Game Design: A New Approach to Difficulty Levels

Original author: Alex Vu
  • Transfer

The problem of difficulty in games has been discussed for a long time. Various alternatives to the traditional choice of difficulty levels at the beginning of the game were proposed, analyzed and implemented. And despite the fact that they correct the mistakes of the traditional approach, many other inalienable problems and difficulties arise within them. Therefore, I would like to propose another alternative - this is rather not a mechanical solution that requires implementation, but another approach to creating design complexity.

I want to emphasize that this approach has been quite successfully implemented in many games, and I will mention them below. But, in my opinion, he wasn’t so deployed to become the main design philosophy. I assume that this situation has developed due to the lack of a clear and thoughtful approach to creating design complexity.

But first, I will try to briefly list the most common complaints about the traditional scheme with levels of complexity and its alternatives.

Difficulty levels

Imagine that you have launched a completely new game for you, which first of all asks you to choose the difficulty mode that suits you, giving you a choice of several options. And frankly, she is not very good at presenting information so that you can make such an important decision with confidence. This is how many games of the past have difficulty implemented, and in modern games such an approach is quite popular.

This is what this approach is usually criticized for:

  • It is not a good idea to ask the player to make such a choice at the very beginning of the game. Selecting the difficulty mode before the game starts - it means making a serious decision based on very scant information (for example, a brief description). After a player chooses a difficulty, he will most likely be forced to play it until the end of the game.
  • Even if the game allows the player to change the difficulty level later, this idea is still bad in itself. First, the immediate process of choosing the level of complexity in the menu is not the choice that the game should strive to give players. They should not figure out the pros and cons. They are not obliged to analyze the risks and rewards resulting from the choice of one of the options. On the whole, the players do not cope well with the choice of the ratio between short-term convenience and long-term interestingness of the game. They don't know the game well enough.
  • This approach destroys the whole point of development through the discovery of more and more powerful tools that expand capabilities and help in the gameplay. It contradicts the gameplay style conceived by the game designer. Most importantly, the player will feel that he is being convicted for refusing increased complexity.

There were several solutions to reduce the impact of these problems. Mark Brown told about them in detail in one of his videos . However, none of them is able to completely solve them, while maintaining a sense of immersion in the game.

Dynamic complexity change

The idea of ​​dynamic change in complexity (Dynamic Difficulty Adjustment, DDA) is based on the player’s “state of the flow” theory, in which the player is completely immersed in the game and its complexity seems ideally correct. Slightly increase the complexity - and the player will feel the annoyance that impedes immersion. Slightly reduce the complexity - the player quickly gets bored, and, as you might guess, will lose immersion in the game. Therefore, as designer Andrew Glassner said in his book Interactive Storytelling, games “should not ask the player to choose a difficulty level. Games must adapt themselves during the gameplay to provide the player with a constant degree of difficulty based on his changing abilities to perform various tasks. ” In other words, in games, a performance evaluation system must be implemented, as well as a dynamic complexity change system, so that it adjusts itself to infinitely different and constantly changing player characteristics. Read more about the technical details of the DDA can be found in the 2005 article "The Case for Dynamic Difficulty Adjustment in Games" by Robin Hanike.

However, despite all the advantages of the “flow state” theory, the DDA system was not without its own drawbacks:

  • Some players, learning about the system DDA, penetrate to her dislike. Especially when it is impossible to turn off the DDA, the players begin to feel that they are “led by the hand” and the game does not respect them as an adult person who is able to cope with difficulties and improve themselves.
  • Players can, and will learn to exploit DDA, pretending they are playing worse than they really are. And very often, the DDA system should not show itself for a while in order not to adapt to the player’s showy skill level.
  • DDA impedes the ability to learn and improve. As the player improves his skills, the difficulty increases to match his level, thus eliminating the possibility of positive results. If a player cannot see that the game gives him feedback on his skill, he will never be able to find out if the change in his approach to the game process was effective.
  • DDA can create absurd situations. One of the most popular examples of the erroneous work of DDA is the effect of “sticking” in racing games, when opponents accelerate and slow down without any visible logic to adapt to the player's skill.
  • DDA is incompatible with some types of tasks. If the task in question is related to numbers, then DDA works easily. However, if the task is symbolic and has pre-created elements that are clearly visible to the player, and has only one or several solutions conceived by the designer, then DDA cannot work.

There are many interesting approaches to the implementation of DDA with its own nuances, which I will not discuss here. Although I can imagine that there can be many ways to make DDA functional and invisible enough for the player with the help of algorithms and implementation features, but let's talk about the basics.

Organic difficulty in games

It seems that there are many other terms that call this approach, but in this article I will designate it as “Organic Complexity”. The video game industry has been experimenting with this tool for about a decade.

The basic idea of ​​organic complexity is that the game does not ask the player to choose or change the desired difficulty through commands in the GUI, and does not automatically adapt to match the player’s skills and progress. Instead, the game allows the player to interact with it in various ways that simplify or complicate the gameplay. They can take the form of tools, decisions, strategies, sequences or input methods, and so on, which often have some trade-offs.

Such a system is implemented in a variety of games, including Dark Souls.From Software, to which Extra Credits dedicated an entire episode . I recommend to see it all.

In Metal Gear Solid VAfter completing each mission, there is a points rating system that evaluates a player’s skills based on several factors, such as secrecy, killing, accuracy, passing speed, the number of mission tasks completed and the tools he uses. A player gets negative points for mistakes, for example, when enemies notice him, an alarm goes up, a player gets into, and so on. There are other factors that are not so obviously associated with a decrease in points. The player at any time can start a full-scale war, throwing grenades at each enemy or calling a support helicopter for an airstrike on the enemy base. He has the tools to do all this, they are fairly easy to use, and when applied, the worst thing you can get is rank C (if the player has passed the mission), as well as a slight decrease in income.

Another example of such a system can be found in XCOM: Enemy Within. The game has a “cheap” tactic that can provide an almost guaranteed victory - it is enough to have a unit with the Mimetic Skin skill to safely detect enemies, which will allow the sniper to shoot them one by one from the other end of the map without any response. This strategy is extremely effective in terms of almost every aspect of combat mechanics; the only risk is that the gunner can bypass the flank, because of which he instantly loses invisibility. The real problem with this strategy is that it is incredibly boring: snipers can just shoot, and in each move it is enough to make only a few shots, except for reloading. This strategy is better suited for newbies and for people who have made mistakes and want to get out of the vicious circle. On the other side of the spectrum are players who understand well how the game and AI of each alien unit work. They are confidently approaching the enemy with minimal armor. Because for them, fighting is not protection from enemies, but manipulation, “pushing” enemies to actions that players expect from them (for example, nobody needs armor if enemies attack only a tank; nobody needs to take good shelters when enemies are too afraid to bypass the flanks due to the Opportunist ability of the Overwatch character, and so on).

From the above examples, several important considerations follow:

  • Difficulty must be created not only on the basis of the game mechanic. She must consider the aesthetics or elegance of each of these mechanics.
  • Punishment does not always have to be tangible or significant, if it is enough to make it clear to the player that he deviates from the intended gameplay. A good analogy is physical pain. It is not the pain itself that causes harm to the body, but a physical injury. Pain is simply a signal from the body, telling you that what is happening now is very undesirable and you should avoid it further. But remember that the choice is still yours!
  • It may not be very correct to place people on a line chart of "game skills", where people are simply divided into "softcore, not very familiar with video games" and "hardcore, always striving for difficulties." By itself, this idea is absurd, because players on such a chart will constantly move down and up, even in the course of a single playthrough. Some people grasp learning faster than the game can predict in the course of their tutorial. Some people, due to reasons in real life, need to abandon the game for some time, after which they return to it, having lost some skills.
  • Instead of evaluating a player’s skill and trying to adapt to it in all aspects, games should evaluate the player’s interactions with the world, using the spectrum between efficiency and aesthetics of the gameplay (or what I will modestly call ludo aesthetics ).

Spectrum "efficiency-ludo-aesthetics" (SEL)

In the efficiency spectrum of Ludo-aesthetics, complexity exists only at the lowest technical level. Both edges of the spectrum simulate what the player would like at a certain point in a game with certain conditions. In this spectrum, game design is created taking into account the interactions, campaigns and strategies of the player. Each of these aspects has its own degree of efficiency and ludo-aesthetics. They are set not only by the mechanics or skill level of the player, but also by how they are experienced and perceived by the player.

Efficiency is related to how well a player can advance in the game and achieve his goals, using the available tools and strategies that he can formulate. Efficiency mainly consists of how easy it is to use these tools and how much they help the player in achieving the goals set by the game. Players seeking to this end of the spectrum and remaining on it, usually looking for the most effective ways to achieve the goals of the game (of course, these methods include the passage of the game in the easiest way).

Ludoestheticsassociated with the perceived aesthetics of the player data of the above tools and strategies. Players who reach this end of the spectrum are not necessarily looking for the most effective ways to achieve the goals set by the game. Rather, they seek to seek additional inalienable advantages derived from unconventional passing. These benefits are:

  • External attractiveness : visual and sound attractiveness of using the object or the object itself. It can be represented by any entity that the player is able to distinguish in the game: a character with an excellent visual design, a cool gun, having attractive graphic or sound effects, and so on.
  • Competitiveness : prestige. The term speaks for itself. There is always a share of players striving for more and more trials to declare themselves to the world. They can even artificially limit themselves to complicate the trials.
  • Enhanced sense of satisfaction comes from a lot of tests that can exceed the goals conceived by the designers in the game. People who set themselves difficult obstacles can expect tremendous satisfaction in achieving their goals.
  • Narrative fantasy : players may aspire to entities that are not effective and not productive in terms of gameplay, because they are better suited to the narrative (in games that contain some degree of ludonarrative dissonance). Or they can add to the narrative an extra layer of depth and tension, thus improving it. In fact, they sacrifice the optimality of the gameplay for the sake of enhancing the narrative fantasy.

Design for ludo aesthetics

The meaning of creating ludo-aesthetics design is NOT to create more and more complex tasks to match the increasing skills of the player (however, it cannot be said that this approach has no advantages). Rather, it is to encourage players to aim for the right edge of the spectrum.

Here are some tips to achieve this.

Creating more depth

Depth is related to the amount of space in which a player can make an interesting choice using a set of data from a game of tools. More details about how Depth differs from Complexity can be found in the Depth vs. episode. Complexity Channel Extra Credits.

In fact, complexity is the number of elements that make up a game, and depth is the degree of interaction between these elements. The very nature of ludo-aesthetics should work with deviations from the standard approach designed by designers (that is, from the “game by the rules”). Therefore, the more such elements “speak” with each other, the greater the chance that ludo-aesthetics will arise, because then the player will be able to find more diverse ways to control each element.

[Also worth reading: Design for Theorycrafting ]

Depth is a very necessary requirement for the emergence of ludo-aesthetics, including as a concept. Without solid depth, the window of opportunity for ludo-aesthetics becomes much smaller or disappears altogether.

Creating patterns that imply the probability of gameplay ludo-aesthetics

To add depth, it is not enough just to add more objects to the game and make them as mysterious as possible. It is also necessary to add hints that there is something more than the eyes can see, thus stimulating the player to explore additional possibilities. What kind of depth does this add? And how to inform the player about its availability?

Below is a conceptual model of a set of tasks used in video games.

Each task is represented by a failure window and a success window. These windows can be spatial, temporal, symbolic, strategic, or combinations of all of the above. These are spaces in which a player enters, behaving in a certain expected way. The black line indicates the player's interactive maneuvers: how to move on and which direction to turn in order to overcome a series of obstacles without getting into the window of failure.

Suppose we have a situation in a 3D platformer: the player is standing in front of the pit, and there is a narrow platform above the pit. In such a situation, we can assume that the window of failure includes all the sets of behaviors that will lead the player to fall into the pit, and the window of success includes those that will lead the player to a safe landing on the platform.

Now consider the same task model, but this time with a slightly modified layout.

As you can see, the sizes of the windows of failure and success remain exactly the same, but the positions of the windows of success are changed so that they coincide in some way (but they are not perfectly aligned so that it is not too obvious). You may notice that inside the windows of success there is a narrower window in which the size of the player’s maneuvers remains extremely minimal. Entering this window provides the possibility of an uninterrupted gameplay flow; A thoughtful and manageable set of behaviors will allow the player to “rush” through obstacles with almost no effort. It is in this window that Ludoesthetics appears.

Of course, such a scheme has many drawbacks: in a real situation it can be extremely difficult to understand that such a window exists. And in order to stay within such a narrow window, the player during the game process must be extremely accurate and / or smart. It is possible to consider such a window of an unbroken stream as a deliberate “weak point” of an obstacle, where a single concentrated attack can destroy all obstacles in one fell swoop. But the process of detecting such a weak point and providing a final blow with great accuracy can require a lot of trial and error, which can be overly monotonous and / or difficult.

Master Spy Example

Ludoesthetics often manifests itself in the form of speed passing games (speedrunning). Passing at maximum speed is not the main goal for most games. The design of games is rarely created taking into account speedrins, and most players should not go through the game at high speed in order not to miss something. Therefore, speedrunning has always been an intentional complication of the game for those who seek to get more pleasure from their favorite games.

However, there are a few exceptions. And we can find the aforementioned uninterrupted stream window on levels from Chris Truitt's Master Spy game.

In this game, you take on the role of a super spy trying to penetrate carefully guarded buildings, palaces and fortresses with a huge number of different enemies, dangers and traps. The player has no tools other than the invisibility cloak that allows you to slip past certain enemies, and at the same time slowing down the character twice.

In the example shown above, the player’s goal is to get the key card on the other side of the wall, slightly to the right of the starting point, and then escape through the white door above the starting point. And although the invisibility cloak allows you to avoid the guards, it does not help anything against dogs that can smell a hero even invisible. Once the dogs are on the same floor with the player, they rush to him at great speed.

To pass the level, you need to perform the following procedure: first disguise, then fall from the first cliff behind the first guard, then quickly exit the stealth mode in order to gain speed, because the raincoat is useless against running dogs. Then, before the player is touched by the first dog, he must move forward to the right, and then quickly jump up. We continue to jump until we get the key card, avoiding the second and third dog. We put on a raincoat, climb the ledge with three moving guards, and then finally jump up to get to the end point.

However, as can be seen in the above entry (which the speedrunner named Obidobi made), as soon as the player reaches the ledge with three moving guards on the right, the guards turn the other way and begin to move away from where the player is, essentially saving him from the need to include masking and reduce speed by half. And just before the player reaches the white door, the guard on the right almost touches the wall, which means he will soon turn left. This is such a narrow window of success that if the player did not begin to move immediately after the beginning of the level and did not remain without disguise at the end, then everything would have ended in failure. The level is created in such a way that it can be completely solved without spending a single moment and without performing unnecessary actions.

So play is much more difficult? Yes. Was such a scheme absolutely necessary? Not at all. But the designer created the level, expecting people to go through the game with a speedran, and if the player looks for ways to get through the game quickly, he will very quickly discover this continuous flow window. From the discovery of such an opportunity, the player gets great pleasure.

Possibility of use

It is usually very easy to approach design too logically and forget about balance, for which design is needed first and foremost.

In this case, it is important that designers provide the obligatory possibility of using all tools that create ludoesthetics, even if their use is extremely niche or it is very difficult to use them. Aspects that do not play any role and mean nothing NOT aesthetic. Suppose we have an RPG, and one of the players goes out of the way to create an unconventional character, because he sees in his configuration some potential for the future, but after the configuration has been fully constructed, he discovers that the meta game has changed, and the window of opportunity for This configuration is long over. This will mean that the entire volume of the depth you have added and the ludoesthetics that you conceived, allowing the player to go this way, is completely useless and completely corrupted. Therefore, always provide the ability to use everything that you add to the game.


Organic complexity and SEL are not the only and not mandatory alternative solutions to the problem of complexity in general. Rather, they represent a shift in the entire paradigm from the idea that games should find more and more complex ways to serve players with different levels of skills, to a design philosophy in which players are given integrated tools in the context of games, allowing them to set their own bar of complexity without destroying immersion in the game and not making players ashamed. It is not enough for the players to remain at the same level of complexity for the entire game, just as it is not enough to dynamically change the difficulty according to their skill. Best of all, in my opinion, to allow the player to choose the complexity of the dish to your taste. The designer simply has to make the cooking process and the gameplay coincide.

Reference materials

  1. The Designer's Notebook: Difficulty Modes and Dynamic Difficulty Adjustment (2008) by Earnest Adams. Taken from
  2. For Robin Hunicke
  3. Cognitive Flow: The Psychology of Great Game Design (2012) by Sean Baron. Taken from
  4. Depth vs. Complexity (2013) by Extra Credits. Available at
  5. The True Genius of Dark Souls II (2014) by Extra Credits.
  6. What Makes Celeste's Assist Mode Special | Game Maker's Toolkit (2018) by Mark Brown.

Also popular now: