Skill Levels : Pro and Con

[svero]

Im interested to hear people's throughts on skill levels. That is... coming into a game and selecting "EASY", "NORMAL", or "EXPERT" skill. For instance, would a hit like Zuma have sold better or worse offering that? What are the reasons to do it? To not do it?

For my part the biggest reason to do it is that there is no one good skill level that applies to everyone. Some people will come into your game with skill they had from playing other games and some won't etc... Obviously there are different approaches one can take like auto-adjusting skill, but Id rather confine the discussion to offering skill levels. It's not something we see very often these days. Why not?

[TheMysteriousStranger]

Personally I would have liked an easy level in zuma. I'm too old and slow to complete those last few levels.

It really depends on the style of game. Action games I'd say require a difficulty level because not everyone is capable of great coordination and reflexes. Puzzle games though don't need it because the gameplay won't suffer just because you can't move the mouse accurately at great speed. Then there's the other option of making already challenging games REALLY chellenging, deliberately. Make it so on "oh my god my eyes are bleeding" difficulty, the player gets given massive score/reward then they have a reason to go back and try that level after completing normal level. This works well on games where the primary driving force is to get that great score and knock off "bored housewife" from the top of the online score.

...just don't get me started on the whole how to implement difficulty levels. Games have been struggling with that problem since the dawn of time*



*may be slight exaggeration

[Tertsi]

I'm also very interested about this.

In a game like Zuma they aren't in my opinion needed because the more skilled player is rewarded with higher scores and while he won't have trouble finishing the levels, he can play as fast and accurately as he can anyway.

Then there are games like Hammer Heads which either have a too poor auto-adjusting difficulty or could very much use more difficulty settings. The tough cookie mode (available only after completing normal) of it seems to be just as easy if not easier than the normal mode, but that's just a design flaw.

[Glen Pawley]

I'll be calling my difficulty levels 'Normal', 'Challenge', and 'Fiendish'. That way, I still have difficulty levels, but I avoid any disparaging connotations with calling a difficulty level 'easy'. Also I hope there will be a psychological incentive to playing the higher levels because it sounds like a worthy challenge, not just redoing the same thing at lower success odds.

[Nexic]

Whilst I always use Easy Medium Hard and Insane (usually unlockable) I think I might change to something like Normal, Hard, Insane. I think putting easy in is risky, because a lot of people will just go for that, and as a result find the game too easy and get bored.

Plus it's also bad in that you could lose a sense of achievement after beating so many levels or completing it.

[Coach]

Generally I don't like them at all. The more settings and options, the worse. You never know from the beginning which options and which difficulty mode that will offer you the most enjoyment and therefore it feels like I have to waste a lot of time just to figure it out. As there also is a difficulty progression from level to level, it is very hard (impossible) to get it right. I also like to know that I share the playing experience on even ground with others.

In my Pipe Mania/Dreams derivative, Soda Pipes, I used three difficulty modes, Slow, Medium & Fast which you could chose from before the start of each level. To advance to the next level you have to complete the level at medium or fast speed. Slow is thus basically a practice mode. Fast is for those who wants an additional challenge after completing the level at medium speed. As a guideline I made the fast mode on the last levels very, very hard but I could complete them - occasionally...

I think this worked very well. As the game demands a mix of creative puzzle solving and fast logical thinking, you could use the slow level to figure the puzzle out and then move on to the medium level for the added challenge of thinking under stress.

[sgm]

For instance, would a hit like Zuma have sold better or worse offering that?

It probably would've sold much worse.

In Zuma, you essentially start off on easy mode. Later, you repeat the same levels, but with more ball colors to match. The easiest mode in that game is the most addictive, I'm assuming most people would try normal first.

If you release several games, it's in your best interests not to include hard difficulties. People will try out more games instead of playing the same one over and over once they're done. With only a couple games, difficulty levels that provide hours of extra fun might lead to more word of mouth sales.

[princec]

I hate skill levels. I never know which one to choose and I always feel as if I'm missing something if I don't play on the hardest setting. I always prefer a game to simply be "right" in the first place.

Cas

[svero]

I hate skill levels. I never know which one to choose and I always feel as if I'm missing something if I don't play on the hardest setting. I always prefer a game to simply be "right" in the first place.

Cas

There is no "right" because people come into the game with varying skill levels. So what's "right" for me is not "right" for say.. my sister.

[princec]

Game design's wrong then

Cas

[svero]

The other related question is... Suppose I've been playing my game every day to test it. I naturally get much better at my own game and so by the time release comes along I'm kind of an expert player and it's difficult to judge what's easy and what's hard. So I don't mind trying to find an easy level by using playtesters and so on. That part isn't so bad. The hard part it seems is ramping UP the skill level. Someone whos a complete dunce at a new game might after 5 levels have built up enough skill that if you ramp the thing linearly they're bored. Very very tricky this stuff...

[Phil Steinmeyer]

A few comments:

My last game (Bonnie's Bookstore), had 3 difficulty levels. Jason at PopCap suggested I rename them (had been easy, medium, hard or something like that) to (normal, hard, genius), so nobody felt stupid taking the lowest level (good idea).

It was a word game, and because there are such vast differences in people's vocabulary, it was probably a good idea to have multiple difficulty levels.

But I'm semi-opposed to explicit difficulty levels for more typical action/puzzlers, for the reasons outlined above - you never feel you're making the right choice (as a player).

I am intrigued by the idea of self-adjusting difficulty. If the player loses on the same level a few times, ease it up. If they're breezing through the early levels using only 30% of available time (or whatever your limiting resource is), toughen up subsequent levels.

The idea being to try to make it so the player always feels challenged, but can still win each level.

BUT, that has problems. If the player knows about the auto-difficulty adjustments, they may feel cheated (you mean if I do BETTER at the early levels, I'm punished for it?), and/or try to game the system.

It also creates problems in games such as mine that have global high scores systems. (i.e. comparing high scores is tricky when the playing field isn't level).

So for now, I've just got a single difficulty level, no adjustments. It seems to be working ok for the beta testers.

[stiill]

Game design's wrong then

Cas

Then I think nearly every game on the planet has a broken game design. :P

[soniCron]

I haven't yet followed up on a recent post about difficulty levels, but here's the though summed up:

A combination of all three is likely the best course of action:

• Asking for the player's choice before ever playing the game is unfair - try asking after the first few levels: "Would you like the game to be more: Easy / Difficult" Or, be more clever and offer path progressions that imply an easier or more challenging course.
• Both the beginning and the end of the game should be on par with the player's ideal level of difficulty - it gets harder because the player is more advanced, not because it's supposed to actually get "harder" for the player. If they like it perceptively easy in the beginning, they won't like it perceptively hard at the end.
• Stay in touch with the player's progress, and make decisions based on their successes and failures. If the player likes it easy and they start losing a lot, make it easier. Vice versa if they like it hard and begin doing well.

In my opinion, the best difficulty level is adjusted by the game so that all players almost lose, but really win (almost) every time.

[Sillysoft]

In my last game I found difficulty levels to be essential. While testing it out with users, there was a huge difference in initial skill level. Some people breezed through and some people failed miserably.

Difficulty levels address this, so I think they are pretty clearly a good idea when you have a game like this. Then it comes down to fitting them into the game in an optimal way. i.e. start on easy, but make sure they know there are higher levels if they are breezing through so they don't get bored and leave.

[Anthony Flack]

The frustrating thing about renaming "Easy - Medium - Hard" to "Normal - Hard - Insane" in order to avoid offending people who are playing on "Easy" is that it misleads people. "Easy" is not "Normal". "Easy" ought to be for people who are either very bad at videogames, or people who really don't want any challenge. Calling "Easy" "Normal" is suggesting that that should be the default... and I think that it would be a mistake for most people to default to a true "Easy" mode. "Easy" should be easier than normal. And if you are too proud to play on "Easy", then use the damn Normal Mode!

[Ronkes]

There was an article about this on GameProducer.net a couple of days ago.

I responded to this on my blog Casual Game Design.

And here is Chris Bateman's take on it.

[svero]

The frustrating thing about renaming "Easy - Medium - Hard" to "Normal - Hard - Insane" in order to avoid offending people who are playing on "Easy" is that it misleads people.

Totally agree. I dont like it for the same reason.

[Game Producer]

Totally agree. I dont like it for the same reason.

I'd also like to point out that you could have one "normal" difficulty level, but then add extra difficulty for those who want to achieve "expert status". Like... to be able to solve a level on could simply walk from door A to door B - that would be easy. But, if the player decides to collect coins, find secret treasure, kill the dragon and save the princess then he would achieve "expert status".

This would make it easy to pass the levels (especially nice when you are stuck with some one stupid level and couldn't care less about slaying the dragon )... but it would also give challenge for those who want it.

[svero]

I'd also like to point out that you could have one "normal" difficulty level, but then add extra difficulty for those who want to achieve "expert status". Like... to be able to solve a level on could simply walk from door A to door B - that would be easy. But, if the player decides to collect coins, find secret treasure, kill the dragon and save the princess then he would achieve "expert status".

This would make it easy to pass the levels (especially nice when you are stuck with some one stupid level and couldn't care less about slaying the dragon )... but it would also give challenge for those who want it.

I like these sorts of bonus goal schemes as well, but they aren't always appropriate for every game. Also they can make level design a nightmare.

[michalczyk]

An interesting and difficult question. I think one should start with the end and ask: who is the target audience? There is a large difference in how people perceive words across age and cultures. Further, each word has its own emotional association with it. What that is, is individual. Then there is the environmental context in which the word has been invoked.

For example, a typical insecure male westerner in his early twenties, playing at a public place with a few new friends, will most likely choose NORMAL. As that would be the safest thing to do. On the other hand, a "here I come" attention seeking personality would likely choose HARD here. The thing is that the above choices are likely to be quite different if they were invoked in a private and safe environment - as is home for most people.

There is a lot more to it, but IMO, if your game is small, casual quick fun and meant to be played by anyone anywhere, it's best to go with the norms (EASY, NORMAL, HARD, or something similar) or not have any difficulty settings. The more targeted your game is the more accurately one can pinpoint the need for such things and the actual word usage. Remember the difficulty settings in Doom and Quake games?

[KNau]

I'm a little late throwing my 2 cents in but it looked to me like some people missed the whole point of "flow" in game design and how it applies to the problem of difficulty level.

First, the easiest solution has been a staple of game design since the early arcade days - start at "stupidly easy" and get to "ridiculously hard" as the game progresses. Shifting the challenge progression to a level design task eliminates the need for a difficulty selection screen or an extended tutorial.

Secondly, let the user progress at their own pace. Again, this is a decades old design law, used especially effectively in RPGs. If the player doesn't feel ready to go fight the goblin horde they can always hang around town killing rats and green slime to build up their skill level and get equipment. The key here is not to cap their improvement potential. If they want to become a level 20 knight before leaving town they should be able to do so.

Combining the two techniques above puts the user in total control of the difficulty level at all times. If things get too hard they can take a break and play in one level / location until they have the resources to move on.

[MrGoldfish]

I like games that rather than give you an 'easy', 'normal' and 'hard' settings give you a range of sliders that allow you to customise the game experience to yourself. Like setting starting amount of money, or number of lives.

No one knows what your customers like more than the customers themselves so this scheme works well.

[soniCron]

No one knows what your customers like more than the customers themselves so this scheme works well.

I find players (or, most people, for that matter,) don't have a clue what they want. They know what they think they want...

[michalczyk]

I find players (or, most people, for that matter,) don't have a clue what they want. They know what they think they want...

I totally agree. After 10 years in the game industry, to me, this is probably the most annoying aspect of it. Having to deal with clients who don't know what they want.

[Anthony Flack]

the easiest solution has been a staple of game design since the early arcade days - start at "stupidly easy" and get to "ridiculously hard" as the game progresses. Shifting the challenge progression to a level design task eliminates the need for a difficulty selection screen or an extended tutorial.

That kind of curve is designed to eventually eliminate the player. It's very much an arcade game model. For home games, it's more appropriate for the curve to be much more gentle - giving the player their desired level of challenge all the way through, and getting just a little harder and more complex to keep pace with the player becoming more familiar with the game. Most of your customers should be able to get to see most of the content; you don't want them being eliminated or forced to give up halfway through.

Also, better players often dislike having to play through loads of ridiculously easy levels.

I like games that rather than give you an 'easy', 'normal' and 'hard' settings give you a range of sliders that allow you to customise the game experience to yourself. Like setting starting amount of money, or number of lives.

Trouble is, it's difficult for the developer (who is intimately familiar with the mechanics of the game) to set up these variables in a fun way. It requires loads of trial and error. For a rookie player of the game, it's basically impossible. They're almost certain to unwittingly set the sliders in a way that makes the game suck. If you can find a player who can tweak your game to be better than you made it, don't give them sliders - give them a job contract...

Letting players set "penalty" variables like lives is a bit problematic, too. Given the choice, nearly everyone will just set it to the maximum anyway.

[papillon]

Letting players set "penalty" variables like lives is a bit problematic, too. Given the choice, nearly everyone will just set it to the maximum anyway.

Well, for a certain kind of game, you can make setting certain kinds of difficulty higher return higher rewards / different content. Normal/Nightmare/Hell difficulties in the Diablo games come with different loot, I believe, so there's an incentive to try it at the harder levels.

Obviously this doesn't work for every game.

[Anthony Flack]

In fact, even for the different dificulty settings, I think it's good to give an extra little reward for each increase in dificulty. Not much; you still want to give out 90% of the game to even the most hamfisted customer... but a couple of little bonus features for completing the game on hard is nice.