This presentation by Vlambeer’s superstar dev Jan Willem Nijman is a good one, highlighting a great many of the tools he personally uses to create the house style of gameplay that defines the Vlambeer experience for many of its fans.
It is, however, a little (for lack of a better term) unscientific, using subjective terms such as “it feels good” to describe a certain effect’s result in relation to player feedback. In this writing, this shall be expanded upon.
Thus, “The Psychology of Screen Shake”.
Psychology is indeed where the crux of player feedback lies. In an interview with Rock, Paper, Shotgun Jan describes the way gameplay hinges on input from the player:
The problem with (and cool thing about) games is that we need player input before our systems become of any value. Game enters player > player puts input into game > game reacts > reactions enter player > player thinks about reactions. That last bit is where it becomes valuable. What we do at Vlambeer is spend a lot of time working on the “reactions enter player” part.
This describes how a game must first present itself and the challenges it contains to the player (“game enters player”) to prompt the player to interact, before the game ‘reacts’ in some way to facilitate an iteration of this process; the game displaying information about itself for the player to then react to and interact accordingly is an ongoing process. How we as developers can make sure that these interactions are perceived positively (in terms of good game ‘feel’) by the player is a matter of psychology.
We, as human beings, are creatures with a finite and rather limited life span; a fact reflected in many games that have ‘lives’ or any other similar penalty conditioning for failure to perform expected game tasks. In the real world, there isn’t any solid empirical evidence to support afterlife and reincarnation theories so to many people, this one shot is all we have. You only live once; and that once has to really matter.
As I’ve stated before (and this is by no means a rare assertion), games’ strength are their interactivity. That a player has some meaningful control over their game world is a large part of the appeal. They can do the thing, if the game allows them to.
In a game, the entire world that is known to a player at any given moment is the screen. This window into the virtual spaces of the game’s universe represents all that the player knows they can achieve in the moment; places to go, enemies to vanquish, obstacles to overcome, items to acquire. It is everything in that one moment. When Jan Willem talks of large explosions from arcade games “feeling” really good in his games, it’s possible that the reason for this is due to how much on-screen real estate the conflagrations command. A larger effect invokes a feeling of much greater change, courtesy of the power transferred from the game itself to the player’s fingertips.
And players love to be seen to have control over their game world and leave their mark; which is why some fighting game players have pet ‘win pose’ attacks they like to perform when an opponent has been felled but inputs are still possible on their end; which is why first-person shooter players ‘teabag’ defeated player foes, to make their presence felt in a large way on the recipient’s end.
And so, the screen; this epitome of all the player knows at any moment in time. Affect the entire screen, and the player knows that they have changed the world in front of them. Many games do things that alter the screen in some way; in Doom, gunfire lights up every visible sector by two units to emulate light from a weapon’s muzzle flash; the King of Fighters series tends to reward hits from Desperation attacks with a white (or sometimes red) screen flash on connected hits; the Street Fighter series (from Super… II Turbo onwards) covers the entire screen with a radiant ‘starbust’ effect when finishing with one of its Super Combination Attacks; Tempest 2000 thought absolutely nothing of filling the entire screen with particle effects any time a player did anything beneficial.
There is a whole gamut of effects out there that alter the entire observable game world in some way by manipulating the totality of contents seen on the screen.
But why screen-shake in particular?
I offer the answer; because half of the player input cycle Jan Willem described is about the game information ‘entering’ the player. Information must be conveyed, and clearly.
A screen flash not only represents an accessibility hurdle for those who have epilepsy, but for the duration of the flash obscures practically everything in the game world. There is a difficult balance to be had in making the flash less rapid and potentially triggering, to not being on screen so long as to impede a player’s persistence of vision with regards what the game world looked like beforehand (and will do after).
A starburst (or similar) overlay has the same problem in the latter case, although is thankfully used very sparingly.
Throwing particles around the screen (or even blurring it) can obscure all-important information the player may need to consider whilst considering game contents.
When shaking the screen, however, nothing is changed. Not really. None of the contents are manipulated in any real way, just offset at slightly different X/Y coordinates per frame. This makes it not only a powerful effect, but a safe one. The player perceives that they have performed an action that changed the entire world, even if it didn’t.
And this concept of informing a player of the changes they have made to the world resonate with a lot of Nijman’s points here: permanence for example is nothing but leaving a mark on the game world, in a less transient sense than a mere screen effect. Enemies’ entire bodies flashing on damage, large bullets, added ‘weight’ to camera moves. To make a player’s every action – no matter how insignificant – provoke feedback in disproportionately large ways, that’s how game ‘feel’ is enhanced. By informing the player that their interactions, nay, their actions matter.
That’s the psychology of screen shake.