Ask Producer: How Difficult Levels?


I am an indie game developer (on my own) and i an just hoping to release my first independent game. I have finished all the programming and I am just making the levels. My question is, to get good appeal, how hard should i make it and how should the difficulty progress?

I think designing should take into account at least 2 or 3 modes from easy to difficult (or from normal to difficult). In Hightailed game there’s total of 4 difficulty modes in the latest version. One is extremely easy, two others are bit harder and the last one is made as difficult as possible.

Making game progressively more and more challenging is a good approach – the same that you can use for player to learn the rules . There’s excellent article that points out Ten ways not to make a casual game and gives lots of excellent insight on how to design your game. I really recommend checking it out.

Summary: You could make the game progressively more difficult. Or you could have 2 or 3 difficulty modes (if possible). Don’t make it too difficult in the beginning.

Edit: Excellent information in the comments:
Casual game designer says he disagrees (and also agrees if you read the whole entry) with me, and I must say that he has a point in his approach. Difficulty modes should not be the automatic answer. In Hightailed it works okay because the difficulty levels change how intelligently AI makes his moves. I’d say the same approach will work in games like chess very well.

Thomas added: “The best way to tune the difficulty I think is to get all sorts of different people to play it and tell you how they found it :)”

Ken points out the importance of testing, testing and testing. He also gave a link to usability testing.

Juuso Hietalahti


  1. I havnt read all of the replies, so maybe I write things people already said.
    But I have a few thoughts.
    More is not always better, this is a bit of thinking not much of your player’s personality. If a player choose the easy option, and its too easy for him. He might not enjoy the game. However, he might not be aware he is not enjoying the game only because its too easy, and if he would have choose a different difficulty, he would enjoy a lot more.
    So you might not want to give a “baby” difficulty.
    My second thought is:
    In virtual fighter, the computers learns how to fight against you slowly but surely. After each opponent you defeat.
    This is great, because you start the game from easy and it gradually becomes harder until it becomes insanly difficult.
    A variation of this could be, a game that learns how you play and adjust the challanges of the game according to the player’s abilities.
    This could be very difficult to achieve, but just might be the “perfect” game.

  2. I got people like my mum to test Easter Bonus and also a load of office workers. I just sat there and made notes and asked them to speak aloud saying what they were thinking good or bad. As a result of the testing I made the demo levels a lot easier and also improved the hint system.

  3. I’ve also added another twist on our Hightailed development version: ‘expert status’ (which many games seems to use). You get extra points for doing some additional tasks (collecting coins or bonus items). You don’t necessarily need to collect the extra items to be able to solve the level, but if you getting them makes the puzzle bit more difficult – without the expense of getting user jammed.

  4. I completely agree with the recommendation to test.

    There’s one additional thing I’d like to point out. Hitting the right difficulty isn’t just a matter of tweaking. Sometimes you need to take a completely different approach. Maybe you can place bonus items in such a way that the player is guided through the level. (It’s amazing how much level design can impact your game.) Maybe you need to change the design of your game altogether: take stuff out, put new stuff in.

    Once you have a game design you think will be fun, it’s tempting to say ‘this is it, I’m just going to tweak it now’, but sometimes you need to bite the bullet and redesign certain aspects of the game. Getting the learning curve and the difficulty level right is tricky, but it is also worth it: player’s will enjoy your game more.

  5. Thomas has it right on the money, usability testing is extremely important to tune levels. Not only tune them, but also redesign them. Microsoft has a team of over 20 PHD and Masters graduates who focus solely on testing a game to its fullest. The game’s division is their largest division too, to give you an idea, the Windows division has around 10 (I believe).

    Test, test, and retest, then test some more. Usually around the beginning of Alpha or even earlier is a good time to do it. There is a whole art to it so here is a link


    Nielsen is the name you want to search for, he is one of the major contributors to the field. Basically though, if you want to test sit in a room 1 on 1 (many observers often annoys or intimidates the tester) then watch them play. Offer no help or advice unless they are obviously stuck and it is hindering testing. Testing usually shouldn’t last more than 2 hours. I work with children mostly so we have only about a 1-hour block of time with them.

    Other than that, take notes. Its amazing what you learn about how a new person sees your game.

  6. Difficulty can be a really hard thing to guage on your own. We had thought we’d designed a nice progression with the levels in Cavemen. Then a number of reviews pointed out that the easy beginning levels dragged on a bit too long, and that they wanted more of a challenge.

    The best way to tune the difficulty I think is to get all sorts of different people to play it and tell you how they found it :)

Comments are closed.