Do delays after losing and before restarting serve as a defense against burnout, or are they addiction machines?

Question

In games like Super Hexagon or even Everwing, there's a very brief but noticeable delay from losing to starting over: waiting for the restart button to appear, pressing it, etc...

Is this a defense against player burnout? If there's no pause that separates one game try from next, the game might feel never-ending and burn out the player quickly.

Or is it a way to keep the player addicted? I often suspected this in Everwing, because that game's hardly shy about its avaricious nature. Perhaps there's a psychological effect in making the player wait before they can play again — like building the craving part of a habit?

Or perhaps a lack of this delay is to keep the flow going, so to speak? Ori and the Blind Forest does something similar in its Ginso Tree level. It has very fast restarts, no button or anything, and deaths do not restart the music that's playing in the BG. The result is that deaths do not break the flow of the game, and despite them the whole level feels seamless.

So, are there any studies on this? Barring those, anecdotes? I couldn't find anything, anywhere.

Answer 1

While it's possible that some delays are deliberately crafted for the reasons you've described, I suspect that in general, they are simply the result of the time it takes the application to perform any reset related activities: loading textures & assets, clearing buffers, reading & writing to save game files and so forth.

In the event that designer / developer is creating an intentional delay, it's most typically based on 'feel' &/or play testing feedback. Often, studios face enough challenges just getting a game to release stage. Things causing obvious & correctable friction (bugs, poor controls, framerate) usually receive priority. Thus, if players say it loads to slow or the designer notices that players are recklessly repeating mistakes, they may attempt to decrease or increase reset times accordingly.

Generally, only the largest of AAA studios hire professional expertise for tuning this level of a games psychological impact. There are some exceptions: anything from the gambling industry and anecdotally, I've heard that some mid-tier mobile developers tinker with this sort of thing (sometimes via AB-testing).

If you're interested in the psychology aspect of this, you might try looking up things related to athletics, attention & performance to repeated tasks (sometimes experiments use games to simulate other activities) and psychological refractory periods.

Answer 2

In my opinion, the primary reason (apart from technical reasons) to consider having a delay before loading or restarting is:

To give dying some impact

A short delay

... is good if you're playing some simple game where you typically die in a minute. Death will rarely have an impact - you die, you restart, a minute later you're back where you were. That's really not a big deal.

You should, in general, avoid delays (intentional or not) with such games.

Death will, however, have an impact when you've reached some epic high score, or barely missed it. In such cases it could make sense to add a small delay to give the player some time to process it. This could also help with retention, in that they're presumably more likely to stop playing if you give them some time to think between games, and stopping on a high will leave a better impression of the game. One way this is (unintentionally?) done is by asking whether players want to post their score to social media.

A long delay

... can be useful for a game that tries to be more immersive.

A game with no suspense is generally not immersive - it's hard to care if you don't have some "what will happen" / "will we win" moments.

You can potentially do this with good story-telling, or having some other consequences, but the easier way and more common way to do it is with the threat of death.

The problem is that death typically doesn't really mean anything in a game - you just reload and keep going. Thus there needs to be some consequence to dying. Some games have perma-death, other take away things you earned. If you have neither of those things, you need to have some way to make the player believe that them dying is something they should actually try to avoid, otherwise they'll just stop caring when they die, and stop fearing the threat of death, which could remove all suspense from a game.

Thus you add a delay between dying and the player getting back to where they were.

In-game versus out-of-game delays

You have out-of-game delays (e.g. loading times) and, in the case of loadable games, in-game delays (how long it takes to play the game until back to where you were).

These two are usually (or should be) closely linked. If you separate those two, that can lead to problems in itself:

• A short time to restart with a long time to get back to where you were in-game means players won't really be given time to process their death and decide if they want to continue, and may be frustrated by the long trek to where they died if they didn't really want to keep playing.

• A long time to restart with a short time to get back to where you were in-game means players will want to restart often, so they'll be frustrated by the long waiting time.

To make players think

... could be another purpose for a delay, but this applies to different types of games (generally).

Consider some story-focused decision-based game.

Such games are meant to be highly immersive, and death may not really come into play there.

Yet you still wouldn't want players to just instantly restart or load.

You want to give them some time to process and think about what just happened.

Answer 3

This is a question with many answers. Aside from the ones in the other answers, some possibilities are:

• To allow a player to get their bearing. In many games on the more immersive side (or games that put you right in the action after spawning), suddenly respawning can be jarring. A few seconds of waiting time can help players: 1) realise they died 2) prepare for being teleported to a different location 3) prepare to re-engage with the game.
• Preventing the player from doing the wrong contextual action accidentally. If you're busy smashing the "attack" button when you are killed and you respawn into a conversation with a friendly NPC, it'd be annoying if you accidentally killed your friend because you couldn't react to dying in time.
• To balance the game. In team shooters, you want a kill to have an impact on the overall game. If someone you kill immediately is able to respawn, potentially very close to the place they were killed, there is a higher chance of any game ending up in a stalemate, where getting a kill does nothing but push one of the opponents back a bit. Capture the flag scenarios are especially vulnerable to this.

Answer 4

In a game that doesn't require touch input in any particular place, showing the "new game" button immediately runs the risk of the user starting a new game when they did not mean to.

Super Meat Boy! doesn't have a restart button, it just assumes you want to restart until you quit. For me, in that game it serves the purpose of allowing my brain to reset for the next attempt.

Geometry Wars 3 has a Quick Restart button that is only visible for a second or two during the initial Game Over screen. This button is not used for anything else during the game, so as soon as you die you can instant restart if you want. Otherwise, you go to the stats page with the "exit to main menu" and "restart" options.

Answer 5

Having the player watch their character die in an elaborate death sequence and then having them repeat content they already passed is a form of punishment. Punishment and reward are essential tools in game design. They are your main methods to guide the player to improve their gameplay. When you punish the player for playing badly, you teach them to recognize their mistakes and avoid them. So a little punishment once in a while can be useful to keep the player engaged.

But too much punishment can quickly turn into frustration. So you want to adjust the intensity of your punishment accordingly to the severity of the mistake.

The majority of games, especially in the AAA sector, are designed in a way that the player will die very rarely. Most challenges are designed with the goal that most players will beat them with their first attempt. You only want to kill the player if the player makes a gross mistakes or refuses to put serious effort into your game. So if the player manages to get themselves killed, the punishment should be quite severe. That way you convince the player to take the game serious again. So show the player an elaborate animation where their character dies, do a slow fade-to-black, followed by an equally slow fade-in of the words "You died", put them into the loading screen (don't bother to check which assets are already loaded - just dump the whole level and make them suffer through a complete reload) and respawn them at a checkpoint which was a few minutes ago.

On the other hand, there are games where the player is supposed to fail over and over again until they succeed. When you design a section in such a game, then you expect that even the most skilled player will fail at least a few times until they finally figure out how to beat your section. You often see such games in the Indie sector. When you are creating such a game, then death is not a punishment. It is your way to tell the players that they haven't figured out the solution yet. So you want to keep the death sequence as short as possible to minimize the punishment effect and let the player retry ASAP. Examples:

• Super Meat Boy
• I wanna be the guy
• Road to Ballhalla

Answer 6

Devil's advocate here: It's an addiction mechanism.

If you spend a lot of time on a game you become invested. This 3-5 seconds isn't enough for you to quit or to put it down and also forces you to wait before quitting. This builds both time played and holds attention to the game for short periods of time to pad the playtime and hold interest. Playing again is the reward at the end of a 3-5 second timeout.