The camera serves two masters

A camera looks like a technical choice and behaves like a design one. Widen the view and the player sees the threat early, and also loses the precision to hit it. Tighten the view and the aim gets honest, and the ambush becomes a cheat. Flatten the world to two dimensions and the gap between two ledges is a number you can trust. Leave it in three and the same gap is a guess about depth.

The second master is the picture. A dramatic angle is a worse angle for jumping, almost every time. Games that try to pay both debts in one frame end up with a camera that fights the player in the exact second the jump matters. The honest fix is to pick the master for that moment. Gameplay owns the camera during the verb. The picture may have it during the breath between verbs. A camera that owns both at once usually fails the verb.

A game camera works for two masters at the same time, and the first one is the gameplay. Widen the angle and the player gains peripheral vision. Zoom in tight and precise aiming gets easier. Flatten the scene to two dimensions and judging the gap between two characters becomes trivial in a way that 3D rarely allows. A linear game can live happily on rails. A game built around exploration usually has to hand the camera over to the player.

The second master is aesthetic. A very wide shot shrinks a character into insignificance. A tight one makes the player feel boxed in. First person reads as immediate and embodied, while a high vantage point reads as power and detachment. Film worked this out long ago: distance, angle, and perspective set tone, mood, and power dynamics before anyone speaks a line.

That hands designers a tool that is useful and a little dangerous. A camera can quietly nudge the frame toward a better picture. Shadow of the Colossus does it while you ride the horse. It slides the view so the action lands on the rule of thirds, instead of pinning the rider dead center. An image is stronger when the subject sits slightly off to one side. The camera stays playable and composes the shot at the same time.

It can also be taken out of the player's hands almost completely. The classic Resident Evil games ran on static pre-rendered backdrops, so every room looked like footage from a different security camera. Part of that came from the original PlayStation's hardware budget. The rest was control. The developers decided exactly what you saw and the moment you saw it.

A fixed camera lets the designer choose what the player knows, second by second. The licker reveal in the original Resident Evil 2 is the demonstration. The first angle uses the leading lines of two walls to drag your eye to a window, where something strange slides past. The next shot seems to stand where that thing was standing, which turns Leon into prey. Then the view cuts high and he looks small.

From there the sequence tightens by rationing information. Leon has to run toward the camera and around a blind corner, so the road ahead is invisible. One angle frames a broken window and a decapitated zombie while hiding the route forward. Another lingers on a bloody puddle, a dripping ceiling tile, and more broken glass. You creep toward the edge of the screen because the camera has made that edge the threat.

This is a very particular flavor of horror. The room is not so much shown to you as withheld from you. Composition, blind spots, and hard cuts push you to picture whatever waits outside the frame. The price is real, since you cannot simply look at what you want to look at. So is the payoff. The view itself becomes the frightening thing.

The 2019 remake of Resident Evil 2 traded that away on purpose. It dropped the fixed angles and parked the camera just behind Leon's head, which makes every encounter close and immediate. The staged licker reveal loses its theater. A zombie lunge gains weight. You end up nearer the character, nearer the attacker, and nearer the instant of contact.

Bringing the camera close to a character generally makes that character easier to relate to, and up to a point that holds. The 2013 Tomb Raider reboot leans on it hard. The jittery close-up ties you to Lara's fight to stay alive, framing her body language, refusing to back off during the worst moments, and staying with her through flooded chambers and climbs that turn your stomach.

That camera even runs a dedicated physics system so it can bump and jolt along with her scrapes. Animation carries the rest of the load, frequently turning Lara's face toward the lens so fear, strain, and relief stay legible. Back the camera off and the action gets clearer, but the link to her body and her expression goes with it.

The cost is peripheral vision, plus the character eating a larger slice of the screen. A shooter can absorb that, because hitting a distant target matters more than seeing everything immediately left and right. A camera parked behind the shoulder keeps some of first person's directness while still showing the hero, which supports cover systems, traversal, and identification.

Action games make the tension sharper. The original God of War games pulled way out and handed the player a wide read on the battlefield, which suited a combat system full of crowds, sweeping attacks, and area-of-effect magic. Kratos looked distant, closer to an action figure than a man, but the fight was easy to parse.

God of War on PlayStation 4 inverted the priority. The camera sat tight behind Kratos, binding the world to his point of view and serving a more human story. Narratively that is the right call. For combat it costs. Enemies arrive from outside the frame, and judging the gap between Kratos and a target while swinging the axe is hard. Throwing the axe works beautifully, because that action behaves like a third-person shooter. General melee is where it strains.

Camera distance does not only change what you can see in a fight. It changes how the fight feels to be in. The older God of War games could make you feel powerful and predatory, because the camera presented the battlefield as territory to dominate. The tight camera in the newer one can make the same player cautious, even backing out of the melee to avoid being hit by something off screen.

Close cameras are not disqualified from action. Hellblade and Dark Souls both hold the player close, and both are built on deliberate movement, reaction, and usually one serious enemy at a time. Put a crowd on screen, as Devil May Cry does, and a wider perspective earns its keep, because reading everything at once is the skill being tested.

The God of War team did an enormous amount of clever work to make the new viewpoint survive combat. Arrow indicators mark where off-screen enemies are standing. Atreus calls out threats behind you. Enemies outside the frame turn down their aggression so the visible ones can hold your attention. Kratos snaps magnetically toward enemies just out of reach, papering over the depth-perception problem the camera created.

They also capped how high a juggled enemy can travel so it never leaves the frame, and added a lock-on button so nobody has to wrestle the right analog stick. A GDC Vault talk walks through the years that went into fitting the combat to the view. The result is genuinely impressive, and the sheer volume of support is the tell. Choose a camera for intimacy and the rest of the combat system pays for it.

There is an easier route than making one camera solve every problem: let it change. Cameras already move during play. Most third-person shooters zoom in the moment you aim, trading width for precision. Spider-Man throws the camera far back while you web-swing, shrinking the hero on screen so the route ahead stays readable. Vanquish hangs back during a rocket slide to sell the speed.

God of War does make small moves of this kind, pushing in when the axe flies and easing out a little when Kratos runs. Other games commit much harder, reshaping the view around what the player is doing. Batman: Arkham Asylum is the clearest case, because it accepts that a complicated 3D game does not owe itself a single permanent viewpoint.

Walking Arkham's corridors, the camera rides close behind Batman's back, keeping you attached to him and to the building. Start a fight and it swings dramatically back so the whole scrap is visible, automatically locking on to whoever he is hitting. The way it sweeps around can feel frantic, but the intent never wavers. Frame the fight, and keep what matters on screen.

Bayonetta pulls a related trick. Certain duels rotate the camera toward a side-on view, borrowing the readability of a 2D fighting game for as long as the fight needs it. Neither game treats every moment as the same kind of play. The view follows the information the player needs.

One game can easily need half a dozen cameras, and Batman runs through most of them. Crouch and the game reads it as stealth, pulling back for peripheral vision on nearby threats. Take cover and it hands you a fixed angle that sees around the corner while still framing him well. On a gargoyle the field of view opens wide enough to take in the room. A vent switches to first person. Some Scarecrow sequences flatten into two dimensions so platforming is all you have to think about. Poisoned hallucinations tilt into a dutch angle to show that he is off balance.

Rocksteady's insight is that a many-sided 3D game has no one-size-fits-all camera. The view is allowed to shift and morph to fit whatever is happening. You still feel close to Batman, because the default spends most of its time behind him, but the game has the sense to give ground when you need more information.

God of War had the same option available in a stronger form: hug Kratos while exploring, then open up once enemies show. Nothing about that requires breaking the continuous-shot presentation. The camera can travel within the same shot, keeping the cinematic ambition and dropping most of the combat cost.

The broader lesson is not that close cameras are a mistake. It is that distance, angle, and control should answer to what the player is currently trying to do. A camera built for story scenes can be a cage in combat. A view that makes traversal thrilling can ruin precise aiming. A horror camera that hides information is brilliant in a corridor and unbearable in a brawl.

None of this is an argument against strange cameras. If nobody had pushed on the viewpoint, there would be no Resident Evil 4 with its pulse-pounding action and no Resident Evil 7 with its body horror. Pressing against the assumptions of a perspective is where the good surprises come from.

Received wisdom says first-person platforming is a terrible idea, yet Mirror's Edge works anyway, helped by a wide field of view and responsive full-body animation. Fighting games are supposed to be side-on, yet ARMS uses long-range attacks and the shifting distance of incoming fists to let the player judge space from somewhere else entirely.

Tony Hawk's Pro Skater followed the skater, and Skate found a different feel by following the board instead. Firewatch gets character across through expressive animation, even though a first-person camera usually buries the player character's personality. What these share is support. The camera choice is backed by the mechanics, the animation, and the feedback built around it.

The camera is one of the strongest tools on the design table. It can carry cinematic intent, and it can make a game feel intimate, frightening, expressive, or fast. It turns bad the moment it makes playing worse, and no document will settle that argument for you. Build the view, get the game running, and play it yourself. If the camera fights you while you play, it is wrong, however sound the reasoning behind it looked. The camera should fit the gameplay, not the other way around.

In the Flockbay app, build one jump and lock the camera where the gap is readable, even if the screenshot suffers. Play the jump. Then try the dramatic angle. Keep the one you can land. The picture can have a different shot.

An AI platformer maker and a 3D platformer template will offer you a cinematic camera. Refuse it on the jump.