus The Jupiter Incident about half an hour in because the story and dialogue was annoying.

    Even games like CoD have enough of a story for me to enjoy them. Not a grand story, but a 'dogs of war' story that made you like the characters such as Price and Mactavish, made you want to hear more from them.

    The only game with a shocking lack of story that I enjoyed was Crysis 3, snce it is so sexy to look at that I didn't care. Even and so, toward the end it was getting old and repetitive so I tried to dig up the story a bit and that just made it worse, I limped to the finish line and got out of there.
     
  31. GarBenjamin

    GarBenjamin

    Joined:
    Dec 26, 2013
    Posts:
    7,441
    It's no more presumptuous than to think that gamers care about story. I think a lot of folks around here just see things in such blanket all-encompassing terms. And there is no real one answer. Some people love the story above else. Some love the graphics above all else. Others couldn't give a hoot about either and place gameplay above all else.

    It only takes about 5 minutes of googling to find a large number of people representing these different groups of gamers.

    So, in a way everyone is right. The story is very important to gamers. Gamers don't care about the story. Good graphics are very important to gamers. Gamers don't care about good graphics. Good music is very important to gamers. Gamers don't care about the music. And so on. All of these statements are true because for each of them a certain percentage (and very large number) of gamers will agree.

    What it really comes down to is folks here are saying that one aspect of a game is most important to them and another is the least important. Which is fine. Just focus on what is important to you and you will reach that segment of players who also place the most importance on that same thing. :)
     
    Kiwasi, frosted and JoeStrout like this.
  32. Billy4184

    Billy4184

    Joined:
    Jul 7, 2014
    Posts:
    6,008
    I wasn't trying to make a generalization, but I felt that @JoeStrout's comment was. I'm not saying story is important to everyone. It depends on the player, and the type of game. But to say that "players don't care about story" seems a little too much of a broad stroke to me.

    I think sometimes as developers we forget that in games, there are things that people say "I like this game because of this and that" and there are other things that are prerequisites, that the players ingest without being fully aware of it, and which would negatively affect the experience if they weren't there. Some people say for example that graphics is not important, but if you made the same game with the Standard Assets or similar it wouldn't be much of an experience. I think story - perhaps even generally speaking - falls into the same category, in that most players would feel a lack of a story even if they don't particularly care about one that you put there.

    But yes, players who actually, consciously care about a story are probably even a minority.
     
    Teravisor, JoeStrout and GarBenjamin like this.
  33. frosted

    frosted

    Joined:
    Jan 17, 2014
    Posts:
    4,044
    I think that 'story' also has a lot of meanings.

    Bethesda games are absolutely dripping in narrative, even if I can't remember the actual story in any of these games. It's funny, I actually don't remember a single dialog or cinematic or anything. I do remember searching through some bombed out shell of a building looking desperately for scraps of something useful in Fallout. I do remember the sense of epic scale and sense of adventure and wonder in Elder Scrolls games.

    Darkest Dungeon has a laughably thin 'narrative' but the atmosphere and mechanics tell an immensely vivid story. Mount and Blade has no pretense of story at all, but I've told stories about my adventures in this game.

    I am not a fan of strong narrative, of plot or dialog in video games. But a game without a world, and a game that doesn't completely place me in the context of that world is a game I have no interest playing.
     
    GarBenjamin and Teravisor like this.
  34. Billy4184

    Billy4184

    Joined:
    Jul 7, 2014
    Posts:
    6,008
    I think this is a crucial point. It's like those books on 'social intelligence' that tell you that a person doesn't remember what you said to them, they only remember how you made them feel. It's the same with games I think, people subconsciously drink up all of the grand story and cinematics and all that - not to mention great graphics and finely tuned controls - and it makes them feel really good about themselves, and then you ask them afterward about the game and all they can say is "it was really fun".

    I was a bit surprised to see how much flak Destiny took for having a lack of lore and story, since it is pretty much an Unreal Tournament shoot-em-up, but then I realized that all the players probably expect a grand story, they expect that feeling of grandeur and being part of galaxy-shaking events, even if all they really do in the game is point and shoot and probably couldn't really care less about where the orb came from and where it goes.

    That said, I'm someone for whom story is really important and I sometimes sit down and watch hours of playthrough of games I haven't even played simply because i want to see the story without having to bother with the grind.

    Anyway, so that we don't get too far off @tiggus question ... well I don't have too much experience with procedural generation but I think that, as I've said many times before, the shape and form of gameplay need to be handcrafted, but not the details. Nobody cares whether a wrinkle map is handpainted or generated, nobody cares if even a village is generated, but they will probably notice if their interaction with npc's is generated, or if the gameplay/mission is generated, if the level design is generated.

    In my opinion, once you have made a fully functioning game with block primitives, everything you do on top of that can be generated without too much of a negative impact, as long as it is generated to a good standard.

    I must say that I'm a little bit concerned about No Mans Sky because while they do procedural generation to a good standard, there is no handcrafted gameplay as far as I know and I think that either they will have to address this quickly or risk people rapidly losing interest. It isn't as if it's like Minecraft, where you can create whatever you want so handcrafted level design and gameplay is not necessary. You're really just a spectator to a procedurally generated universe with limited interaction and while that is all very nice, it won't be nearly enough to hold people's interest I think.
     
    Last edited: Apr 3, 2016
    JoeStrout likes this.
  35. Deleted User

    Deleted User

    Guest

    It's not presumptuous, it has been proven again and again with successful game releases that have amassed masses of fame and fortune (Final Fantasy series (well the early ones), The last of us, Elder Scrolls, Witcher, Baldurs gate, Mass effect, GTA, I could list SO many) .. But that’s not to say that games without a story line aren’t as if not more successful then story based games. Context is always a must, but some games don't even need much of that.

    What is presumptuous is telling people that all “gamers” don’t care about story lines, which is simply wrong. There’s a massive market for plot based games, so if that’s what you want to do go for it.

    Taking preferences out of the equation, figures don’t lie. Story's are expensive in games, if they didn't matter then AAA are making a major mistake.. But let's face it, they are "AAA" for a reason.! (Because they sold tons, if you didn't guess).

    The problem with story based games is it can be detrimental if not done right. Then again making poor strategy systems in an RTS can have the same impact..

    End of the day it comes down to one thing, gamers want good games. As a developer you have to somewhat pander to the genre you're aiming for, story ain't your thing.. Fine, don't make an RPG.!
     
    GarBenjamin likes this.
  36. frosted

    frosted

    Joined:
    Jan 17, 2014
    Posts:
    4,044
    I think @RockDyne really said it well:

    Players will play good games. Narrative, environment, mechanics, story, graphics, these are all tools that go into different games. If you manage to mix together a solid brew there are players who will play it. The more you can mix together, the more that you may appeal to a wider and wider audience.

    Really great games, especially top quality AAA, can crank each of these to 11. It can have huge environments, top quality mechanics, a great story all told through beautiful graphics. Small timers need to pick their battles and let some things slide in order to crank one of those nobs as far as you can go.

    But at the end of the day, players will play good games.
     
    GarBenjamin and Billy4184 like this.
  37. Steve-Tack

    Steve-Tack

    Joined:
    Mar 12, 2013
    Posts:
    1,240
    @Billy4184 I agree, No Man's Sky is going to be a real litmus test in terms of how far you can realistically take procedural generation.

    The guy who did Spelunky recently wrote a book about developing the game and gets into the philosophy of procedural generation. He likes games that are designed to be played all the way through in 30-60 minute chunks; he compares that experience to going to the arcade.

    The game I'm working on is a "play once" type of game. I've been tempted to look into something like Dungeon Architect to procedurally build underground areas on planets (it's mostly a spaceship game). I suspect I'd spend more time trying to build the rules than just draw out what I really want.

    I do disagree slightly with whoever said Diablo is a bad example of procedural generation. I've done some grinding on greater rifts, and it'd be that much more mind numbing if there were only a limited number of layouts.

    In terms of story content, I'm surprised at the variety of players out there. I did a pretty elaborate mod for the original Deus Ex long ago, with a number of narrative bits along with a linear story that slowly unfolds. The mission system makes it crystal clear what the current objectives are, so it's easy to bypass the story and just get on with it. I tend to see story elements as providing context and not much more. But many players were quite invested in every bit of narrative content in there and gobbled it all up. It's something I try to remember while building my current game. Even though *I* don't like tons of story details in my games, many do love that stuff.
     
  38. Billy4184

    Billy4184

    Joined:
    Jul 7, 2014
    Posts:
    6,008
    Yep, context is everything, which is what story provides. Every game needs the 'mentor' figure to tell you that you're the one, and the universe needs you, and all that jazz. Even if the end result is just point and shoot.
     
  39. frosted

    frosted

    Joined:
    Jan 17, 2014
    Posts:
    4,044
    I think the point is that no. Not every game needs any of this. Some games do, many do not. The games you like may need this, on the other hand, I tend to really avoid anything remotely of the sort.

    All games need context, but how a game delivers the context can change really dramatically from game to game.

    That's fine. It's even great. It means there's plenty of space for different games that can appeal to different players. The more niches out there, the more games the market can support.
     
  40. Deleted User

    Deleted User

    Guest

    So the conclusion as always is whatever, keep on working..
     
    JoeStrout and frosted like this.
  41. frosted

    frosted

    Joined:
    Jan 17, 2014
    Posts:
    4,044
    LOL!
     
  42. RockoDyne

    RockoDyne

    Joined:
    Apr 10, 2014
    Posts:
    2,234
    Yes. I think it's mostly in two camps of worldbuilding/backstory (immersion) and plot (engagement). Worldbuilding is hard to screw up if you actually put the building elements in the world and let the player explore to find it. Plot on the other hand is almost always FUBAR. The point of plot is to provide context and reason to what the player can, might, and should do, yet is usually used to tell the player just to do one thing.

    There is also the issue that people don't perceive plot if it isn't explicitly stated. The common perception is that story is the thing that interrupts the gameplay, so people don't notice story beats when the player still has control.

    Rule 34 bro. The internet is such a massive echo chamber that not only can you find people that share your fetish, but that someone was also passionate enough that they have already made porn about it. So from the groups you mentioned, which would be allegorical to foot fetishists (where anyone outside that fetish absolutely does not understand)?
     
  43. Billy4184

    Billy4184

    Joined:
    Jul 7, 2014
    Posts:
    6,008
    You're reading too much into what I said. I'm not saying it needs to be explicitly a white-bearded gentleman who presents you with the Sword of The Chosen One or any of that, I was just making a particularly colorful point. I am saying that IMO every game needs the player's actions to have context, to have meaning in some way. And players probably don't even consciously notice it very much unless it is not there.
     
  44. frosted

    frosted

    Joined:
    Jan 17, 2014
    Posts:
    4,044
    What would you call a game like Civilization and do you think it'd fall into either of those camps?
     
  45. GarBenjamin

    GarBenjamin

    Joined:
    Dec 26, 2013
    Posts:
    7,441
    Yeah that is the whole point. There are so many people in the world and a lot of them are online. You can find large numbers of people out there who will back up any preference a person may have. I've read through countless discussions online of people posting about not caring about story, not caring about graphics and also read many discussions on completely the opposite. They won't play any game that doesn't look superb. They won't play any game that doesn't have a strong story.

    Basically, I have read so many different (and often very strong) stances on these things the only conclusion I think is reasonable is all of these points of view are valid. Obviously not for everyone but for gamers as a whole. There is a group who love the 2D pixel art games and prefer them over the latest AAA 3D games. There is a group who love stories and those who love stories to unfold via cut scenes. There is a group that don't want stories and skip every cut scene as soon as it starts.

    It just depends on what a person believes and what they look for. If a person sets out to prove stories in games are important they can search and find evidence to back that up. If a person sets out to prove stories in games are not important they can also search and find evidence to back that up. Same for graphics style and quality. Music. Heck even game play. A person could most certainly find evidence that gamers only want to play FPS games or RPGs.
     
    syscrusher and frosted like this.
  46. Deleted User

    Deleted User

    Guest

    @GarBenjamin

    Only thing that matters to me are sales figures, if certain types of games (in a specific genre) do well it's obvious there's a user base for it.
     
  47. frosted

    frosted

    Joined:
    Jan 17, 2014
    Posts:
    4,044
    Last edited: Apr 3, 2016
  48. GarBenjamin

    GarBenjamin

    Joined:
    Dec 26, 2013
    Posts:
    7,441
    Yeah I am not even sure what is being compared here really. I'm not saying story doesn't matter. For me playing a game I tend to think of it as producing a story through my gaming experience. If the devs want to provide some framework to help mold that story that is fine. But not necessary for me and I usually click past all of the cinemas as soon as they begin.

    I think it is very easy to look out there and say gamers must want huge 3D RPGs or FPSs with top notch graphics, "big" stories, epic music and cut scenes because after all that is what sells in the millions and is what the AAA are making. I think it goes hand-in-hand. As good as Shovel Knight and Undertale sold I would expect had they been produced by a big AAA company we'd see them selling many times the number of copies they have.

    There are always exceptions of course. Even when games are fantastically popular I don't think you can rely on that as proof of what gamers want. I know that sounds absurd but is all the gamers want really just Flappy Bird? Or Minecraft? Spelunky? Castle Crashers? Terraria? Candy Crush? Or maybe good ole Farmville is the answer.

    I agree there is value of course in looking at such things but I still believe there is no real answer. People want what they want and it varies from person to person. And success of the game depends on many different things. Having a huge amount of money and contacts available is a big factor in the AAA game companies sales numbers.
     
  49. Deleted User

    Deleted User

    Guest

    That list really doesn't sound right..

    https://en.wikipedia.org/wiki/List_of_best-selling_video_games#All_platforms

    It does make me often wonder if it's worth going to all the trouble making an RPG when Diablo 3 is one of the best selling games of all time..
     
  50. Billy4184

    Billy4184

    Joined:
    Jul 7, 2014
    Posts:
    6,008
    Yeah it varies from genre to genre. Some games have hundreds of pages of lore and backstory and people go and collect them and it's really important to them. In some games it doesn't matter. Same with controls, in a flight sim, you can need a hundred page manual to be able to fly the plane, in other games it's point and shoot. Games are different, but there are genres with expectations from the players who play that sort of game. As a dev you need to know who you're making the game for and not make blanket assumptions based on one sort of game.

    IMO the only thing that games have in common, in terms of story, is that there has to be some kind of context, which is able to make each of the player's actions feel meaningful. If you can cast the player character immediately into a known role, such as 'underdog' or 'world-saving hero' you don't need to do too much in terms of story. But you can't just go and make a random game about wandering around with no identity and no destiny. RPGs for example are often all about the psychological feeling of becoming powerful from a weak starting point. You need something for the player to relate to so their actions in the game feel meaningful.

    Also, don't forget, you're making a game for a player and not a developer. It might be fun for you to tinker around with mechanics but players may be looking for more of a personal experience. Again, you have to accurately gauge what players of your genre want.
     
    Last edited: Apr 3, 2016
Thread Status:
Not open for further replies.