The latest season of Respawn Entertainment's battle royale game Apex Legends introduces a beloved shooting style made iconic in the 80s and 90s by director John Woo: dual-wielding pistols. Ever since Chow Yun Fat slid off that railing in Hard Boiled, game developers have tried again and again to capture that particular blend of style and action, but in a medium that has all sorts of shooter genres (top-down, first-person, third-person, etc.), it's a surprisingly difficult mechanic to get right.
Why? Well, just look at the “On the knees” system in Apex Legends to find out. According to an email chat with some Respawn developers, implementing it required “monumental” technical work, a task more strenuous than the “art” of developing the new Season 22 map “E-District.”
While developers like lead level designer Steve Young and map designer Garret Metcalfe were able to experiment with creating new locations designed around the new character leveling systemBR Systems Designer Eric Canavese and his colleagues had to go back through over 5 years of content and code to bring basic pistols like the P2020 and Mozambique into the dual-wielding era.
Apex LegendsThe “akimbo” system opens up new design possibilities
According to Canavese, Respawn originally conceived the Akimbo weapons as unique weapons that players could pick up. This way all the animations, design, and code requirements could be designed like any other new weapon.
While “evaluating” that idea, Canavese said the team saw an opportunity to develop an entire dual-wielding system, which could breathe new life into lower-tier weapons and make them compelling options throughout the game.
But even with a more open system aimed at capturing John Woo’s action fantasy, Respawn’s solution highlights the constraints that first-person game developers find themselves in. Grabbing another gun increases the rate of fire of the given weapon, enables automatic fire, and restricts flank fire to a smaller area.
You can already see a missing piece of the fantasy: the player’s weapons are locked in one direction, eliminating the ability to aim them at different targets. This limited system also doesn’t allow players to dual-wield different weapons like they can in the Halo and Wolfenstein series. These aren’t terrible compromises, but then you run into the hardcore development challenges the team faced.
Apex Legends It's a five-year-old game, and while some core features (like supply pods) have been updated over time, there's legacy code and design direction from five years ago that still needs to be accounted for. Respawn's high-quality first-person animations are deeply ingrained in the weapon and character designs, and holding a gun in each hand is starting to create animation headaches.
For example, the adrenaline-fueled speedster Octane has custom animations that allow players to see him use his stims in many different states of motion. Players already push his animations to the limit when they activate his speed boost while reloading a weapon, performing actions that would theoretically require a third hand to complete. If Octane were to join the team with two weapons, he would need special animations to juggle his stims and his pistols.
All of these animations become even more complicated on ziplines, where players seemingly glide with one hand, but need both hands to shoot.
Then came balance. Giving players two pistols with the same stats wasn't enough. Canavese said they needed to be “completely re-evaluated” in their base form to ensure that when you doubled them up, it felt like a big power spike (but not pure large).
Going all-in on this overhaul has had a huge impact: It’s transformed two weapons that were “largely irrelevant” prior to Season 22 into powerful options.
There's something to be said for having weapons in a multiplayer game that never quite find their niche, but have their moment of glory when a team has the resources to build new systems for them. It doesn't just make that weapon more fun now, it validates the decision to create it in the first place. Sometimes in a live service game, it takes about five years to make good ideas stick.