Difficulty, and Accessibility

Watching, reading, and listening to people play games (mine, and others) makes me think about accessibility, difficulty, and how these things should affect me personally, as a game developer.  I think now that I've come into contact with such a broad range of games, I can make some conclusions that I might not have been able to before.  I've now played the gamut, from games like Crypt of the Necrodancer, TGM's "Death" mode, VVVVVV, Super Meat Boy, and others that really require you to learn, be smart, and adapt, and punish you harshly when you fail...to games like Journey, FEZ, To the Moon, even Botanicula, that try to be more universal, in the sense that there is no "failure", no real "skills" needed, no need to really understand the workings and mechanics and physics of the system.

Often as we develop games, it's impossibly difficult to separate ourselves from our skills and knowledge.  This comes on multiple levels, of course.  At the extreme end of the spectrum, you've got the guy who makes a game jam game for LD and thinks that everyone who plays his game will just know that "you need to bring the red thingy to the blue square to win" and in order to do that you need to use some random combinations of keys on the keyboard that are never given to you anywhere.  But slightly less obvious than that is when we gear our games to our own preferences and skills.

When I made Rhythm Gunner, I made it in classic Super Crate Box style--one hit and you're dead, and you need to really get used to the game's tight controls and mechanics before you can start excelling.  Well, people liked it...but also didn't like it, simply because of the difficulty.  As a result people scored it much lower overall, much lower than Labyrinth which I didn't think was particularly awe-inspiring.  The thing is...I loved the HECK out of Rhythm Gunner.  After I made it, I was actually addicted; I kept trying to beat my high score, eventually getting to a score of 121 points.  As a comparison, the highest other score that was reported was 44 points, and the majority of players scored below 10 points.

So was Rhythm Gunner a mistake?  Well, heck no!  To me, personally, it's the most exciting, interesting, and exhilarating game I've made for LD, and I still have fun playing it to this day.  It appeals to me for the same reason that TGM does--it has a high skill cap, it requires constant focus, adaptation, and reaction, and the difficulty combined with action contributes to being in flow state.  So what ended up being the least popular of my games is actually my favorite.

And here's where I must remind you of a quote from the composer of VVVVVV:

"It's better to write for yourself and lose your audience, than to write for the audience and lose yourself"
--Souleye, composer for VVVVVV

So do I regret making Rhythm Gunner the way I did?  Nope, you bet I don't.  I did the best that I could given the 48 hours of work time that I had, and it came out awesome.

Now, that being said, if I had, say 72 hours to work on Rhythm Gunner, then yes, yes, yes, there are several things that I would change about it to address the learning curve issue.  Things like:

- An easier level with only 1 or two weapons.  Having 4 right off the bat is too hard for people to digest.

- A practice mode that lets you try out weapons before the enemies start spawning.

- A completely separate mode that doesn't rely on rhythm?  Because let's face it...some people just can't get rhythms at all!

- A "softcore" mode where you have multiple hits before you die?

I think one thing about Rhythm Gunner is that the "perceived difficulty" is very high because you die immediately after getting hit.  To me this isn't something that I actually want to change for the same reason that Super Crate Box features instant deaths--when the game punishes you with instant death for being hit, you become hyperaware of bullets and enemies since your brain is able to make that association.  If I changed the game so that you take 5 hits before you die, none of the first 4 hits really feel as punishing.  To me, it decreases the amount of time where I feel like I'm "in the zone".

Anyways, what I wanted to actually address is "target audience" and accessibility.  Rhythm Gunner is a good example because not only is it a punishing 2D platformer, it's also a rhythm game!  So already I'm imposing two requirements upon players before they can really get "the intended experience" out of it--you need to be decent at Super Crate Box, and then you also need to be decent at music games.

Would it be better to make Rhythm Gunner accessible to people who are really bad at rhythms?  Well, sure, I mean, I =could=.  As I said, I could make an alternate game mode where it's just a plain ol 2D platformer with no rhythm requirements.  But that's not Rhythm Gunner.  If I'm going to reduce it down to that, why not just play Super Crate Box? (which is better designed and has more weapons)

So what I'm trying to get at is, yes, you need to make your game learnable (providing a big wall of text is bad NOT just because people are stupid and don't read, but because learning by doing is just better and more intuitive), and yes, you need to provide a difficulty ramp, and yes, you need to think about the methods in which your players will learn the mechanics of your game (my new game, Melody Muncher, has probably my best tutorial system ever, yay), and yes, (unfortunately?) you still need to think about systems that will reward your players to make them "feel good".  But, maybe you shouldn't go so far as to destroy what makes your game....your game.  If I'm making a VVVVVV or Super Meat Boy style game that is designed to be fun BECAUSE it's punishing, hard, and force you to retry levels again and again, what should I do about people who have bad coordination and don't like difficult games?  Should I make an easy mode where you can take multiple hits before you die?  Should I make a "god mode" where you can't die at all?  Well, no...that wouldn't actually be a fun game.  Imagine playing Super Meat Boy, except you automatically win each level no matter how poorly you do.  Is that a fun game?  No, it's actually really pointless.

So you need to decide what the point of your game is.  "Unfair" difficulty is bad.  So is inconsistent difficulty, and difficulty that's inherently unnecessary.  But many of us play games =because= we like analyzing and overcoming challenges.  And in that sense, difficulty is =vital=.  So you need to understand and separate where your difficulty is important vs unnecessary.  (This is a point where the Mega Man games are so well-designed)

And, sadly, you can't cater to everyone.  Not at all.  Because even those games that I mentioned, like FEZ and Journey--you'd think that anyone can enjoy them, but that's actually not true!  Some people are going to try and play FEZ and get stuck with basic platforming.  And some people are not going to "get" flying in Journey, or maybe even walking!  Navigating a 3d space using a game controller is something we've been doing for 10 years, but what about someone who has never done it before?  I think it's a worthy and humble cause to make your game more accessible, but at what cost?  If you try and please everyone, you are bound to fail.  Think of the old parable about the man and his horse and his son...

In the end I think it's a better service to both yourself and the world if you make a game that some people can play and enjoy a lot, rather than try to make a game that anyone can play and be unimpressed with.

That said, I am pretty happy with the accessibility of Rain.  Unfortunately, there was still at least one person who failed to complete it because they thought it was a screensaver and not a game ("I kept wondering why did that Right arrow show up on top of the girl?").  See?  You can't get everyone...