Let me tell you the difference between the classical approach to creative tasks and the generative approach. Classic - you have a task, you think and give a solution, that is, you take over most of the process. Generative approach - I give the task to a computer and it generates many solutions. Modern generative algorithms know how to limit themselves, evaluate their own performance, weed out the bad options and work with the good ones. Generative approach in the 90's and 00's was unknown to anyone, now with the boom on AI it is a trendy topic. It is the second approach that we use and will continue to use.
Combinatorial - means that there is a combination at the generated content. What does content production look like in games? First I made digging, then building, geology, bomb, architecture. That's how content grows as we write aspect after aspect. It's a handmade process
When you're adding new features to the game, that's linear content growth, we take each aspect and put variety on it. For example: Different levels were made for digging, different types of blocks, different bombs, different architecture for building. So the content grows in breadth, but it's not explosive growth.
There is a polynomial growth of content, when I do some things that in combination with each other will give a large number of variations. For example, we're developing the diggeroid parts system right now, there are batteries that are responsible for how much energy the robot carries. There's a default small battery that has a little bit of energy in it. We can make two parameters. The first one is the size of the battery - medium, large, very large. Depending on the size, you get a different game. Large batteries weigh a lot, so they have a lot of energy, but they have a lot of mass, which makes the robot go slower, so you need to either put a powerful motor, which will consume more energy, or settle for a slow speed, it's a matter of balance. The second parameter: what technology is used to make the battery. There are ordinary ones, there are those that last a long time, never wear out and there are batteries that carry a lot of charge, but they heat up the hull and our depth limit is reduced, so they need a cooler. By multiplying these two axes, I get a lot of content. There can be several such axes, how many variants I accumulate on each, that's the variety I'll have. Each axis forms variants, when multiplied, they form a lot and so we can generate a lot of content.
Let's move on to the concept of self-regulating. The main idea is that our game developed according to the principle of evolution, there was no gameplay design and balance, because in Mines the balance finds itself, and players adjust to any circumstances that are there, the main thing is to create elements that balance everything, for example an open market. It is imperative to add feedback to the system that will regulate elements of the game so that there are no imba's.
So we've covered all three points, that's the idea we had in 2019, but modern technology tells me that we might be able to make ERZ exactly the way we wanted.
Combinatorial - means that there is a combination at the generated content. What does content production look like in games? First I made digging, then building, geology, bomb, architecture. That's how content grows as we write aspect after aspect. It's a handmade process
When you're adding new features to the game, that's linear content growth, we take each aspect and put variety on it. For example: Different levels were made for digging, different types of blocks, different bombs, different architecture for building. So the content grows in breadth, but it's not explosive growth.
There is a polynomial growth of content, when I do some things that in combination with each other will give a large number of variations. For example, we're developing the diggeroid parts system right now, there are batteries that are responsible for how much energy the robot carries. There's a default small battery that has a little bit of energy in it. We can make two parameters. The first one is the size of the battery - medium, large, very large. Depending on the size, you get a different game. Large batteries weigh a lot, so they have a lot of energy, but they have a lot of mass, which makes the robot go slower, so you need to either put a powerful motor, which will consume more energy, or settle for a slow speed, it's a matter of balance. The second parameter: what technology is used to make the battery. There are ordinary ones, there are those that last a long time, never wear out and there are batteries that carry a lot of charge, but they heat up the hull and our depth limit is reduced, so they need a cooler. By multiplying these two axes, I get a lot of content. There can be several such axes, how many variants I accumulate on each, that's the variety I'll have. Each axis forms variants, when multiplied, they form a lot and so we can generate a lot of content.
Let's move on to the concept of self-regulating. The main idea is that our game developed according to the principle of evolution, there was no gameplay design and balance, because in Mines the balance finds itself, and players adjust to any circumstances that are there, the main thing is to create elements that balance everything, for example an open market. It is imperative to add feedback to the system that will regulate elements of the game so that there are no imba's.
So we've covered all three points, that's the idea we had in 2019, but modern technology tells me that we might be able to make ERZ exactly the way we wanted.
