I have an HTML game I built with Phaser 3, and I can't decide what to do with the controls. I've changed them several times, and now I feel like I've hit a sweet spot between usability and fun. However, I still see people struggle.

The game involves a boy with a slingshot that can be pulled back and fired to knock the enemy's head off. There are two virtual buttons on both side of the screen that are tapped to turn the character. I'm considering having the character turn automatically, so that the slingshot is the only mechanic involved and the game can be picked up and played without instructions. However, I feel like this will destroy the whole point of being fast to turn and shoot the enemy.

Should a simple game like mine require instructions, or is usability more important?

  • 4
    \$\begingroup\$ Is this all controlled through mouse? Having to turn the character and operate the slingshot at separate times sounds like annoying movement to me if it's all done via mouse. \$\endgroup\$
    – JMac
    Commented Jun 13, 2019 at 19:21
  • 2
    \$\begingroup\$ @JMac On the desktop version I had keyboard inputs to turn, but I’ve decided to go with the auto turn and I’m getting a lot better feedback now. \$\endgroup\$
    – Cannabijoy
    Commented Jun 14, 2019 at 0:06
  • 2
    \$\begingroup\$ Why is there a dedicated button for turning the character? Since this is a slingshot game then I presume you click, hold, drag, and release the mouse to fire a shot, right? If so then click-n-drag to the left should fire to the right and click-n-drag to the right should fire left. \$\endgroup\$
    – MonkeyZeus
    Commented Jun 14, 2019 at 13:05
  • 1
    \$\begingroup\$ @MonkeyZeus You would really need to see the game because it’s hard to explain. I’m trying to avoid linking my game on here, but if you click on my username I have my website on my profile. \$\endgroup\$
    – Cannabijoy
    Commented Jun 14, 2019 at 22:21
  • \$\begingroup\$ It's Castlevania I, where the zombies come from both directions, and instead of being able to move: all you can do is turn to face left or right, and you have an Angry Birds slingshot. @MonkeyZeus - has it right: automatically turn to the opposite direction I'm pulling. If you make me turn, all of sudden I'm pulling in the wrong direction. (which btw, that bugged it out and turned into a shooter. I'm on desktop) \$\endgroup\$
    – Mazura
    Commented Jun 15, 2019 at 2:13

5 Answers 5


Are your test player having trouble discovering the controls or using them?

If they have trouble discovering them, you can add a prompt that explains them. To make it unobtrusive for players that don't need it, you can have it appear if the player stays there doing nothing.

Which, brings me to the next thing: probably all your game needs is a practice/creative/free mode, where players feel free to experiment with the controls without any pressure. That will get them used to the controls in the case that the trouble is using them and not discovering them.

While a game can do well without instructions or tutorials, they could help you reach more people (because not everybody is keen on experimenting without an example or guidance).

Usually I would say that you should let the player configure the controls. However, I recognize that that does not make sense for every game and input device. With that said, consider if you need sensitivity options.

There is also an argument for allowing for a simpler control scheme which will allow players who struggle to beat the game (a kind of assist mode, you can let the player know that it is not the intended way to play, yet it can allow them to enjoy the game anyway if they are having trouble). And this can be implemented without a lose of depth if the full controls are required to get high-scores.


If you are going for a tutorial, the player does not need to notice it. Examples such as NES Mario (which Philipp explains, you might also be interested in Super Mario 3D World's design philosophy) and Megaman X (which is f■■king genius) are great case studies. A more recent example is Portal, which is tutorial for the most part.

See How to Design a Good Game Tutorial.

On the other hand, if you are thinking about tweaking the difficulty, consider that "Hard" does not have to mean (although it often means):

  • Punishing (you fail, you lose a lot of things)
  • Tedious (it takes a lot of time to get things)
  • Inaccessible (hard to understand or control)

Instead, "Hard" can mean challenging. That way, Hard can be fun. Separate challenging from difficult, and you can increase challenge by adding rewards that require more skill, without making completing the game harder.

In fact, who says difficulty is a single dimension concept?

Thus, if you are increasing accessibility, you are not necessarily making an "easy mode". The game can remain hard (challenging). I want you to consider the concept of an assist mode.

To be honest, it is a bit of a branding issue. You see, some games lie to you (have hidden mechanics) as a mean to tweak difficulty. Usually to make it less punishing, however sometimes to make the content last longer (more tedious).

Difficulty in games has lowered in some parts of the industry, and accessibility is blamed for it (even though it is not the only factor). Also, some people want challenge※ and some people want to relax※※... there is no perfect difficulty.

※: Some games keep you engaged the whole way, they often throw unexpected twists to the player, have your adrenaline rush, your heart pumping, and huge dopamine rewards. Yeah, too easy might not be good.

※※: Some games allow you to get in the zone, archive a semi-meditative state as your mind is freed while you body just does the motions - rhythm games are good at this, also crowd-combat fighting games (no, repetition does not have to ruin the game)... Other are fun because they act more like sandboxes, they are more toys than games, where the fun comes from finding creative and amusing ways to do things. Yeah, too hard does not help either of that.

And then there are games that mock the player for picking an easy mode, failing or quitting.

  • \$\begingroup\$ I have a play button and a "controls" button that opens a new page with a picture of the controls. I have also told people what to do, and they still do all sorts of strange things. I'm thinking about just adding the autoturn, and making the enemy faster. I want everyone to enjoy the game, but I also really like your idea of making an "easy" mode. \$\endgroup\$
    – Cannabijoy
    Commented Jun 13, 2019 at 8:54
  • \$\begingroup\$ @Cannabijoy extended answer around "easy" See also affordances. \$\endgroup\$
    – Theraot
    Commented Jun 13, 2019 at 10:31
  • \$\begingroup\$ *kin --> keen "Kin" means "relatives" (roughly). \$\endgroup\$
    – jpmc26
    Commented Jun 15, 2019 at 18:29

There are some games which pulled off the "silent tutorial" thing pretty well. The basic rules are:

  • Introduce game elements one at a time
  • Introduce them with as little distraction as possible
  • Introduce them in a way that it is impossible or at least very difficult for the player to not do what they are supposed to do and discover the mechanic on their own.

A good example for a silent tutorial is the first level of the original Super Mario Bros for the NES. From the perspective of todays gamers, the game mechanics seems obvious. But remember the situation of their players back then. For most of their audience it wasn't just the first platformer they ever played but in fact the very first video game they saw at all. So they had to expect that their audience came to the game with absolutely zero knowledge. That meant a good tutorial was crucial. Did they open the game with a long text explaining to the player what to do? No, they just threw them from the title screen into the game:

  1. The player starts at the left side of an empty screen. There is just one thing the player can do: press right to start walking. This teaches the player the very first rule of the game: "move right to progress".
  2. Then the player encounters the first enemy walking right towards them. Most players actually run into the goomba and lose on their first run. This teaches the player the second rule of the game: "Do not run into any creatures you meet".
  3. At the next try, the players will discover that they can jump. They will jump over the enemy. This teaches rule number 3: "Your primary method to solve problems in this game is by jumping over them".
  4. Interestingly, the most convenient pattern to avoid that gooba makes the player jump into the first question mark block and receives a coin. So the player already learns another mechanic: "Jump against the question mark blocks from below and good things happen to you".
  5. There are more question mark blocks for the player to apply that knowledge. They are grouped with the rock blocks which do not react at all to the player jumping against them. This teaches the player "There are different blocks in this game, and they react differently when you jump at them".
  6. One of the question mark blocks releases a mushroom. The player does not yet know that the mushroom is good for them. It might just as well kill them like the goomba did. But it moves in a way that it is very difficult for the player to avoid it. Most players do touch it, and are surprised that it doesn't just leave them alive, it actually does something good: Their character grows. This teaches the player: "Not everything that moves hurts you. Mushrooms are good".

And the game teaches all that in just the first minute of gameplay without a single word of explanation.

There is a lot more the first level teaches the player in ways they don't even realize they are being taught, including the existence of secrets, common jumping patterns, koopa shells, fire flowers, stars and many more. Analyzing that whole level and figuring out why the designers placed things the way they did can be an enlightening experience in tutorial design.

I am looking forward to learning how to play your game.

  • \$\begingroup\$ Hey thanks for the answer! Sorry I just now realized I didn’t respond. This is a really great answer and I love the way you break Mario down. I’m definitely going to reference this when I build my platformer. This makes me view games in a whole new light. \$\endgroup\$
    – Cannabijoy
    Commented Jun 15, 2019 at 6:52
  • \$\begingroup\$ Another good point about 1-1 is that it tends to rely on antepieces as well, introducing a lot of design decisions in a safe manner to let the player see them and figure them out before there's any actual risk to failing. It also led to a lot of other games building on it for their own tutorials, such as how Metroid includes an impassible obstacle a couple screens to the right of the starting area to teach you that "unlike Super Mario Bros., this game lets you scroll the screen right and left." \$\endgroup\$ Commented Jun 15, 2019 at 16:45

Although I like the other answers, I think nobody mentioned the most important: It all depends on your target audience.

Do you target people who are gamers? They play games regularly and your game would be familiar to them? Then most likely you wouldn't need anything more than a simple screen hidden behind a button that explains the basics. (Just in case someone misses something)

Do you target a broader audience? Maybe children too that have no experience in games? Then yes, you'd need some sort tutorial, instructions, otherwise you risk the user getting overwhelmed and giving up.

Having said that, I'm a person that hates the tutorials that "grab the user by their hand" and ask them to click on buttons while everything else is disabled. I like when playing the game is the tutorial by itself, like the original Super Mario, where the stage itself was teaching you what is an enemy, what is a block, what happens when you interact with them, and learning how to jump. It's up to you how you want to approach it.

  • \$\begingroup\$ I have a tutorial, but nobody seems to read it or understand it. Even if I tell them how to play, they do everything other than the controls, which makes them frustrated. I think my only option is to add the autoturn and make the enemies faster. \$\endgroup\$
    – Cannabijoy
    Commented Jun 13, 2019 at 9:00
  • \$\begingroup\$ It may be. In the end, you can't make the users do anything. \$\endgroup\$
    – bob
    Commented Jun 13, 2019 at 19:06
  • \$\begingroup\$ @Cannabijoy: To me it sounds like your users do not understand that your controls are controls, e.g. it sounds like they do not understand they can be used/clicked and are not just decoration. So maybe changing the design of the controls would improve things. For example, instead of having just arrows, draw obvious 3D buttons around the arrows. Animating them (pulsate or "wiggle") to draw attention may also make it more obvious to do something with them. \$\endgroup\$
    – DarkDust
    Commented Jun 14, 2019 at 13:27
  • \$\begingroup\$ @DarkDust Yes that was definitely a problem. That’s really from my own stubbornness, because I refuse to put viewable buttons on my mobile games. The art is pretty cool, but it’s very minimalistic and that’s the atmosphere I’m determined to convey. \$\endgroup\$
    – Cannabijoy
    Commented Jun 14, 2019 at 22:57

this can be a very general question and depends on game design. most of players hate hints and tutorials but most of the time its needed to add the pace of learning in the game. you have to find the best way to introduce your game to the player. sometimes a cheetsheet in loadings is enough some times you have to give the data step by step. think about your design

  • \$\begingroup\$ I have showed people how to play and they still can't figure it out. It's like they do everything but what they're supposed to do. I think I'm just going to settle for the autoturn and maybe add more enemies to make it more fun. It's a very unique game, and I really don't want to spoil it with bad controls. \$\endgroup\$
    – Cannabijoy
    Commented Jun 13, 2019 at 8:56
  • 5
    \$\begingroup\$ I made very bad experiences with "cheat sheets". The problem with showing them to the player at the start of the game is that the player can not yet contextualize the actions described in it and thus easily forgets about them. If you decide to go for this nevertheless (which is understanable because it's a pretty cheap way to explain the game) at least make sure the player can access it again at any time. \$\endgroup\$
    – Philipp
    Commented Jun 13, 2019 at 10:44
  • \$\begingroup\$ Agreed, cheat sheets should be visible or accessible from the pause menu, or from some other ingame equivalent if you don't intend to have a "pause" feature. \$\endgroup\$ Commented Jun 15, 2019 at 16:54

How much in-person user testing have you done? You can theorycraft all day long on why might be happening but there's nothing like seeing it for yourself what they are stumbling on. Then you can tweak as needed and repeat the user testing (with new players.)

Even you you use heat-mapping for remote testing, in-person you can see the player body language, where he is looking and often they will voice out loud what they are thinking. Most big cities have game development meetup and they are the perfect venue for that.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .