Mirrors Edge, and it’s sequel, are some of the most natural successors to the Half-Life series. Given that Half-Life is held up as perfection in shooter design, and Mirrors Edge is a) often not considered very good and b) not even a shooter, this is perhaps a contentious statement, but it’s one that makes more sense to me as I think about it.

I’ve been replaying the Mirror’s Edge series, you see: for my next coursework project i’m building a similar free-running game, and there’s a lot to understand in how and why Mirror’s Edge does what it does. As such, apologies if this post is a bit of a ramble, but I wanted to talk about the design and level design of Mirror’s Edge, if only for the purposes of marshalling my own thoughts.

Mirror’s Edge Doesn’t Care If You Know Where To Go

It’s one of the defining features of the series: the bold splashes of red, used entirely as a gameplay device. The game calls it Runner Vision, and while it’s clumsily used as a representation of Faith’s in-character instincts, it’s almost genius as a game design tool. Compare Mirror’s Edge to Prince of Persia, Assassin’s Creed, or Tomb Raider – you could roughly group them all together as “free running” but they’re worlds apart. Minute-to-minute the challenge of PoP, Tomb Raider, et al, is in planning your route: inspecting a wall covered with grooves and vines and scrambling up it. These games often switch neatly between third-person combat and this kind of traversal, and as such get lumped uncomfortably into “Third-Person Action-Adventure”, with “Action-Adventure” as useful a definition of playstyle as “Romance” or “Combat.” I like to think of them as environment navigation games – clumsy phrasing perhaps, but the fundamental challenge is to figure out your route from A to B. Mirror’s Edge just tells you how to get there.

It doesn’t care, you see, and nor should you: that’s not the challenge. The challenge isn’t in what actions to take, it’s in taking them perfectly. You can happily move at walking pace if you want, in the same way as you can play Half-life using only the first pistol. That’s not the point: the point is optimisation. It’s finesse, it’s beauty. Runner Vision is instructions, yes, but the general feel is”yes, sure, do this, whatever, but-“. There’s always more than a few routes to take and Runner Vision is rarely the fastest. On a second-to-second basis, you’re constantly rethinking, optimising, and customising: what if I wallrun and jump-crouch over that fence, instead of vaulting straight over? Is it quicker to hop that pipe or slide under it? Often you’re not moving more than half a meter or so from the suggested path, but it’s your route, and it’s better.

There’s some element of finding a path, too. Catalyst offers challenges where your goal is to find your way up to a billboard, or to a visible but inaccessible collectible, and these often fall flat because Mirrors Edge has so few movement tools: connecting them together to “solve a puzzle” is usually fairly trivial, except when areas are deliberately obfuscated. The challenge runs work better: the star system is brutal, and glorious: getting the full three stars requires not only running your route completely optimally – no mistakes or stumbles at all – but also finding the best route, which is typically easy enough, cutting corners on a grand scale.

Mirrors Edge’s Vocabulary Is Very Small

I mentioned above that Runner Vision as a representation of Faith’s instincts feels a bit janky to me, and at least part of this is because very rapidly, Mirror’s Edge teaches the player a whole bunch of their own instincts.  Both games have a relatively small moveset, and it’s easy to learn the visual cues for these moves quite quickly. I can springboard from an object of this height, can jump this high and wallrun that far. The levels are very uncluttered (the setting here works in their favor well – you couldn’t set this game in the rubble of City 17, thematically similar as they are) and you can spot an object, pick out it’s purpose, and plan your moves around it, all at the speed of thought.

There are visual cues, too, as well as the runner vision. Walkable walls often have gentle shoe scuff-marks on. These work less well, I think – it’s part of the vocabulary again, but a clumsier and more obvious part, and one that only makes a certain amount of worldbuilding sense if you don’t think too much. I’m not sure they’re necessary either, once you know the scales, but I suppose I can’t begrudge DICE for wanting to be as clear as possible. At least it’s not as bad as Tomb Raider or Uncharted, where everywhere you go someone has painted white stripes on all the climbable rocks.

This is all in the name of speed, and the whole point is that you never need to stop and look around to tell where to go next. Here it comes, our first Half-Life 2 comparison: it’s all over this. Light, shadow, and density of detail are valve’s stock-in-trade, but they also use carefully placed collectibles, sharp movements, and well-placed enemies to keep the player moving forwards on instinct. Look at the end of Episode 1, where the player is fleeing a Strider through a maze of containers, and how easy it is to put your head down and just run, the game subtly pulling your attention sideways to turn you in the right direction. Look at the airboat or car sequences (especially the chopper chases), and look at how clear your route is at all times, without anything so gauche as a minimap or directional indicator.

Like above, it’s all about intuition: the player just follows the red lines. They don’t need to know what it is they’re knowing.

Mirrors Edge Handles Pacing Very Well

This is the next major comparison to the Half-Life series I can think of. Half-Life 2 was built on a simple maxim: push the player to the edge, then keep pushing. Then, later, break. Over and over again you can feel it happening to you: the battles last almost until the point of exhaustion, then keep lasting, pushing you through the pain barrier and on, and on, until you reach that moment of calmness and stillness, and it’s all the more welcome. You can see this in all of valve’s combat-heavy games. The Left 4 Dead series explicitly tracks player state so as to dynamically increase or decrease the enemy flow, or even adjust the map on the fly to extend the route. Play through Left 4 Dead 1, and note how long so many of the levels are, how they extend to a punishing slog, so that every moment of respite, every stumble into the saferoom, is truly earned.

In short: Mirror’s Edge does this too. It’s less good at pulling or pushing you onwards, so you can stop at any point and take a breather, but in those scenes where you are being moved forwards, or the few extended fights in the original, you can spot the pattern originally: push you to the edge, then keep pushing. Here’s the problem with the comparison: combat is so miserable that every moment of that “push” is frustrating and wearying.

Mirrors Edge Is All About The Setpieces

It’s wierd, to look at yer Calls Of Duty and yer Battlefields and their bombast – blow up the eiffel tower! Nuke america! – and track it back to HL2 blowing up a factory tower as you go past. The Half-Life games were replete with these setpiece moments, the sharp sideways twists that stay with you long after you finish playing. Some are merely cool things to look at, some are changes in gameplay, others are merely sharp surprises. A highly incomplete list: The room full of explosives, blowing up the tank with the mounted gun, the grenade in the vent, the flight from the scanners (another great quasi-freerun moment), the explosive introduction of the manhacks, the destruction of the factory chimney, shooting down the first chopper, the massive jump into Black Mesa East, the holdout on the rooftops, the first snipers, the lighthouse defense, the razortrain on the bridge (hell, I would say the whole bridge counts), the antlion boss, the garbage crusher, the turret holdout sequences, the first strider, the other rooms full of explosives (they’re a recurring theme), and, stopping here because I’ve been going on too long, the moment when you get your powered-up grav-gun. A shooter without these moments would be dull and unmemorable: one with too many would be unfocused (Call of Duty feels like it wants to be a different game every five minutes).

I’ve criticised Catalyst for having too few of these moments, but it has some nice ones. Watching the mall detonate in front of you is straight out of the Valve playbook, as is the slow arrival into the arena of the first Sentinel.  The first game uses far more and better: catching the train, climbing the storm drain, chasing Jackknife, the jump between cranes (which illustrates another Valve trick – have someone impressed when the player does cool things. In ME1, your handler spits out his coffee in surprise when you make the jump. In Catalyst, Icarus says “cool, now up next…”), the helicopter ride, the fight on the shipping containers. To me, these are an integral part of the experience for both series, and something that no other games really do as well.

Mirrors Edge Is All About The Feel

This wasn’t a fun conclusion to come to, since I know that this will be largely outside my grasp as someone who can’t animate. But: the feeling of Mirror’s Edge is vital to it, more than any other game I can name. If you land poorly, Faith staggers, if you run fast, you can hear her breathe harder. It’s dedicated to putting you in the body of this free-runner and the physicality of the situation. I’ve seen people complain that it’s first-person because it makes it harder to judge the jumps. Firstly, no it doesn’t, are you high, what the hell, and secondly that’s like complaining that the Mona Lisa is a painting not a sculpture, because you want to see the back of her head. I would be arguing for the first-person perspective even if it did make things worse, because the physicality is so crucially important.

---

And now for the game I’ve been obliquely mentioning since I named it in the title: Neon Arena. Neon Arena is the game i’ll be making for the next part of my gamedev course: a first-person free-running game, built in UE4, and set in the arena of a futuristic challenge: players go into the arena to earn fame, fortune, and more fortune, by lasting as long as they can as a wave of drones hunt them down. The scoring system is cash: Players earn money when a drone can see them, but drones can stun or tag the players, slowing them down and knocking them out, forcing them to keep moving. Beacons appear in the level, and reaching a beacon collects money (while failing to reach one in time causes the end of the game).

While I expect it’s outside the scope for the current game, I’d also like to add a “showboat” button, to hit for a context-sensitive bit of pointless acrobatics – a backflip, a taunt, a cartwheel. Performing this while seen by a drone earns you more money from adoring fans, and chaining them together is more effective. Fighting the drones could also be an option too: smashing a drone costs money, but if another drone sees it the fans might enjoy it enough to pay: maybe even earning you a profit if you do it in a cool enough way, like hopping off the roof onto a drone and smashing it against the ground, or slamming it against a lamppost as you sprint past. Frankly, that just sounds fun by itself.

We’ll see how much of that is actually possible: but that’s what these posts are gearing towards. I hope to write more about the development process as I go through: we’ll see how much of that is actually possible too.