Mechanical Drama in Jamestown

Video Games

Jamestown Screenshot

While I’m not typically a shmup player, lately I’ve been enjoying a great indie shooter called Jamestown. It piqued my interest with its colonial Martian setting and beautiful pixel art. However, the game’s lasting appeal rests in the strength of its peculiar Vaunt mechanic.

When activated, Vaunt grants the player a seemingly arbitrary list of benefits: a brief bullet shield, followed by increased damage and a score multiplier for as long as the energy meter is kept filled. In practice, this mechanic gives Jamestown its own particular systematic rhythm of tension and respite. I’d like to use Vaunt to explore the idea that a game mechanic can have an inherent dramatic arc similar to those created by traditionally authored stories.

When I speak of a dramatic arc, I mean something along the lines of the three-act structure commonly used in storytelling. Strongly authored media (films, books, plays, etc.) map very nicely to this kind of structure, as does the authored content in video games. For instance, a level in Jamestown is composed of a series of increasingly intense enemy waves followed by a climactic boss fight. The pacing and intensity of this level design is determined entirely by the game’s creators.

Three Act Structure

When examining a game’s effective dramatic arc, we must of also take into account the player’s agency. The abilities that the player has control over can increase or decrease dramatic tension. For instance, a bomb move that kills all the enemies on screen gives the player a breather, thus creating a moment of dramatic respite. A slow powerful attack creates high tension (as the move charges) followed by low tension (as the hit connects.) These types of mechanics create a parallel dramatic arc authored by the player. The two arcs superimpose each other, interacting in a sort of wave interference that heightens some moments and dampens others.

What would the dramatic arc for the Vaunt mechanic look like? It begins with normal moment-to-moment gameplay, with its mild variations in intensity. Some high-tension impetus for activating Vaunt would then occur: a mid-wave difficulty spike or simply the filling of the energy meter. In response to this event, the player hits the ‘B’ button to activate the bullet shield and begin Vaunt mode. The momentary safety of this shield creates a brief nadir in dramatic tension.

Jamestown - Begin Vaunt Mode

After the shield has expired, Vaunt mode remains active as long as the player continues to replenish the energy meter by killing enemies. As I mentioned previously, remaining in this mode grants a score and damage multiplier. These multipliers increase the benefit of skilful play, which in turn increases the innate tension of moment-to-moment gameplay. Furthermore, surviving the duration of Vaunt mode grants an additional score bonus. The dramatic tension increases parabolically as the potential benefit of the survival bonus accumulates over time.

Vaunt mode can terminate in three different ways. The most desirable outcome, illustrated in the graph below, is when the player runs out of enemies to kill and the energy meter expires gracefully, granting the full survival bonus. In terms of dramatic tension, this would be a plateau followed by a resumption of normal gameplay. If the player instead becomes overwhelmed during Vaunt mode, she has the option to press ‘B’ again to activate a second bullet shield. This choice ends the mode prematurely, only granting half the survival bonus. This outcome would be represented graphically by a second brief peak in tension. Finally, the player could die during Vaunt mode and lose out on their survival bonus entirely.

Jamestown - End Vaunt Mode

These graphs reveal some general ways that a game mechanic can increase or decrease dramatic tension. When they provide momentary safety, increase performance pressure or ransom potential reward, they have a strong dramatic effect over time. We can also see how the mechanical tension interacts with the authored dramatic arc. The encounter design provides the high-tension impetus for activating Vaunt, and the level’s difficulty strongly influences which way the mode will end. The nature of Jamestown’s Vaunt mechanic works in concert with its strong level design to create an exciting gameplay experience.

The MDA framework asserts that game mechanics create an aesthetic response in the player. When we observe this response over time, we see patterns of dramatic tension and reprieve that match those found in storytelling. We often praise the narrative power of heavily-authored games, but the best video game tales often emerge from system-centric games like Dwarf Fortress, Civilization and EVE Online. If we acknowledge that game mechanics have inherent dramatic arcs that superimpose the authored content, then we can begin to analyze mechanics in terms of their storytelling potential.

→ 2 CommentsTags:  · 

The Principles of Programming in SpaceChem

Programming, Video Games

SpaceChem screenshot

SpaceChem is a remarkable puzzle game about fake chemistry. The game challenges you to build a factory in order to transmute the given input molecules into the given output molecules. While chemistry is the theme, on a mechanical level it has more in common with programming. The methods used to tackle challenges in SpaceChem are akin to real techniques used by computer programmers. I’d like to elaborate on these manifold similarities, as well as explore how games like SpaceChem could be used to promote procedural literacy.

The player commands two circular “waldoes” by laying out paths and instructions for them. The waldoes follow the path and sequentially execute any instruction they come across. They can grab, drop, rotate, sync, bond, fuse, request input or dispatch output. The waldo’s analogue in computing is the processor, a hardware component that sequentially executes basic operations defined by machine code. Like processors, waldoes are the engines that drive the control flow of a SpaceChem factory.

When a waldo grabs a molecule, it gains the ability to perform instructions directly on it. In other words, a grabbed element is more readily and rapidly available than one lying elsewhere on the grid. Conceptually this is analogous to storing data in registers, a form of computer memory that is accessed very quickly and that the CPU manipulates directly. Just as a waldo can only grab one molecule at a time, computers have very few registers and must therefore rely on caching.

If grabbed molecules are like data in registers, then molecules left on the grid are cached. The cache is a larger, cheaper form of memory, but it is slower to read and write. Data must be written from the cache to a register in order to be manipulated directly by the CPU. The amount of memory in SpaceChem’s “cache” is governed by the area of the grid (8 x 10). Each coordinate on the grid can therefore be considered a unique memory address. This analogy is enforced mechanically: a factory “crashes” if two atoms collide on the grid, since you can’t store two values in the same memory address.

SpaceChem screenshot

Each factory has two waldoes, and they must be properly coordinated as they move through time and space. This coordination is facilitated by the sync node, which tells a waldo to wait until its twin has also hit a sync node. Parallel waldo management is akin to parallel programming, and they share the same perils: deadlock, starvation, race conditions, etc. The waldoes are like threads operating on shared memory space, and sync nodes are functionally similar to semaphores (operations that tell threads to signal and wait.) Operating two waldoes simultaneously in SpaceChem forces the player to confront the same shared resource problems as parallel computing.

SpaceChem and programming involve similar challenges: laying out simple instructions to achieve a complex result while managing limited time and resources. Like a good software specification, each puzzle is clearly presented as a black box defined only by its inputs and outputs. The player lays out instructions, starts up the factory, observes errors and corrects them, iterating until the puzzle is solved. Solutions can then be further optimized to take less time and use fewer instructions. SpaceChem and programming are engaging, flow-inducing activities because they have an identical inner loop: implementing, debugging and optimizing.

At this year’s GDC, Michael John asserted that programming is 21st century literacy. If computer programming is currently considered esoteric knowledge, it’s because our general education is not preparing students to think about problems in an algorithmic or systematic manner. Ian Bogost calls this procedural literacy: “the ability to reconfigure basic concepts and rules to understand and solve problems.” SpaceChem may not be an ideal game for the classroom (it’s far too difficult, for one), but it strongly suggests that the best way to learn about and engage with complex systems it to play with them.

→ 1 CommentTags:

Starhawk Announcement

Video Games

Starhawk Screenshot

If you follow me on Twitter or elsewhere, you may be aware that I moved from my native Montreal to Austin, Texas last year to work for a new studio called LightBox Interactive. We’ve been requisitely tight-lipped about our project until this last Friday, when we finally unveiled Starhawk to the community.

Starhawk is a third-person shooter for the PS3, and a spiritual successor to Warhawk. The game is set on the lawless frontier of space, where the rush to mine rift energy incites conflict between the scrappy Rifter union and the post-human Outcast. These two factions form the basis of the online 32-person multiplayer. The single player campaign follows Emmett Graves, a gunslinger pariah hired to protect the small mining town of White Sands.

The core of the game is characterized by fast-paced action and air/ground vehicle combat on huge open maps. The most iconic of these vehicles is the Hawk, which can transition from a nimble airship into a staunch mech. It also features a new system called Build & Battle, which allows players to quickly call down fortifications, turrets, vehicles and reinforcements in real-time on the battlefield. This distinctive system unlocks a multitude of strategic options and gameplay styles.

Starhawk is currently set to release in 2012, and you can follow its development on the Starhawk website. You can also check out a profile video of our studio and interviews with the project leads on G4. I’ll also add the usual disclaimer that opinions expressed in this blog are always my own and do not represent my employer.

→ 1 CommentTags:  ·  · 

The Squirrel Keys

Video Games

Deadly Premonition

I’m still not quite sure what to make of Deadly Premonition. Everything you’ve heard about it is true: it has atrocious combat, compelling mystery, laughable graphics, memorable characters, inappropriate music, etc. The game is full of fresh ideas and has stuck with me for months, but I still wouldn’t feel comfortable broadly recommending it. I would, however, like to examine one particular section of the game that I think was both absurd and oddly well-designed.

The sequence I’d like to highlight occurs very early on in the game. Having arrived in Greenvale the previous night, FBI agent Francis York Morgan (“please, call me York”) heads to the Sheriff’s Department to begin his investigation into the murder of Anna Graham. He is greeted there by mild-mannered deputy Thomas MacLaine. Visibly flustered by the small town’s recent tragedy, he has lost the key to the evidence locker and requests Agent York’s aid in retrieving it. He further notes that the key can be identified by its “Southern Flying Squirrel” key holder.

At first glance, this seems like a fairly typical time-wasting fetch quest, albeit one that eventually reveals a humorous twist. However, the hunt for the squirrel key rather cleverly addresses a number of high-level design considerations. Namely, it serves a purposeful role in establishing character relationships, aiding player navigation and setting a tonal precedent.

Right away, the reasons behind Thomas’s request tell us a lot about him and his relationship with the other characters. His reaction to Anna’s death shows that he is kind and sensitive, and the fact that he lost the key demonstrates that he is mildly incompetent (or perhaps just inexperienced) as a police deputy. It is strongly suggested that petitioning York’s aid is motivated by a desire to avoid confrontation with the Sheriff, George Woodman. This hints at the nature of their relationship: Thomas is meek and very intimidated by his grouchy superior.

Deadly Premonition

After receiving Thomas’s request, the player is allowed to freely explore the Sheriff’s Department. The building is rather large, with several offices on the first floor and a prison and firing range in the basement. Depending on the player’s exploration path, she may first stumble upon a “Pointy-Tailed Flying Squirrel” key in an inauspicious storage room. However, when this key is returned to Thomas, he politely explains:

…Ah. You don’t know your squirrels, do you. This isn’t the right key holder. This is a Siberian Flying Squirrel. It’s closer to a land-based squirrel. The Siberian Flying Squirrel is larger than the Southern Flying Squirrel, and has a standing tail while the Southern has a hanging tail. But we’re looking for a Southern Flying Squirrel right now.

There are, rather absurdly, five types of squirrel keys strewn about the Sheriff’s Department. One key is hidden in the locker room, another in the kitchen and yet another outside the firing range. Furthermore, as items they are only identified by description, such as “long-tailed”, “grey” and “striped”. Agent York, who frequently voices his urban prejudice, presumably lacks the country knowledge to discern their exact species. The Southern Flying Squirrel key (described as “curvy-tailed”) is found in cell #5 of the basement jail, which is the furthest room from the starting point.

Having the player complete the same fetch quest up to five times may seem ridiculous. However, remember that this sequence takes place during the player’s first visit to the Sheriff’s Department. It’s a location that plays a crucial role in Deadly Premonition, and the player will have to return here very frequently during the investigation. Obliging the player to initially explore the building from top to bottom ensures that he will be able to comfortably navigate the space in the later sections of the game. If the player ever wants to talk to the Sheriff or interview a suspect in custody, he’ll know exactly where to go. Even jail cell #5 plays an important role towards the end of the game.

Deadly Premonition

Finally, the absurdity of this sequence helps to establish a whimsical tone running in parallel to the more serious otherworldly murder investigation. Deadly Premonition is an open-world game, and Greenvale is full of colourful characters who ask York to perform optional odd jobs for them. These side quests are as strange and as varied as fetching ingredients, competing at darts, chauffeuring an old lady, answering medical trivia, and catching a legendary fish. The player may feel compelled to disregard these excursions in favour of concentrating on the serious task at hand. The fact that a trifling chore like the squirrel key hunt is integrated into the main storyline suggests that it is perfectly acceptable to indulge the local populace’s strange demands.

Deadly Premonition’s squirrel key hunt is remarkable because it subtly conveys a great deal of information about the game, the town and its inhabitants in a short playable segment. It’s emblematic of the best kind of tutorial: one that is completely integrated into the core game and teaches through play.


Combat Flow in Steel

Video Games


Few online multiplayer games have become as habitual for me as Team Fortress 2. I’ve returned to it every few months since its release back in 2007, usually to plumb the strategic depths of a handful of familiar maps. In August of 2008, the Heavy update introduced Steel, a community map created by Jamie “Fishbus” Manson. Its complex layout and objective structure immediately intrigued me, and after hundreds of matches it has become one of my all-time favourite multiplayer maps.

Steel is an asymmetric attack/defend control point map. The red team begins with all five points under their control, but the blue team only needs to claim control point E within the allotted time to win the game. While E is open for the entire match, capturing it early is a daunting challenge. The control point is perilously situated right outside the red team’s spawn point and across a bottomless chasm. To make matters worse, E’s capture timer is twice as long as any of the other control points.

To facilitate their assault on the final control point, the attacking team can unlock strategically advantageous layout changes by claiming four optional control points (A through D). Bases shift, objectives change, new routes open up while others are closed off; every change has a significant impact on the high level flow of combat. While other control point maps tend to split each stage of combat into a segregated segment of the map, Steel cleverly folds five stages into one large contiguous layout. Each sequential pairing of A-D and E effectively functions as a distinct stage of the match. This iterative repurposing of the combat space is a large part of what makes Steel unique and noteworthy, and is a something I believe is deserving of detailed design consideration.

To demonstrate this, I’ve annotated the following overhead maps to visualize a loose concept of “combat flow”. The lines represent optimal or strategically valuable routes from spawn points to objectives, and tend to represent where combat will take place. By analyzing these player flow patterns, we can better appreciate how each phase of the match demands that players redefine and reinterpret the combat space.

Phase One – Control Points A & E

Steel - Capture Point A

The first optional control point available for capture is A. It is situated in the southwest corner of the map, inside a small shack atop a gently sloping hill. The defending team can reach A by simply stepping through the western gate of their spawn camp. They have a 60 second setup phase to fortify the control point undisturbed.

To capture A, the blue team has a choice of two main attack routes. The spawn camp’s main exit opens into large wide corridor winding around to approach the point from the west. Curving around the enemy spawn point in this manner affords a certain degree of safety. However, this route has several drawbacks: long sight lines make defending Snipers a serious threat, a blind corner could conceal an enemy sentry nest and Spies often lurk in the open plain’s nooks and crannies.

The alternate northwestern exit of the blue team’s spawn sends players onto an elevated plateau. This point overlooks the capture point, making it an ideal shooting range for Demomen and a rapid launch point for Medic ubercharges. To dissuade blue players from poking over the ridge, astute defenders will often place a sentry gun on the elevated platform in the back corner of the base.

Control point E is open for capture throughout the match, though it is exceptionally difficult to do so early on. It’s located on a small circular platform surrounded by a bottomless chasm, a gap that can only be crossed by double jumping (Scouts) or rocket jumping (Soldiers/Demomen). The path to E is quite long, winding past locked capture point B. Furthermore, the defenders can quickly respond to an attack on E by exiting through their spawn camp’s eastern gate or sniping from the western window.

Control point A can be a tough nut to crack, so an all-out assault is often not be the best approach. Skilled attackers will apply calculated pressure at E in order to pull some of the defensive force away. This allows their team mates to make a push for A while the enemy forces are distracted. The tactic of splitting the defender’s focus in this manner is crucial to offensive success in Steel.

Phase Two – Control Points B & E

Steel - Capture Point B

Claiming control point A unlocks two nearby doors: one through the back of the shack leading to D/E and the other opening a short underground route towards B (not shown on above map). The first path permits the attackers to stealthily assault E, though the efficacy of splitting the defenders in this phase is greatly reduced due to the close proximity of B and E.

Control point B is a rectangular yard in a courtyard surrounded by tall buildings. A small wooden fence blocks the western approach. Since the point isn’t immediately visible from the defender’s spawn point, it can be quite difficult to recapture it once the red team starts pushing. An alternate path from the defender’s spawn camp leads to an east-facing second storey window, an ideal camping spot for defending Snipers and Demomen.

The blue team can mount their attack along two parallel paths. Travel on the lower path is slightly longer, sloping upwards into the underside of B. The route bisects a man-high wall, which can function as loose cover for both sides. There’s also a ramp leading to the roof of a small shed, but this detour rarely provides an offensive advantage. The higher path leads to a ridge overlooking the entire courtyard, an advantageous position that carries the risk of high visibility. There is also a windowed hut on this ledge, just opposite the one available to the defending team.

Phase Three – Control Points C & E

Steel - Capture Point C

If the blue team manages to capture control point B, two major changes occur: the door blocking the eastern path towards C slowly opens and the red team’s spawn camp is shifted to the far northern side of the map. Though the defenders continue to spawn right outside E’s northern approach, the attackers have effectively gained control the entire southern half of the map. You may notice that the blue team’s camp is now far away from current objectives; this is a key element of the attack/defend asymmetry, but the disadvantage can be negated by the strategic construction of teleporters.

The area around C is characterized by a strong contrast between high ground and low ground. The control point is located on an elevated balcony nestled into the back corner of the base. The red team also controls a small ledge just south of their spawn camp overlooking C. Both of these positions are high above ground level, offering long sight lines for Snipers and elevated sentry nooks for Engineers. Fighting in the pit below often devolves into a messy brawl, somewhat favouring Pyros and Scouts.

The blue team controls an opposing elevated ridge on the southeastern side of the base, which is a great place for Engineers to set up camp. Large rock crenellations offer convenient cover for Soldiers, Snipers and Demomen to bombard C’s defences from. The western path is also a strong candidate for offensive fortification, as it allows quick access to both available control points. From here, blue players can move between C and E almost as quickly as red players, splitting the defender’s focus very effectively.

Phase Four – Control Points D & E

Steel - Capture Point D

If control point C is captured, three bridges extend across the chasm surrounding E. This layout change has two important consequences. Firstly, whereas control point E was previously available only to Scouts, Soldiers and Demomen, it can now be captured by any of the nine classes. Secondly, a major central route now connects the north and south ends of the map. Both of these factors make E significantly easier to capture, spelling doom for uncoordinated defenders.

The red team can immediately reach D and E through the spawn camp’s western and southern doors respectively. With the final control point now vulnerable, the defenders must be vigilant about concurrently defending both points. This is usually achieved by moving between bases as threats arise, though a cleverly constructed sentry gun can cover both approaches at once.

The blue team can now take advantage of a door that was unlocked in the first phase, allowing them to traverse the long northwestern path from A to D. The winding eastern route previously used to attack B and C now becomes the principal attack vector for E.

There are also two minor attack routes that I omitted from the diagram for clarify purposes. Firstly, any blue players left straggling at C at the end of the last phase can sneak westward by passing in front of the red team’s spawn camp. Secondly, there is a window overlooking E from A which is ideal for Snipers (you may recall that this nest was previously under the red team’s control in phase one.)

Phase Five – Control Point E

Steel - Capture Point E

Occasionally the blue team will manage to capture all four optional control points before capturing E. Claiming D blocks off the red spawn camp’s southern exit, forcing defenders to take the longer western route to get back into the fight. The three bridges to E also sprout guardrails, protecting players from being pushed into the abyss. The blue team can continue their assault via the roads from A and B, or attempt to intercept respawning red players at D. These layout changes put the red team at a severe disadvantage; their only hope is to make a noble last stand and try to run out the timer.

I suspect that these diagrams and descriptions will fail to evoke the real experience of playing Steel. Beyond the inherent limitations of projecting a 3D space onto an overhead map, there are multitudes of small clever design decisions that these high-level illustrations fail to capture. However, I hope these annotations do reveal how Steel cleverly creates five distinct combat scenarios superimposed over one layout. This unusual map rewards those who can reinterpret the changing space around them. Failure to attack or defend Steel is often more fundamentally a failure of adaptation and imagination.

→ 4 CommentsTags:  · 

© 2007-2015 Matthew Gallant, hosted by A Small Orange, powered by Wordpress.