Wednesday, October 15, 2014

The Eight Rules of Roguelike Design

An article appeared on Gamasutra recently titled The Eight Rules of Roguelike Design by John Harris.  I thought it would be fun to analyze Lone Spelunker in the context of that article.

Unfortunately, most of the article focuses on item design.  Lone Spelunker doesn't really have items in the same way that other roguelike games do.  There is equipment, to be sure, but gameplay doesn't revolve around finding equipment.  Everything you need, you carry into the game with you.

However, I still think Harris was on the right track with most of his suggestions.  I think all you have to do is replace "item" with "situation", and his advice not only still holds for roguelike design, but actually becomes more on-point.  The items are only one element in the complex alchemy that makes a roguelike challenging, surprising, and enjoyable.  So the underlying concepts that make items better can also be applied to the situations the player finds himself in.

For instance, you could easily change this rule:
Item masquerade rule - Items should be difficult to identify even for spoiled players, due to similarities between items of the same type. this:
Situation masquerade rule - Situations should be difficult to identify even for spoiled players, due to similarities between situations of the same type.
This rule would still apply to items, but it would also apply, say, to different monsters they face, different room templates, different room features, different traps, etc.  The core idea here, that there should be interesting and surprising elements to gameplay even for veteran players, applies just as well to the entire game world as it does to the items.  (Similarly, his comments on monsters can be expanded to situations, as well.)

In Lone Spelunker, your starting equipment is your only equipment - the interesting bits are provided by the environment.  To meet this rule, then, the game would need to provide surprising twists and turns in the passages, rooms, etc., even for people who think they are familiar with how they work.  Thankfully, I've got ways already coded into the cave generation algorithm to ensure a bit of this.  You may start seeing patterns to how the caves work (which isn't all that unrealistic, since you see patterns in cave development in real life), but you will be surprised at times where you find passages and where you find yawning chasms ready to gobble you up if you step errantly.

Let's look at some of the other "rules" Harris puts forth:
No beheading rule - Provided reasonable play, the player's character should not be killed or harmed too greatly and permanently in one attack.
Lone Spelunker has no monsters that attack you, so it meets this criteria trivially, if we're going to be technical about it.  But I believe what Harris is getting at is that you shouldn't be suddenly presented with something that erases a lot of progress you've made for yourself.  If you're careful, you ought to be able to survive because things shouldn't be able to just walk up and kill you in one blow.

In this sense, Lone Spelunker both fails and succeeds this test.  It fails because, like a sudden beheading monster, death can come swiftly and suddenly in the game.  You'll be traipsing along the cavern floor, happily taking in the scenery...and bam - you've stumbled into a 150m chimney.  That can feel like a sudden beheading.

But it's also important to note that if you are careful where you step, you cannot be killed at all.  As long as you aren't careless where you move, you can guarantee for yourself that there is no "beheading" moments coming your way.

In other words, the game has no power to make anything deadly to you.  Every challenge may be overcome without regard to any items you've found, experience points you've gained, stats you've built up, or anything like that.  If you die, it was by your own hand.
Reducing grind rule – The mere act of spending time should deplete some finite resource, forcing players to keep those resources up, while also limiting the amount of "grinding" he can do.
I'm not sure I agree with this rule.  I agree that grinding can be a chore, but adding a finite resource that gets depleted over time only serves to punish people who do grind - it doesn't remove the need to grind from the game.  In other words, it doesn't make the game less grind-y; all it does is limit how much you can feasibly do it.  If the impetus is there to grind, that's the problem.

The reason people feel the need to grind in roguelikes is explicitly because of many of the other things Harris mentions - such as all the focus on items.  You grind because you know that the potion over there on the other side of the lake in Brogue could be the thing that saves your life seven levels below.  You leave stuff behind at your own peril.  (And what happens if you leave behind one of those ever-depleting resources, like food?)

A better approach is to reduce the impression that if you don't get every single item, you're screwed.  Lone Spelunker solves that trivially by having no items at all.

Race you can't win rule – The game's monster difficulty should increase slightly faster than the advancement of the player, so as to force him to rely upon items and tactics.
In general I agree with this, although again, this is too specific.  Why call out only the monster difficulty – why not difficulty in general?  What about traps?  What about challenging level features?

As with the ones regarding items, this rule doesn't really apply to Lone Spelunker because the game has no monsters.  But there is difficulty.  Terrain can get more unpredictable.  Stakes can be higher by making drops steeper and deeper.  Lone Spelunker does a bit of this in its generation, but ultimately, when the threat of failure comes from your own carelessness, simple cave size is the main indicator of difficulty.  The bigger the cave, the more likely you'll end up tumbling over an edge at some point because you will have far more opportunities to take careless steps.

So these are only some of the "rules".  Does the fact that several of the "rules" don't apply to my game mean it's not a roguelike?  I don't think so.  Maybe it's blurring the lines a little, but I don't think random items and monsters are necessarily required for a roguelike.  If the items aren't random, or if the terrain is deadly enough without monsters, it's still a roguelike, to me anyway.

Besides, Lone Spelunker is solidly in the "Procedural Death Labyrinth" category.  Though it breaks many of Harris' "Rules of Roguelike Design", I think it still follows the spirit and impetus of those rules - minimal grinding, unpredictability, constant challenge, reasonable chance of life, etc.  These are all things that improve the roguelike experience, whether they do so with monsters or items or traps.

Or stalagmites.

1 comment:

  1. I have to agree that the change to situation makes them more relevant.
    The only one I don't agree with at all personally is No beheading rule.

    That's not to say every single game should have instadeath features in it, but some of my favorite roguelikes are the ones that do. It's a thrilling experience to not know if an unidentified scroll/potion/gear is going to be incredibly useful or set me on fire and if a new enemy is going to be simple or smash my face in. Opposed to "I wonder if this potion is good or only just okay" there's not much threat or excitement if everything is only mildly dangerous at best, it does of course depend on what kind of game you're trying to make. I just don't agree that it's flat out bad game design.