November 25, 2016 at 17:13As we came up with and made our first Android game. Part 2: Levels
Hello World! We are two newly minted developers of mobile games, former classmates, graduates of Kazan Federal University, Aydar and Ed, continue the story about the development of our first game "Syncomania" on Android.
For those who have just joined, briefly recall the contents of the first part :
In this article, we will talk about working on game levels and some principles that we adhered to. Perhaps they will be useful to you.
As Terry Pratchett wrote: "The vital ingredient in success is not to know that what you have planned is impossible to fulfill." Probably, thanks to our naive and romantic approach, we achieved the result.
The story of the first level: “Everyone, even the smallest, should have a personal living space, so each hero got a separate room. And one of the heroes was lucky - the only way out is right in his room. At first, the lucky one behaves selfishly and seeks to leave the level as soon as possible. But with experience, he realizes that such behavior is not the most effective. For optimal passage, you need to work in a team and first think about other heroes . ”
Variability of passage is a feature of the game, which we are sincerely proud of. All our levels have alternative solutions: short or long, obvious or elegant. Each has its own world record - the minimum number of moves known at the moment. We encourage the search for the shortest solution with the help of generally accepted stars, as, for example, in Angry Birds. But, in our opinion, the opaque method of evaluation causes bewilderment and irritation. In “Syncomania” we tried to make it visual: the player gets three stars if he passes the level in the minimum number of moves. With one star, too, everything is obvious: just save all the heroes, not paying attention to the records. But for what merits give two stars? We needed a threshold value. In order not to set it for each level separately, we introduced a simple formula:
threshold = 1.5 * minimum number of moves.
Thus, if the world record is 20 moves, then for 29 the player receives two stars, and for 30 - only one. Subsequent tests showed that many people sometimes lack a couple of moves to the perfect passage. We decided that they need to be motivated, and introduced another gradation: if a player passed the level by 1–4 moves worse than the world record, then he gets 2½ stars.
Since the number of levels is limited (currently 96), we have slowed down the rate of consumption of content using the completion mechanics. A new chapter, consisting of 12 levels, opens after the player collects a given number of stars. In this way, we increase replay value by handing extra hours of enjoyment (and shamelessly hoping for in-game purchases and advertising clicks to appear).
Variability does not make sense if a three-star solution can be found at a glance. Therefore, we try to come up with bait for players: one or more walks seem obvious, but lead to ineffective or losing conditions. For example, the “Angle” level is visually very simple, but almost no one passes it the first time. Most seek the path of least resistance - immediately to the exit in the upper right corner. Eating our bait, the player gets into a stalemate situation:
While testing the initial levels, we noticed that the symmetrical arrangement of objects looks harmonious, and randomly scattered traps and walls create visual trash. Then we decided to use symmetry everywhere. But, as you know, there are no rules without exceptions. For example, at the Candle level, we deliberately deviated from symmetry and set up a lonely wall, hinting that it should be used during the passage ( although this may be bait ).
At an early stage of development, we tried to find out how small changes in the arrangement of objects affect the method of passage. For example, we have about 10 modifications of the Four Rooms design, and different tricks are used in each of them. At the Compression level, the player learns to bring heroes together to pass between traps.
In "Simulators" will have to adapt to the behavior of enemies.
And at the “Castle” level, you need to help the locked hero get out of the room.
We apply the principle of similarity at levels with new mechanics, so that it is easier for the player to navigate in a familiar environment.
At first, we gave the levels bizarre names, such as Besame Mucho or Imitation of Crab Meat. But, when tests of the mobile version began, it turned out that long names did not have enough space on the screen. It was necessary to change the words or even abandon them. We decided that levels would be better remembered if they had a “talking” name. For example, the level at which traps look like a monkey's face, we called "Macaque."
By the way, the working title of our game was “One 4 all” - “one for all” (with a hint of four), and we came up with the current “Syncomania” (attraction to synchronization) before launch.
It seems to us that people better absorb material if they come to a decision on their own. We do not tell the player how to interact with the new mechanics, but encourage him to find his tricks with simple examples. In particular, at the “Section” level, the player is confronted with a seemingly hopeless situation. More precisely, there is a way out, but it is surrounded by traps: if you bring out one hero, you kill another. The game makes you think: “The developers must have messed up something - this level is impassable. But before that, everything was logical. ” And then the player draws attention to the walls placed, independently discovering one of the main tricks: changing the distance between the characters. Joy is in discovery.
Ed: “While we were writing this article, my wife passed this same level in 24 moves and broke a record that lasted more than a year! During development, I also liked to set records at new levels (after all, the first passage by default was the best) . "
In the last article, we said that we do not have a stable automated way to detect the shortest passage. Therefore, we set most world records ourselves or with the help of our first testers. Usually, having scored three stars, the players calm down and move on to the next level, but some show unconventional thinking and open up unexpected decisions that we have. All new records with the passage are sent to us through Google Analytics, and we change them with the next update (and the champions are motivated, and the game is developing). So, our most loyal player Andrei Vladimirovich has already set a dozen records. In addition, we decided that the champions would be pleased if the levels at which they set world records were marked with a special golden color in the menu. But if someone else improves their record,
Having developed most of the levels, we were faced with the problem of ordering them according to increasing complexity. This is a non-trivial task, if only because the search for a solution is subjective for different players. In addition, the same level is easy to complete in the forehead, but damn difficult to pass optimally. As a result, we decided to streamline them in search of an obvious solution, so that the players did not stumble on the first pass, but if they wanted, they could plunge into optimization at the initial stage. To introduce people to new mechanics gradually, we divided the levels into thematic chapters: Traps, Streams, Enemies, etc. At the beginning of each chapter, we place the levels easier, and by the end we check the player for strength. So we formed a curve of complexity with fluctuations.
At the stage of organizing, we threw out a lot of boring levels, but testing showed that we also need to add something. For example, we added the first chapter last, on the advice of our testers. It turned out that the newcomers did not have time to master the tricks of the game, and they needed more simple levels for swinging. It became clear that we were too immersed in the game, and our gaming experience cannot be used to assess the complexity. Then we tried to imagine ourselves in the place of the newcomers who first launched our game, and created concise introductory levels using only the walls.
Aidar: “The work on the first chapter brought me great pleasure, since severe restrictions opened up scope for creativity. But still this is an introduction, and the most interesting part was to develop the latest chapters for experienced players . ”
Creating levels, we did not think through new mechanics and puzzles in advance. During programming and testing, they showed themselves. For example, once for the sake of testing, we put a box on a conveyor belt from threads. According to the laws of the game world, the box moved along the course and, stepping off the conveyor, clogged a narrow corridor for us. “Hmm, what a nuisance - it must be used,” we thought. So in our turn-based game, a kind of timer appeared: you need to have time to solve the riddle in N moves, otherwise the output will be walled up with boxes.
Experimenting with two timers, we directed their flows to a free cell. Boxes reached the end and mutually destroyed in a collision. This came as a surprise to us, since in other cases the game mechanics did not allow pushing the boxes together. After a little consultation, we decided that this was not a bug, but a feature . As a result, the Skull level was created where you can destroy boxes in this way.
Having tried numerous types of interactions, we came to the idea that for many centuries served as the engine of science and progress: “Why don't we mix all this and see what happens?” As a result, we got the most confused at the first (and second) glance level - “Whirlpool”. However, despite the frightening appearance, this is by no means the most difficult puzzle in the game. Given the small number of steps that do not lead to loss, this level can be completed by a simple bust. Even our automated solver handled it - in just 16 moves.
However, in most cases, it will not be possible to find a solution by exhaustive search - with each move, the number of states grows exponentially. The game should use a creative approach, and it will have to be fully demonstrated at the most difficult level, according to testers, “Prisoners”.
Last level story: “Terrorists are holding hostages in two small cells. The heroes have no choice but to move up: a step down - and the hostages will be killed. In addition, insidious villains (yes, this is us) mined the entire upper limit of the level. It would seem that the player faces a dilemma: to sacrifice hostages or destroy heroes. But even if by some miracle (spoiler: boxes) the player manages to eliminate the enemies, he will notice that the room with the exit is surrounded by streams directed outward. And this part is the insidious highlight of the level . ”
To be continued ...
In the next part we will talk about the development of the game interface.
For those who have just joined, briefly recall the contents of the first part :
- we came up with a turn-based puzzle about four heroes ( depicted as luxurious white circles );
- all heroes ( and their enemies, made in rich red ) are controlled and move synchronously;
- heroes may die when they collide with each other, stepping on a trap, stumbling upon an enemy or falling under a box;
- the goal of the game is to deliver all the heroes alive to a single exit ( green box );
- It is advisable to find a short solution.
In this article, we will talk about working on game levels and some principles that we adhered to. Perhaps they will be useful to you.
As Terry Pratchett wrote: "The vital ingredient in success is not to know that what you have planned is impossible to fulfill." Probably, thanks to our naive and romantic approach, we achieved the result.
Remuneration
The story of the first level: “Everyone, even the smallest, should have a personal living space, so each hero got a separate room. And one of the heroes was lucky - the only way out is right in his room. At first, the lucky one behaves selfishly and seeks to leave the level as soon as possible. But with experience, he realizes that such behavior is not the most effective. For optimal passage, you need to work in a team and first think about other heroes . ”
Variability of passage is a feature of the game, which we are sincerely proud of. All our levels have alternative solutions: short or long, obvious or elegant. Each has its own world record - the minimum number of moves known at the moment. We encourage the search for the shortest solution with the help of generally accepted stars, as, for example, in Angry Birds. But, in our opinion, the opaque method of evaluation causes bewilderment and irritation. In “Syncomania” we tried to make it visual: the player gets three stars if he passes the level in the minimum number of moves. With one star, too, everything is obvious: just save all the heroes, not paying attention to the records. But for what merits give two stars? We needed a threshold value. In order not to set it for each level separately, we introduced a simple formula:
threshold = 1.5 * minimum number of moves.
Thus, if the world record is 20 moves, then for 29 the player receives two stars, and for 30 - only one. Subsequent tests showed that many people sometimes lack a couple of moves to the perfect passage. We decided that they need to be motivated, and introduced another gradation: if a player passed the level by 1–4 moves worse than the world record, then he gets 2½ stars.
Since the number of levels is limited (currently 96), we have slowed down the rate of consumption of content using the completion mechanics. A new chapter, consisting of 12 levels, opens after the player collects a given number of stars. In this way, we increase replay value by handing extra hours of enjoyment (and shamelessly hoping for in-game purchases and advertising clicks to appear).
Deceiving appearance
Variability does not make sense if a three-star solution can be found at a glance. Therefore, we try to come up with bait for players: one or more walks seem obvious, but lead to ineffective or losing conditions. For example, the “Angle” level is visually very simple, but almost no one passes it the first time. Most seek the path of least resistance - immediately to the exit in the upper right corner. Eating our bait, the player gets into a stalemate situation:
While testing the initial levels, we noticed that the symmetrical arrangement of objects looks harmonious, and randomly scattered traps and walls create visual trash. Then we decided to use symmetry everywhere. But, as you know, there are no rules without exceptions. For example, at the Candle level, we deliberately deviated from symmetry and set up a lonely wall, hinting that it should be used during the passage ( although this may be bait ).
At an early stage of development, we tried to find out how small changes in the arrangement of objects affect the method of passage. For example, we have about 10 modifications of the Four Rooms design, and different tricks are used in each of them. At the Compression level, the player learns to bring heroes together to pass between traps.
In "Simulators" will have to adapt to the behavior of enemies.
And at the “Castle” level, you need to help the locked hero get out of the room.
We apply the principle of similarity at levels with new mechanics, so that it is easier for the player to navigate in a familiar environment.
Names
At first, we gave the levels bizarre names, such as Besame Mucho or Imitation of Crab Meat. But, when tests of the mobile version began, it turned out that long names did not have enough space on the screen. It was necessary to change the words or even abandon them. We decided that levels would be better remembered if they had a “talking” name. For example, the level at which traps look like a monkey's face, we called "Macaque."
By the way, the working title of our game was “One 4 all” - “one for all” (with a hint of four), and we came up with the current “Syncomania” (attraction to synchronization) before launch.
Self learning
It seems to us that people better absorb material if they come to a decision on their own. We do not tell the player how to interact with the new mechanics, but encourage him to find his tricks with simple examples. In particular, at the “Section” level, the player is confronted with a seemingly hopeless situation. More precisely, there is a way out, but it is surrounded by traps: if you bring out one hero, you kill another. The game makes you think: “The developers must have messed up something - this level is impassable. But before that, everything was logical. ” And then the player draws attention to the walls placed, independently discovering one of the main tricks: changing the distance between the characters. Joy is in discovery.
Ed: “While we were writing this article, my wife passed this same level in 24 moves and broke a record that lasted more than a year! During development, I also liked to set records at new levels (after all, the first passage by default was the best) . "
Highscores
In the last article, we said that we do not have a stable automated way to detect the shortest passage. Therefore, we set most world records ourselves or with the help of our first testers. Usually, having scored three stars, the players calm down and move on to the next level, but some show unconventional thinking and open up unexpected decisions that we have. All new records with the passage are sent to us through Google Analytics, and we change them with the next update (and the champions are motivated, and the game is developing). So, our most loyal player Andrei Vladimirovich has already set a dozen records. In addition, we decided that the champions would be pleased if the levels at which they set world records were marked with a special golden color in the menu. But if someone else improves their record,
Difficulty curve
Having developed most of the levels, we were faced with the problem of ordering them according to increasing complexity. This is a non-trivial task, if only because the search for a solution is subjective for different players. In addition, the same level is easy to complete in the forehead, but damn difficult to pass optimally. As a result, we decided to streamline them in search of an obvious solution, so that the players did not stumble on the first pass, but if they wanted, they could plunge into optimization at the initial stage. To introduce people to new mechanics gradually, we divided the levels into thematic chapters: Traps, Streams, Enemies, etc. At the beginning of each chapter, we place the levels easier, and by the end we check the player for strength. So we formed a curve of complexity with fluctuations.
At the stage of organizing, we threw out a lot of boring levels, but testing showed that we also need to add something. For example, we added the first chapter last, on the advice of our testers. It turned out that the newcomers did not have time to master the tricks of the game, and they needed more simple levels for swinging. It became clear that we were too immersed in the game, and our gaming experience cannot be used to assess the complexity. Then we tried to imagine ourselves in the place of the newcomers who first launched our game, and created concise introductory levels using only the walls.
Aidar: “The work on the first chapter brought me great pleasure, since severe restrictions opened up scope for creativity. But still this is an introduction, and the most interesting part was to develop the latest chapters for experienced players . ”
Examples
Creating levels, we did not think through new mechanics and puzzles in advance. During programming and testing, they showed themselves. For example, once for the sake of testing, we put a box on a conveyor belt from threads. According to the laws of the game world, the box moved along the course and, stepping off the conveyor, clogged a narrow corridor for us. “Hmm, what a nuisance - it must be used,” we thought. So in our turn-based game, a kind of timer appeared: you need to have time to solve the riddle in N moves, otherwise the output will be walled up with boxes.
Experimenting with two timers, we directed their flows to a free cell. Boxes reached the end and mutually destroyed in a collision. This came as a surprise to us, since in other cases the game mechanics did not allow pushing the boxes together. After a little consultation, we decided that this was not a bug, but a feature . As a result, the Skull level was created where you can destroy boxes in this way.
Having tried numerous types of interactions, we came to the idea that for many centuries served as the engine of science and progress: “Why don't we mix all this and see what happens?” As a result, we got the most confused at the first (and second) glance level - “Whirlpool”. However, despite the frightening appearance, this is by no means the most difficult puzzle in the game. Given the small number of steps that do not lead to loss, this level can be completed by a simple bust. Even our automated solver handled it - in just 16 moves.
However, in most cases, it will not be possible to find a solution by exhaustive search - with each move, the number of states grows exponentially. The game should use a creative approach, and it will have to be fully demonstrated at the most difficult level, according to testers, “Prisoners”.
Last level story: “Terrorists are holding hostages in two small cells. The heroes have no choice but to move up: a step down - and the hostages will be killed. In addition, insidious villains (yes, this is us) mined the entire upper limit of the level. It would seem that the player faces a dilemma: to sacrifice hostages or destroy heroes. But even if by some miracle (spoiler: boxes) the player manages to eliminate the enemies, he will notice that the room with the exit is surrounded by streams directed outward. And this part is the insidious highlight of the level . ”
To be continued ...
In the next part we will talk about the development of the game interface.