Introversion's Defcon is getting AI Bots. And it learns and adapts over multiple games. And, interested individuals will be able to build their own AI models. Until now, the game has featured only a simple, limited bot. You can read a very interesting article about the construction of the Defcon AI by the AI author, Robin Baumgarten. He discusses choosing between fun AI or realistic AI, and other design issues associated with the bot's construction. At Baumgarten's site, the author also has available two PDF files on the project: a massive 130 pager thesis and an in-depth article about the design and construction of the AI. I read through the article and, though much of the information was over my head, Baumgarten put significant though, time, and consideration into the AI design. The problems he faced I think are quite similar to those of game designers. I'd never realized how intricate, complex, and interesting designing AI is; it's amazing.
I really enjoyed this presentation. Paul Bennett is a creative director for IDEO, a design consultancy. The ideas he presents in this video are nothing short of genius. Bennett covers a menagerie of ideas, and I think many of them pertain to game design. From 5:30 to 7:00, Bennett discusses "looking with your peripheral vision." Cliffy B once said that the games industry is too "insular," that we are too inwardly focused and don't draw ideas enough from other sources.
Both Cliffy and Bennett have good points. Design is like life: the more diverse your interests or experiences, the more dynamic and enriching will be your life and ideas. Paint, build cars, sweep and scull, play games, and each individual experience will be better, for experience will enrich the others. Diversifying yourself lends you multiple perspectives of everything you do. Just like design. If we only play games, only read about them, only design them, then our games will suffer from, we will be drawing ideas from what is a very anemic pool. Bennett asks us to our peripheral vision. And part of using our peripheral vision has to do with taking initiative, seeking out new experiences and learning from them; then we will have a bigger pool of ideas, an array of perspectives that we can use to conceptualize and perceive our designs.
From 7:00 to 8:30, Bennett discusses a single design challenge, which is a two-handed input-device for nurses. Again, what Bennett has to say is directly applicable to game design, mainly, I believe, objective player observation and prototyping. Our ideas may not pan out exactly how we would like them. I know prototyping is nothing new for game development, but there is no harm in stressing the matter. Bennett is saying that we need to objectively and attentively observe our players, and their use of our design systems, before we charge headlong into something that, in the end, may not work as well as we had planned. Even before prototyping, we need to pause in our design process to review what we have done so far, and reassess the value and contribution of each design aspect.
Successful?
At 9:20, Bennett is discussing how we use the world around us, specifically how, for convenience, we "wrap the tea bag string around the cup handle." Bennett says: "We're sort of using the world around us to create our own design solutions. . . . This is people designing their own experiences, you can draw from this." He continues about how we "co-opt our environment," how we take advantage of the environment itself or objects therein for our personal use and benefit. We want to take advantage of these observations, to use natural human responses to create better game experiences. I'm sorry, I don't have any specific examples. Part of a comfortable gameplay experience is ease, natural understanding of functions. We aren't making virtual reality, not yet. Players are kind of two steps removed from video games, separated by the screen and the controller. We need to work with these steps, so instead of barriers they become tools. Is a person more connected to nails with a hammer or with bare hands?
Another aspect of this topic is player expectations. Players, having lived in the real world, have expectations about how their in-game worlds should function. In a recent interview with GameTrailers, Will Wright discussed Grand Theft Auto IV, specifically the AI: "The behavior of the people is so lame compared to the graphics. You know, thats the weakest link by far. We have these people that look like fairly reasonable humans, but they act like staplers." I actually think we've made admirable leaps in expectations over the past few years, for example, with Oblivion or Mass Effect. Aside from NPC AI, however, simple things like menus are expected to function a certain way. The best way to find out how people expect these to function is to playtest, adjust, and playtest some more.
Part three, and the most important part, of Bennett's presentation is "unthinking situations." Essentially, Bennett talks about designing for your audience, not what you assume your audience to be. To discover who your audience really is, Bennett says, "put yourself in their shoes." He says its good to "re-frame the ordinary," or innovate." Innovation starts with an objective view of the intended design, or, a subjective view, as long as that view is from the eyes of your audience. I think sometimes its easy to forget about the players. Its so easy to think about how we're going to play the game, maybe we even describe well how the game will be played. But if from the get-go we didn't think, in a sort of selfless manner, of the other players, the ones who will be playing the game, then we the design may fail where we were so confident it would succeed. We need to remember that a human is behind the controller. We must take them into account: their emotions, their mind, their reasoning, the dexterity of their fingers. The game was designed well when players think that its so easy to play, it must have been easy to make.
My brother recently pointed me to an interesting Gamsutra article entitled "Where's the Design in Level Design." Written by Tito Pagan. The article comes in Two Parts: One, Two. The first part is admittedly more interesting than the second, which focuses more on how to develop levels rather than the design aspect. A good summary article comes from the first page, actually, where Pagan writes:
A well-designed level takes into consideration a whole set of requirements, such as user interaction and navigation, which are inherent to the purpose they serve. How will the spaces control and direct the player throughout the explorative and interactive experience? What sort of directional and responsive feedback mechanisms will be provided to assist the process? How will all of the elements tie together to form a cohesive environment that is well understood without compromising aesthetic appeal? The level designer must also consider the impact of particulars such as sound, space, lighting, pace, and scale.
The article is worth reading, the first part at least. Pagan talks about wall orientation and window design that is interesting enough in itself, not to mention its application to game design. Check it out.
When designing a game, particularly a role-playing-game, which comes first: the story or the gameplay? Think about any RPG, is the story more important or the gameplay? Regardless, which aspect is the primary design point?
In an interview with IGN, Eisuke Yokoyama talks about the development of Final Fantasy XII: Revenant Wings for the DS.
We did not set out to make a game that was connected to FFXII. In the beginning it was supposed to be a 'My first Final Fantasy' experience - that was the overall plan for this game. We had already created the battle system. It was ready to go, but at a certain point during the development and planning process, Toriyama's team started talking. FFXII had just been released and it was really popular, so it was more a matter of timing that it turned into this kind of game.
I was surprised to read this. I actually don't think there is any one order to design an RPG, or any game, for that matter. Whichever comes first comes first. If you come up with a great story that would work well with a video game, congratulations. But I think its probably equally effective to come up with a gameplay system and then build a theme and story around it. Both aspects are very important, however. Whichever comes first, in the end both need to be given equal consideration and development. An RPG needs both to be truly great. Personally, I almost always design the gameplay and control scheme before even thinking about anything else. But who knows, maybe one day that will change if an interesting story and world suddenly pop into my head. Both approaches are entirely feasible. Which foot do you start off with, gameplay or story? And do you think one approach is superior to the other.
Yesterday I came upon this 1up post summarizing recent happenings in the MMO world. The primary topic of discussion was Vanguard: Saga of Heroes and its development process at Sigil Games. 1up linked to two interviews at F13.net, an MMO-focused blog. The interviews are with an anonymous Sigil ex-employee and Brad McQuiad, CEO of Sigil games. Each interview casts very different lights on the development of Vanguard. One thing both agree on, though, is that development did not go smoothly. Apparently,development of Vanguard didn't kick in until a year before release, even though it was in development for four. The employee attributes this to the lead designers, including Mcquaid. Mcquaid himself attributes development difficulty to publisher funding and support. After reading both interviews, I think there is validity to each person's argument. I would highly recommend both interviews, they're enlightening. But the reason I wanted to bring this up is because, honestly, it was jarring. Its easy to get caught up in design and forget about development. Game development is the practice of game design. I love game design, but development is what causes all of those great ideas to come to fruition. Design is only half the battle to making a great game. Check out those interviews. The 1up story is really good also.
I want to apologize for the lack of updates as of late. I've had finals these past two weeks and couldn't take time to post with all of my other obligations. But now, I'm free as a bird. I will be working a full time job all summer, so most posts will come at night.
Anyway, back to game design. Yay!
You may have noticed that Starcraft II was announced saturday. The originial Starcraft is my favorite RTS and I really couldn't be more excited for the sequel. Blizzard is approaching the design of Starcraft II cautiously, knowing full well the popularity of its prequel. Their goal is two-fold: retain the significance and uniqueness of Starcraft's play style while also improving upon and adding to the game's design. What this means specifically is that Blizzard is not making Warcraft III with a Starcraft skin. There are several primary differences between Blizzard's more recent RTS and their 1998 space-opus. As this GameSpot report points out, Warcraft III is focused on micro-managing big battles whereas Starcraft was more about economic build-up and mass-army control. Blizzard wants to bring this RTS design into Starcraft II. Additionally, they are apparently very focused on creating a counter system, where each unit has another unit, unit-type or ability they are specifically weak against. I think its fantastic that Blizzard isn't just making Warcraft IV in disguise. They are trying to create a genuine sequel that sticks to the greatness of the original Starcraft.However, they are bringing in several new features that have been developed since Starcraft. One in particular I would like to bring to your attention is a newfound revolution around elevation and unit movement. Blizzard has only shown a very small slice of the inevitable mass that is Starcraft II. But we can already see some common threads woven through out the game's design. The environment is now a much bigger factor in than it was in Starcraft. First, units are no long revealed when attacking from without fog of war. In the original Starcraft, enemy units would be shown, and therefore vulnerable, when attacking from a higher position. Now, units may attack from a height above their foe, and remain hidden while doing so, while also mainting advantages in both attack and defense. Almost every revealed unit thus far takes advantage of this adjustment.
The Colossus is a new Protoss unit that looks very similar to the alien ships in the newest War of the Worlds. The unit walks on four legs and shoots laser beams on enemies far below its bird's nest. Starcraft has multiple levels of terrain that are normally connected by ramps. Of course, there are also walls and platforms. The Colossus is easily able to walk over terrain dips and rises, thanks to its long legs. While this is only speculation, the unit may also be able to walk across smaller buildings, like bunkers or photon cannons. Anoter Protoss unit, the Stalker, has been given the blink ability seen with the warden form Warcraft III. The Stalker is a newer, dark templar version of the dragoon from the original Starcraft. Blink allows the Stalker to instantly teleport anywhere within its range of vision, including over walls and valleys. Last, the Reaper is a new Marine-light unit for the Terran. The unit wields dual-pistols and sports a jetpack. This allows it to hop over and down walls with ease, much like a grasshopper.
Ah....the Zerg Rush. Classic.
Of course there are and are bound to be many units that cannot simply jump across levels of terrain. This gives those that can a distinct advantage, at least in terms of mobility. But what these units have in movement skill they may lack in other areas, such as defense. Its interesting. Blizzard has presented themselves with something, elevation levels, and then created a solution, units that can overcome the elevation difference. Were these two facets designed to complement one another? Elevation is very important in Starcraft, and obviously is so in Starcraft II as well. I think that Blizzard wanted to create solutions for players to get across terrain easily. This seemingly small change is going to destroy so many strategies from the original game, while also opening many new ones. I like what Blizzard is doing with Starcraft II, what do you think? Zerg Rush It Do you think that Blizzard's focus on elevation is a mistake? Why or why not? Do you think their should be a focus on micro-manegement as opposed to mass army control?
this public service announcement brought to you in part by free time.
Letting the Player Lose When's the last time you played a game where, no matter how hard you tried, you lost? This question is the basis for designer Ben Schneider's article recently featured on Gamasutra, entitled Losing For the Win: Defeat and Failure in Gaming. Schneider poses an argument that losing is games is good. Not making the game impossible to beat, but, designing a game so that the player has no control over winning a situation. He says this is good for two reasons: One, it can be used to drive a story forward. And two, it can make winning later on all that much more satisfying. Schneider makes an excellent argument, this article is well worth reading, and the concept even more worth pondering. Schneider does not think this is easy to pull off however. He thinks incurring regret in players is the greatest potential pitfall of designing a game where the player loses. He makes one statement in particular which is really quite profound:
Books and movies have a huge advantage in not incurring regret in their audience. Their “players” have no agency; as much as they may dislike a twist in the plot, it’s not their fault. As game designers, we must reckon with regret. Our players have to do more than like the story; they have to accept each turn of events and roll with them, and never wonder if they should have gone back to get it right.
THIS is the beauty of video games. The power to involve people. Players are a part of the game; more then a participant, they are the core that all else revolves around. A cliche perhaps? "With great power comes great responsibility." As game designers, we are able to exert from complete to little control over a player's actions. Sometimes one way serves players better, other times the opposite-we must discern the benefits of each for individual situations. Players understand winning as a means to a game's completion. Schneider wants them to understand losing as serving this function as well: for games to be designed in such a way that losing both advances and enhances the story, and also has an effect upon the player.
Thinking of Loss In what whays would losing in a game be better then winning? What types of ways can a player lose (eg. automatic cutscene)? Which do you prefer and why?
Introversion's Defcon is getting AI Bots. And it learns and adapts over multiple games. And, interested individuals will be able to build their own AI models. Until now, the game has featured only a simple, limited bot. You can read a very interesting article about the construction of the Defcon AI by the AI author, Robin Baumgarten. He discusses choosing between fun AI or realistic AI, and other design issues associated with the bot's construction. At Baumgarten's site, the author also has available two PDF files on the project: a massive 130 pager thesis and an in-depth article about the design and construction of the AI. I read through the article and, though much of the information was over my head, Baumgarten put significant though, time, and consideration into the AI design. The problems he faced I think are quite similar to those of game designers. I'd never realized how intricate, complex, and interesting designing AI is; it's amazing.
Posts
