Repetition versus raw uniqueness

If you look at most games, you'll find that they re-use the same assets over and over again. It's an efficiency of development and an attempt to make the most out of time spent creating assets. So you'll see pieces of the environment repeated thousands of times, each enemy appears multiple times but is really the same animation and the same behavior AI and so on. Every bullet re-uses the same sprite. Every explosion pulls from the same animation. etc. It's really become an exercise in 'compression' of the amount of time and effort it takes to build something. We're in the habit of building 'game generators' based on a handful of unique parts re-purposed and re-used over and over again.

This I think has led to a very deeply ingrained style of game - which spans all genres - where everything has become predictable and repetitive. At least, in most types of games. One exception that comes to mind is perhaps graphic adventure games where every scene is unique and characters tend to appear only once. But you think of like 2d games - platformers, shootemups, etc... massive re-use of the same imagery and enemies. Or also 3d games with the same textures being re-used on wall segments and the same enemies showing up hundreds of times in different places. While this can still often provide a lot of variation and excitement, it's still quite far removed from a raw highly unique game experience. Obviously budget has something to do with this because games with higher budgets seem to have more custom content and less obvious re-use.

I also feel that it contributes to one of the major problems that game developers have, which is how to make your game unique. When I pick up yet another game which has the same basic attack patterns, the same basic enemy movements, the same basic re-use of the same enemy sprite 1000 times over moving in the same way as all the others and looking exactly identical, you have to admit that adds up to some boredom. If you keep seeing the same thing over and over again you get bored and your brain gets used to what it's seeing and that makes it less engaged and less aware and less stimulated. But we all know the power of seeing something 'new and different', the brain perks up and gets excited and there is a greater tension and interest.

Obviously there is a problem here of the sheer amount of time and effort it might take to make a game which doesn't re-use anything more than once. Games that do this tend to be very artsy games where the whole thing is custom graphics and there's no real sign of repetition. That takes a long time to develop. But it's also engrossing. A lot of this I think goes back to the 'tradition' of trying to get 'more game' out of 'less resources', such as when computers/consoles had very limited memory or resources and so you really had to find ways to re-use the same stuff in different ways in order to make 'larger' game experiences. This trend has continued on and on even though memory and resources have massively increased, and everyone is still trying to produce a game which milks the resources for every ounce we can get out of them... for reasons of development/cost which has nothing to do with what the user wants.

We seem to be worried about people having seen and done the same game and the same style of repetitive use of assets over and over again, which numbs the audience and stops them from paying attention for very long.... then why not 'fix' this by spending more time making truly unique, 'un-compressed' games? Games where you spend that extra time making several different versions of the same enemy, or where each one looks a little bit different, or has its own unique character? Instead of like, here comes another wave of 50 of the same sprite moving in exactly the same pattern as each other. It seems it is almost taboo or unheard of to tackle a game in this way since it's generally not what developers do - there is such a big trend toward repetition, re-use, re-appearance of the same stuff over and over again, just because it is 'easier' to develop this way.

I guess I'd like some feedback on this idea because I'm considering trying to make my game highly unique in this way, not using the same sprite or the same exact AI or behavior/movement for any enemy, so that everything has a unique and unpredictable quality. What do you think, pros & cons?

 

I see what you're saying although I think the root cause has nothing to do with trying to reuse assets. People might think it is and even say that is what they are doing but that is a symptom.

The reason is because people keep focusing on the graphics. Every time hardware gets more power and more space instead of taking advantage of it to make much more content that is unique they instead bump up the quality level and make the same amount in much more time or even less content in same time.

This higher def content requires more space to store the assets, more power to render the assets and most important.... more time to create the assets to begin with.

This is one of the reasons I am much more interested in creating low res games. Because I can make much more content in the same amount of time it would take to make one or maybe a few HD assets. Working my way toward that goal slowly and steadily.

Bottom line is if people stopped pouring all of their time into making every asset the greatest HD possible they could spend that same time making several unique assets with less detail.

But that has never happened. New hardware more power for decades has always been the same ole thing... "hey now we can make higher detailed assets and fill up all of our time making the same or even less amount as well as maxxing out this new hardware and really making it chug!!"

 

I wouldn't say it's never happened. Minecraft stubbornly ships with 8x8 pixel textures, still to this day.

And that brings up what I was going to suggest to @imaginaryhuman's post: what about procedural generation? This can provide a lot of variety without needing a gazillion hand-coded assets. Yes, the textures are reused, but in such a game the textures aren't important — what's important is the layout. And that's unique.

I still play Minecraft on occasion, and for me the most fun part is at the beginning of the game, when I'm exploring a section of a new world. It's always fresh and different. Soon I dig in and start fortifying the village or whatever, and soon thereafter I start to get bored, because it's not new anymore. (Yes, I've tried wandering off... but I'm trying to keep this reply from doing that right now.)

Of course procedural generation eventually wears thin too; we may not have seen this specific pattern, but we've got a decent mental model of the sorts of patterns the code can generate, and they no longer hold much surprise. But that has to do with how deep the procedural generation is. You could put more time into that, instead of more time into unique sprites, and I think get more bang for your buck.

 

Right "never" say never. But you see what I mean I am sure. I don't know if it has ever happened with a AAA game. Possibly but very unusal if it has. And these days a lot of Indies seem to have the same kind of mindset pushing for max graphics quality instead of max content.

Mindcraft was as it should be IMO. He used lowdef graphics so he could actually build his game. Some Indies focus on low res 2D and 3D for same reasons but in general in game dev people are always trying to push the graphics quality up hoping to stand out I guess.

 

I do see how this ties into procedural content and how perhaps procedural methods could be used to 'vary' or 'remix' content to make it appear to have more variety.

 

The funny thing is that gameplay absolutely wants repetition. So much about system mastery, and ultimately learning, is pattern recognition and experimentation on top of that pattern recognition. Players generally want patterns and perceivable consequences, where their input has largely consistent and clearly defined rules. Just imagine if hitting the X button was just as likely to cause the player to explode into a million giblets as it might reward them.

There is plenty of room for grey here. Modular constructions in the hands of an idiot can lead to forty miles of beige corridors, but in competent hands could recreate all of Paris. At the same time it's not that unrealistic for things to look identical. Walk down your block and count how many houses are the same, or see if you can spot any discernible differences in the lampposts.

It's way more complicated than just saying reusing assets can/will lead to boredom. Lego's wouldn't work otherwise.

 

Yes good point and I was waiting for someone to talk about this... whether/how repetition is actually needed in the game, whether familiarity is needed, how much etc?

Something that also came to mind are the 'mega texture' technologies which allow a game to basically unlimited texturing rather than repeat. .... is that not a move toward this and considered an improvement?

 

Not entirely sure what you're talking about with mega textures since the first thing that comes to mind is idTech.


Think about how well the scientific method would work if you don't have a control group, or without running multiple parallel tests. You can't discern the signal from the noise, or worse yet, make false assumptions. You end up going to great lengths to inform the player about what they are making decisions on (they'll probably ignore it), rather than letting them figure it out themselves.

Just from an artistic perspective, think about leitmotifs. There's the very deliberate repetition of elements for the purpose of applying meaning.

 

Sometimes I wonder what games people are playing. Outside of arcade games, I really don't see that much repetition at all.

Take your average FPS game. Sure there is plenty of asset reuse. On average there are only a dozen or so weapon types and about the same number of enemy types. But the game achieves variation through placing the same enemies in different contexts. That way a player can start to develop a feel for the game rules, but still gets new experiences.

A game where the rules are constantly new at each encounter sounds horrible. How is one supposed to learn to play the game of the rules keep changing? Might as well simply go outside and experience real life.

 

Well obviously the game must have SOMETHING that is consistent throughout otherwise you're not going to be able to accomplish anything. There has to be some dynamics, structure, progress, familiarity to some extent etc. I'm not saying to just throw totally random objects at the screen. But I am pointing out just how ingrained this model of development is that everything considers it 'normal' without questioning it. You could have consistent rules and stuff without having, like you said, only a handful of actually unique enemies. There is a lot of gameplay in encountering the same thing in many different places especially if that thing is quite expressive or varied, but I would like to push it even further into the care and attention to detail department.

 

Have you played any of the Fluxx card games? The whole pretext of them is that the rules change constantly. They are quite popular & have spawned a lot of reskins for different cult themes e.g. Firefly, batman

 

Never liked them much. But a lot of people do. So I guess there is a case for a game with random rules.

 

I feel like the mould has been broken a lot of times, but as soon as it's broken the next 10 big games are all the same gameplay from start to finish, and the masses eat it up. So whether right or wrong, it doesn't seem like it's going to change anytime soon.

 

You do have a point. The mass of humanity loves to basically 'flock together' and follow each other and not step out of the 'herd' in fear of danger etc... just like animals,.... so most of the time they just act unconsciously and without very much awareness or personal decision making. "Sheeple" I guess... for every unique new game or genre there are hundreds of copiers who pretty much stay within the formula and add and remove bits and pieces or change the visuals. The challenge is to be the disrupter and be the first and be the one who gets all the early attention for that new idea... which may be short lived. One thing you can do though is to create something that is 'hard to do' that's original making it much more difficult for it to be copied.

 

Repetition and reuse isn't a bad thing. Its arrangement that is the real issue. As players we want to achieve different emotional outcomes in the games we play and repetition can either help accomplish or kill that objective inside us. Take a puzzle game, you will find huge repetition here but the best puzzle games use repetition to their advantage to change minor things each time so that you are constantly trying to re-evaluate to solve the puzzle. Mastering these escalating challenges is the draw.

MMOs, in my opinion, are where repetition is used terribly. They need to fill the game world with stuff but the game world is so large they need a way to fill it in a time effective way. On top of this MMOs need to create a time sink for the players so the use of repetition is a cheap way to get the player to spin their wheels. The by product of this is that when something is achieved, it took a long time and there is a meaningful feeling of work/reward that some people actually strive for.

Like anything, moderation is key. Too much repetition and player's fun changes to just work. Too little repetition and the player will feel unsatisfied. When we find good gameplay loops in games, we want the chance to enjoy them more than once. Once a gameplay loop overstays its welcome, that's when its a problem.

In summary, reusing assets isn't bad. Reusing game mechanics and level designs isn't bad. What is bad is if the reuse isn't purposeful done in order to advance what the game is trying to achieve in terms of user satisfaction.