Guerrilla Games: From Killzone to the Machines

Studios that build a reputation on one genre rarely leave it, and the ones that try usually stumble on the way — the muscle memory built across a decade of shooters doesn’t disappear just because a new brief says “open world” on the cover page. Guerrilla Games is the exception worth studying, because the Amsterdam studio’s entire second act, built around a robotic megafauna and a bow, is now more commercially and critically significant than the linear military shooter series it spent its first decade perfecting.

A studio born from a national industry still finding its feet

Guerrilla’s founding in 2000 came at a moment when the Netherlands had no established tradition of large-scale console development to draw on — unlike the UK’s home-computer-fed pipeline of studios or the more mature Japanese and American console industries, Dutch game development was still a comparatively young scene. That context makes the studio’s rapid rise to first-party, console-launch-tier status within a single console generation more notable than it might first appear: Guerrilla had to build an entire studio-level production discipline largely from scratch alongside its actual game, without a deep domestic bench of console veterans to hire from in the way a UK or Japanese studio of the same era could.

The Killzone decade

Founded in 2000 and acquired by Sony Computer Entertainment in 2005, Guerrilla spent its first console generation building Killzone (2004), a PS2-era first-person shooter positioned deliberately as a first-party answer to the console-defining military shooters of the period. The sequel, Killzone 2 (2009), is the entry most closely associated with the studio’s technical reputation — a visually dense, corridor-driven PS3 shooter built to show off the console’s hardware in the same way first-party shooters at Sony were routinely tasked with doing. Killzone 3 followed in 2011, and Killzone Shadow Fall arrived as a PS4 launch title in 2013, giving the studio the distinction of having provided a hardware-showcase shooter at more than one console generation’s launch.

What that decade actually built, underneath the specific Killzone storyline, was a studio-wide competency in large-scale, technically ambitious world rendering under a tight, linear authorial structure — corridors and set-piece arenas dense enough to sell a new console’s graphical capability to a room full of skeptical press. That competency would turn out to be exactly the foundation the studio’s next act needed, even though nobody outside Guerrilla could have predicted where it would get pointed next.

The pivot nobody saw coming

Horizon Zero Dawn (2017) is one of the more startling genre pivots in modern AAA development: a studio with thirteen years of linear military shooter experience released an open-world action-RPG starring a huntress armed with a bow, set in a post-apocalyptic landscape roamed by robotic animals, with no prior Guerrilla project suggesting the studio had this game in it. Directed by Mathijs de Jonge, with a world and narrative built around a central mystery about the collapse of a machine-dependent prior civilisation, Horizon Zero Dawn succeeded largely because of what the Killzone years had actually taught the studio: the robotic machines that populate Horizon’s world are essentially Killzone-grade technical showpieces, repurposed from enemy soldiers into enemy wildlife, with the same rendering ambition pointed at an entirely different kind of encounter design.

The bow-and-arrow combat at the centre of that encounter design borrows directly from the studio’s shooter background too — precision aiming, weak-point targeting and ranged positioning are all skills a decade of first-person shooter development had already refined, simply relocated from a human enemy roster to a bestiary of mechanical predators with visible, targetable components.

The engine that carried both eras

One piece of continuity runs underneath the apparent discontinuity between Killzone and Horizon: Decima, the proprietary engine Guerrilla built up across the Killzone years and then substantially reworked for Horizon Zero Dawn’s open-world demands. Decima’s later, publicly documented use outside Guerrilla’s own games — licensed to Kojima Productions for Death Stranding — is a useful outside confirmation that what the studio built to solve its own rendering problems turned out to be general-purpose enough for a completely different developer’s completely different genre. That’s a smaller-scale version of the exact same pattern as the Killzone-to-Horizon pivot itself: a tool built to solve one specific, narrow problem turning out to have transferable value once someone was willing to point it somewhere else. The engine’s evolution from corridor-shooter renderer to open-world-weather-system renderer mirrors, almost exactly, the studio’s own evolution across the same years.

The family-friendly detour

LEGO Horizon Adventures, released in November 2024 through a collaboration between Guerrilla and Studio Gobo, is the smallest and least-discussed entry in the studio’s recent history, but it’s a telling one: a co-operative, family-friendly reimagining of the first Horizon game’s story, built for an audience the mainline series was never designed to reach. That a studio built on hardware-showcase military shooters now has a LEGO spinoff of its own IP in its catalogue is as clear a signal as any that Horizon has become Guerrilla’s primary identity rather than a successful side project — secure enough as a property to support a spinoff built for players who will likely never touch the original Killzone games at all.

A PlayStation Studios sibling with a similar arc

Guerrilla’s transformation reads even more distinctly next to Insomniac Games’ rather different studio history, another first-party PlayStation developer with a multi-decade record, but one built on running several established franchises in parallel rather than retiring an old one for a new one. Where Insomniac’s stability comes from never letting one series crowd out another, Guerrilla’s defining move was the opposite: a studio willing to let its founding franchise recede almost entirely in favour of a single new property, betting the studio’s future on one pivot rather than spreading that risk across a portfolio. Both strategies have worked, within PlayStation Studios’ broader stable, which says more about how much room a platform holder with enough first-party studios can give any single one of them to take a genuinely large risk than it does about which approach is objectively correct.

Forbidden West and the sequel’s harder problem

Horizon Forbidden West (2022) faced the classic open-world sequel challenge: a bigger map, more systems, and the risk that scale alone would substitute for design intention. The full read on why the machines deserved a better menu argues that the game’s combat and creature design remained as sharp as the original’s, even as the surrounding progression and inventory systems grew in a way that occasionally worked against the clarity the first game had established — a common growing pain for a sequel asked to be bigger without a correspondingly clear idea of what the extra size was actually for.

A remastered version of the original, Horizon Zero Dawn Remastered, followed in 2024, giving the studio’s breakout title updated presentation matching its sequel’s fidelity — a housekeeping release rather than a design statement, but a sign of how central the Horizon property had become to Guerrilla’s identity within seven years of the first game’s launch.

What Killzone actually got right, underneath the reputation

It’s worth resisting the temptation to write the Killzone years as merely a warm-up act for the studio’s real work, because doing so undersells what those games achieved on their own terms. Killzone 2’s weighty, deliberately unresponsive-feeling gunplay was a specific, debated design choice — weapons with real recoil and momentum rather than the snappier, more forgiving feel most competing shooters of the era favoured — built to make every firefight feel like it had physical stakes rather than pure reflex arithmetic. That design philosophy, prioritising the feeling of a mechanic over its immediate accessibility, is the same instinct that later let Horizon’s combat design ask players to read an enemy’s exposed weak points rather than just spraying damage at a health bar. The throughline isn’t only the technology. It’s a studio-wide comfort with combat that asks for patience and observation rather than pure speed, expressed first through a rifle and later through a bow.

What the pivot actually proves

The Killzone-to-Horizon transition is worth studying specifically because most studio pivots this dramatic fail, and the ones that succeed rarely do so by accident. Guerrilla didn’t abandon its shooter-era competencies when it changed genre — it relocated them. Precision ranged combat became the bow. Hardware-showcase rendering became a living, weather-driven open landscape. Linear encounter design against human soldiers became modular encounter design against modular machine anatomy, where different components of a single enemy could be targeted, disabled or exploited independently, in a way that’s a direct descendant of a shooter studio’s instinct for weak-point design, applied to a wildlife simulation rather than a battlefield.

That’s the actual lesson in Guerrilla’s history for any studio contemplating a similar leap: a genre pivot that survives contact with a real audience usually isn’t inventing a new set of competencies from nothing, it’s finding the genuinely transferable skill underneath the genre’s surface trappings and pointing it somewhere the studio’s prior work never suggested it could go. A military shooter studio and an open-world RPG studio don’t look, on paper, like the same organisation. Guerrilla’s twenty-five-year record says they can be.

The clearest proof is how little friction the transition actually produced once it happened. Horizon Zero Dawn didn’t read like a shooter studio’s tentative first attempt at something unfamiliar — it read like the game the studio had quietly been building toward all along, without anyone outside Amsterdam realising that’s what all those Killzone years had actually been rehearsing for.