The XX Show #11 – Game Difficulty.

Join Nate and Troy as they analyze and discuss the many aspects of video game difficulty settings. If you think you know what they’ll be discussing, you might be in for a surprise. There is a lot more to game difficulty that one would imagine. Come and listen to what they think.

eXplorminate Music by MangaDrive

Show notes:

  1. What constitutes game difficulty? – 1:00
  2. How does difficulty impact the A.I. and how it plays the game? – 9:00
  3. Does game difficulty slow down the player progression? – 18:10
  4. Do players want to play difficult games? – 24:45
  5. Do developers want to make difficult games? – 31:48
  6. Does the marketplace support difficult games? – 38:06
  7. What’s the future of game difficulty? – 47:26


16 thoughts on “The XX Show #11 – Game Difficulty.

  1. nice XX-show.
    the part about Thea, i realy second that. even if failing you archive something. that was what made the difficulty of Thea a nice challange, rather than a dull difficulty to chew on.

    would like to point some more out, but my english is not good enough to discuss such a topic i’m afraid :)

    Liked by 2 people

  2. Good discussion guys. As you pointed out, different players have different play styles and play games for different reasons and have different expectations from games. Also developers vary in their outlook and development policies. The AAA corporate developers make games to make profits for shareholders and bonuses for senior management, so they will design their games differently to a company run by game enthusiasts. The ideas Troy put forward about machine learning and being able to reset it etc made a lot of sense to me for those players who want a challenge; but as was also mentioned a lot (maybe even a majority) of players dont want real difficulty. They want a fun experience and they want to win.

    Liked by 1 person

  3. The thing about difficulty that frustrates me is the often huge delineation between winnable and unwinnable difficulty levels towards the higher end. For example, I can beat Civ 5 at Emperor almost every game, regardless of Civ, but I get absolutely throttled at Immortal.

    Sliders definitely help nudge the difficulty up without creating that artificial winnable/unwinnable line of demarcation. The one game that does the best at this is AI War Fleet Command. Tons of different AI personalities that have different strengths and abilities, large difficulty spectrum (1-10 with a number of them broken down into fractions like 9.1, etc.) customizable resource bonuses/penalties to production for both the AI and player, and other things to really customize the experience.

    The other thing I really like about it is that there isn’t that one “meta” strategy that is a requirement to win at the higher difficulties. The game gives you enough tools to let you craft your own strategy, and if it’s good enough and if your execution is sound, you can beat higher difficulties with various approaches.


  4. I get the points make in the show about devs not considering good AI a priority when making a game. That said, why not make hooks into the game so that 3rd parties can take a shot at developing good AI? I think that good games would get good support from the community and there are very creative people out there and crowd sourcing can work. Indeed they could even require any AI code to be public so that it can be built on. I personally would love to try my hand at developing AI for certain games but with very few exceptions this just isn’t an option without special support from the company/devs. Indeed there are even universities and classes that might be interested in doing this class assignment. Overall I feel there is a lack of creativity and overall laziness in the market regarding game difficulty.


  5. It not cheating for the AI to have a handicap. As long as it is understood at the start what they are. If I get the letter H in a game of horse from a superior play, I am not cheating. I would be cheating, if I announce I have make Horse, after 4 times earning a letter. We did not agree that I get that free H.

    In most games the AI knowing the map, such as Civ3, is not a means to give the ai an edge. It was done to reduce the need for a lot of code to have the Ai mange it pathing. Thus saving the need for more powerful PC and of course saving cost for the devs.

    At higher levels the AI get extra advantages, again not cheating. If you do not want to spot the AI those boost, then there are levels where they get none or even give the player some.

    To me cheating would be, if the AI played differently against the human. When its code checked to see that it is dealing with another ai or a human.

    Liked by 1 person

  6. Nate you kept saying “Distant Worlds too” in this show and I kept hearing it as “Distant Worlds 2” and I kept having heart palpitations as I thought there was some new information about the new game.

    Liked by 1 person

  7. I always thought AI cheats to increase difficulty, not using better algorithms for more difficulty:

    1) AI always know where your cities are without exploration.
    2) AI always know where the nice planets or nice resources are without exploration.
    3) AIs gang up on you not analyzing diplomacy status or how they like you previously.
    4) AIs are giving advantages in resources…


    1. Let me elaborate a bit. In some games, the AI is hindered by maluses or less optimized algorithms, so increased difficulty let the AI play using the same tools and abilities that a human player uses.


  8. I think dynamic difficulty would be ideal. You don’t want to get beaten hopelessly but you also don’t want your opponents to be complete pushovers.

    Liked by 1 person

  9. The observation that 1-2% beat games at max difficulty is very true. But the games that are recognised as “difficult” almost always sell far better than those that are tagged with “brian dead AI” or “stupid easy”. The list is endless. Latest classic example is DOS2 selling vastly better than PoE2 Deafire when DOS2 is much, much more difficult. Cuphead mentioned. Factorio, Rimworld. Stellaris wiping the flooe with NuMoO.

    So what is going on here? I suggest there’s something more complex happening here and devs that say “Oh man, forget the whining hard core crew” are paying for that decision big time, Obsidian certainly are at the moment. Deadfire was released a month ago, DOS2 Sept 17. DOS2 is still in top 40 outside of sales. Deadfire dropped out of top 100.



Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s