Every once in a while, I check out a batch of interesting-looking Flash games. Most of them lose their charm after a few minutes, but here and there an occasional gem crops up. Continuity is one of these gems.

The game’s gotten plenty of publicity so it doesn’t really need my attention, but I did want to talk a bit about its execution. Combining platforming with a tile sliding puzzle is a clever genre-mashup, but it’s Continuity’s overall package that grabbed my attention.

The first level contains only closed off tiles, with one tile cleverly serving as the controls layout, and the other taking the role of filler to keep the tile sliding mechanic consistent.

The visuals are sparse and abstract, but also very clean. Two music tracks accompany the action, one in the platforming section and one in the tile sliding section, and a quite a few sound effects are used for feedback. Level progression follows a nice, gradual curve, and every time a new concept is introduced (e.g., switching tiles in mid-jump or using multiple keys), the complexity of the layout is scaled back.

And then there’s the interface.

When I first saw screenshots of Continuity, I was half-dreading switching from keyboard controls to mouse controls every time I wanted to move one of the tiles. I pictured dragging the cursor to the appropriate square, clicking the mouse button (or worse yet, holding it and physically sliding the tile), watching the tile glide to its destination, moving the cursor back to the tile with my character, once again clicking the mouse button, and finally going back to the keyboard-controlled platforming.

The empty spot is in the center, and any of its 4 adjacent tiles can be slid into its position at the touch of a button.

But no, Continuity surprised me with yet another elegant design decision. The spacebar toggles between the platforming and the tile sliding, while the arrow keys control all movement. This includes the player avatar during the platforming sections, and the tile sliding during the puzzle sections.

The platforming is fairly straightforward with the left and right arrows dictating direction, and the up arrow serving as the jump button. Keys and doors are also picked up/activated automatically, removing the need for any extra input.

The tile puzzle uses the arrow keys as well, but in a slightly different fashion. At any one time, there are a maximum of 4 tiles that can slide into the single empty slot. If the player wants to fill the gap with a tile that’s underneath it, he simply presses up and the tile slides into its position. This creates a new empty gap, and the arrow keys get re-mapped to its adjacent tiles.

Some of the later levels employ rather complex tile-edges that make for numerous valid combinations.

It’s an interface that’s custom made for the requirements of the puzzle section, and it’s very intuitive. The player never has to select the tile he wants to manipulate either, which speeds things up quite considerably. This is especially important since Continuity’s gameplay requires lots of tile sliding.

I don’t think a lot of people take notice of such things, but that’s the way it should be; if it works well, it shouldn’t really stick out. It did for me, though, and it made me spend a lot more time with the game than I do with typical Flash offerings.

Then again Continuity has the polish and depth that many similar titles lack, so that shouldn’t be too surprising.

A little while ago I dug into Scary Girl for not being a very fun game. This brought up some discussion about what actually makes a good 2D platformer, so I decided to expand on the topic. Below is a list of what I see as three common aspects of many classic platforming titles. These point are not the only things that made those games great, but they’re a shared base that appears again and again.

Sonic CD saw the addition of the dash move. It left Sonic more vulnerable than his spin dash, but it lasted longer and was a great way to instantly build up speed.

The moveset is a nebulous term that encompasses all of the player character’s abilities and properties. These include the standard run and jump, but also other mechanics such as sliding, and the rules of various behaviours, e.g., how much time does it take to accelerate to a maximum velocity, how does the character react when he’s hit, etc. Temporary powerups such as Mario’s Starman and permanent modifiers like Mega Man’s boss weapons also fall under this umbrella.

So what makes a great moveset?

Well, let’s start at the beginning. First of all — and it’s hard to believe that this needs to be explicitly stated —  the moveset should be very clear and accurate. There’s nothing worse than running over a collectible and not picking it up. Well, unless it’s getting hit by an enemy that’s clearly not touching you. This kind of stuff is incredibly frustrating, and it makes the player feel cheated by the game.

Mega Man's wall-slide and wall-jump mechanics added a lot of gameplay elements to the famous series.

As soon as we’re sure that we’re not actively pissing off the audience, we can build a connection between the player and the game itself.

To start off, the interface needs to be quick and responsive. Input should have an immediate effect on the character in order to foster a sense of full control. Granularity and different control techniques, i.e., pressing, tapping and holding, are also important as they provide a level of precision to the movement.

It’s important to note that the majority of 8-bit and 16-bit games actually ran at 60 frames-per-second. Sure, many of the animations were composed of only 2-5 frames, but the actual motion of the sprites was very smooth. This not only aided the physics, but also created a very dynamic sense of movement.

Yoshi's Island has one of my favourite movesets of all time.

Now precision in platformers is often associated with pixel-perfect leaps that — if not properly executed — result in game death. While that is sometimes the case, precision is an ever-present facet of these titles that’s experienced at virtually all times, e.g., jumping up to a moving platform, dashing through a tight tunnel, firing shots at floating enemies, etc.

So how do we actually make the moveset fun?

Well, there’s something to be said for vicariously living through a speedy, agile ninja that performs maneuvers one would not likely do in real life. However, what I consider even more important to the “fun factor” is the integration of the moveset with the various facets of level design itself.

On largely aesthetic level, it helps a great deal if the game is composed of various zones that each have their own unique look. Of course that uniqueness is often accompanied by numerous interactive objects that add variety and help with the pacing, but there’s one small detail that occasionally falls through the cracks: the separation of foregrounds and backgrounds.

Donkey Kong Country has experienced a bit of a backlash over the years, but it was a stellar platformer with clearly outlined levels.

This might seem like a relatively small issue, but if not handled correctly, it can confuse and frustrate the player. Confusion is rarely a good thing, and pretty art is a poor consolation for jumping on phantom platforms.

Now as far as the environments themselves, it’s not a coincidence that they’re often filled with all sorts of slides, bridges, trampolines, ladders, etc., In a way, they’re simply playgrounds for the player, both literally and figuratively. They cater to the moveset and enhance the flow of the game.

Bonk's Adventure was relatively slow-paced and straightforward, but it included some truly bizarre and entertaining levels.

Smart playgrounds also funnel the player into using his various abilities. Once again, this is to provide a wide array of experiences by fully utilizing existing resources. The funneling itself can be subtle — indicating a path through a series of collectibles — or forced — requiring the player to scale a wall in order to proceed.

A great example of organic playground elements in Akumaj? Dracula X. The long water slide ends with a leap onto dry land, quickly followed by a boat ride with the ferryman.

When designing levels, a guiding approach also helps with creating specific setups for how the player enters and leave a specific area. For example, a path of collectibles can lead to an isolated spot that contains a useful powerup. Getting there requires a series of leaps and wall jumps that take the player through the lines of collectibles and deposit him at the desired destination. On his way back, the collectibles are gone, removing the need to retrace steps and making the descent itself easier and more enjoyable.

A good example of incentive-funneling in Super Mario Bros 3. A flight-leaf powerup is followed by a straight runway that ends with a path of coins leading up into the sky (where more collectibles await).

Picture this scene: your character is jumping through the air, an enemy is homing in on him, and a stray missile explodes and destroys a brick wall. A snapshot of this scene reveals multiple states: jumping, homing in, and exploding/crumbling.

States are basically logical denominations that encompass various behaviours and properties of “game objects.” They give standard enemies and end-level bosses a rudimentary intelligence, and challenge the player to decipher and exploit their patterns. This imbues them with personality, and once again encourages the use of various techniques to defeat them.

Every Mega Man game is filled with enemies that all have unique, state-driven behaviour.

However, states are just as important to the levels themselves as to their inhabitants. Even though it’s easy to visually separate enemies and interactive objects, they’re largely the same thing. Springs, conveyor belts, swinging vines, breakable walls, etc. fill out the playground, and they do a tremendous job of turning a static image into a living, breathing environment.

On the initial playthrough, states encourage experimentation and improvisation. Despite the fact that they’re entirely predictable, they also enhance the replay value. Learning the mechanics and patterns of a game guarantees that the player — at least to a certain extent — will continuously get better at it. In some cases, this can even lead to the creation of speed runs.

And if you really doubt how vibrant a game can be largely through its use of states, click the play button below.

Time the occasional jump, and you can blast through this level at top speed.

Now as you start putting all three of these principles together, chances are you’ll also foster those intentional and emergant moments that players tend to remember.

What moments, exactly?

Well, it’s stuff like using pinball bumpers to dart around in Sonic the Hedgehog while avoiding the boss’ attacks. It’s exploring all new underground caverns in Super Metroid thanks to a special ability. It’s running through a level of Super Mario Bros. without stopping while piranha fish sail above you, but never quite touch you.

The flow created by intelligently combining movesets, levels and state mechanics is a large part of what makes platformers fun.

And as a final point, the combination of all three of these aspects can actually be seen in the intro to Indiana Jones and the Raiders of the Lost Ark, and who didn’t like that?

Yes, it's pretty.

The game’s aesthetics are definitely its high-point, somewhat aping the twisted surrealism of Beetlejuice. Scary Girl even does some interesting things like the animating dialogues — something that I’ve been meaning to throw into a game ever since reading demian5‘s When I Am King.

As a platformer, though, it’s simply bad.

I also found myself scratching my head when my character wasn't picking up the collectibles she was overlapping.

The issue with Scary Girls is what’s endemic to so many indie games: an overabundance and over-reliance on physics.  For every Armadillo Run, there’s ten titles like Pac-Man physics. This particularly hurts platformers as the whole genre relies on “tight” controls and precise movement (even in the easy games).

Yes, Mario, Sonic and Mega Man all had physics,  but they weren’t realistic. The algorithms behind those games were MIN/MAX-ed to attain a certain “feel,” and the level design reflected that. There was usually no need to involve mass, the Coriolis effect, or the actual trajectory of a human jumping ten times his own height in an earth-like environment. Instead, the physics were meant to be fun and intuitive, and the architecture of the levels supported them and the player’s goals.

You need to jump to get the cross of collectibles, but hardly any of the possible trajectories will achieve that.

Sure, N was quite a departure from that, but it wasn’t your typical Flash platformer either. It had a very zoomed out view, a high resolution, lots of different surfaces, etc. The game still wasn’t my cup of tea, but it was aware of its strengths and used them to build unique and entertaining playgrounds. Most physics-based platformers, though, seem to occupy a space somewhere in between N and nostalgic games like Mario, and they’re rarely any good.

Scary Girl’s second stage is the ubiquitous underwater level, and, naturally, it’s even slower and floatier than the on-land action. It also uses tank controls, i.e., left/right to rotate, forward to advance, and it’s a mess. Even though you have to dive, the buoyancy of the water is constantly rotating your character to face up. What’s worse, there’s an air meter, water currents, and painfully slow step-like diving movements. It’s pretty much the complete opposite of fun.

Despite its good looks, the game’s an awkward struggle with no flow. I doubt I’ll ever play it again.

Arsha and her sidekick.

Monster World IV is something of a semi-official sequel to Wonder Boy 5: Monster World 3. It’s a cutesy, large-sprited side-scroller in which the player takes on the role of a young girl named Arsha.

The game is mostly a linear platformer with some rudimentary puzzles and RPG elements, but its highlight is the cute little sidekick Pepe. It’s easy to initially assume that Pepe will help you fight the various enemies you encounter, but he never actually attacks anyone. Instead, he can be used to help Arsha traverse the game’s environments. This might not sound like a big deal, but the there’s lots of variety here:

If Arsha jumps while holding Pepe in her hands (done by pressing the A button to summon him), he will attempt to hold her weight and fly onward. This is effectively a gliding mechanic that’s very similar to Yoshi’s in Yoshi’s Island.

---

If Arsha is holding Pepe while in mid-air, she can also pull down on him and perform a second jump (this effectively becomes the game’s double-jumping mechanic).

As a side note, the second jump’s animation is a somersault, which is a bit different from the standard leap of the first jump.

---

In the Handera Volcano level, Arsha can throw Pepe at mini lava geysers. If he flies over one, he’ll settle in its opening and plug it up. When he turns red, Arsha can jump on him and be launched high into the air.

---

Once thrown, Pepe will stop to press any buttons in his path.

---

In Rapadagna’s secret gardens, a tree exists that allows Pepe to grow each time he snacks on one of its fruits. Although this doesn’t seem to be mandatory — at least at first — it changes Pepe’s physical shape and slightly alters his attributes.

---

Although Arsha can no longer walk while holding Pepe during his grown-up phase, the initial one can easily serve as a shield from fire-based hazards.

---

The fire pictured to the left is actually too big for Pepe to blow out, but he can extinguish smaller ones that are peppered throughout the levels.

---

Arsha is not the best swimmer, and whenever she falls under water, calling Pepe will make him dive in and rescue her. Pepe’s not too fond of water either, though, and he’ll always shake himself off afterwards.

---

While standing on a raft, Arsha can call on Pepe to give her a boost. When Pepe lands in her arms, he’ll automatically start blowing air, propelling the boat in the opposite direction.

---

With a slightly grown-up Pepe, Arsha can stop her frantic sliding in the Ice Pyramid by calling on her weighty sidekick. When Pepe lands in her arms, he will almost immediately stop the slide.

---

If a thrown Pepe passes by a hidden passage, he will immediately stop and point it out. This is sometimes the only way to find entrances/exits, but it’s mostly used in just one zone.

---

In the Ice Pyramid, Pepe can be frozen solid, creating an ice block that Arsha can push around in order to get to otherwise unreachable ledges.

---

Apparently Pepe is also hot-blooded as he can melt right through certain ice blocks that block the player’s path.

---

Pepe will not hesitate for a moment when it comes to sacrificing himself in order to save Arsha. Pictured to the left is the aftermath of Pepe taking a big missile to the face that was heading straight for his companion.

---

Eventually Pepe will turn into a somewhat creepy, gigantic rodent whose song cleanses the entire Monster World.

---

Pepe will also sacrifice himself during your fight with the last boss where he’ll (at least temporarily) get turned to stone.

---

And aside from Pepe, here are the rest of Monster World IV’s significant bits:

• A very colourful, vibrant take on an Arabian theme in an anime style.
  

  The a magic carpet ride to Aegis Island.

• Arsha is a mostly silent protagonist, only speaking during the game’s ending sequence.
• A HUB town allows you to upgrade items and enter numerous other areas — an approach that would become very popular in later 3D platformers.
• A magic genie, one of Arsha’s other companions, can return her to the main HUB at any point in the game. An animation of him carrying Arsha is also played when first entering a zone.
• Arsha’s health meter is composed of two rows of hearts; the red one is her armour rating, while the blue one represents her hit points.
  

  The boss of Handera Volcano.

• Whenever ten blue tear collectibles are obtained, the game pauses and the tears spin out and encircle the player. They then gradually move towards the HUD and combine into an extra blue heart. An interesting side note to this is that the blue tears stop appearing in levels after the player has collected the maximum amount of blue hearts. This is a somewhat casual mechanic as it gives the player plenty of chances to get the highest health rating.
• The hitboxes of enemies aren’t always harmful. In fact, Arsha will often simply bounce off of an enemy (especially when she’s jumping) if she touches one during its non-attack phase. A similar kinetic rebounding mechanic is executed when an actual attack connects, with the reaction being based on the circumstances of the contact, i.e., did Arsha attack successfully, did she get hit, how strong was the hit, was she blocking the attack with her shield, etc.
  

  The screen-blitting that follows walking through doors.

• The first enemy you encounter in the game can’t actually hurt you, while the last ones can attack you with a stream of projectiles from a screen or two away. Enemies gradually become more aggressive, block more of your attacks, gain new abilities and dish out more damage.
• Enemies often have a pre — and sometimes a post — state and animation to their attacks. For example, the pig warriors wind up for their 360 spinning club attacks, and once finished, they stumble over and remain dizzy for a short amount of time. When fighting these enemies, Arsha can quickly move forward and slash her sword to interrupt the wind-up, or wait for the spin to end and attack the stunned enemy.

• 

  Entering the Stream Sanctuary.

  The NPCs are aware of their own proximity to Arsha, which is used for some nice effects. For example, when first embarking on your quest, the townspeople will turn around to face you and wave goodbye as you pass them by. An interesting change up on this is when you later visit Rapadagna — the NPCs will not only turn to face you, but also follow you for a limited distance. This actually gives off a strong vibe that they’re suspicious of you, following you around to make sure you don’t do anything bad.

• In order to reach Aegis Island, Arsha travels down a cloudy runway and eventually boards a magic carpet. This initiates an on-rails, auto-scrolling section where the magic carpet serves as a floating platform. Arsha has no control over it, but she can still jump around and attack approaching enemies, and even temporarily settle down on fluffy clouds that dip under her weight. A nice touch here is that during this segment, Arsha has a unique idle animation. It shows her slightly crouched and bracing her face, giving off the illusion of an unstable platform and strong winds.
  

  Arsha on her way to raise the Mute Tower.

• At the end of the carpet ride, the Aegis Island castle comes into view in the background. This initiates a scripted sequence where Arsha flies “into” the screen, with the castle scaling up to meet her.
• When Arsha falls vertically, the camera follows her at a speed slower than that of her decent. Since these vertical falls are never too long, she rarely goes off-screen, but it can happen. This adds a sense of range to the actual fall, which is further accentuated by Arsha’s automatic animation change — if she’s in a falling-down state for an extended period of time, her animation switches to a somersault.
• Towns consist of various “layers” that can be traversed by pressing up to walk through doors. This transition involves a new screen buffer being blitted over the previous one via an expanding circle. Once this layer circle takes up the whole screen, control is returned to the player.
• Non-town levels also have doors, but these simply teleport the player around in a single layer. They do, however, all look different and initiate unique walking in/out animations. Some are even directly connected to the gameplay, i.e., they require a bomb to blast through, or a key/button combination to unlock.
  

  Blasting through a rock-sealed door.

• As Pepe eats fruits and grows, the ineraction animations between him and Arsha gradually change. For example, Arsha will hold onto just his hind legs when he’s small, but will latch onto Pepe’s whole body when he gets bigger.
• For dramatic effect, when bosses die they freeze in the current frame of their animation, cycle through some palettes, and finally explode in a shower of coins and body parts.
• The HUD is very responsive to the in-game action, which is especially evident with the coin purse. Not only does its numerical value scroll up as you collect money, but the purse itself bounces up and down as long as the numbers are changing. This is also consistent in all parts of the game, with the same effect being employed when you buy and sell items.
• The coins that fall out when you open chests or defeat enemies have the same standard collision detection as most moving entities. This is even used to hint at a secrets, e.g., in Aegis Island, a certain chest spills out coins that travel through a phantom wall, indicating that Arsha can do the same.
  

  The end of Monster World.

• A magical doorway in Aegis Island leads Arsha to a Lilliput land where all the tilesets and enemies are scaled up. Aside from being a cool aesthetic trick, it also plays into gameplay mechanics since the boss you encounter in this area can’t be hurt by the “miniature” version of your character. In order to defeat it, Arsha must push it back with her attacks far enough to go back through the Lilliput door. At this point, the giant blob will follow Arsha but stays shrunk, allowing her to finally defeat it.
• In a somewhat melancholy moment, when you finish the game, the magic genie tells you that this is the end of the series and it’s now time to go outside and play.