Introduction
This will be a series of articles about game design, using the popular games DOTA and League of Legends as illustrative examples. The specific words, "Design is an Optimization" was coined by the design director of Riot. It means that to him, design is an iterative process of optimization.
Thus if some aspect of DOTA or LOL seems inferior or different to the other, it is because a conscious choice has been made by its designers – the trade-offs have been examined, the benefits judged more than the costs.
Design Evolution History
DOTA was a custom Warcraft 3 mod developed in 2003 by Eul, and has since grown and evolved organically through the years as a one man project. Leadership passed from Eul to Guinsoo in 2004, and then to Icefrog in 2005, who has remained the sole developer till today.
The game design of DOTA is driven through the feedback of the playerbase, shaped by the unique quirks and mechanics of the Warcraft 3 engine. Icefrog is a famously reclusive figure – he rarely interacts with the community, and no one knows his real identity. Icefrog is seen as the mysterious game designer prodigy, who consults no one on his design decisions, only descending from his mountain perch every few months to reveal the fruits of his labours (a new patch, maybe new heroes) without explanation. Today, Valve is working on developing DOTA2, which is the same game but with improved graphics.
LOL is developed by Riot Games, and it was designed by a team of experienced and talented developers, many with storied histories and credentials. Zileas (currently Design Director at Riot Games) was a key developer in World of Warcraft and Warcraft 3, as well as being a competitive Starcraft player himself. LOL’s designers make a lot of effort to engage with the community, explaining their design philosophies and debating game balance. Their vision is clear: with rigorous design, and learning from the missteps of other major games like DOTA and WOW, they aim to create a game that both attracts new players and yet retains depth and strategy to engage the competitive scene.
DOTA was never monetized, and this shows in their game design, which is aimed at satisfying veteran players. LOL on the other hand generates substantial revenue for their developers, and their design is aimed at creating a low barrier to entry, making it easy for new players to join and start playing.
Without further ado, let us get started!
Complexity versus Gameplay
This question is a core idea in game design, so we’ll start by covering this concept.
Complexity is never good in itself. It’s a negative trait you accept in order to gain other, more positive traits. It’s like asking whether a 500 page novel is better than a 300 page novel – it’s a false question. The real question is always going to be, which one told a better story?
You will see this analogy played out over and over in game design.
What game designers really want to achieve is something called “gameplay”. It’s the interaction between players that creates the “game”, and hero skills require some complexity in order to make that happen. For example, say you had a skill that you could cast on an enemy hero to deal damage to it. That’s kind of boring: it’s low complexity, but also generates very little gameplay. By increasing the complexity of the skill, we can create more gameplay. For example, if the skill had to be targeted on the ground, and there was a 1 second travel time – now the attacker has to predict the movement path of the victim to land the attack, and the victim is also given a chance to dodge the skill. The game immediately ramps up and become more engaging to BOTH players, at the cost of only a small amount of complexity.
Where to draw the line is where it gets contentious. How much complexity do you really want to trade for gameplay? As you scale up the complexity you get less and less gameplay in return, and that comes with all sorts of negative effects that we will go into later – Burden of Knowledge, False Choice, Unclear Optimization, and many more. Each concept will get a blog post on their own.
The Invoker from DOTA is an example of a hero that is deemed “ok” in DOTA but too complex to exist in LOL. Using a system of elemental orbs, he has to invoke spells into existence before casting them, and he can only have two spells active at once. He effectively has 14 different spells which do vastly different things, making it very difficult to play him and more importantly, also very difficult to play against him.
Trade off #1: Burden of Knowledge
Overly complex mechanics can create a “Burden of Knowledge” problem when the gameplay it tries to create relies on the victim understanding what is going on. It's basically like being introduced to a game by a friend, and you think you understand the rules and you get a few turns in, and then suddenly your friend goes "Aha now you've lost, due to this obscure mechanic I never explained to you!"
The poster boy for illustrating this effect is the Bloodseeker’s Rupture in DOTA. If he casts Rupture on you, for the next 5 seconds, if you move, you continuously take damage. In theory, this skill is supposed to create interesting gameplay – it’s supposed to give the victim a choice of whether to stay still and fight, or decide if it’s safer to run to a better position and accept some damage in return. It’s an interesting mechanic, and needs great player judgement to use, especially when multiple heroes are involved in the clash – you might be better off running towards your ally who has a heal spell, for example, or moving slightly to dodge an area target damage effect another enemy was going to cast on you. Considering there are over 100 unique heroes in DOTA, it's quite possible that new players who do not understand what is going on may attempt to flee or chase the enemy, and thus continually take damage until they die, without ever understanding what it was they were doing wrong.
Riot would never release a champion with this level of Burden of Knowledge, despite the interesting gameplay it can provide. There are less costly ways of achieving the level of gameplay and interaction they want. Again, we come back to the origins of the games themselves: DOTA is optimized for veteran players, while LOL is optimized for new players.
Notice the focus is on the person playing against the character design in question. Complexity can be acceptable if it only impacts the person that chooses to play that character, because that is his choice to make. Some players will deliberately seek out complex heroes to play, because they enjoy them. However, complexity is a bad thing if it impacts a person playing against that character: you don’t get to choose your opponents, and if you face something with a high burden of knowledge (like the Bloodseeker) and don’t know how to deal with it, you will have a very frustrating experience.
Skills that follow this design philosophy can be found in both games. Both Nidalee’s Javelin Toss and Mirana’s Sacred Arrow follow this pattern. They are both “skill-shots” - you fire the projectile, and it flies in a straight trajectory until it hits something - with a unique twist, in that they gain effectiveness the further they travel. This means it’s trivial to land the shot at close range, but it will be ineffective. Players can potentially spend months mastering their use – figuring out the maximum range, trying to predict enemy movement patterns, making the best tradeoff between range and actually being able to land it. But from the victim’s perspective, there is nothing complex about this interaction – all they need to know is being hit by an enemy projectile is generally bad (true for all enemy projectiles) and they should get out of the way if they can. Hence these abilities do not create a Burden of Knowledge problem for the victim, and are regarded as examples of well designed skills which create a lot of interesting gameplay.
The next post will deal with the concept of "Power without Gameplay".
This will be a series of articles about game design, using the popular games DOTA and League of Legends as illustrative examples. The specific words, "Design is an Optimization" was coined by the design director of Riot. It means that to him, design is an iterative process of optimization.
Thus if some aspect of DOTA or LOL seems inferior or different to the other, it is because a conscious choice has been made by its designers – the trade-offs have been examined, the benefits judged more than the costs.
Design Evolution History
DOTA was a custom Warcraft 3 mod developed in 2003 by Eul, and has since grown and evolved organically through the years as a one man project. Leadership passed from Eul to Guinsoo in 2004, and then to Icefrog in 2005, who has remained the sole developer till today.
DOTA has a playerbase of about 20 million players
The game design of DOTA is driven through the feedback of the playerbase, shaped by the unique quirks and mechanics of the Warcraft 3 engine. Icefrog is a famously reclusive figure – he rarely interacts with the community, and no one knows his real identity. Icefrog is seen as the mysterious game designer prodigy, who consults no one on his design decisions, only descending from his mountain perch every few months to reveal the fruits of his labours (a new patch, maybe new heroes) without explanation. Today, Valve is working on developing DOTA2, which is the same game but with improved graphics.
LOL has an estimated playerbase of 35 million players
LOL is developed by Riot Games, and it was designed by a team of experienced and talented developers, many with storied histories and credentials. Zileas (currently Design Director at Riot Games) was a key developer in World of Warcraft and Warcraft 3, as well as being a competitive Starcraft player himself. LOL’s designers make a lot of effort to engage with the community, explaining their design philosophies and debating game balance. Their vision is clear: with rigorous design, and learning from the missteps of other major games like DOTA and WOW, they aim to create a game that both attracts new players and yet retains depth and strategy to engage the competitive scene.
DOTA was never monetized, and this shows in their game design, which is aimed at satisfying veteran players. LOL on the other hand generates substantial revenue for their developers, and their design is aimed at creating a low barrier to entry, making it easy for new players to join and start playing.
Without further ado, let us get started!
Complexity versus Gameplay
This question is a core idea in game design, so we’ll start by covering this concept.
Complexity is never good in itself. It’s a negative trait you accept in order to gain other, more positive traits. It’s like asking whether a 500 page novel is better than a 300 page novel – it’s a false question. The real question is always going to be, which one told a better story?
You will see this analogy played out over and over in game design.
What game designers really want to achieve is something called “gameplay”. It’s the interaction between players that creates the “game”, and hero skills require some complexity in order to make that happen. For example, say you had a skill that you could cast on an enemy hero to deal damage to it. That’s kind of boring: it’s low complexity, but also generates very little gameplay. By increasing the complexity of the skill, we can create more gameplay. For example, if the skill had to be targeted on the ground, and there was a 1 second travel time – now the attacker has to predict the movement path of the victim to land the attack, and the victim is also given a chance to dodge the skill. The game immediately ramps up and become more engaging to BOTH players, at the cost of only a small amount of complexity.
Where to draw the line is where it gets contentious. How much complexity do you really want to trade for gameplay? As you scale up the complexity you get less and less gameplay in return, and that comes with all sorts of negative effects that we will go into later – Burden of Knowledge, False Choice, Unclear Optimization, and many more. Each concept will get a blog post on their own.
The Invoker uses 3 orbs in combinations of 3 to invoke 10 different spells
The Invoker from DOTA is an example of a hero that is deemed “ok” in DOTA but too complex to exist in LOL. Using a system of elemental orbs, he has to invoke spells into existence before casting them, and he can only have two spells active at once. He effectively has 14 different spells which do vastly different things, making it very difficult to play him and more importantly, also very difficult to play against him.
Trade off #1: Burden of Knowledge
Overly complex mechanics can create a “Burden of Knowledge” problem when the gameplay it tries to create relies on the victim understanding what is going on. It's basically like being introduced to a game by a friend, and you think you understand the rules and you get a few turns in, and then suddenly your friend goes "Aha now you've lost, due to this obscure mechanic I never explained to you!"
Bloodseeker's Rupture spell does damage to the enemy based on the distance they travel
The poster boy for illustrating this effect is the Bloodseeker’s Rupture in DOTA. If he casts Rupture on you, for the next 5 seconds, if you move, you continuously take damage. In theory, this skill is supposed to create interesting gameplay – it’s supposed to give the victim a choice of whether to stay still and fight, or decide if it’s safer to run to a better position and accept some damage in return. It’s an interesting mechanic, and needs great player judgement to use, especially when multiple heroes are involved in the clash – you might be better off running towards your ally who has a heal spell, for example, or moving slightly to dodge an area target damage effect another enemy was going to cast on you. Considering there are over 100 unique heroes in DOTA, it's quite possible that new players who do not understand what is going on may attempt to flee or chase the enemy, and thus continually take damage until they die, without ever understanding what it was they were doing wrong.
Riot would never release a champion with this level of Burden of Knowledge, despite the interesting gameplay it can provide. There are less costly ways of achieving the level of gameplay and interaction they want. Again, we come back to the origins of the games themselves: DOTA is optimized for veteran players, while LOL is optimized for new players.
Notice the focus is on the person playing against the character design in question. Complexity can be acceptable if it only impacts the person that chooses to play that character, because that is his choice to make. Some players will deliberately seek out complex heroes to play, because they enjoy them. However, complexity is a bad thing if it impacts a person playing against that character: you don’t get to choose your opponents, and if you face something with a high burden of knowledge (like the Bloodseeker) and don’t know how to deal with it, you will have a very frustrating experience.
Mirana's Sacred Arrow stuns for a longer duration the further it travels
Skills that follow this design philosophy can be found in both games. Both Nidalee’s Javelin Toss and Mirana’s Sacred Arrow follow this pattern. They are both “skill-shots” - you fire the projectile, and it flies in a straight trajectory until it hits something - with a unique twist, in that they gain effectiveness the further they travel. This means it’s trivial to land the shot at close range, but it will be ineffective. Players can potentially spend months mastering their use – figuring out the maximum range, trying to predict enemy movement patterns, making the best tradeoff between range and actually being able to land it. But from the victim’s perspective, there is nothing complex about this interaction – all they need to know is being hit by an enemy projectile is generally bad (true for all enemy projectiles) and they should get out of the way if they can. Hence these abilities do not create a Burden of Knowledge problem for the victim, and are regarded as examples of well designed skills which create a lot of interesting gameplay.
Nidalee's Javelin Toss does more damage the further it travels
The next post will deal with the concept of "Power without Gameplay".
