Frederick Raynal stopped by Virtools headquarters on October 21, 1998 to talk Game Design with Bertrand Duplat and Frederique Krupa. PART ONE: Game Design PART THREE: The Future of Game Design	Game Design Tools and Methods How did you get into game design? I've always like to construct things. When I was little, I like to construct board games, establish the rules. When I was 12, I took a correspondence course in electronics, because it allowed me to construct little gadgets. So I just fell into it. My early games were things that made an LED blink. I added a speakers, etc... So I put all these things into a box and called it my computer. For me it was always about constructing games, even if at first you went beep and a light blinks... Through electronics, I met people that were already working on the first computers, until the day that I was able to buy myself a ZX 81. It was the first computer readily accessible, so then I really got serious. So my first programs were games comprised of 10 lines of code. My first game was called Laser. I even made the box by hand which I've kept, a line would go from the side of the screen and when it neared the center, you pressed on a key to try to get it. I taught my friends how to do little things, how to make a little adventure games. After that came characters. When I turned 18, my dad saw that I was getting seriously into electronics. Being a merchant, he decided to start selling computers in the early 80's. He was there for the arrival of the Spectrum and the Atari. I decided to stop studying Advanced Math in college, but I did finish high school. Back then, I wasn't going to learn about programming in college. There are some notions of higher level math I would have really liked to have learned, but now there are people that do that much more efficiently than I could. I don't regret my choices. At my dad's shop, I also continued to program, and my first professional programs were Minitel emulators, the first for CPC 464 for Digitelec, followed by a program to transfer data by modem, a LAN server, with a dedicated language entirely programmed in assembler. Really crazy stuff, and at the same time I was doing my first commercial game, Robix 500. I made it in a week to have some fun. We packaged it by hand with my dad. It was little game, we sold around 80 copies. Popcorn was my first big game. During the day, I worked in the store and at night I worked on Popcorn. The graphics at the time were in CGA on a PC 1512. I worked on the gameplay and on the design. It was my first game that became relatively well known, complete with keyboard interruptions, and other aspects that perhaps didn't work so well, but it was the first game that I was able to really get out there. My big weakness - although I did everything including design the box - I would just put the games on the shelf, since the creation was the pleasurable part. For a long time, I thought everyone did the same. I didn't think about bothering to try to sell it. I could have started selling my games from the age of 16. They were largely sellable at that time, but it just didn't occur to me. Popcorn was released as freeware. The game went around the world. We received tens of thousands of letters from everywhere telling us it was the best game. Compared to the PC version of Arkanoid, which was really dreadful, Popcorn was 10 times better. We acquired quite a reputation in the process. At that point I was drafted to the Army, which probably saved my life. I was going crazy, turning into a real hermit. For two and a half years, eating and sleeping by necessity, communicating only with my machines. It was fantastic programming experience, but I couldn't talk to people anymore. How did you make your first games? You mentioned there was an early division between programming and graphics, programming and game design... Each game was developed from the point of view of gameplay and technique. I worked in Assembler for a very long time. At Infogrames, I developed their graphic libraries, working out routines that ran a hundred times faster than theirs. Afterwards, they gave me more interesting responsibilities. I've always approached game design from a technical standpoint. Now it's less direct, no longer: "how are we going to display that on-screen". Now it's more: "We are going to do a visual like that, we are going to manage these elements this way." My best games, I've noticed are the one that come out in one shot. There's a spark and the whole idea is created very quickly, sometimes in 10 seconds. They are detailed later, but the initial ideas come very fast one after the other. Afterwards, the group aspect of development is a little more tedious and time consuming, but it allows the details to be worked out. The goal of the game, the gameplay, the technique have to come quickly. You were one of the first to design games in 3D. What are the changes in the evolution from 2D to 3D? No, there were others working in 3D before me, things already quite evolved. Virtua Fighter came out at about the same time. There weren't many 3D characters, Robocop came out at the same time as Alone, and was quite different. Alone's environment was 2D, hand drawn. We built a wireframe model that was digitized in wireframe that was then filled in by hand by my wife. What has changed going into 3D, now that it's an absolute standard? Is the transition from pure 2D difficult? There are some negative points to the transition. I regret the fact that 3D doesn't make strong use of the viewer's imagination, now that the tendency is to be ultra realistic. Hopefully this period will go by quickly, because we are at the technical level where we can do Realism. We are going a little overboard to make the world more realistic. 3D removes the need for imagination. It would be nice to do things that change a little. I do think we will continue to see 2D games though, even if some elements are in 3D, games that are more like cartoons. 3D environments all pretty much become the same in the end. You know how to relate to it instinctively since we exist in a similar volumetric space. We are already familiar with constraints of reality. 2D games allowed for more freedom from those constraints. But even with cartoon/fantasy 3D, you are still face the same constraints. So we will have to be submitted to the ultra realism phase for the next few years. Right now, I think it must be awfully intimidating to try to start off in game design, but in a few years, when we will have gotten over realism, games will go back to a more cinematic role and offer more options for newcomers. What would you recommend to someone who was getting started in game design? I would tell them to learn to program, and after that, they will see. Which programming language should they learn, C or C++? Good question, You have to learn C++ because it teaches you a way of looking at things. Already just the notion of Objects.... You have object in the program and objects in the games. On the other hand, the heaviness of C++ - that it is easily portable and everyone can recover your sources - is really the inverse. If you really want to master your code, you have to learn C. Start with C++, then code in C. During the Conception/Prototyping stage, how do you work? I make sketches and notes for myself. My tools are paper, pen and my PSION to make things legible for everyone else. After I program bits to see what happens, to test to see if I like something. Creation is good for that stage. When we want to attack a new project, it takes so much time for us to put in place the rendering engine to make things move. This is to say that we can't make a game in one evening. But with Creation we can get things functioning, even if you don't have the real graphics. With Creation I just used cubes and basic forms, because a game doesn't need the graphics to make the gameplay function. It permits you to immediately build something very quickly to have an idea of the way objects are to be placed and communicate it to the graphic designers to make something a little more evolved. It allows you to have something up and running quickly. I used it to build little games in a weekend. What is your personal process for conceiving a game? You mentioned you like your ideas to come in a flash and that if you had to labor over it, it wasn't a good sign... The game needs to fall in place very quickly. Personally, if a game works well, but a link is missing, I may work on it for a few days or a week. For Alone and LBA, it was a matter of seconds. The global vision of the game was a snap. How has your role evolved over time? Are you more programmer, game designer, project manager? I would like to get to the role of game conceptualizer and spend more time coming up with ideas. That said, I like to program as well, so I still want to be involved. I can't do it all, but I like to keep my hands dirty. So really, I am more a project manager and conceptualizer. But I'd like to become a conceptualizer with project managers, so I can make more games. I have ten games that I want to do, but only two will see the light of day, because they will have expired by time we can get started, seeing how long it takes to develop a game. Who participates on a game design production team? It depends really. We do all aspects of game development in-house. Some companies subcontract the graphics for example. As for internal game development, everyone gets involved. First we have the creator who works on the game's conception with the game designer, writers and the programmers. We now have a lead programmer since I can no longer cover all the bases. There are script writers depending on the style of game, musician who works on music and special effects, scenario programmers who work on the scenario code, not the engine - they implement the game design. Graphic designers are used to make the backgrounds, models, animations, prerendered sequences, supporting materials, textures, and publicity screenshots for marketing purposes. Each person has a role but is also responsible for giving ideas and feed back on the game? Exactly. We also have to be able to say Stop. When we start production, you put a halt to new ideas to be able to finish the game. Of course, we don't refuse to throw out things if they don't work. It's the French way of doing things. Americans work more rigidly. Everything is put on paper first, then its pure production. We allow ourselves to make necessary changes. Are there different ways of conceiving and developing games? Yes. Ubisoft, for example, works another way. They have a pool of programmers, a pool of graphic designers, and a group of game designers. They command work from each other, especially the graphics department. What's nice is that they all work on all the projects, everyone keeps busy, and it's very efficient. But then your sense of personal accomplishment can be diminished since you don't really see your contribution to the whole. We work more like a groups of musicians, each with a particular role. This works well for us, but other places subcontract out entire sections of game development. What are the principal steps in game design? And how long do they take? We take about a year for a sequel or 18 months with a team of 12 people. It's pretty average. We make games that are pretty expensive, because since Alone in the Dark, expectations have been pretty high. So we have always had to put in the whole package, but aside from Alone costing $400 thousand, our games cost $1.5 to $2 million. When I worked alone, I tackled the programming of the engine, the game design, and mixed the steps... Now we have the conception stage, the prototyping stage, and the production stage. We mix alot of the conception and prototyping, and dedicate around four to six months. The rest of the time is production. What's the general evolution of game development? It's not with 2D to 3D. At first, you could do a game all by yourself, doing the programming, graphics and music. Then the job split into programming and graphics. Then it evolved to programming, graphics and music. Then programming, graphics, music and scriptwriting.... So with time, the teams became bigger, and the programmer relinquished more of his original tasks... Even with Alone in the Dark, I did just about everything: the engines, the compilers... Sure I worked 15 hours a day. Now it seems difficult to conceive it all. What tools have you used to help you create your games? Well, historically we created all of our own tools, from the rendering engine to 3D modeler, until not so long ago. I think the shift came about five years ago with the release of 3DStudio. It was the first public tool that allowed the people to get into 3D modeling. Aside from that we bought compilers as well, and after 3D studio, we also started to buy music tools. It was simply faster to make our own specialized tools. Our tools focused on a type of technology that we were aiming for. For LBA, I made a tool not unlike Creation that permitted with our rendering engine, the display of the scene, integrating the user interface that allowed them to build the scenario, control parameters... It was the tool that allowed us to conceive the game. Do you use a Level Mapper? Is it unavoidable? Generally speaking, we are moving away from tools for specific games. Our levels are drawn on paper first, by graphic designers, and programmers code them in script. So for you, the evolution of game design tools is to go from specific to a little more general? It's why Creation appealed to me, it is really a tool that corresponds to our vision of the situation. You have the interpreter, the scenario coded in another language (in this case a a graphic language), and the rendering engine. From there you can still change the engine if you want thing rendered differently, but you will still need to develop the scenario's language. How do you see Creation in this evolutionary process? How do you see it for yourself? Well Creation wasn't quite ready when we really had a pressing need for it - the fact that it can manage just about anything. The moment when we start a new project - when the graphic designers get to work because they have tools like Softimage or 3DS Max - they can start to work while we still don't have anything yet from the programmers. So we can begin working out the game design, what the game is going to look like, see the camera's viewpoint, etc.. While we are in that early phase, where the rendering engine takes a long time to put in place, we can start the game design and have the designers and animators operational. Creation allows us to see if the first phase, the first contact works, then we can start to develop the gameplay. It entails programming with Dev; more specifically, with Dev's SDK if we are after something specific like a special way of managing the joystick. So the first and second phase can be worked out in Creation before things get rolling. That's really interesting. NEXT: Frederick Raynal on the Future of Game Design This article may not be reproduced in any format without the authorization of Virtool S.A.