Devlog #7
Hello guys!
Today we're going to talk about level design!
There are a lot of good tutorials on how to design levels. Only most of them are based on the 10 golden level design advice.
The future designer is introduced to the fact that he should set landmarks, leader lines and use many other techniques of hidden player guidance.
As Split is largely a logical game, I will focus on this logical aspect. When a "room" is created (it does not necessarily have to be one room) in which we are dealing with a puzzle, we must first create a prototype of it. It takes a long time to set up all the lights, effects and decorations, so there are some facilitations.
Before anything looks like a game, something like a virtual mockup is built, which is called Blockmesh or Blockout. It consists in setting simple shapes so as to mark where the player will be able to move. Often you can move your character in such an unfinished world in order to be able to measure individual distances or heights of elements on which you can jump.
Before anything can be built, we're going even lower, and actually I'm going down a few levels until we climb back into the blockout.
The first is an idea or research. If we already have a very specific idea of what the puzzle should look like, we can go to prototyping, but if there are only some general assumptions or only some part of what will be in the game is designed, you will need to get some inspiration from other developers. Before we started designing anything to play, we had a whole list of titles: Portal, Portal 2, Q.U.B.E. 2, Relicta, The Turing Test, The Talos Principle, The Witness and Antichamber, and that's just from those 3D titles. In addition, you can find a lot of interesting speeches or articles about the mindset of people creating puzzles.
To go on to the next part of the article, we should remember the kindergarten classes because one of the most useful materials for prototyping is plasticine.
What is the advantage of designing in plasticine over creating it right away in the engine? At first glance, the virtual environment is more scalable because we have a limited amount of raw material in the real world. In addition, when we set something in the virtual world, we can immediately start the game and test. When it comes to pace, when it comes to making something in plasticine, then we have to move it to the engine anyway. However, in practice, we will never do puzzles consisting of 1000 rochs, only a few or a dozen. What's more, translating pieces of plasticine is much faster than moving individual elements of the world with the mouse. Thanks to this, we can test 15 different systems and choose one of them, which in the engine will be set by us so that it matches the appropriate distances. Putting everything in the engine so that it fits for the tests is a chore. In addition, it is usually the case that the person who creates the puzzle and the one who will stick it together in the engine are two other people. So what the puzzle designer does, he creates some diagrams, then he writes them on paper and gives them to the person who then plots them onto the geometry of the world. Thanks to prior research, the puzzles will often require a small amount of changes to the engine itself.
What is the example puzzle about? You have a kind of number pattern here, there are holes in some places. The boxes are positioned so that you can only jump on elements next to or diagonally. Before we even enter the fields, there is a written order on the walls that you will have to remember. If there are numbers 3, 6, 2, 3 consecutively, it means that first we have to jump to the tile with number 3, then 6, 2 etc.
Each subsequent puzzle using the same theme should be more difficult. At the beginning it consists of 3 elements, and it can be solved in one move, the next one has 6 elements, and it can be solved in three moves. The next puzzles as they develop will use previously known mechanisms.
What cannot be easily conveyed in plasticine is the geometry of the levels. It seems possible, but it requires a lot of raw material, and the material itself has a certain load capacity, so you cannot make too large pieces of plasticine hanging on plasticine. Instead, geometries can be drawn on paper to fit the layout of the puzzle, or you can find other material. Doing multiple levels on paper doesn't show the transitions between floors well, and some more complicated ones even require you to make extra rolls, so it's better to build it in the engine itself. Of course, there are also materials that lend themselves to the smooth creation of geometry. These are mainly various types of blocks. You can design many floors in them, they are durable and easy to make platform elements.
The combination of the geometry of the blocks along with the appropriate spacing of the improved elements gives a very large field for creating spatial puzzles that would not be achieved by any of these materials separately. Such a prepared fragment of blocks connected with properly arranged mechanisms is then converted into a blockout. If it is not ready, it goes back to the prototype phase again, if it is good enough, it goes on the road by adding visual appeal.
Finally, let me just mention that in most 2D games, physical design is not needed, because they have such an uncomplicated process of moving to the digital world that programmers prepare an editor that allows you to click all important elements easily and super quickly. Mostly this is used in puzzles with a limited map size or based on connecting layouts.
We would also like to invite everyone to our Discord where we will be happy to answer all your questions about the current work - DISCORD
Today we're going to talk about level design!
There are a lot of good tutorials on how to design levels. Only most of them are based on the 10 golden level design advice.
The future designer is introduced to the fact that he should set landmarks, leader lines and use many other techniques of hidden player guidance.
As Split is largely a logical game, I will focus on this logical aspect. When a "room" is created (it does not necessarily have to be one room) in which we are dealing with a puzzle, we must first create a prototype of it. It takes a long time to set up all the lights, effects and decorations, so there are some facilitations.
Before anything looks like a game, something like a virtual mockup is built, which is called Blockmesh or Blockout. It consists in setting simple shapes so as to mark where the player will be able to move. Often you can move your character in such an unfinished world in order to be able to measure individual distances or heights of elements on which you can jump.
Before anything can be built, we're going even lower, and actually I'm going down a few levels until we climb back into the blockout.
The first is an idea or research. If we already have a very specific idea of what the puzzle should look like, we can go to prototyping, but if there are only some general assumptions or only some part of what will be in the game is designed, you will need to get some inspiration from other developers. Before we started designing anything to play, we had a whole list of titles: Portal, Portal 2, Q.U.B.E. 2, Relicta, The Turing Test, The Talos Principle, The Witness and Antichamber, and that's just from those 3D titles. In addition, you can find a lot of interesting speeches or articles about the mindset of people creating puzzles.
To go on to the next part of the article, we should remember the kindergarten classes because one of the most useful materials for prototyping is plasticine.
What is the advantage of designing in plasticine over creating it right away in the engine? At first glance, the virtual environment is more scalable because we have a limited amount of raw material in the real world. In addition, when we set something in the virtual world, we can immediately start the game and test. When it comes to pace, when it comes to making something in plasticine, then we have to move it to the engine anyway. However, in practice, we will never do puzzles consisting of 1000 rochs, only a few or a dozen. What's more, translating pieces of plasticine is much faster than moving individual elements of the world with the mouse. Thanks to this, we can test 15 different systems and choose one of them, which in the engine will be set by us so that it matches the appropriate distances. Putting everything in the engine so that it fits for the tests is a chore. In addition, it is usually the case that the person who creates the puzzle and the one who will stick it together in the engine are two other people. So what the puzzle designer does, he creates some diagrams, then he writes them on paper and gives them to the person who then plots them onto the geometry of the world. Thanks to prior research, the puzzles will often require a small amount of changes to the engine itself.
What is the example puzzle about? You have a kind of number pattern here, there are holes in some places. The boxes are positioned so that you can only jump on elements next to or diagonally. Before we even enter the fields, there is a written order on the walls that you will have to remember. If there are numbers 3, 6, 2, 3 consecutively, it means that first we have to jump to the tile with number 3, then 6, 2 etc.
Each subsequent puzzle using the same theme should be more difficult. At the beginning it consists of 3 elements, and it can be solved in one move, the next one has 6 elements, and it can be solved in three moves. The next puzzles as they develop will use previously known mechanisms.
What cannot be easily conveyed in plasticine is the geometry of the levels. It seems possible, but it requires a lot of raw material, and the material itself has a certain load capacity, so you cannot make too large pieces of plasticine hanging on plasticine. Instead, geometries can be drawn on paper to fit the layout of the puzzle, or you can find other material. Doing multiple levels on paper doesn't show the transitions between floors well, and some more complicated ones even require you to make extra rolls, so it's better to build it in the engine itself. Of course, there are also materials that lend themselves to the smooth creation of geometry. These are mainly various types of blocks. You can design many floors in them, they are durable and easy to make platform elements.
The combination of the geometry of the blocks along with the appropriate spacing of the improved elements gives a very large field for creating spatial puzzles that would not be achieved by any of these materials separately. Such a prepared fragment of blocks connected with properly arranged mechanisms is then converted into a blockout. If it is not ready, it goes back to the prototype phase again, if it is good enough, it goes on the road by adding visual appeal.
Finally, let me just mention that in most 2D games, physical design is not needed, because they have such an uncomplicated process of moving to the digital world that programmers prepare an editor that allows you to click all important elements easily and super quickly. Mostly this is used in puzzles with a limited map size or based on connecting layouts.
We would also like to invite everyone to our Discord where we will be happy to answer all your questions about the current work - DISCORD