sithys

Actually yes! There is an oft-told story about Mario 64 that is described well by EuroGamer, which I will quote here: "in the opening stretch of development all that existed was a moving cuboid, which soon enough was animated and became Mario - more or less as he is in the final product. Depending on which source you read, Miyamoto spent months and months just playing with this character in an open space, or a garden, or early level prototypes, with nothing else to do but move"

If you are going to do something to the player to make the player feel the emotion that the main character is feeling, to create a resonance between the game play and the narrative, then you cannot allow the player to bypass this emotion by any means. The mechanics in this case need to be consistent and absolute. Restarting the level over and over again, or using a cheese strategy, or memorizing the exact order of input for the entire level that has the maximum probability of success should all be impossible. If the designer intends to create an experience then the gameplay must actually create that experience. We must judge designs on their effectiveness at creating the intended experience and not on the intent. When I mean a transparent report, I mean a quick table with rows of information at the start of the map. You could even add flavor text, and I think it would fit in well with the theme of the game. If you could imagine a UI that shows the enemies in the map in a long list with the boss at the top and below that enemies carrying dangerous weapons, with maybe some color coding, and when you select an enemy in this list you get a small bit of flavor text like a local guide telling you how he met that soldier in training years ago and he always carries around a blue gem... This setup would save the user a bit of time having to iterate through enemy inventories at the start of the map, it takes away none of the challenge of formulating a strategy, and it even adds some narrative. What is there to lose? So yeah, I think when I say transparent you are saying quick, so that's embarrassing.

Growth rates are not exactly a life or death game feature. Players are not often going to experience frustration and feel the game is cheesy and opaque because of growth rates. I have gone through entire Fire Emblem games loads of times without ever restarting a chapter because of a bad level up. You cannot say that a strategy, approach, or rule of thumb can be applied to all aspects of a design and ignore the actual emotional impact of a design decision.

I am going to look past your strawman and ad hominem arguments for now. In your last point you say that that I have a double standard but that is not the case. The standard is universal. Either the task is fun in of itself or it is not. Running around in fog of war can be made fun through a set of techniques that require fog of war. Does that make sense? Look at a game like XCOM which uses fog of war in every map. The maps are mostly random so you can't actually look up the solution online, the player is forced to think of possibilities and minimize risk with limited information. The player is never asked to repeat tedious chores to acquire information that was designed to be critical to success, nor is the player allowed to continue through the game if they fail to acquire the skills the designer intended the player to acquire. I am not saying it's a perfect game, I am simply saying that the design has more effectiveness in creating specific emotions and behaviors in the player. Would you please describe your strategy for approaching enemies with varied and opaque AI behavior for the first time?

An interesting question. I would approach this question by theorizing about two different universes, one where the designers decided initially to show growth rates and then later removed them, and in the second universe like in our own they do not show them but decide later to show them. So in universe A the player gets a new unit and is able to see exactly what the stats will look on average by the end of the game, and whether it is worth it to invest in a new character or use a stronger character now. Then one day they take away this ability and the player must use their intuition. Malcolm Gladwell talks about several experiments in his book Blink in which human subjects began sweating after 7 cards were revealed from two different 52-card decks when one deck was slightly better on average than the other. Human subconscious probability processing can be quite good (even if conscious probability processing is awful, because it is). So in universe A the behavior of the players basically remains the same except for a slight delay of maybe 3-2 level ups. Opposite to this, in universe B player behavior basically remains the same except there is a little less delay in the decision making. I think complexity is added but little depth is added or removed. I predict that play testing would reveal that the information might be a little overwhelming for new players without being meaningful, but veterans would enjoy having quick access to the information.

You are comparing apples to oranges. Basically you are saying that fog of war exists and therefore the UI does not need to be transparent. Why even have a damage calculator when you are about to attack an enemy? The player should be expected to do the math in their head! Why not just make the entire thing text-based and in binary, the ultimate fog-of-war. If a designer deliberately obstructs a piece of information from the player, then the experience must be designed around the fact that the information was hidden from the player. This is a very different design problem from transparency. In this case the hidden information is what the designer intended and the acquisition of the information must in itself be a fun task. If you made a game that had fog of war all the time then that game would still need to be fun. I am saying that the acquisition of information in Fire Emblem is in itself not fun. Having to look through every enemy inventory before the match starts is a somewhat meaningless exercise. A transparent report at the start of the match would suffice without taking away from the game, the player still feels tension when they are presented with the challenge of actually dealing with the dangers in the map. The fun is not in being shocked by the danger but by trying to theorize how to deal with it. I can't stress this enough. Nobody plays fighting games because they are physically difficult to interact with. Learning to play a fighting game is like learning to cook while having to sprint 20 miles and back to the grocery store each time you need an ingredient. There is a significant physical barrier that prevents the player from ever making a meaningful decision. And that is another issue altogether. In general, if a computer is good at something, chances are it's not a very interesting problem for a human. It doesn't matter if it's clicking buttons with precise timing or paging through inventories over and over again.

Could you provide some context and explain your rationale behind at least one of these examples?

I have never encountered a single ambush spawn that discouraged turtle strategies. The only thing I have ever experienced from ambush spawns were repetition of content I already beat. Would you like to give a specific instance where the ambush spawn is the only/best disincentive to turtle?

Early game lightning tomes with +Hit in FE10. Really, if there are any characters that miss a lot and I need them to do consistent damage at certain points in the game, the Forge becomes a good option.

All ambush spawns accomplish is forcing the player to replay through unchallenging content that they are more than qualified to move past. The only response to being "warned" of an incoming ambush spawn is to turtle up and hit the end turn button over and over. That doesn't sound like an interesting or compelling challenge. There are much more efficient ways of creating much more interesting challenges for the player to experience.

You are saying exactly what I am saying. They are two completely separate problems. One problem is important, one problem is not important. You need to make the unimportant problems transparent so that the player is not distracted from the important problems. Restarting the game over and over again just to figure out how the AI behaves is not an important problem. Your vision of Fire Emblem is one of guard-rail driving. You turn your brain off and drive straight until you hit the guard rails and repeat that process over and over again until you reach the destination. It's not a healthy approach to driving and it's not a healthy approach to designing cars and roads. Much of the tension is not caused by punishing the player for wrong moves, much of the tension is caused by punishing the player just to punish the player. I am a western player, and so I would be interested in your perspective on what an SRPG truly is. Enlighten me. By the way, not all red men want to kill the green men, only a subset of the red men want to kill the green men. Because they are programmed to. And you can't see code. Furthermore, you are viewing the level from the perspective of a person who likely has years of experience playing a series that was on the verge of being cancelled because of low sales. Experience with the product does not warrant ignoring well-understood design principles.

You can have mental effort, brain training, improvements to skills, abilities, perceptions, intuition, and you can gain experience, dive into alternatives to solve a problem, etc, while also having transparency. Ambush spawns are an extreme example of lack of transparency, because they literally add nothing to the game and they do not accomplish any of these goals, at best they punish the player but there are much better, more efficient ways of challenging the player. Games must always present problems to the player. The designer must make sure that the problem presented to the player is the correct one. Some problems are meaningless. They add nothing to the experience. Such problems should be eliminated. Opacity of the mechanics and the UI and the level design are such problems. When a person turns the stove on they ARE presented with a problem. The problems of how to cook the food they want to cook. The designer of the stove is responsible for making a product that gets out of the way and lets the user solve the actual problem that they are actually interested in solving, making food. The same is true for game design, the problem in Fire Emblem is not one of data acquisition but of data processing. The skill the player is interested in learning is how to respond to changes, not how to identify changes using an opaque UI.

If you think the average player is smart then you would be dead wrong. Don Norman dedicated an entire chapter to the topic called "Human Error? No, Bad Design" in his landmark book The Design of Everyday Things. The chapter "Knowledge in the Head and in the World" further supports all of my arguments. The human brain did not evolve to map controls to burners on your stove at home, and despite the obvious nature of this fact your typical household stove is so poorly designed that even after years or decades of use people still turn the wrong burner on. And for what? I would ask the same question of many games. Your player is going to be bombarded with numerous questions when they encounter a new mechanic for the first time, and the only way to answer those questions is to try over and over again. And for what? What does the player gain? Nothing, the emotions of surprise and conflict are perfectly achievable using transparency. Your average human DOES actually need to be physically shown how things work. That is an observable and irrefutable scientific fact. You can refer to the aforementioned book if you depute this. A player cannot make decisions, and therefore they cannot experience conflict, surprise, or any of the emotions the designer wants them to experience, if they cannot first perceive and then create a mental model of the mechanic. In the example of Super Mario 3D World, the player is introduced to a novel mechanic in a safe place where they won't lose a life if they fall off the flip switches. What would the designer gain by omitting this safe environment? Absolutely nothing. What does the designer lose by including this environment? Absolutely nothing. The player is still presented with a challenge once they have created the mental model required to interact with the game. The player still experiences all the good emotions, surprise, conflict, intensity, change, and all at no cost. I am going to repeat this again and again and again in every post in this thread if I have to, there is a difference between perception and decision-making. It is the responsibility of the designer to make perception trivial so that the greatest number of users can reach the decision-making cognitive stage, because as Raph Koster points out in his brief book on game design, the human brain evolved to enjoy games that model reality in an insulated environment to prepare players for the real world decisions they will need to make.

Can you give some examples from Skyward Sword so I can follow? I played the game but it has been a few years. I remember not liking the whole dowsing system.

Once again you hit the nail on the head with this post. The player has so many questions. All of these questions can be answered simply by introducing the player to the mechanic in an insulated environment and by demonstrating a burning house in a scripted sequence. I agree with the design philosophy that the game should naturally show the player what to do instead of tell them with a tutorial. Show, don't tell. Not showing (and not telling) is not a valid solution, it's even better to tell than to do nothing at all. The player shouldn't have to restart the level over and over again just to answer questions about how the game behaves when it is very easy for the game to demonstrate new mechanics using simple techniques that were available as early as Sacred Stones, as was demonstrated.

Absolutely perfect. Thank you very much for putting together this post. It perfectly demonstrates the right way to introduce something to the player. Mark Brown talks about this as well in his video on Half Life 2: In the example of the swamp level in the first part, I don't think I agree that it is obvious that the bandits attack the prisoners. From the let's plays I have watched the players are consistently surprised and frustrated by the way the mechanics is introduced. Furthermore, the only way I found to stop the enemies from killing a prisoner is to use an unintuitive strategy that involves placing Vika in specific spots while untransformed. If feels cheesy and stupid to succeed and fail at the subobjective. Why have a subobjective that creates negative emotional experiences no matter the outcome? A better way to introduce the mechanic would be to start the player further from the prisoners, but have one single prisoner nearby. Have a bandit that does very low damage just out of attack range but who will always move into attack range and attack but not kill the prisoner. The player is not given control until the bandit is given a single scripted move at the start of the level, and the bandit always has low enough speed so that you can reasonably kill him with any character. This setup is guaranteed to introduce the player to the mechanic.

Things are more obvious in retrospect. If you were to playtest this mechanic I guarantee that only a tiny, statistically insignificant percentage of the population would take the hints and predict that they need to kill the soldier on the first turn. The first order optimal strategy to saving all the houses in this level is to immediately restart the chapter after realizing that the top houses cannot be saved. When you have a first order optimal strategy like that, it indicates that there are flaws in the design. When I mention that a more elegant solution would be to have the information present in the game world instead of in the UI, it's because I feel that if information needs to be in the UI then the design of the world needs improvement. Providing information in the UI should be a last resort. In Fire Emblem, it can't really be avoided because of the precedent set by earlier games. People who are familiar with the games will be able to use the UI even if it is poorly designed. Here you have redefined the word "strategy." Strategy is defined as the "the art of planning and directing overall military operations and movements in a war or battle." Strategy does not necessarily entail being a scout or understanding what a scout does, though good military strategies make use of scouts. A strategist does not need to scout the battlefield, he has help to achieve that particular, relatively uninteresting task. There is a clear separation between information acquisition and information processing. The first is tedious and the second is engaging. You could look at a well-designed spreadsheet of information and have fun coming up with a strategy. If the designers added a new UI that provided important information, such as equipped weapon and stealable items in color-coded rows, where selecting a row highlighted the enemy in the scenario map in a split UI, it would go a long way to providing information to the player. You could even put the text "Scout's Report" at the top of the interface for flavor. If you take nothing away from reading what I am saying, please understand that there are TWO phases to cognition in my example, perception and decision-making. They are absolutely independent, and designs that fail to help the player perceive information should always be improved. Perception is NOT strategy.

Clearly? What makes you say that it is clear? If you sat down with a random sample of people from off the street and did a formal scientific experiment to determine the clarity of the player's implied task, how do you think people would do at taking the non-existent hints? I design and program user interfaces for a living and from a programming perspective most of my suggestions are trivial. A heat-map would not be hard to implement or use, though I will admit it is a somewhat inelegant solution in that it doesn't solve the actual problem, it just treats the symptoms of the actual problem. A more elegant approach would be to make the relative threat of the enemy units much, much more apparent using all known psychological and perceptive techniques, though there are practical limitations to how detailed and varied you can make the assets in a game. Better enemy threat indicators would be an extremely efficient solution terms of time and money for the developer, which is why I mention it. The player can check the inventories, yes. The question is not if the player can check the inventories, the question is whether the player should be expected to check the inventories. Checking inventories is not in itself a fun or even an engaging task. Good game design requires that all activities that the player engages in are by themselves fun. Jonathan Blow talks about this in several of his interviews, if you strip a game down, take away all the graphics and aesthetics, the music, the narrative, the Skinner Boxes and the loot treadmills and the XP systems and the stat systems, is the game still fun? The goal of the designer is to start with a fun experience and then add all the trappings that make it even better, not start with a fundamentally boring experience and trick the player into wasting time with it using psychological hacks.

Well it is compelling, I understand that. But it needs to be compelling in a responsible way. In my example it is literally impossible to respond to the soldier on the top level once he starts burning down the houses up there. It doesn't matter how compelled the player feels to rush up there and save the houses, there is an archer with 7+ crit chance standing guard over the area. The only way to save the houses after turn 1 is to restart the level and kill the soldier on the first turn. Is that what the game designer intended? Surprise is good, tension is good too. That's the whole point of adding mechanics into levels. Tormenting the player is not good. Disrespecting the player's time is not good. If a game designer thinks to herself "how can I create tension and surprise in the player's mind without frustrating them," then that designer is more likely to create the type of experience that the player actually enjoys. It's a matter of intent. If the designer instead thinks: "how can I make this as hard and punishing as possible?" then you are going to end up with a very different experience. My point is that it's not black-and-white when it comes to designing mechanics and that surprise is not a bad thing. Surprise can be a very very bad thing. Consider ambush spawns. You fight your way to the end of the level and are about to face the boss, then enemies randomly appear on the enemy turn and kill all your healers. SURPRISE! The benefit of surprise is not in the moment of shock but in the moment where the player must stop and re-evaluate their assumptions. The player must engage in root cause analysis, they must shift their mental paradigm, they must see the true goal that they have been unable to perceive up until that point. The goal is to make the consequences of a surprise meaningful while making the shock just powerful enough to create an emotional response without punishing the player. Why would any designer want to punish the player for being surprised?

A prequel to the Tellius saga would be cool, where you play all the legendary heroes and do all the stuff you hear about in the story.

I am playing through Radiant Dawn again right now for the first time since 2012. One thing that I notice is that a lot of the levels can be extremely difficult for very stupid reasons. Let me give a few examples: Sometimes there will be packs of enemies that stay perfectly still at all times, while other units will move only when attacked, while others will move if you cross a certain threshold, while others will move all the time. Sometimes several of these 4 types of units will be nearby, making it difficult to make decisions about how to advance. This is especially true for bosses. Lots of maps have unique mechanics that are only used once. For example, there is a map where you must run around putting out fires in a village. On said map, there is a soldier just above the starting location that you need to fly up and kill with a flying unit, however the player is unlikely to know this until they see the soldier starting fires. Enemies with special weapons, such as hammers or horseslayers, can appear in other packs of enemies with normal weapons. The player can sometimes be surprised by the death of a unit because of these weapons. All of these examples have one thing in common, the game is not being transparent. The player must do some math or search the inventory of every unit that might attack in order to gather information. From a game design perspective, some of these activities might be consistent with a vision of the player as a strategist and therefore might be desirable to some degree. The questions arises as to whether or not this experience is always a positive thing for the player and if not, how this experience can be remedied. So the first question, is this a good thing? Fire Emblem: Radiant Dawn was widely criticized upon release for it's extreme difficulty. People who play Fire Emblem already, fans of the genre, might take the lack of transparency for granted. Ask yourself, if you gave Fire Emblem to your grandparents and it was the first video game they ever played, do you think they would enjoy it? Would your average random person on the street? Why or why not? Game designers must always ask this question: what do I expect from the player? Fire Emblem expects several core abilities from players. The ability to solve puzzles involving long-term and short-term goals and prioritize objectives. The ability to keep units safe while meeting combat objectives. The ability to recognize strategic advantages. The ability to gather information about the current state of allies and enemies. The ability to respond to sudden changes. These are just some high-level expectations, and this list is not exhaustive. The issue of transparency affects all of these player expectations in some way, so transparency in my opinion is is one of the most important features that Fire Emblem is lacking. Needing to restart the level because of a mistake is reasonable, needing to restart the level because of an opaque user interface is less so. So for the second question, how can more transparency be added? In Awakening, bosses that don't move appear with a range overlay that indicates that they don't move. There is also a button that causes the range of all enemies on the map to appear in purple, so you know if you are safe from danger (except ambush spawns). This is a step in the right direction, but I would like to see even more transparency to aid the player, especially when advancing on an group of enemies at the limit of movement range. Imagine, instead of a simple purple overlay to show danger, there was a blue-to-red heat map of danger that would highlight extremely dangerous spots and less dangerous spots. It would be probabilistic and not guaranteed to be completely accurate because of the AI's movements, but it would help the player quickly identify danger spots they might have overlooked. Will that make the game too easy? Well no, what happens if that red dangerous spot is the spot you need to be in to kill the thief headed for the chest? Interesting conflict and triangularity do not need to be reduced by improving transparency. So before I tackle the issue of new game mechanics, it might be a good idea to watch Mark Brown's video on Super Mario 3D World: If a map introduces a brand new mechanic, it should probably do so in a way that is safe for the player. The player should be relatively insulated from punishment if they don't grasp the mechanic right away. In the above example, I would have redesigned the village so that the player starts with some enemies nearby that they are encouraged to fight immediately. At the end of the first turn an enemy soldier just outside of the player's range would run into the player's range and start a single house on fire. The player would be able to react and thus understand the core mechanic. Then once the player understands the mechanic you throw in a twist that makes the player think. I might have placed some enemy magic users on the top level to encourage the player to send a flier up to fight them, and then spawn the soldier on the top level on the following turn instead of at the start of the level. So for the last example, I would redesign the UI to provide information about bonus damage on or above the units. Since most enemy units do not deal bonus damage, the UI would not become cluttered. If a unit has a hammer, a little red icon of an armored knight with an explanation point, or some other icon, would provide enough information to the player to quickly identify the threat. Once the threat has been identified, the player must decide how they want to dealt with it. Dealing with threats is interesting, but if the threat isn't transparent and the player loses a unit as a result, the player will feel that the game is too cheesy. So what are your thoughts on transparency? What UI changes or level design changes would you make to improve the player experience? I want to hear from you.

Radiant Dawn by far. Echos of Daybreak brings back many good memories of this game.

If I woke up and discovered that I had never played any fire emblem games, but I still somehow knew what one to start with, I would start with FE7 for sure. It had all the features from previous games which had withstood the test of time. Some of the design decisions made in FE7 made the game difficult in exciting (and sometimes very frustrating) ways, especially in the levels with "other" units that you need to recruit or save. So if you can get through FE7 without quitting the series, then you should be prepared for the crucible that is Radiant Dawn. Some of the design decisions were questionable though, be sure to read up on how to recruit Canas because he is the only dark magic user in the game and he can only be recruited in a side quest that only activates under certain conditions. I didn't even know Canas existed until I bought the players guide (this was back in the early 2000s) and I was very confused at all the vendors selling dark magic spells that nobody could use. They did a lot of things right in Awakening, I will admit. I wish all the previous games had the distinct purple and red overlays for enemy ranges. The level aesthetic design in Awakening is also extremely polished. Personally, I would save that game for last as a dessert. It plays way differently compared to other games in the series, and going from Awakening to FE7 would be a pretty steep difficulty shock.

Thanks for replying DragonFlames. I appreciate the participation! I was afraid that this thread would fall off the page without any replies. I don't remember having many issues with the Dawn Brigade chapters in Radiant Dawn, but I do remember that that endgame chapter basically required using Nailah to kill the thief if you didn't want to restart the chapter 20 times waiting for the stars to align. Maybe it wasn't the best example of balanced triangularity, but it is a good example of very imbalanced triangularity. It is important that the expected value of the various outcomes be the same. In this case the XP reward of using your weaker units is not nearly enough to justify the effort. I don't think the challenge of playing through the game with weaker units can be considered a reward. The fact that arenas are nearly risk-free is a critical oversight by the designers. The arenas should give less XP and have a higher chance of killing a unit. The risk of using the arena would create the difficult decision scenario that designers seek to create. As a player you would want to use the arena when a unit is about to level up and you can just do a match or two, which carries less risk than grinding. I believe that a fair challenge should be part of the game by design and the player should not have to adhere to arbitrary self-imposed rules to create a good experience for themselves. Also, I think the reward for disadvantageously attacking a unit is in some cases balanced and in other cases you have no other choice. I find myself doing it in situations where it is the optimal move. Doing so often opens up a safer position for your units. Yes, you are getting the theory right, but in this case I think this suggestion might create perverse incentives for the player. Also to respond to your comments about the magic/javelin/throwing axe, those weapons are not in the slightest bit balanced. The hit-rate for ranged melee weapons should be much lower so that there is more risk in using them. Mages should have a lower chance to dodge. They should be glass cannons, not super units. Just a few numbers tweaked here or there would suffice in most cases. At the same time it is important to remember that the enemy placement and stats are designed under the assumption that the player has a certain power level. The enemy usually has lots of units. If you equip a javelin to attack one with range, then get attacked 4 times by melee units that are using standard weapons, then you are at a significant disadvantage while defending against those 4 attacks. So there is some risk even in the base game, but I will admit it could be improved. I noticed other areas where the designers of FE7 use triangularity. On the pirate ship level in Eliwood's story the entire right half of the map is covered in mercenaries, perfect for farming XP with Florina. However, there is a single archer in the pack and one mercenary with a lancereaver. So the player must manage the risk imposed by those two elements while farming. The player is incentivized to stop and plan out a strategy for mitigating the risk, whereas without those risks the level would be significantly more monotonous.

Welcome to Elements of Fire Emblem, a series of topics about Fire Emblem from a critical perspective. In this first topic I want to talk about triangularity, a concept described in Jesse Schell's book on game design. Triangularity is a word that Schell uses to describe "balanced asymmetric risk" in his chapter on game balance. The player is one point of the triangle, a low-risk low-reward choice is the second point, and a high-risk high-reward choice makes up the third point of the triangle. To quote Schell: "One of the most exciting and interesting choices for a player to make is whether to play it safe and go for a small reward or take a big risk to try for a big reward." He also says that monotonous games "can quickly become exciting and rewarding when you add a dash of triangularity." Enemy thieves and brigands provide a source of triangularity in many Fire Emblem games. The player is presented with a choice, play it safe or take a risk. The guarantee that everyone will survive the chapter is the small reward, the contents of the chest or village constitute the big reward. Only when the player risks being caught out of position is he or she able to chase down a thief or brigand before they make off with the treasure or destroy the village. Even if the player theorizes a good strategy the risk still ends up being higher than playing it safe, if only slightly. I also want to talk about a specific chapter from Radiant Dawn, the Part 1 Endgame. Consider these two questions: What other aspects of Fire Emblem embody the quality of triangularity? What mechanics would you add to Fire Emblem to specifically create the quality of triangularity? I will update this first post with people's responses to the first question and I will try to reply with an analysis of your answer to the second question. But feel free to comment however you like!