Controlling difficulty in a randomly generated game.

Started by
30 comments, last by RalemProductions 10 years, 1 month ago

I will be brief to start, and add a longer version later for those who are still interested.

Brief Version

I am developing an old fashioned space shooter with mutating aliens. Most play testers find it very difficult to play despite adding detailed, in game hints. I need to know why my clues aren't being picked up. I will return the favor if you have a PC demo available for a game that you need feedback for.

You can download the game (It Never Ends) here:

http://www.misterdonovan.com/

Long Version

For nine months now, I have been working on a potential iOS shoot'em up game that features mutating aliens. It's called "It Never Ends." I was hoping to create a short game, with high replay value, by partially adapting the theory of Natural Selection. The core concept of the game is to modify your ship to meet the ever changing alien threat.

Even though it has place holder graphics, the basic game is playable. Unfortunately, unanimous play testing shows that the game can become so insanely difficult that it can discourage some people from continuing.

I have given the player two powerful tools to counter this: the nuke and a retreat option. Unfortunately, I seem to be doing a poor job of explaining how to use them despite complete access to "Intel Notes" and in game hints.

  1. The Nuke will completely destroy all incoming waves and forces the aliens to redesign their ships. It does, however, take away some of Earth's limited resources.
  2. The retreat option allows you to save the resource value of your ship and redesign it to face them again. However, the remaining aliens will destroy more resources.

Armed with this info, could you please test the game and offer feedback/suggestions concerning the game's difficulty? If you are currently designing a stand alone PC game, and have a demo, I will offer you feedback for your game in return.

Advertisement

Well, I downloaded this to try to troubleshoot the design, but I couldn't stay alive long enough to encounter the problems with evolution.

Edit: I thought I'd add that I really like Space Invaders and Galaga, and moderately like Asteroids, which are all similar games to this. But I don't like hard games which are also dexterity games because I know from experience that they aren't going to become satisfying to play because my natural dexterity isn't high enough for me to ever get really good at it.

I want to help design a "sandpark" MMO. Optional interactive story with quests and deeply characterized NPCs, plus sandbox elements like player-craftable housing and lots of other crafting. If you are starting a design of this type, please PM me. I also love pet-breeding games.

Downloaded it, tried it, but quit after less than a minute. Sorry to say that, but I found it boring.


Sorry to say that, but I found it boring.

Thanks for your constructive input... sad.png


I am developing an old fashioned space shooter with mutating aliens. Most play testers find it very difficult to play despite adding detailed, in game hints. I need to know why my clues aren't being picked up.

There are two issues.

First: Is my game too hard ?

Maybe, maybe not. Take a look at flappy bird ( or one of its 1000 clones) and you will see, that even hard games have an appeal to (casual) gamers. You need to consider the target audience. Once you start adding new features to 'streamline' your game, you will have issues with your target audience. Making it too complex will shy away casual gamers, whereas making it too easy will shy away gamers who seek a certain challenge.

Second: Are my game features not obvious enough ?

This might be an issue which will be hard to solve. The core problem is, that new features (instead of just making e.g. the enemies weaker) will introduce new level of complexity. At a certain threshold, more players will have trouble to understand the handling and benefits of certain features. Good tutorial might help, but eventually you must think about your target audience (again :) )

Ideas ?

Well, you have an space shooter, which should target more casual gamers, therefor this

  1. The Nuke will completely destroy all incoming waves and forces the aliens to redesign their ships. It does, however, take away some of Earth's limited resources.
  2. The retreat option allows you to save the resource value of your ship and redesign it to face them again. However, the remaining aliens will destroy more resources.

is too complex in my opinion.

I would reduce the complexity by removing the retreat option and pimping the nuke. The nuke should be time-bound. Eg have a nuke symbol on your screen which reloads over time (maybe faster if you kill enemies). Once it is up, it starts to blink and a sound like 'Nuke Available' will tell the player, that something powerful is up to use. After using it, start the reload again. The player will automatically learn how to time and use the nuke after a few tries, there's no need to explain it.

I've really a hard time with the game complexity in my game too, and I'm fighting it by reducing its complexity a lot. Complexity should be only for gamers who rock your game and need new challenges. For the rest, you need to make your game easy enough to play through it without an deep understanding of the complexity of the game mechanisms. To smooth it out, try to introduce different difficulty levels, the player should win at least at easy, at mid with some practise and at hard if they truly understand the game (complexity).

Well, I downloaded this to try to troubleshoot the design, but I couldn't stay alive long enough to encounter the problems with evolution.

Edit: I thought I'd add that I really like Space Invaders and Galaga, and moderately like Asteroids, which are all similar games to this. But I don't like hard games which are also dexterity games because I know from experience that they aren't going to become satisfying to play because my natural dexterity isn't high enough for me to ever get really good at it.

Thanks for your time. I was initially going to ask if you liked games of this type. So your edited response is very useful. Unfortunately, this game is a hybrid, and I'm failing to communicate that to people. A large part of the game requires the player to analyze the alien design, and then create a counter strategy. I never really wanted it to be a full on twitch game, but I can't argue with the majority of testers: It still requires a lot of dexterity.

The earlier demos were actually harder! The player ship had a generator and battery which could become depleted during battle. The idea was to include power management during combat. Play testers told me that they hated it, so it was removed. My next step was to create a learning mode, but I think I need to tone down the difficulty of the entire game first.

Please let me know if I can test a game for you.

Downloaded it, tried it, but quit after less than a minute. Sorry to say that, but I found it boring.

Thanks for giving it a shot. Could you answer these questions?

Was it boring because it was too easy or hard?

Do you like games in this genre? If so, could you tell me their names?

I tried your game. I will send feedback soon.

Thanks.


First: Is my game too hard ?
Maybe, maybe not. Take a look at flappy bird ( or one of its 1000 clones) and you will see, that even hard games have an appeal to (casual) gamers. You need to consider the target audience. Once you start adding new features to 'streamline' your game, you will have issues with your target audience. Making it too complex will shy away casual gamers, whereas making it too easy will shy away gamers who seek a certain challenge.

Second: Are my game features not obvious enough ?
This might be an issue which will be hard to solve. The core problem is, that new features (instead of just making e.g. the enemies weaker) will introduce new level of complexity. At a certain threshold, more players will have trouble to understand the handling and benefits of certain features. Good tutorial might help, but eventually you must think about your target audience (again )

Thanks for your thoughtful and thorough feedback! It helped me get my goals in line.

I think a large problem with my approach is that I didn't intend to make a casual game. So, people are expecting a casual game and getting something else. They probably feel betrayed as if it was a bait and switch. My game is a hybrid between a space shooter and a strategy game. In retrospect, it was probably a bad choice because I think people tend to like either reflex games or strategy games; not both at the same time.

My game is definitely complex. Unfortunately, it also seems to have a dexterity/reflex requirement. Again, these two elements are unlikely to appeal to fans of either genre. I think I'm going to scale down the dexterity element and emphasize the strategic element.

I will give some thought to a tutorial, but I was intrigued by your "Nuke Pimping" concept. Instead of making it optional for the player, I'm thinking about keeping track of how many aliens get by the player (damaging Earth). When thirty get by, it's clear that the player is having difficulty, so Earth automatically sends a nuke up to protect itself. This makes sure that the player doesn't miss the nuclear option. It will also have its normally intended effect of scrambling the alien design; hopefully to something the player can handle.

Thanks again for helping me with my goals.

I will be sure to test Gnoblins and send you feedback.

You don't explain that your shot takes time to recharge well and there is no good indicator. I saw there is one of the ship but that's meh. I can barely see it

Same with the shields. I can barely see it.

The ships fly too fast to begin with

the laser weapon limits movement, lasts too long, and is far too powerful

The shots regular shots are too fast

The shots are too slow to recharge and too weak

There is no indicator for other things that can get damaged

The "waves" mean nothing because there is nothing that acts as a wave

The AI has no pattern what-so-ever

The game's difficulty starts at the second when you find yourself about to die in most of these games...which if I had to rank it is about an 8 or 9/10 in difficulty as most games purposely do this and it's the result of you messing up somewhere.and not a result of the design of the game.

You might want to consider instead of randomizing the AI of every ship, have waves have the same AI.

You should also the from the same base and then evolve from there rather than just making it randomly constructed rather than saying "here's a random challenge" try to beat it...and oh yeah it's far too fast. Start it slow and an easy pattern then let then have the enemy have their stats slightly modified from that slow base, working to some ultimately random formulation of the enemy.

This allows a beginner or anyone for that matter to ease into the game, figure out what's going on and then crank up the challenge over time.

Another thing you might want to do is consider a reward system for collecting resources where you can have extra lives, more shields, different type of weapons, etc...

Another thing you might want to consider is the idea of changing the firing mechanism from wait for it to charge to the player charging the weapon. That way they can release lots of smaller weaker shots or few large shots.

You might also want to mention that shots diminish in power further from the point of origin...which makes the charging of shops pointless as past half screen they seem to be too weak to matter...

That's all i can think of

I played until I got finished off. Here's my report.

Opinion: I am a hardcore player. Your game is hard. Flappy bird is deceptively easy.

Explanation:

The amount of time it took to play to the bitter end challenged my ability to stay interested. Reflex-intense games can be boring, while at the same time making them incredibly challenging. This was caused by the momentum controls present in going left and right

In the later levels when I had to really dodge things, the dominant strategy reveals itself, because whatever else I might try would end up getting me killed. The downside is that there is little flexibility in the end-game.

I encountered one apparent bug.

At some point it did not allow me to repair any longer. I could tell it was wrong, because the thrusters were still too slow.

I think I was getting hit by E.M.P. as well, and those bombs that shoved me around raised the difficulty up a lot.

Brief personal analysis:

I devise strategies, and I follow them through. In this respect I kept playing until I found the challenge about 4 turns in when the aliens finally got smaller.

I couldn't bring myself to play it more than once so this is not a thorough analysis. If I did keep playing I'd just end up tuning my reflexes to match a strategy, which isn't going to be a common player trait.

Thanks for playing it to the bitter end. I hope that I can return the favor and test one of your games sometime. My e-mail address is donovan@misterdonovan.com

I wonder if you're aware that every game of INE is different? The aliens' mutations are random, so there is no generally dominant strategy except to use the Nuke when absolutely necessary. For example, some aliens are equipped with shield piercing blasts. In this case, it's pretty pointless to invest in player shields.

There are 20 different alien special abilities, and they carry two at a time. Aside from that, they also have different firing parameters, weapon types, and movement patterns. Even though the game is usually hard, there are times when it is laughably easy (like stealth ships on fire).

You didn't exactly encounter a bug regarding repair, but you did help me out with a flaw in game design. You may only repair your ship six times. After that, there are no more spare parts. I need to put in a notification to let the player know.

I was wondering if you used the Nuke player special, or the retreat ability?

Also, what was your final score?

Thanks again.

This topic is closed to new replies.

Advertisement