|
|
|
| XAVIER
BOISSARIE GAME
DESIGN > |
GAME
DESIGN What is a game? That’s a question that many psychologists, sociologists and philosophers have worked on.... As a game designer, a game takes a mechanism that exists in real life and removes its overwhelming complexity. Life is full of extremely complex rules, ambiguous terrain, people that evolve continuously on their own... A game is much simpler. The rules and terrain are defined, and they rarely change from the beginning to the end of the game. The problems and objectives are also clearly defined. It is easier to capture someone’s interest and motivation in a game than in real life, because it’s simpler and clearer. From there, we can create certain phenomena like the immersion of the player in a game, that in certain extreme cases can cause players to become almost hypnotized by their game, not see time go by, and spend all their energy resolving objectives defined by a game that they have mastered and extremely well understood. But what about the pleasure aspect that may not be very evident in real life? It is exactly the same mechanism as in real life. Motivation and action is related to the pain/pleasure mechanism, and the compensation for the player is pleasure. The reward or the pleasure can be positive, like in real life, or it can be negative, in rare cases. Pleasure can be immediate gratification, the pleasure of playing, reacting, the sensation of a flow of images or of action. Or pleasure can be deferred based on a long term objective, and we can experience pleasure from the progression towards that goal. What is primordial in a game? A terrain, rules and a goal are primordial. Atmosphere is important initially to attract the player and to make that person feel comfortable in that universe, and so if it isn’t attractive, the player may not get into the game. Now, there are a lot of examples of games with a frustrating graphics quality that are really motivating for the player. For example, with Tetris, you can have a very simple game that is really captivating. So the principal element is to have a clear objective, clear rules, and an understandable terrain. This can come into conflict with a program that is too loaded down and that will hide the trail. For example, if you have too many visual elements, and all don’t have equal importance in terms of information. There can be a bit of gameplay that is “Hide and Seek,” but in most cases, I avoid visual competition with things that are not necessary for my game. Instead, I try to privilege useful information. What is the role of gameplay and narrative development in games? There are consumers that need to be immersed in game, to go into a universe, and for those people, narrative development is very important. The extreme of this type of game are linear action adventure games where you take the role of an actor, or games like Riven or Myst where there is a real effort on detailing the universe. On the other hand, you have a family of consumers that lend little importance to narrative development but are more into gameplay itself, the interest of the game. What’s really important in a game like that is the hierarchy of the information. I personally put gameplay first. How do you define and develop gameplay? What’s important is to put pressure on the player. That’s primordial and that’s planned, whether intentionally or innately. I think a lot of game designers don’t even pose themselves that question. There are two domains where pressure can be exerted in a game. The first one is in regards to the goal of the game and the player’s progress in regards to that goal. That’s a positive pressure or positive motivation. And the second one is in regards to the short term problems and obstacles, or negative pressure. Surviving an attack for example. Those two kinds of pressures need to be carefully dosed. Too much pressure, the player no longer believes in it. Not enough, the player gets bored. That’s why games are progressive, and the scores and stats are kept so that the player can see how his performance improves. What about the balance between positive or negative pressure? In a Shoot’em Up, it’s purely negative pressure. If you do nothing, you die. In a purely constructive game, say Sim City (but without the potential of destroying your city), it’s only positive pressure. More often than not, it’s a balance between the two. It’s not a judgment value, its a question of motivation. Within positive pressure, there is a whole variety of different types, and in negative pressure, there is a wide variety of different problems. Positive pressure is always linked to potential problems! For example, you have a delicate system in balance, and your role is to fix the problems that threatens your system, like a captain on a ship. There is a mix of positive and negative pressure and motivation in regards to progress. If I fight against someone and win, I get more points, more territory, more glory... That’s also widely used in games. So a game is really a fabric of these different kinds of motivation working together? You have to make clear choices regardless, because if you have too many, the player may lose sight of the objective. So you need to put the emphasis on a particular point. What about games that have ambiguous or multiple goals? Well, you get closer to real life. If everyone can choose their own objective, open games like that result in purely individual experiences. It is less a game that you can share with your friends or that you can play on a network for that matter. The games that seem to work best are ones with very clear goals. Life - the game where you can do anything! Some people criticize game because they generate fewer emotions relating to human rapport or character identification? The critics probably weren’t looking at the right kind of games. In a game, all emotions are possible, but one game will not supply all of them. If you want human rapport, you can go play on networks or the internet. They allow you to incarnate a character, define its characteristics and engage in role play. These role playing games are controlled by people. A character is really someone behind his or her computer. Of course we can always want more in regards to what exists. When there is enough people who want more from a kind of game, then a new type of game will soon emerge. Do you see new genres evolving? All games were pretty much invented before the invention of video games. What changed was the framework and interface of the game. The motivations and game structures preceded video games, like manhunts, hide-and-seek or puzzles. There are some that exist that are not exploited by video games because they don’t lend themselves to that media. Realtime 3D presents some interesting potential in video games particularly with games based on physical simulation. It’s is not very frequently used, since it’s CPU intensive, but it’s used in sports games to control the ball and its interaction with other elements. There is still room in that domain. Frederic Raynal recommended that there should only be one innovative aspect to a game at a time, either in its interface, gameplay, aesthetic... I whole heartedly agree. It’s a very good rule to respect. Players can’t learn an entirely new language - visual, narrative, interface, technique - each time they play a game. So you need innovation but not everywhere simultaneously. There are games like Populous, Doom, Warcraft, and Tomb Raider that were downright revolutionary in multiple domains and were commercially successful, but there is a danger to being too innovative. It’s easier communicating to the public and the press about one area of innovation rather than multiple ones, since that gets downright confusing. |
||
| GAME
DESIGN > SIERRA SANGRE > PROCESS > |
SIERRA
SANGRE What is the purpose of this prototype? The objective of this prototype, tentatively titled Sierra Sangre, is to demonstrate the potential of the software - a game intended for a large audience - where we demonstrate the validity of the game’s concept, its gameplay, its points of interest, its ergonomics and the pleasure it provides. This is a speculative demo created to sign a deal with a game publisher or distributor. Is Sierra Sangre in an established game genre, and if so, what genre? What innovations do you bring? It's an action/adventure game in the widest sense. In a lot of action/ adventures, the characters are sometimes a little stiff. So in this demo, we are trying to show characters that have a certain richness - with the capacity to evolve, even on their own - with their own activities and objective. Action/adventure characters are usually stereotyped. You find them at certain places. They have certain professions. They are either enemies or friends. Sierra Sangre has a certain aspect that we find in role playing games. In other words, a character can be both friend and foe, it all depends on the types of relationships we develop with that character. Did we help him at a moment in time? Or did we get in the way of his objective? So there is an openness in the relationship between a player and characters in this universe. What game design issues did you resolve with Sierra Sangre? The first issue is background: the concept, the environment of the game - creating an atmosphere that a player wants to dive into. The next one is interface and gameplay: how the player is going to interact with the software, its ergonomics, how you access and control the characters, manipulate the camera, organize the actions of the game. The third aspect is scenario: the objectives of the player and all the things that will define his motivations, and the goals, the pitfalls, the problems, things that will get in the way of his goal. With this demo, are you able to develop the third aspect, the scenario? We tried to cover all three in parallel, and effectively, the narrative aspect is the one we see last, because we need all the elements in place to gauge the interest of the game. So the scenario is often developed last and requires quite a bit of work and fine tuning. In certain types of games, like action games, the objective is relatively self evident. If enemies attack you, and you do nothing, its over... You need to act to survive. The objective is self-evident, so narrative development in this case dresses up the objective. The objective is already present for the player, so it doesn’t pose a problem in a demo. In dialogue sequences, that’s a little less true. Dialogue takes place in relation to a specific objective, so there is really a hierarchy of objectives. Local short-term ones, like, "I’m thirsty, I’m wounded, I need to cross this river, etc." Things that need get resolved quickly. Then there are long term objectives where the player needs to put in place searches for certain objects, resolve an enigma, construct a relationship with a character, bring him something to obtain something else. Short term objectives are easy to put in place in a demo, but the long terms ones, require demos of bigger amplitude. When you get to the point where you can introduce a long term objective into your demo, you almost have a finished level in a game. In Sierra Sangre, there are mostly short term objectives, but there is one long term goal. It’s an obstacle course, where you need to find clues. What about character development in Sierra Sangre? Well I worked quite a bit on character personality, but I can’t go into that very far since its proprietary. Then there’s work on the AI (Artificial Intelligence), for example, determining a character's effectiveness in regards to a particular objective. After the conception of the AI, which is purely theoretical, we construct the behaviors and spend time fine tuning them, to increase or decrease a charcter's efficiency. You often have to decrease it, because certain characters - engaged in destructive behaviors - can be much more efficient than any player. For example, if I assign a character a behavior to attack from behind, well it can be a real pain to play in a game where all of a sudden you are killed from behind. The majority of shoot’em ups avoid attacking from behind because its annoying for the player. You don’t want destructive characters to be too efficient so as not to destroy the gameplay. So the goal is to create characters that are intelligent, but not too intelligent? It’s amazing how much more efficient the computer can be than a player, even if the characters have the same chances, the same displacement speed, the same speed of the draw, the same weapons, the same precision in target range... In a Shoot’em Up like Doom, where the precision is linked to the direction or pace of the character, precision can be uncertain, but if the player can use a mouse, precision is increased dramatically. Even if we have the player with increased precision, and the adversary with decreased precision, we can still create characters that are really effective at resolving a particular objective. So you need to find equilibrium in the system, especially if the system progresses over time. So in reality, AI is really a simulation of intelligence? A game's system is never as complex as reality, so a character’s intelligence is task specific? Yes it’s an expert system, not a unpredictable behavior, though many times behaviors can be interpreted as intentions that do not really exist. There is a logic behind the AI that is the logic of the person who programmed the behavior. Reasoning is external to the characters, accomplished by the program’s conceiver. But a behavior is AI in the sense that it resolves a problem or finds a way to be efficient regarding an objective. There are lots of methods for constructing AI. I use a method where I fraction everything that a character can do in terms of behaviors that are called based on local criteria. So based on the perception of reality by a character, and based on a certain number of criteria, a character will choose to act this way or that way. It’s a very classic system. A character is an automaton with multiple skills, but a character’s intelligence, or rather the performance of the character, is tied to its capacity to adapt to particular circumstances or its understanding of a particular situation. In Sierra Sangre, what is the situational understanding of a character linked to? It’s physical surrounding, the distance of a particular objects or characters? Yes, the character perceives his surroundings and processes the information. AI is very much inspired by reality. Certain characters have behaviors that are much more complex that others. Some have very simple repetitive actions where they spend their time throwing rocks, while others spend most of their time reacting to the central character. Is the idea to give all characters equally complex behaviors? Let's say some characters are integrated and well developed, and other still need work. Once you have the first one though, it goes much faster. Not that they are all identical, but there is a base structure on which we can easily put variations. The characters are all pretty different, so they will have different responses to similar situations. The characters are very emotive, showing grief or joy depending on their success. How did you achieve that, and how does this relate to the Demo’s AI? First part is a trade secret, a fabrication technique. Their emotiveness reflects the internal life of the characters. We put a lot of work into the individual personalities of the characters. Based on their personalities, you need to relate to them accordingly. Is the character friendly? Is the character surly? The scenes result from the behaviors of different characters, but it remains coherent because each character has an objective that is pretty well define. There are still a lot of surprises, even for us. With individual criteria and objectives, each character based on his personality will react to a situation differently. There is internal monitoring, totally transparent, that occurs, and even with established rules, we can’t always predict exact outcomes. So we know there will be surprises, but not exactly what type of surprise. In addition to that, a character’s response will also be influenced by a player’s relationship to him. If you shoot his hat, he won’t be happy. If you give him a glass of water when he is thirsty, he will be thankful. There is a strong logic behind it, and the characters have a "relationship memory," so you can make relationships evolve over time based on the player’s interaction with that character. Was there a strong effort to make the characters facially expressive through morphing of the mesh? We use a lot of facial animations because our game relies heavily on behaviors and dialogue, but that might be superfluous in a lot of games... They make our characters more lively. We used a system that required more polygons on the face, but that wasn’t excessively costly either, since we used Gouraud shaded characters. How many polygons are dedicated to the characters? It varies with a character's distance from the camera - there is mesh swapping that ranges from 300 to 700 polygons per character. We are still testing it, but it will improve performance if it is applied to every character in a scene, since they have the most intensive polygon usage in the scene. Since it is based on the distance of the camera, if all the characters are simultaneously nearby, that may slow down the frame rate, but that situation has been programmed to not happen in our game. All characters will vary their distance to the camera. Then they attack from behind, because it’s fewer polygons to calculate! |
||
| GAME
DESIGN > SIERRA SANGRE > PROCESS > |
PROCESS What was the structure of the team to produce this demo? We had a very small team. A graphic designer working on paper, a graphic designer/ animator working in 3D, a writer and a director. My role is director, but it is really the role of assembler. The writer and 2D graphic design only participated at the start of the project. Compared to a traditional demo making team? It’s a reduced team size. We would have needed two graphic designers and two programmers, full time. So technically, you programmed this demo? The low level programming was accomplished by Dev, while the higher level programming was by us. We mostly used Creation's low level building blocks, aside from the "Unlimited Controller" to manage character animations, which is very well done. How many hours were spent on this demo? Like many heads of projects who work weekends, I’d rather not know. What was the budget? I can't divulge that information, but evidently we saved money with fewer team members working on the project. Can you compare the procedure for preparing Sierra Sangre with Dev to a traditional production process? Well the conception stage, where you begin working out the structure is still on paper. That includes: - The structure of the demo - Obligated milestones - Scenario - Artistic intentions But the construction phase with Creation’s environment is a bit freer, since it is not a tool dedicated to just one application or one particular type of game. If production is less restricted, you really need to develop your own production methodology for your project. Someone with very little method may find himself floundering a bit in an environment that is wide open, but it is the same situation that you can have with a 3D modeling/animation software, or with any creation tool for that matter. Compared to the traditional process, Creation allowed us to move more quickly to the PC allowing us to experiment during production. How did Creation change your work as a game designer? Creation gives game designers much more direct access to the fabrication. Speaking from personal experience, I understand coding from a distance in terms of constraints. To work with programmers, you have to understand their language and their constraints. So my work was all on paper, all about creating the rules, schedule of conditions... Now I intervene directly on the game. It allows me to understand the constraints in manner that is much more upfront than the traditional game production method. It also allows me to get new ideas. So for me, it was an enormous change. You try out your ideas directly? Exactly. Once you have integrated elements into an environment, when you get an idea, you can immediately evaluate it. You have fewer consultations to get it done. In fact it really reduces the exchanges necessary between the various professions that make up a game design team. So you also reduce chances of miscommunication. Is it critical to have a plan in place so that when you are creating the demo, you know exactly where you want to go, or is experimentation important during the working process? Every time you implement a new element, you are experimenting, that’s the interest of Creation. We also have the possibility of integrating an element and testing it in real time, and although it’s not recommended, you can even edit the script while it is in run time! You can immediately see the effect of your modification. With Creation the development phase is less sequential. In traditional production, generating code takes much more time. There is less room for experimentation, and the process is much more sequential. A certain number of tasks have to be accomplished before we can see something. The way we currently work, our graphic designer tests all the elements, models, textures, and animations in Creation before I integrate them into the project. If you have a developer coding specific behaviors for a project, he can easily work on them on his machine and test them out before integrating them into the project. How do you feel about the statement that you need to come from the programming camp to be a good game designer because you need to understand the constraints? I am in accordance with that, and so that is why I will never make a good game! (Hearty chuckle...) No, the ‘programming camp’ isn’t really meaningful, because programming is not a sect. Between the programmers there are different camps. They can be obsessive about different aspects like clock speeds, rendering speeds, hardware, cpu management, physical simulations, AI, collective intelligence, artificial life, camera movements, special effects, etc... So there are different programming cultures, and some are fascinated by gameplay and gameplay mechanisms. Peter Molyneux comes to mind. He isn’t a developer who is all about graphic performance. He works on rather simple and effective methods of gameplay and works with others on the other domains. Programmers make better games because they get their hands dirty. Does a tool like Creation give designer access to this programming culture? Yes! Creation covers the major bricks of C++, and the elements that are missing will come soon. Creation's construction principles will also facilitate the work of a large number of programmers who work on higher level programming (developing levels and filling out the environments), and they will be able to get to that stage without having to learn C or C++. With Creation there is a simplification of their work with better efficiency. I know a guy who works in C, C++ and Assembler, but he finds it easier to do certain tasks in Creation. Of course the cerebral elasticity of the programmer also comes into play, because after he has been formatted by a computer language, will he be ready to go to another world? If he already has all his tools calibrated, he may be reticent about going to another method of development. But someone like you who is not a programmer, but is in high level programming of this demo, was it critical that you had the programming base before you could make this demo? No, but you need a working method to know how to get through it. There is no predefined way to work in Creation, so that base is important. What is your method? - Avoid specific cases, like in traditional programming. Try to make the greatest number of things as generic as possible. - Then its a question of readability. Organizing things to be clear. Creation accomplishes that very well. Everything is there that you need to organize your project, to find what you are looking for quickly. Once you have a demo, what is the procedure to continue its production into a full fledge game? Well the demo has allowed you to evaluate a certain number of elements, including feasibility, scheduling, staffing and tasks of each team member... So if you lack the capacities to continue development in-house, you need to present the demo to people who will be able to finance the game development like a publisher, though you can have quite a few different financial partners involved. Usually you have development and publishing tied together to facilitate development. Publishers are critical for the marketing of the game. Can there be commercial outlets for independent designers that are different from those of video games, or is it going to a be an epiphenomenon? The commercial value of a product can be tied in certain ways to its media impact, its capacity to move the public. Afterwards, it is a process of selling on the part of the creator. If there is tool that allow creators to produce content that will move the public, with no commercial contracts behind it, evidently the media is going to jump all over it. So if you are asking if there is a motivation on the part of the creator that is not based on financial renumeration, I would say that the power to move the public is already a great reward. >> Back to the Top >>> Back to Insider's View Index This article may not be reproduced in any format without the authorization of Virtool S.A. |
