Gamasutra: The Art & Business of Making Gamesspacer
Defining Boundaries: Creating Credible Obstacles In Games
View All     RSS
September 3, 2014
arrowPress Releases
September 3, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Defining Boundaries: Creating Credible Obstacles In Games

July 1, 2008 Article Start Page 1 of 5 Next

['Why can't I jump over that wall?' In this intriguing design article, Sidhe's Griffiths (Gripshift) examines the usability-related issues - and solutions - for frustratingly invisible and unbreakable barriers in games.]

Throughout our everyday life, we come across barriers and boundaries which prohibit our movement. We instantly recognize them and know what we need to do when we come face to face. Some barriers we can easily circumvent; a door, for example, temporarily blocks our way until we open it and step through to the other side.

Games are the same. When we are in the game world we come across barriers which block our path and force us to turn around or open them up and step through to the other side. The problem with barriers in games is that they are sometimes not that distinctive.

What happens when we cannot tell the difference between a "friendly" barrier (one which we can interact with in some way) and one which is so unmovable that a nuclear missile wouldn't make a dent? More importantly, how does the player feel and what do they do to overcome this?

This article looks at the various techniques used by game developers to blockade our paths and, from a usability perspective, what it means for gamers.

The Issue

Gaming technology is incredible. With new hardware advancements, developers are bringing us bigger and more exciting worlds. We often have the opportunity to roam for miles or to actually walk through entire cities, exploring seemingly every nook and cranny.

But what happens when there are areas in the game that developers don't want us to get to? Usually there is some kind of barrier that halts our progress or, alternately, an element of the story of the game which explains our inability to continue onward.

These boundaries are needed for a number of reasons, not least of all due to the fact that no matter how good our technology is, we simply cannot create a truly open-ended world. It is just not possible at the moment.

Boundaries also help us learn areas of the maps, because if everything was open to us at once then it would likely be simply too overwhelming to take in. By opening things up slowly to the player, we are introducing a gradual learning curve.

Another use of the barriers is to simply stop us from trying to go down certain areas of the game which simply don't exist. We can't have the players running everywhere because this means we have to create massive environments when it's simply not needed. So to overcome this we simply create barriers that halt the player's progression in certain areas.

The Answer: Human Computer Interaction

However, it is with these barriers and boundaries that a problem lies with game design usability. Unbeknown to players they actually rely heavily on two key principles in the field of Human Computer Interaction, or HCI: visibility and affordance.

In HCI texts, visibility and affordance applies to the controls that users can see. Donald Norman defines visibility as "Controls need to be visible which implies that there is good mapping between the controls and their affects. For example, the controls on a driver's dashboard have good visibility whilst video recorders have not." 1

Affordance is a technical term that refers to the properties of objects and how they can be used. A door "affords" opening. However, in HCI, what is more important is perceived affordance. This entails what the person thinks can be done with the object. Does the door suggest it should be pushed or pulled?

So how does this translate into games and game design specifically? Well for our discussion, we are looking at the game-world itself - the actual environment - and therefore we need to ensure that the players are always able to see their path through the game.

We should never have a situation whereby the player ends up at one point just sitting there, scratching their heads and unable to figure out where to go. This is not the same as a game which requires the player to figure out where to go.


1 - Preece, J. et al. Human-computer interaction. Addison-Wesley Publishing Company, Inc., MA, 1994

Article Start Page 1 of 5 Next

Related Jobs

Sony Computer Entertainment America LLC
Sony Computer Entertainment America LLC — San Diego, California, United States

Lead Script Writer
Mobilityware — Irvine, California, United States

Senior UI Artist
Zindagi Games
Zindagi Games — Camarillo, California, United States

Software Engineer
Zindagi Games
Zindagi Games — Camarillo, California, United States

Lead/Senior Designer


Joel McDonald
profile image
Good article, although I would have liked to see you go more into some concrete solutions to each of the obstacle types that you presented. Take doors, for example. You mentioned how they can go wrong, but how can we as designer correct the situation? Say we are designing an open world set in a city. It's obviously not time/resource-feasible to make every single door have affordance since this would force designers to have to flesh out the interiors of every single building. It's equally not feasible to remove all the non-interactive doors because that would break the immersion of the city. The remaining alternative seems to be to visually distinguish interactive vs. non-interactive doors in some way, but even this isn't a *great* solution since the a) player still must learn this in some way and b) sometimes the game narrative doesn't afford a good explanation for the visual discontinuity between doors.

Dave Bedrosian
profile image
Great article.

Responding to Joel, I think visually distinctive doors is a good solution. I think the original Thief did this really well: functional doors look fairly new while nonfunctional doors look old and decrepit. Sure, it took a few tries to recognize the difference, but the game was very consistent: there was never a case where a decrepit door just happened to open to access a "secret" area (at least as far as I know!).

By contrast, as the author describes, while playing Max Payne, you simply have to try every door, which isn't terrible, but does get tedious...

Mike Scott
profile image
I can't help but think that in real life, doors are doors, and they don't "indicate afford" through looking differently from other doors. However, if the lights are on, is someone home? If there's cars in the parking lot out front, is the business open?

It seems as though there could be myraids of clues left for someone to determine which door they should try. I would imagine genre would be the best indicator of how to handle the situation. Horror might have noises beyond that door, or odd lights, or even self-turning doorknobs for ghost horror. Gory horror might have "drag marks" under some of the doors, or bloody handprints/footprints to indicate unlocked doors, and undisturbed pools of blood for locked ones. Magic might have a "twinkling glowy outline" to indicate it's a "sealed" door.

It's odd though, the "magically barricaded" door wasn't liked by the author of the article. If nothing else, the "spoiling of what was coming" should build suspense. If you're rolling your eyes, the game isn't working for you and no amount of door treatment will save it anyways.

The Gears of War reference seemed a little odd too. I often found myself frustrated by trying to go places that seemed accessible but weren't. In multiplayer, I was blocked by a broomstick leaning against the wall. In single player I was blocked from stepping onto/over shin-high "obstacles". My biggest complaint for the game was the lack of realistic obstacles for those that stray from the beaten path.

Halo 1 had a great system that kept everything very clear in the fact that there was no real "barrier" over the top of the maps, but things like an intermittent blast of light would cause a vehicle that was attempting to escape the level to stall temporarily.. thereby getting it to slide back down into the level.

Flying vehicles seemed to have an "operating ceiling" where overhead clouds often blocked vertical progress.

In Halo 2, you had to work to get outside of the primary map area, and from there, you could seeingly go almost anywhere, with few exceptions. All barriers semed realistic, and in-fact, if nothing else the affordance seemed unrealistically compliant at times. For example, jumping on-top of a street-lamp to jump on top of a set of fabric stormshutters before accessing an out of the map location.

Halo 3 changed things though, where a soft barrier blocks progress and restricts "out of level areas" That's the true barrier reached in that image shown within the article. By the time you attempt to access that top rock, you are obviously on a 'mission" to escape the area. You wouldn't run into that barrier if you were attempting to play the game rather than get out of the level.

I found the concept of "suitcases" blocking character movement to be very appropriate. Or do you often go to the airport and climb through/over people's luggage when a path around them presents itself?

Having said that, it really depends on what your character/gamers state of mind is at that point. If he's running from head hunters.. I'd imagine he'd trip over suitcases, not be blocked by them, but short of that, why would the character not prefer to walk around them, rather than be rude to the other characters in the storyline?

In any event.. I really did like the article as a whole. I'm partially nitpicking, I suspect, but as an active gamer myself, I had a differing perspective to some of the elements the author presented. My own "take" on barriers and affordances is to compare them to real life counterparts. Do I approach a door and know it's locked or unlocked prior to touching it? If I see a log on the ground in an open field, do I expect to be able to step over it?

Obviously, you have to remember that as a game, it should have more positive elements than negative ones. More open doors, even if they are blocked immediately beyond them, are required if you don't have a way for the player to recognize blocked doors from a distance. This way the player doesn't feel that their character is stupid. The character lives in this world, and should know how to navigate. If you need to, just use less doors, period.

If it doesn't really HURT to make something accessible, why block it? I see that a lot. A room will be fully realized, but it's blocked off with something like a trashcan.

Giveth, don't taketh away from the player.. They're here to be entertained, not frustrated at their character's inability to navigate, and fit through what visually appears to be an acceptably narrow corridor that you determined was actually a dead stop.

profile image
Iím going to have a really good time here! First off to the author, your statement that, ďÖwe simply cannot create a truly open-ended world. It is just not possible at the momentĒ is WONG! Take a look here at these Infinity, Quest for Earth trailers:

I want to be PERFECTLY CLEAR here. The ONLY reason that open ended worlds, or to put it in laymenís terms, entire planets for the player to explore without boundaries, are not being created, is because of COST. Even that is a cop-out now because with clever coding a simple, base planet can be created that could be used for 1000ís of other planets by simply only varying each one slightly. Terrain is easy to create, and easy to randomize. Plus the planet itself provides the visibility blocking necessary to keep the game running smooth. After all the player can only see a mile or so in any direction. So lower poly models and lower res textures can be used in far distances, and everything beyond that is simply not rendered. The only area rendered at any one time is the 1-2 mile radius the player is standing in the center of. Simple, and no need for fog or barriers.

Think about this for a momentÖ What can we expect to find on an earth-like planet? Some variation of flora depending on the environment, and some logical variation of fauna based on environment. All a game developer has to do is create a few different flora and fauna types for a each environment type, then logically populate some base planet model with them. This would also work for entire cities where every building is accessible. A room is a room is a room. It has four walls, a floor, and a ceiling. There will be some architectural details, and different models may be in the room in some logical fashion. All that has to be done is design a base room, a few different architectural styles to match any building architecture, and a large assortment of textures from brick to plaster to wall paper. Code the game so that a building with certain specs is randomly populated with rooms matching that spec and items to give the room some life.

Ultimately barriers are NO LONGER NEEDED except in limited capacity. They are a carryover from days gone by used as a way to make a game as cheaply as possible for the maximum amount of profit. Along these lines are stupid things like fog, and cutting stuff off too close to the player. Even the least powerfull computer of the last two years can handle a mile or two of terrain populated with flora, fauna, houses, etc. The insides of buildings donít need to be in separate cells like they are in Morrowind or Oblivion either. If the player canít see the inside of a structure, donít render it until he or she can. If they can see partly n through a window but itís still far away, render a low-res version of the room and textures until they get closer. Itís all almost laughably easy.

Now letís talk about Halo 3 for a momentÖ Iíll use the multiplayer maps as an example here. To increase the interest of players who play this game online all sorts of barriers have been imposed in EVERY SINGLE multiplayer level. I know because I have tried to get out of them all. I understand why the barriers are there, as I said it is to ensure that online players are happy. Halo 2 was apparently, according to some, ruined because gamers found superbounces and ways out of the multiplayer maps and would use those against other players online. For online play I understand and support the existence of barriers. But those barriers should be removed from all but the outer edges of a map in offline maps. They should be a simple option the player can turn on or of, depending on the preferences of those playing with him. This too is easy to code. Something simple like, ďif online barriers=true, else barriers=false.Ē Just a simple switch that is activated when a player selects the online multiplayer option. Easy. I watched the making of the new maps including Ghost Town. They put all sorts of barriers right inside the map, and they admitted it, to make sure that nobody would have the advantage. But all these barriers over the tops of the buildings should not be there for offline play. In Valhalla there is no need to have the death barriers behind the red base as the mountains provide a natural barrier. In Blackout there is also no need to have barriers right at the tops of the buildings, some of which get in the way for a player jumping from one location to another in the level. You donít need any barriers inside the building on Last Resort. The stupid spaceship sitting thee on a landing pad just out of reach in Rats Nest is an obvious affront and insult to curious players, and no barriers at all beyond the fence are needed in The Pit. Itís painfully obvious to see that Bunjie was upset about what players had found and done in Halo 2, so they took overly aggressive measures in Halo 3 against them, but it was all only for the online players, who were whining about everyone using these secrets, instead of taking the time to get off of their lazy rears and learn them for themselves. I donít think any of them understand how long it takes to learn these superbounces and ways out of the level in Halo 2. I spent 6 Ė 8 hours at a time, sometimes fruitlessly, trying to learn them. But you know what having these in Halo 2 did for that game? It made Halo 2 an even batter game than it already was. It is one of the few games that reward the player for out-of-the box thinking and exploration.

Now letís talk about Far Cry, shall we? For my example Iíll be using Instincts/Predator for the 360. Why are there barriers, ON AN ISLAND?!! I can understand barriers outside the island, which should have been placed well out into the open ocean so it becomes obvious to anyone swimming out there that there is nothing to explore and no reason to continue. This worked well, although done a bit differently, for Morrowind. But those barriers on the island itself should have never been there. In fact game developers need to explore giving the player more freedom to explore. How about some rock climbing equipment for that cliff face over there? Or an earnable and developable rock climbing skill? But back to Far Cry, there are invisible barriers everywhere that in every case donít need to be there. Iíve used the editor; the player simply canít go beyond the boundaries of a level. All the multiplayer maps in FarCry should end at that boundary, with nothing going beyond, to provide a perfectly acceptable ending point for a map.

In the end game developers need to seriously look into having entire worlds the player can fully explore, without boundaries. For multiplayer maps on these worlds, a simple, logical marker at the edges of the playing area will suffice. Something that fits in with the theme of the game. Standing stones with magical symbols on them for fantasies. High-tech shield fences for futuristic games. Era appropriate blockades for other types of games based in our preset or past. While we certainly can create entire worlds for the player to explore in a single-player fashion, unfortunately controlled areas are needed for multiplayer. Until we have the technology to put a few million gamers on a planet and let them duke it out in some gametype or other. That probably is beyond our technological means right now. But everything inside that area should be explorable. Every door should open and lead into a room. Every window should be breakable and allow the player to pass through it. Every cliff should be climbable; every roof top accessible if the player is ingenious enough to figure out how to get up there.

The time of invisible walls, not allowing the player to jump or climb, imposed and unnecessary barriers is gone. Itís time to move on, stop bitching about the budget, and do something truly ambitious. The payoff will be well worth it. Players will respond positively to a gaming experience with few if any limitations. Some players will need their hands held, or they will get lost. But these few shouldnít ruin the fun of the many. Give these players all kinds of guidance options from maps to compasses and let them control which ones they use. Let those of us looking for adventure find it, and enjoy ourselves during the journey.

Remember we have the right to play the game we have purchased in any way that we want. Stop messing around with this right, and spend more time pushing the boundaries and removing archaic limitations. Polish up your games, make sure they play well and control well, and that they reward the player for exploring. If playerís use what they have found against others online, so what? The other players will just have to find these things themselves. But you wonít loose enough players to be concerned. If a game is good people will play it, and Halo 2 is a perfect example. Despite players using secrets against others online and in many cases having a much higher skill level, people still played it for years, and may even be still playing it now.

Itís time to think outside the box Ė literally.

- Deathbliss

Mitch Kearney
profile image
Deathbliss, I have to say that maybe you missed one key part of this article - barriers are needed because otherwise the player will become overwhelmed and lost. If I was playing a game where I was given a task, and I was given licence to go walkabout and all I found was random flora and fauna, then the game is over for me. What is a game? It's a set of rules and objectives designed to create a fun experience. If I lose sight of those rules, the objective is no longer defined, then there is no more fun.

The actual reason I'm posting is because there was an example used in the game which reminded me of something that happened when I started to play Halo 3. You know that bit at the start where the soldiers climb over the rocks near the waterfall? Well, it was launch night and I was tired, so I spent the first 20 minutes of play wondering why the hell I couldn't climb over those damn rocks too.

The next day after work things became a little clearer, but its a perfect example of affordance betraying the barrier.

profile image
I can't say I agree with the notion that games simply shouldn't have barriers. If you're creating an open world game where the running theme is "go anywhere" then it only makes sense, but if your game is a straight-forward shooter, then why would you waste resources creating a believable (and pointless) downtown district because "someone got over the big brick wall"? Unless it provides some gameplay advantage to the player, then all allowing players to literally "go anywhere" will do for a focused game will be to get the player lost and frustrated.

As for what the article was actually talking about, it goes without saying that the difference between a barrier that can be overcome and one that cannot should be distinctive. Exploration in GTA4, for example, is limited to things that just "look inviting" for me, although even so I still run into frequent red herrings. I think Morrowind/Oblivion's decision to make everything explorable, but have less stuff overall, was the best way to handle this issue for open world games. The world still felt believable, even if the big cities weren't particularly big if only because it all felt alive. Like a system, and not just a collection of scripted sequences.

Mike Scott
profile image
During Halo 3's testing stage, they used a breadcrumb marker that was "dropped" by a player every few minutes. Each one was a different color and reflected their position on the map after a certain duration of time. Early tests showed that a LOT of people were getting lost on the first level (the jungle level, "Sierra 117"). Many players were ending-up back where they started from. To rectify that issue, designers were forced to add one-way boundaries in the form of cliffs that they dropped down, to ensure that players could see which way they needed to go, and weren't able to backtrack.

(this was documented nicely in a "Wired" Magazine cover article from about a year or so ago)

Barriers are 100% needed for directing game flow.

I too tried to follow the Marines up that waterfall in Halo 3 in that same level because I missed the audio clues that would have told me there were enemies to my righthand side. I agree that it wasn't a good move having your "team" split up at that point and half of them go to where you can't follow, unless there's a distinct REASON you can't follow.(such as an extremely narrow path or something).

The key is to make the environment SEEM to be open, while in reality, limiting the players movement options to ensure they don't get lost. This is what creates the need for these barriers.

Otherwise, there's nothing to stop the player from visiting areas of the map out of sequence, thereby ruining the storyline development where the character needs to be somewhere specific to move the game forward.

The bottom line is, and I beleive the intent of the article, is to forward the idea that these barriers should be created in such a way that the player recognizes there's a barrier at that location with methods that work with the player's immersement into the game, rather than be jarring, frustrating elements that defy what the player has been taught in terms of accessibility.

While I wholeheartedly agree with that concept, it's something that can be taken too far.

If all of the barriers are realistic, it may quickly feel claustrophobic to the player, which puts you squarely back into the mess games were in when the concept of first-person shooters originated.. the rat maze of walls.

There's a balance that must be struck between "seemingly unlimited travel", and claustraphobic mess. You want breathtaking vistas, but you don't want the player managing to get behind the scenes, nor do you want them running into ANY seemingly unrealistic barriers blocking it from them.

Here's an example.. a Beach level setting requiring the player to travel from one end to the other. Say the game storyline requires that the character stay on the island.

My belief is that you should first consider the real world implications of that: If you were on an island, what would stop you from leaving it?

In the real world, a barrier wouldn't be needed, because unless someone's primary goal was to actually leave the island at that point, they're not likely to even want to get wet unless they're dressed for it and the water temperature is within a certain comfort level. So, to those of you with a "casual to moderately eager" desire, no boat means no way off the island. That is, until you're at a point where you REALLY want to go such as a need for survival. To keep you on the island at that point you may need to have knowledge of sharks that may be in the water, or the next island is much too far to swim to. The real life "barrier" at that point is based on your personal limitations, perceived or otherwise.

The goal as a designer is to recreate those perceived limitations, and make them "real" to the player. In games, the player doesn't care about getting wet, and in some instances, doesn't even care about dying "the first time through the game".

That means you need a barrier.

You can follow the old designs which make even superheros unable to swim in waistdeep water, or the Halo designs that allow you to even take a jeep and just drive underwater until you hit an invisible wall, but those are unrealistic to the player.. They're ACCEPTED, but it probably doesn't assist in the immersion level.

So it would seem that the design goal should attempt to recreate a position where the character doesn't want to get wet.. perhaps through realistic movement limitation considerations, or weapons that fail to operate correctly when wet, or other items that must be "dried out" before use.

Vehicles shouldn't be "ok" with anything beyond brief and/or partial submersion, and should have significant movement rate deductions/adjustments whenever they encounter a significant amount of water.

Perhaps movement restrictions become greater and greater the longer the character is in the water, or the more submerged they become? It would effectively become a realistic barrier that's entirely beleivable in game terms and in reallife terms.

Failing that.. say with a "superman" character that can simply fly away (though your game story is weak on the point of trapping them on an island anyways), you need to head another direction, which is incentive-based.

For example, if the tribesmen see you leave the island, they will kill your girlfriend that they're holding hostage. If you leave the island before determining the pirates goals, they are likely to relocate before you return, taking their gold with them. The soldiers that are camped here are patrolling the coastlines in boats. There's a good chance you'll be spotted and shot down.

In game terms, you have to establish those incentives, and typically follow-through on them at least partially.

ie- my favorite game barrier is a Halo 3 multiplayer level ("Snowbound") If you start to leave the gameplay area, large automated cannons start beeping and aiming in your direction.. the message is clear.. and one or two more steps outside the game area will result in them open-firing upon you. Beyond that area is an invisible barrier, but only through significant "tweaking" can you actually reach it before the deadly accuracy of the powerful cannons overcome your character's shields. One of these areas is a field area, but the other ones are part of a canyon path that appears to continue around a bend up ahead, but if you actually manage to get back there and around the corner, it's a dead end.

Ben Younkins
profile image
Reading this, I immediately remembered this:

Perhaps we should give some good examples of affordance? Half-Life 2 denoted non-functioning doors via a lack of a doorknob. If I recall correctly though, they didn't use any extra geometry to show when a doorknob was present, so you had to keep your eyes sharp for the texture details. Any other examples?

Tim Elder
profile image
Firstly, awesome article. It seems like a common sense approach, but so many top notch products still seem to fall short.

I found it interesting that magic barriers were frowned upon. If 'magic' is a part of the game, then surely that is sufficient to justify a magical barrier. So long as it is clear that the barrier is there (both in terms of visibility and affordance), why it has suddenly sprung up, and why it later disappears (eg a trap laid by the bad guy that you now have to fight), then I don't see a problem.

Deathbliss, I like the idea of massive procedurally generated worlds (or galaxies, or the Universe even) but there is one simple problem with them - natural worlds work in a procedural universe, but artificially created items don't. By "artificially created items" I mean things that are created by intelligent entities such as buildings, streets, furniture, vehicles etc. And not just the objects themselves, but how they are placed or laid out. Algorithms just don't capture purpose in the same was as intelligent design.

Take Dwarf Fortress for example. The game world is created algorithmically to start with, creating land masses, elevation maps, weather systems, rivers, lakes, biomes etc. Then it populates the world with colonies of dwarves, humans, elves, goblins and kobolds who all establish their fortresses over time.

The game world is huge, REALLY huge, and the wilderness is vast and can be explored for hours on end (in Adventure mode). The problem is that the only interesting fortresses to explore are the ones that are player made. The algorithmically built fortresses are bland, generic, and look like they were plonked down on any vaguely suitable terrain. They also don't serve any purpose as fortresses - they have terrible entrances (in terms of defence, entrances require intelligent analysis of the surrounding terrain to build properly) which would let the smallest of goblin invasions run rampant. In defence of Tarn Adams (creator of DF), the generated fortresses aren't made to be fantastic, and he notes that he hasn't spent much time on them yet, but any content created by algorithm will be inherently repetitive and generic by the very nature of algorithms.

My point is this - regardless of how much processing power we have, it still takes the time of a human being to properly craft interesting and engaging gameplay environments and experiences. The quantity of the world can be algorithmically generated, but the quality needs a human touch.

profile image
Good points all!

Unfortunately most players have limited attention spans. They must have a clear goal and immediate satisfaction. We need to break them out of this mindset! Has anyone here read the Rama series by Arthur C. Clark? If a game developer was going to make a next-gen version of this excellent series they are faced with a dilemma. Should they give the player full access to any area of the ship they can unlock or get into with the natural boundaries being of course the edges of the ship, or should they put in a bunch of invisible barriers, fake doors, and so forth? Naturally to favor the majority of gamers and their pocketbooks they would choose the later. But Rama in this example should really be created with no imposed barriers. Instead you simply give the players clear goals and logical direction. Then everyone is happy, because gamers like me can explore to our heartís content, and gamers that must be led by the hand will easily be able to see where to go and what to do next.

So it is possible to give the player and entire island to explore in any direction they wish. Nothing has to be approached in some chronological order. Even things that depend on the player being somewhere at a certain time can simply provide them with another alternative should they miss the window. Let players run all over the island, armed with clear list of objectives that they can complete any way they wish. This is the definition of true freedom in a video game, and is an example of real next-gem features.

No, most barriers are not needed at all. An entire world can be give to a player to explore, and as long as the player has a multitude of travel options, clear directions, and certain goals they wonít get lost and they will break out of this hand-holding rut theyíve been forced into by so many games in the past.

Imagine the possibilities of an RPG created this way! A player could start as some unknown character in any of the cities, villages, or towns on the planet. They could then begin to develop the skills they wish to use and become the kind of character they wish to become Along the way they could talk to the townsfolk and pick up on the main story thread. Or they could keep an ear out for some treasure of relevance to their character. Then they could leave the town on foot, in a boat, in an airship, or astride some animal, etc. They could watch the world unfold around them as they travel, get off wherever they like, fight whatever enemies their character may have, gain experience, and continue on in their journey to continue along the main story line or to find another treasure. The player will be completely immersed in believable world with true depth, where with the right skills and equipment they can go anywhere they wish.

This is the goal I wish game developers would strive for, and I hope one day to see a game such as this.

- Deathbliss

Billy Bissette
profile image
The article does not mention another issue with barriers, in that games encourage players to explore them by putting rewards behind some of them. Even if you don't need to enter any of the 50 doorways in an area, games have encouraged players to check every single one of them, because one of them might lead to bonus items. Games have conditioned players to explore all the boundaries, because they sometimes reward players for making the effort.

As for the Lost example of luggage, it isn't even so much that the luggage forms a barrier, but the annoyance that tends to come with such barriers. The player does need to get beyond the luggage, but it is more of a hassle to do so than it likely should be. Even if you can freely walk around the luggage (rather than needing to take a twisty side path through a forest or the like), it is still more of a hassle to walk your character around the obstacle than it realistically would be to walk around it. (And it may also look weird, with your character getting stuck behind a corner of luggage that he should be able to step over, or repelled by an invisible wall that keeps him some variable distance from the luggage, or whatever else might happen in a game.)

And as for Halo 3, I too got confused and frustrated trying to navigate the first stage

Evan Kawa
profile image
-Deathbliss: I think you make a few good points. I agree that as technology progresses it has become possible to create (nearly) limitless physical worlds. I also agree that perhaps some designers today have become accustomed to designing around certain limitations that are unnecessary given new and improved technology. However, I think that if you hoping to see the game that you described in your last paragraph there, you will be holding your breath for a long time.

You'll notice that above I said its possible to create "nearly" limitless worlds. This is because of a few things. First, technically your world can only be as large as your computers memory allows. Even with procedural generation, eventually you are going to run out, or it is going to have to dump something that was stored previously. If it does that, then when you come back to an area you already visited, it will be different and your world falls apart. The bigger problem is that even procedural worlds require man-hours to populate them. Models and textures must be created by someone, and in this example it would require a VERY large number of them to keep the world interesting and different from area to area. If you just re-used the same assets one billion times, pretty soon people are going to catch on and your world is again, broken.

Along those lines is the problem with the depth of gameplay. It's nice to imagine a world where you can do anything you can conceive of to accomplish your goal, but it still requires time and money to have someone code all of that stuff into your game engine. By the time they finished, the engine would probably be so out of date that nobody would want to play it and they would have to re-do it all anyway, lol.

Now, all that aside, you aren't just talking about having infinite terrain to explore, and an endless skill set, but also completely removing the linearity of a story..... Now that IS impossible to do in a game while still maintaining an immersive, believable world. Either the designer would have to make the game so that every action the player could ever make realistically effects the outcome of all future actions, for every possible combinations of actions a person may make.... ever (i.e. that designer would have to be GOD :P) or the designer would have to make it so that any action you take has absolutely no effect on anything in the world, ever.

In the former case, its impossible to create. There is no way any person could conceive of every action a player might try to take and make the game behave in a believable way. If they truly allowed free-reign and the player does something that doesn't effect the future in the way he expects, then your immersion is lost. In the latter case, you could complete all objectives in any order you wish, however immersion is already lost because nothing ever effects the world. The player already knows its fake. This is why some barriers are required for a game to be a game.

The trick is to hide the barriers or to blend them into the world well enough that players don't realize that they don't break immersion. Thats what this article was really getting at. Its not a matter of "can i do this with this object?" its "should i be doing this?" or "would it help me to do this?". If the player thinks that climbing over the rocks will help them proceed in the game, then they should be able to climb the rocks. If the designers don't want the player climbing them at all, then they better make it clear to the player that they cant be climbed. If the barrier is designed well, the player will see it, instantly recognize it as what it is, then move on. The problem comes when the player doesn't recognize it and is frustrated or surprised when they are stopped by it.

Going back to the suitcase example, if i were on a beach after a plane crash, and there was luggage between me and my destination, I would want to kick it aside or climb over it to get to my destination as fast as possible. I would expect to be able to do that in a game given the same situation. Now, if i were in an airport under normal circumstances, i probably would walk around the luggage. If there were an emergency and i had to rush out or something, then i probably would kick them over instead. Its all about the context of the situation. If the barrier doesn't fit within the situation, then you lose immersion. It doesn't matter if a player cant do everything conceivable at any time, because if they are trying to do things that don't fit the situation, you have already lost them. If you are actively trying to escape the world in the middle of a mission, you cant honestly tell me that you would be more immersed if the barriers weren't there at all. If thats the case, then you are no longer playing the same game because you have changed the rules.

I think thats the spirit of this article. Its not about accounting for every possible action at every time. Its about not breaking immersion by giving the player no indication that the action they are trying to take would ever work, and instead, leading them to the correct one.

Mike Scott
profile image
I misunderstood the picture from the Lost game, and didn't realize that this wasn't occuring after folks had pulled all of their luggage out, rather than immediately after the plane crash. As I recall from the show, the characters all tried to be respectful of people's "areas" where luggage was stacked. That luggage still looks stacked to me, but in retrospect, that's not the important thing here.

I get your point, in terms of what the players focus is at that point. However, before judging it, perhaps it's important to know why it was there and what other options were availible? It would seem that they were trying to keep people out of the plane in an area that had few beleivable obstacles to make that happen with.

Perhaps rather than say it was bad, we should suggest an alternative to it? I can see from the image, that the character actually appears to have an area to walk through. THAT I beleive is the crime.. it's not really about whether you can, or would, go rushing through suitcases like an impervious bull in a china shop, it's about the fact that it looks like you can step around them based on the image.

To a player, who isn't personally experiencing any of the shock and drama that the NPCs are, and isn't afraid that the plane may become engulfed in flames, or searing hot, or perhaps weakened and dangerous, or perhaps a charnel of death with sealtbelted strips of bacon clutching melted cupie dolls, and all the horror that implies, etc, they just want to get in there and loot the plane, check out the cool images, etc..

I guess the designers/art team didn't coordinate it very well, and the piles of luggage weren't tall enough in appearance, or severe enough, with perhaps smoking chunks of debris added to gaps in it. They opted to make it look as much like the show as possible.

Unfortunately, the player, by definition, is a sociopathic misanthrope. They don't give a crap.. the game is just starting, they want coins, health or barterable goods, plus possibly weapons, and secret bonuses while the NPCs die around them and cry over lovers/sisters/brothers that were lost in the crash.

The need for a barrier seems pretty strong, but shy of engulfing the aircraft in flames, and then later pretending like it's still intact luggage on it, there's probably not a ton of options availible to keep misanthropes on a beach, away from an openly accessible crashed aircraft.

Unfortunately, the show isn't about being trapped on an island with a misanthrope looter with no problems killing you to get that last soda you've be saving. It's about humanicizing everyone (including the player) in such a way that they are beleived to be.. well.. less misanthropic, and more able to carry-out the duties that the show character did, in terms of forwarding the plotlines, being involved in personal relationships, etc.

At that point you have to trap the player that isn't attempting to play his role for whatever reason, and do it in such a way that it wouldn't make the player character seem "odd" to the other NPCs. Often overlooked, this can be critical of immersement. If the player thinks the NPCs are idiots and not reacting to their hijinks.. it becomes a different game from the one intended. (which CAN be good, but unlikely)

Unless of course your NPC's and storylines are suitably programmed to accomodate the player being a misanthrope who doesn't mind walking through fire, or hitting people, etc, just because he can, or jumping off cliffs, etc... you need to weed that behavior out by not allowing it to happen.

In a situation like that, which must have been full of trade-offs, I can't help but wonder what I could have done instead, when faced with that task.

("Hey?! How come I can't grab the leg from that dismembered corpse and use it like a club?"..)

Neal Holtschulte
profile image
Someone mentioned putting sharks in water as a barrier, but I'm surprised the "death barrier" hasn't been discussed more. All Return Fire ( maps were on islands. Two of the vehicles in the game sank in water, but the other two did not. Once these vehicles got a certain distance from the map, a sub surfaced and fired a heat seeking missile, which was just barely, not quite avoidable. My friends and I loved this barrier. We actually had contests to see who could get farthest from the map. The trick was to sucker an NPC helicopter out to the fringe with you and then fly such that the missile from the sub targets the NPC, then run like hell before another sub surfaced. It was a blast to play and it was just a mechanism for keeping players on the island! I think game designers can and should be more clever about barriers as was done in this game.

I also wanted to say that I prefer clear indicators of where I can and can't go. If you have to color open-able doors differently than decorative doors, so be it. I agree that players, myself included, have been trained to "wall hump" everywhere looking for secrets. Some of us don't want to miss anything, but at the same time it is not pleasurable to hump 50 pseudo-doors to find one that opens. Yes, I could just look up the secrets on the internet, but that takes away some of the fun. Secrets should be discoverable via puzzles or hints, not through brute force searching or google searches.

Gareth Griffiths
profile image
I didn't realise i'd attract so much interest from writing the article. I must firstly thank you all for taking the time to actually read it and then to write such interesting comments.

The idea for the article was to explore the basic kind of barriers that gamers come across in games. These are the kind of barriers that are there simply to limit us in some capacity and as technology progresses, they may become less reliable simply because they do not fit in.

The advent of new technology in gaming is both a blessing and a curse. When I was playing games back in the day on the old Spectrum, it seems I didnít really care if I couldnít reach some area for apparently no reason. It was ďone of those thingsĒ and I would simply seek an alternate route.

However, as graphics become more realistic, are people now starting to think that maybe games should mimic real-life more. A case in point. Iím in my local game shop, waiting in line to buy my game. While there I glance around and see a kid playing one of the Burnout games on a display model in the corner. Suddenly, his car collides with an object and violently turns, spins and ends up in a crumpled heap on the floor. A split second later the car is back on the road, looking a bit worse for wear but still not as bad as when the collision actually occurred. Great fun I thought. However the kid, who was probably no more than 10 years old, actually had the audacity to say ďHmm, I donít think it would really do that in real-life, would it?Ē

This intrigued me but it got me thinking about the whole barrier issue and it may be that as technology progresses more, relying on things such as the ďinvisible barrierĒ just isnít going to cut it anymore. The example regarding the doors is another thing. Sure, if weíre in a corridor in real-life we may have to try every door but when youíre in a game, having to move around your character with a controller in a trial-and-error routing just becomes frustrating. And when the average time spent playing a game is around 30 minutes, do you really want to go and try EVERY door?

However, barriers in a game are certainly needed, even if its just to ensure players are able to learn all the techniques required to play the game. If there were no barriers then the player may deviate from a ďtutorial pathĒ and not get the necessary information. What we need to ensure though is that the manner in which we implement them is done in an appropriate manner.

In response to the comment stating that Infinity: Quest for earth wonít include barriers. I really donít think the player will be able to explore every inch of a planet uninhibited. While there may be billions of planets the player can visit, I am almost certain that when they start walking on the planetís surface they will be stopped at some point. However, if I am wrong, I hold my hands up and wave a white flag but I will be extremely intrigued as to how the developers handle this situation.

Oh, the Half-life 2 comic was extremely funny by the way.

Steve Westhoff
profile image
Reading this article was a breath of fresh air for me. I've always been frustrated by invisible walls you can't pass, and doors you can't bash down (like Half-life 2 comic). And I personally didn't mind the pop-up barriers in Devil May Cry series... the first 50 times anyway.

I agree that for single player games, the game should be "mostly linear" with the perception that it is open world. I'd give the example of Battlefield: Bad Company for this. BF:BC uses the enemy attack boundaries, where "enemy mortar range" is redded out on your map. Even though it is a bit unrealistic in certain geographical spots (not consistent within a radius), it works for me. Aside from that, you can approach your goal from any angle, in any vehicle, and with any weapon. And forget about doors... you can blast through most walls!

Tangent: One thing that I would have liked to have seen in BF:BC's cases of simultaneous multi-goals is benefit/side-effect changes according to the order that you completed or attempted each goal option. For example, say the player was given a two part simultaneous multi-goal of both A) raiding an ammunitions depot and B) taking out a communications center.

Raiding the ammo depot first (A then B):

+ gains you better weapon to make communications attack easier

- reinforcements are called in to the ammo depot you are attacking

Attacking communications center first (B then A):

+ disables reinforcements at ammo depot

- better weapon at ammo depot has been transported elsewhere

In online multiplayer games, I do NOT think there should be restrictive boundaries for in level exploration. Getting "out" of the map should not be an option, however if you can find obstacles to jump and climb on, then they should be able to get there. An example that I can think of is in Call of Duty 4's "Backlot", where a player can jump onto a trash bin, jump onto a ledge, and then jump onto a window balcony that does not have a "floor". There are identical window balconies in that level that you CAN stand on. This inconsistency is restrictive, deceptive, and feels "anti-player".

I agree with Deathbliss in regard to the time invested, exploratory player advantage. Instead of giving frequenters of the game more powerful weapons, upgrades, etc, keep the weapons balanced across the player experience, and leave the advantage to those who seek it out, through exploration of course.

Some of my favorite MP games, in retrospect, ALL my favorite MP games, had secret places you could get to, and I don't think that's just a coincidence.

Call of Duty 4's "Ambush" comes to mind... as if the level designer PURPOSELY put in secret spots. Trash can behind a fence that allows you to climb over, a metal roof that you can run and jump into a blown out hole in the wall to reach a secret roof, etc. I think more level designers should implement these types of spots into their levels.

Paul Cameron
profile image
Love the Article, and agree with both sides of this semi dilemma. Boundarys are needed be it land air or sea. Take for example Pirates of the Burning Seas. The map is fairly small. I sometimes find myself humming "It's a small world after all" as i spend 30 minutes or longer transversing from a port in South America to say Texas. Its frustrating because of the length of time it takes, however it introduces the senses to a tiny bit of what it must have been like in the 1700's always in fear of "something" happening, pirates, etc. Yet it becomes even more frustrating when all I want to do is go from point "A" to "B" as quickly as possible, and now theres a player introduced barrier, that I must cross or go around. A Circle of death I must pass thru or avoid. I can't just turn the game off and finish it tomorrow, I must make my way or lose my ship and all its valuable cargo. I can always weather out the storm IE. red zone(PVP) area in a" safe port" or spend an additional 15 or 20 minutes avoiding the areas . However I cannot if I wish take a trip to Brittain with a load of cargo. I cannot visit far off ports risking ship and cargo sailing around the horn. I cannot walk from port to port. There is no "land" bridge. The world is too small yet vast enough. I find that when the developers of the game spent months and months as well as mega bucks to create a city in the game called Pont De'Pietre a french town , with a huge VAST immense city, They went way overboard, and made that one destination point a vast wasteland for all the gamers. It was an excersize in their developmental abilitys, and a waste of gamers times. It should be sold to World of Warcraft as it is much more of a dungeon than a city, and has no real game value. They would have done better by creating a walkway between towns connected by land with an occasional storyline attack. Finally I wish the game had the ability to sail to all areas of the world. A huge map, vast oceans to explore, when I didn't feel like running the red zone. The point is the game is about sailing in the 1700's. That includes exploration, but not getting lost in a city for hours. but exploring the oceans that encompass the world, in that time.

Brian Bartram
profile image
One "boundary" you didn't cover, though it's possible I missed it being mentioned in the comments, is the "proximity to the mission area" boundary. Missions in open-world games often make use of this logic to prevent players from straying too far from the action area.

As an example I submit "Transformers, The Game (Xbox360)". Despite being open-world, many missions are defined by a boundary visible on the minimap, and straying outside this area causes a mission fail.

The classic "follow mission" works similarly - if the player gets too far away from the AI being followed we give him a mission fail.

I'd be very curious to hear opinions on this type of "boundary" - especially since I'm implementing them in the game I'm currently working on. Seems like the nature of streaming, open-world games has made this approach more common than ever.