Like me, you’re a videogame expert, so you know what the roguelike genre is. I needn’t tell you that it derives from Rogue (1980) and features permadeath, procedurally generated dungeons, turn- and tile-based action, and a fantasy setting. You — like me — didn’t need to look that up on Wikipedia or anything.

Rogue Legacy: old-school toughness couched in modern complaisance. Source

Rogue Legacy: old-school toughness couched in modern complaisance. Source

Rogue Legacy (Cellar Door, 2013) is a more distant descendant of Rogue, applying two of the core characteristics — permadeath and procedural level generation — to an action platformer. Those two characteristics have interesting implications for the hero problem.


You’re not going to explain the hero problem again are you?

Fine, skip this bit. The hero problem is the tension in an interactive story between a narrative of player-character success and failable game mechanics. For example, if the narrative is that a blurry, blockheaded Brosnan-Bond shoots Alec Travelyan to death on the superstructure of a big satellite dish, what do we do with the fact that he just got shot to death himself? What we do is we pretend it didn’t happen and we go again, but (story-wise) that strains our suspension of disbelief.

As Andy Astruc neatly set out on Electric Phantasms this week, the fourth wall works differently in videogames: the line between the diegetic and the real is thick, faint, and of uncertain location. We need to both be and not be Bond, but managing that without disrupting storytelling is tough going.

We don’t want to give up the narrative because we’re talking about stories — we need a narrative in there somewhere. But we don’t want to give up the failable mechanics either, because interactive systems are more affecting when you try, fail, learn, and get better. On top of that, the potential for failure makes success meaningful. These are the tensions of the hero problem, and Rogue Legacy manages to resolve them.


So how does permadeath help?

Permadeath solves the hero problem by insisting that when you died just now, that really happened. The player is not back at the start of the level or wherever, and there is no need to pretend you weren’t defeated. What you saw happen happened, and your engagement with the story is not disrupted. In other words, the narrative of player-character success is replaced with a narrative of possible failure, which is no longer in tension with mechanics of possible failure.

However, taken on its own, permadeath raises two new design problems: a character reset could be frustrating because you may lose a lot of hard work, and restarting all the time could get repetitious. Imagine playing Ghosts ’n Goblins with just one life, and going again and again with a different knight. It would get frustrating. The old arcade game structure is frustrating enough even with multiple lives and credits.


Is this where a procedurally generated world comes in?

This is where a procedurally generated world comes in. If the world is different each time then you’re not replaying exactly the same sequence whenever you die. That takes care of the repetition. However, there is still the problem of the player losing everything they worked so hard to gain. Rogue Legacy takes care of this one, too.

When you die in Rogue Legacy, yes, you’re back at the family manor, where you were last time, but your new character has inherited the spoils of your previous character. The work you did has in a sense been stealthily saved and reloaded, but that terribly gamey system has been dressed up in a narrative that actually holds together without having to mentally undo your failures.

The family manor. Source

The family manor. Source

The idea of an endlessly repeating roguelike evokes images of drudging grinds through flickering ASCII dungeons, but Rogue Legacy is a fresh and generous game, like modern games need to be, without copping out with low difficulty or excessive handholding (in fact it’s bloody hard) — a great legacy for the Rogue games.


Wait — this game isn’t an interactive story!

OK, Rogue Legacy’s not really aiming to be an interactive story, but it represents a great model for one. The real breakthrough is not that it resolves the core tension of the hero problem — any permadeath roguelike gives up a success narrative and so avoids conflicting with failable mechanics — but that it does so while maintaining the established, forgiving, and practical structure of continuation and progression in spite of failure.

This is a neat conceptual blueprint for a story-led game. Hopefully Cellar Door will make a sequel, and it would be really cool if they developed the narrative elements. For now, again, I’m just imagining…


Read all that? Hit the browse page for more videogame analysis!