Hello, I am currently carrying out some research into the relationship between architectural design and game design. I am specifically interested in how architectural theory can be translated and used in game design to produce the gaming environment -' the architecture of virtual cities'. Aspects I am discusing include phenomenology (how people feel within spaces) simulation, landmarks and planning of the city, the notion of navigation and familiarity and how these relate to the player. How do these enhance the game play? What aspects do environment artist consider in creating an environment? In order to develop my research I would like to gain your thoughts and opinions on the subject from designer and player points of view, and let me know of important issues that i have not mentioned. Thanks guys

Hmm... What your doing is quite amazing(if you can pull it off). Generally, game designers don't go into that much detail on architecture. That seems to be more the province of Interior designers and architects etc. I'd suggest YOu find one of those, but they would probably be just as clueless as I am, since what you are doing is midway between the two realms.

Now that I've professed my ignorance, Here's my thoughts, from a players perspective. I think that architecture has several purposes in games, and that the balance of those purposes is defined by the nature of the game. In some games, most notably RPGs and RTSs, Architecture exists to give a sense of identity to it's "creators". Certain groups will have certain architectural styles, and these should add to the mood of that particular group. Icky, creepy aliens should have icky creepy architecture, however you generate that effect. Generally, I think these architectural styles tend to be exagerated, as the player is generally not well versed enough in architecture(or more likely: too distracted by gameplay) to note subtle differences.

Exaggerated architectural differences can be important for establishing the identity of a particular group, especially if the game is fast paced, because other methods have less pronounced of an effect. The phenomenology you mentioned is important here: If the player feels creeped out and disgusted in the alien city, that makes the aliens that much creepier, addign to thier image as the "bad guys". If the player feels welcome and safe in the streets and halls of the mighty republic, that can make it easier to establish them as the "good guys".

The other purpose in architecture that I can think of, which I've touched on briefly, is to creat mood. Obvoiusly, you don't need to seperate identities as much in a game like Doom. You do, however, need to create an atmospher of fear. The base layout is rather poor, for a base, becuse it is "not quite right" It doesn't make sense to set up a base that way, and so the player is creeped out. Add flying skulls, blood on the walls, and the impression is complete.

The placement of dripping, pulsing things is also an important part of the phenomonology you mentioned: in a claustrophobic, confined area, A dripping, pulsing, apparently living mass you've never seen before can really add to clausrophobia, if you have to walk right by it. If theres a satanic ritual in the next room, and if that ritual actually summons an imp, the player jumps halfway out of her skin.

This is very interesting to think about. what are you doing this research for?

What a fascinating topic. Now, the thing that immediately comes to mind is that in conventional architecture, the object of design is to improve people's experiences, to make things easier and more clear. In contrast, the game architect is only partially the player's friend. He may intentionally occlude paths to make the player go around the long way. He may use a limited and repetitive palette in order to disorient the player. He may put the player in situations where the player at once has difficulty seeing the area around him and feels exposed to those around him, in direct contravention of the concepts of prospect and refuge.

But there are some things that carry over. As anyone who played the original Wolfenstein 3D will tell you, it's no fun to be lost. I would say that there are certain challenges in a game which are "legitimate" (need to shoot that guy before he shoots you, and you only have three bullets, and if you run too fast he'll hear you), and certain challenges which are "illegitimate" (the key bindings suck, and every one of those corridors looks exactly the same because the designer was lazy, and why the hell would you make a building so confusing that you're constantly backtracking). A game designer's role is to manage the legitimate challenges (see "flow theory") while removing the illegitimate ones as much as possible. It's in that second category, I think, that one can pull out the architecture textbooks.

One example of this is the ability of architecture to guide people. One who walks towards a properly designed office building finds himself drawn through a main entrance into a lobby. From there, the paths to the elevators, to an information desk, etc. are clear. When he feels himself to face a choice, he is given the information as to what the effects of his choices are; when he feels his path to be obvious, it is because the correct path has been made to appear most inviting. Likewise, a good level design which is intended to support a linear plotline will draw the player around the appropriate locations without making him feel as though he's on rails. A bad level design will send him around randomly, constantly sending him into locations which he doesn't understand or care about because he doesn't yet have the context in which to interact with them. He will walk straight past his next objective because he doesn't realize it's there, and will then become quite confused as to what he has to do.

Great Topic!

I would add that we use architecture to communicate function to the player as well as to generate mood. You'll see this on a large scale in RPGs, where cities are often divided into districts, with the architecture in each conveying villager houses, shops, blacksmiths, etc. RTS titles use this same approach for individual buildings, so players can tell at a glance what each structure is used for.

Level designers also need to be concerned with technical limitations in their engines when designing cityscapes. Cities in RPGs are usually very dense environments in terms of objects, AI, and other players. Thus, city layouts are often planned around to work with an occlusion system to so the engine only has to process what is within the players immidiate area, and not necessarily the entire city. The capital cities in the MMO's WoW and DAoC are great examples of this. The good level designers can create easy to navigate cities even with this restriction, while the bad ones end up with a jumbled mess of buildings and deadends.

Good luck with your research, post again when you have some more insight to share!

You speak of virtual cities, which introduces the topic not only of architecture but of urban planning. A frequent (and naïve) dream of usually rookie game designers is to create a virtual world where you can "go anywhere and do anything." Well, in a real city you don't want to go everywhere, because a lot of places are just boooooring.

Virtual cities are created for the explicit purposes of supporting a narrative and a series of interactive engagements. Consequently, were one able to examine the city as a whole, I suspect one would find the urban planning to be quite fragmented - heavily detailed in areas that carry the symbology you mentioned, depending on the game, such as a public square with imposing statuary in a futuristic or mildly political shooter; presenting labyrinthine pathways reminiscent of the organic evolution of European towns in games with an emphasis on stealth; and so on, with each feature being dictated by the narrative milieu and gameplay emphasis of the game or segment of the game.

Similarly, one wonders how completely modeled large buildings in some games are, given that the story/"mission" only necessitates passing through a given subset of areas. Many games use locked doors or other forms of access restriction inside buildings to "funnel" the player in the desired direction, some quite clever (rubble and debris, a sudden fire). Outdoors, where such "natural" obstacles may be harder to come by, we are often presented with the "invisible walls" problem, where the spatial design suggests that a place is accessible but narrative, interactive or physical (e.g. system memory) constraints make that undesirable and the game therefore brusquely turns you back after a certain point.

I think this is a very interesting area of study, as it can help with planning to reduce the many physical design inconsistencies that lead to the illegitimate challenges Sneftel mentioned. Keep us posted!

One difference between real world cities and virtual ones is the need for utilities.

Real cities are designed with various utilities necessary to support the buildings in it such as powerlines that supply power, pipes to supply water, sewers to remove waste, and roads and railways to move people and goods around.

Virtual cities don't necessarily need these things for them to function (exept insofar as the rules of the game require it). So some buildings can be built without any need to connect them up to utilities to support them.

Also, one thing I've been thinking about for a bit (which doesn't really fit in with archetecture directly) is that in most games that require crafting or making items there generally isn't any waste products produced from the crafting or consumption of it. For example smelting iron ore to make iron bars generally doesn't result in a pile of impurities extracted from the metal, or after consuming a potion there isn't an empty bottle to throw away.

As a result, virtual cities tend to be free of garbage cans, landfills, or incinerators needed to dispose of garbage. Those are generally taken care of by the game automatically removing junk items to free up memory.

So virtual cities aren't really bound by the same rules of necessity that real ones are.

Before I start, I wanna say this is an awesome topic.

I don't think anyone focused on game play with architecture. I find that interacting with the architecture around you is amazing. One thing that kinda frustrates me in architectures in a game is where there's a building but no inside. Just a painted door. Portal algorithms have gotten so advanced nowadays that having tons of buildings with inside rooms is possible. In the idea for a city, I especially like when buildings are connected by catwalks, giving a large than life kind of feel to the game. Not just the walking on the ground kind of feeling.

Back to interacting with architecture, the buildings have to be geared for the game play. If it's an FPS you want the architecture to add tactics and give the feeling that no where is safe in the city. With an RPG for the most part everything can be for looks.

With the idea for creating a mood abandoned houses and dark alley ways are always cool for creating fear in the player.

On the subject of a city though, transportation and how the architecture goes along with it can be amazing. For instance sub way lines, sewers, or monorails in the sky can bring the player away from the ground level. It also gives the player something to look at instead of just searching on one level of the game.

Hi,

Wow what a response, its good to see so many people as fascinated about the subject as me. I am actually a researcher within a School of Architecture, but I am incredibly fascinated with games and their meanings, from both a theoretical and players point of view, and i am really interested the relationship between the two. I want to develop on my skills to eventually design and create my own enviromnents for games. You have all added some really interesting points.

I get so frustrated too when the buildings are just a shell and you cant get inside. However, the game would be huge if you could get inside every building, and would it really add to the gameplay?

One game that is in development, Crackdown, is going down this route. Their city is fully immersive with every element within it able to be used, you can pick up anything of the street and use it as a weapon, and every building can be entered. This type of game, to me is fasinating, and it will be really interesting to see how it works. Is this however too much? Overkill for the player, or will the player take time out of missions to explore and 'live' within thier city?


I think what follows has been briefly touched upon, but ponder over this, and lets discuss.

How should we understand a ‘good’ model for a virtual city?

Is it one which evokes an atmosphere within which we want to ‘hang around’, explore, inhabit? Like tourists, do we search out novelty and authenticity? The representation of typical houses and real streets suggests this might be the case, but how do games ‘play with’ familiarity?

Does familiarity of place, help navigation?

Part of Edinburgh have been directly translated in Project Gotham as a race course, and when I drive around I can stop outside my old flat, with the exact same coloured door and I can see where I work. This is quite an uncanny experience that takes away from the gameplay itself.

So, are more fictional ‘typical’ cities the preferred paradigm?

I think this is where games like Grand Theft Auto- San Andreas work really well, in Los Santos it’s the mix of the same city structure as Los Angeles but refined and some of the iconic features taken out. It has key landmarks such as the Civic Centre and Perishing Square, but there is no Frank Gehry's Walt Disney Concert Hall what does that say about the role of iconic architecture in virtual cities? There must be legality issues but it is still interesting.

Dan Houser, the founder of Rockstar Games, and Aaron Garbut, the art director in Blueprint magazine, no. 226 provide an interesting point when they say:

“I don’t think it’s possible to be completely immersed in a game if you’re constantly making comparison. We are trying to convince people the are on the run from the police – all that they’d be thinking is whether their gran used to live on this street or the next one”

“We could use a real city in the game, but we chose not to, the actual locations do not lend themselves to great design”

So if the actual locations don’t make a great design then what elements, signs and symbols are retained to produce a sense of place and identity without working as distractions?

What Cities would then make a good model? How about Manhattan? It has already been adopted by a multitude of film and games but can it be a good model.


I think that the grid structure of the city gives way for a lot of confusion and repetitiveness, imagine trying to locate a particular street, going up and down, although it adopts well in a driving game. It certainly works for Spiderman, lots of nice tall buildings to swing from. But for an RPG is it not just overload and boring?

Or what about Venice? It has been used in games such as Tomb Raider. It’s more organic in its form. With canals proving to be interesting paths, and the confusing streets becoming legible with the mass of distinct details.


I will stop rambling on, but please discuss. :)

[Edited by - LeanneT on November 4, 2006 6:51:38 AM]

Quote:

Original post by LeanneT How should we understand a ‘good’ model for a virtual city?

Hmmmm. It is important, I think, to decide early on what a "virtual city" is. Is it, as practiced in Second Life, an inhabited multiuser "world" where people "live" and interact? Or is it the backdrop of a game which involves but does not center around the world? If one approaches level design from an architectural standpoint, one risks neglecting other principles of level design in favor of architecture. Just something to keep in mind. It's all a balancing act.

Quote:

Does familiarity of place, help navigation? Part of Edinburgh have been directly translated in Project Gotham as a race course, and when I drive around I can stop outside my old flat, with the exact same coloured door and I can see where I work. This is quite an uncanny experience that takes away from the gameplay itself. So, are more fictional ‘typical’ cities the preferred paradigm?

Deus Ex used painstakingly accurate models of Battery Park (NYC), parts of Hong Kong, etc. I would say the uncanniness of your experience (reminds me of the Uncanny Valley of human representation) is a concern but a minor one; few people's flats have been rendered in Project Gotham. I suppose it could become more a problem if large areas of the world were mapped in this way, but there seems to be neither the ability nor the need to do this.

Quote:

I think this is where games like Grand Theft Auto- San Andreas work really well, in Los Santos it’s the mix of the same city structure as Los Angeles but refined and some of the iconic features taken out. It has key landmarks such as the Civic Centre and Perishing Square, but there is no Frank Gehry's Walt Disney Concert Hall what does that say about the role of iconic architecture in virtual cities? There must be legality issues but it is still interesting.

GTA:SA, I think, wouldn't want a Gehry building poking out of its sun-drenched urban drama. That gets into the more mundane idea of theme; Deus Ex included a hell of a lot more exploding barrels than donut shops. Rendering real-world areas is fine, but other considerations of level design overrule it.

Quote:

So if the actual locations don’t make a great design then what elements, signs and symbols are retained to produce a sense of place and identity without working as distractions? What Cities would then make a good model? How about Manhattan? It has already been adopted by a multitude of film and games but can it be a good model.

In the case of Paris or New York there are certain landmarks which you would assume to be iconic; and, indeed, a lot of games pick them out on the skyline like a postcard catalogue. Personally, though, I don't think it's necessary. You spend a hell of a lot more time looking ten feet in front of you than miles into the distance. The Eiffel Tower can provide a persistent feeling of location, but isn't it sort of a naive way of doing so? The differences between New York and Paris are hardly limited to their landmarks. By the time your level designer has given you a "feel" for the city, you don't need them anymore. (The considerations might be different for a racing game where you are moving past the scenery at upwards of 100 MPH.)

Quote:

I think that the grid structure of the city gives way for a lot of confusion and repetitiveness, imagine trying to locate a particular street, going up and down, although it adopts well in a driving game. It certainly works for Spiderman, lots of nice tall buildings to swing from. But for an RPG is it not just overload and boring?

Certainly overload. A lot of early games which replicated cities (see Gabriel Knight: The Beast Within's Munich) limited the user to a few key areas of the city, connected by a subway-themed method of moving between them. This fit well into the gameplay because it allowed new locations to be "discovered" as they became part of the plot. No searching for Marienplatz; when you know there's something there, it shows up on your map.

In these days of seamless loading, the technical considerations which gave rise to this situation have been pooh-poohed by jaded players and developers. We aren't content to be magically transported around walled areas of a city any more than we're content to look at untextured cubes and imagine them to be buildings. But perhaps there was something to it? The problem with a city is that it needs to be large enough for everyone. World of Warcraft's Ironforge needs to hold dozens and dozens of locations you don't give a damn about because some other class does. The problem with New York is that other people need to dry clean their clothes but you don't (your trenchcoat, apparently, never needing any sort of changing or laundering). I see this as an important disconnect between real-world architecture and game architecture: real-world architecture needs to create individualized spaces, whereas game architecture can BE the individualized space.

Hi there Leanne :) I'm actually an architecture undergraduate student right now (though struggling with it!) and I too had a lot of thought about this topic.

First and foremost, I think it is important to note that as a general rule, it is not necessary to have architecture technicalities (specifically constructional details) in all level and architecture designs. What is important is the surface of the architecture that the player can see, like the walls, windows, rooftops and not constructional details like roof trusses and foundations etc. Besides, good thing about the game environment is that you can alter the physical law and produce a fantasical futuristic technology environment. (e.g. a floating castle)

Basically in a game environment, the player can only use 2 of their senses, that is the eyes and ear (Though the latter is usually given to the sound designers). Therefore, it is not too important to think about touch, smell and taste that we experience through real-life architecture. Because of this, I personally think that the aesthetics of architecture plays a HUGE role in game environments compared to real-life architecture. I tend to find that Art Nouveau, Futuristic, Post-Modernistic or Baroque architecture styles in a game environment to be quite wonderful. Whereas Modern architecture style in a game environment seems a bit boring to me.

And, I agree with a lot of people's perspectives about designing an experience for the player. Architecture in a game, more or less defines the identity and culture of the environment the player is in. And the spatial values of the architecture closely ties in with the gameplay, where it be a storyline or how the game is played.

For example, you may need a large open and inviting space in the middle of a town, so players can gather, rest and buy/sell items or generally just socialise. That's for an RPG game. If you're designing games like Prince of Persia, it will then be important to emphasize on balconies, ledges, columns and poles because the player will be using them a lot (and creatively too).

I think my opinions seems pretty logical and practical. -_-