Reading the article, it seems that his real point is not that Procedural Content Is The Future, but rather that not every game needs level design. Some games get along with no “levels” at all, just interesting mechanics and different starting conditions.
Which is certainly true: think of every card game ever conceived. On the other hand, it’s not exactly a goddamn revelation.
Man, it’s not that crazy. Randomly lifted dialogues and poetry would end up being better than programmer written shit.
This is the design philosophy of Diablo 2 and 3.
Yes, “I put on my wizard hat” type exchanges between NPCs would be epic… ON WEED.
Not sure I follow what you mean here because I can’t imagine a randomly generated level holding a candle to a proper hand crafted level. Good levels need to start from a solid base - a shell, as we call it - which is built upon once it’s decided that the level is fun to play. Building on top of a random set up would be weird for a lot of reasons and very hard to make work.
For something where the gameplay was organic like, say, Mirror’s Edge Speed Trials it could probably work but for anything story based I don’t see it.
Looking forward to seeing what you’re working on at home though, if it ever comes to light!
Ultimately, the goal is hand crafted levels with soul. Those are the best.
For example, Minecraft’s outstanding world generator.
Masturbatory high level formulaic design with no contact with why people buy games.
Replace “procedural” with “systemic” and maybe you have something.
Minecraft is good for how the player uses the system, not because the generated world is of any use.
To an extent, yes, but that’s assuming there is a best place for crates, which isn’t necessarily true. Plus, no one is going to even attempt a 100% procedural AAA game so it’s a BS argument. Also it’s not that hard or risky, given how many games out there do exactly that.
As for saving money, it doesn’t really, unless you’re talking very small projects like roguelikes. For AAA titles the primary expense for content is not level designers but asset creation – meshes, textures, animations, etc. (Clever level designers actually save you money by figuring out how to re-use existing expensive assets in ways that never occurred to you before.) Procedural content would replace cheap level designers with programmers, who are more expensive. And it’s increasing risk, since it’s easy to calculate how long it will take to hand-place crates, but very difficult to estimate how long it will take to come up with a functioning General Theory of Crate Placement.
See, this is where you are wrong, because it does cut down costs. Yes, the assets take the majority of the grunt work. But LD stuff is far less cut and dried. LDs, at the best of times, don’t just pump out a level and it’s done – it involves a lot of hand tweaking and manual adjustment. But that only happens after they build it to begin with – which is probably a solid half of their time, significantly more in a game where they have to populate the environments and build systems around it. Which is funny, because ultimately the majority aren’t even that good at it. Starting from a ‘finished’ perspective allows them to focus on what’s truly important about level design, and thus, they can get to polishing their content much quicker, and they can produce significantly more of it.
I love the idea of procedurally generated content, but the results outside of strategy games so far have been either horribly unbalanced/unpolished (most roguelikes) or so innocuous as to be pointless (Diablo or Torchlight, which have random levels but the randomization doesn’t actually do anything interesting.) Procedural content has its place, either as a starting point (landscape generation) or as a diversion (e.g. a random mission goal to add spice) but I can’t see it replacing hand-crafted content until it becomes substantially better.
Read some of my previous posts to see how it can make things better without resulting in the problems you identify.
Programmers don’t write dialogue in any modern game.
That’s because you belong to the old school of level designers, who are still convinced that the exact path of the player needs to be polished to a sheen, rather than building a game where it’s not the path that’s important, but the interaction with the environment.
The very best level designers look at any given space as a blank slate to be played with, rather than looking for the right space. If your gameplay is so dependent on the placement of a hallway, or a room, or a desk, then chances are you are doing it wrong.
I get the feeling a lot of people in this thread aren’t quite grasping the nuances of what Charles is proposing here.
The difference between procedural and systemic is that procedural is how you develop content (on the fly), while systemic is how the player interacts with it (non linear).
Which is why I say procedural is the silly belief for a magic formula that develops content for no money or time, while systemic is a pragmatic approach to improve games.
What does this have to do with me? I’ve never said it’s a magic formula or that it magically developers content for no money or time.
I think everyone should reread post #8 in this thread.
I disagree that one is ‘better’ than the other.
procedural (or ‘randomness’) gives replayability
handcrafted (scripted or predesigned elements) gives player experience
The game designer needs to decide what approach will lead to the best gameplay depending on how replayable they want the game to be.
games needing high replayability need more procedural content. e.g. civilization - the ‘procedural’ part is the map generator and AI, which gives replayability. The ‘handcrafted’ part are the rules around it, the units and the technology tree. Imagine playing on the same map with the same starting position (removing randomness)? The game could only be played once or twice before a player gets bored. Imagine if the game randomised the technology tree every game, so that you could research combustion on turn 1? Your game would be a laughing stock. More realistically - to randomise the tech tree means dumbing it down a lot to something like “engineering level 1-10” etc - increasing replayability but at the cost of player experience (the original MoO actually had an original compromise which enabled some randomisation of the tech tree with minimal cost of player experience).
Games indended for 1 run through can use much more handcrafted content for a great once-off player experience (insert random bioware title here (no pun intended)). The only reason to put in procedural/random content for these games is for replayable parts (e.g. if your RPG had a lot of random encounters you’d want a decent battle generator).
One further point is using procedural content to create realistic environments. We’ve all played that FPS where you can’t run 20 feet either side of you even though you’re supposed to be in a jungle. So you can 1) ignore, 2) spend more time hand-crafting a bigger world or 3) making/populating a bigger world with algorithms (Daggerfall/Operation Flashpoint(?)). Whether its worth the time and effort I dont know, I guess it depends how immersed you want your customers to feel.
And here I thought that was the current school of level design.
Or does this metric have an “Old Old School” designation?
Charles’ implication is that people still thinking in terms of the current school of level design are already out of date and don’t even know it yet.
Yep. When the shift happens, they’ll be left behind pretty hard. I’m already seeing a big shift, where LDs who understand that the player controls the situation and not the developer immediately make much better levels and situations than LDs who don’t.
Well hopefully this shift will occur before Thief 4 comes out.
Only most randomly generated stuff sucks ass and feels cheap. You don’t save money if you make a game and everyone hates it never buys it.
Considering most procedurally generated levels in games aren’t that amazing or are incredibly simple, I don’t see it magically becoming something great anytime soon. Hell the cost of creating the system to make those levels remotely well has to be on par with just making a bunch by hand.
Maybe you should read post #8 a third time.
I guess I should go see a psychatrist, as I appearently can re-read post #8 as often as I want to, but everytime I do, my mind shuffles the words “Hellgate London” someplace in the post, though I DO know these words aren’t actually spelled out.
rezaf
I think one of the things that came across really strong in the essay I linked in the first post was that… there’s an important difference in the design process between handcrafted and procedural games. So, it isn’t something that will necessarily come across to the player or that will even appear in every game, it’s just that it will be a trend in games. The difference being, in a game where you are hand-making each level or environment or event or whatever, you have to lock down your game mechanics quite early in the process. If it’s a well-established genre, or you have a really clear vision of how you want your mechanics to work, or if the mechanics are somewhat incidental to the point of the game, that’s fine. But if you’re trying to break new ground or experiment or do something that is based on an idea that might or might not turn out to be workable… this is a problem, because you keep having fresh insights all the way through the development process, but if you locked down the mechanics back near the beginning, you either have to just discard those insights, or you have to risk breaking all kinds of stuff in the game as it exists so far in order to accommodate the new mechanics (or the refinements to the existing mechanics) and as a rule, on a project of any size, what’s going to happen is probably the former.
Which means that the tendency (I’m sure there will be exceptions to the rule) will be for games that are linear or story-centric or handcrafted or what-have-you to be built around a core of very familiar gameplay “tropes”, while games that have procedurally or systemically generated “content” will be more free to explore new and different types of gameplay, or at least more experimental variations on existing gameplay.
I agree that this doesn’t automatically make all procedural games better than all handcrafted games. And indeed, there are some very strong examples of innovative games that have handcrafted levels - Brain immediately springs to mind. But as a description of what happens in the development process, I thought this point - about the point in time when the door closes on new gameplay ideas - was pretty compelling.