Keyboards and Roguelikes

The original Rogue.

As work continues on interface and other polish elements, we’ve set aside a little time for extra keyboard controls to control things such as dungeon movement. To those who’ve been waiting for this addition since goodness knows when: now is your moment to rejoice! You’ll have it soon.

There’s always been a case for the addition of directional keyboard controls (countless bus trips spent coding and testing dungeon runs using a laptop and trackpad have demonstrated this admirably). It’s never been at the top of our priority list, though, since it just hasn’t been vital or practical enough to implement until recently – and I mean that in a broader context, oh beleaguered and anecdotal naysayers! And wow, there’s been a lot of debate in this area.

One of the rather interesting points about the now-moot debate on additional keyboard controls has been an argument which I’ll call the Appeal To Roguelikes. In short, the creation of Desktop Dungeons was heavily inspired by the roguelike genre (in particular, Dungeon Crawl: Stone Soup) and draws a lot of design choices, themes and philosophies from those sort of games.

Occasionally, in the matter of keyboard controls and other topics, players will ask us to pursue a particular design decision because “NetHack / ADoM / Blahblah does this and it works for them”. While this argument sounds reasonable on the surface, it doesn’t hold up to further scrutiny because despite its inspiration and theming, Desktop Dungeons is still its own game.

Since its initial alpha release, people have constantly been speculating that DD isn’t really a roguelike – just a puzzle game or Minesweeper knockoff with a similar flavour. I’ve never really been interested in weighing in on this debate (it feels more academic and amusing than anything else, and doesn’t affect the quality of play whatsoever), but I’ve always acknowledged that folks do have a point. There are so many departures from classic roguelike gameplay that it’s probably quite easy to exclude the game from the genre entirely if given a sufficiently purist definition.

Knowing this, it’s a fallacy to assume that everything which works for a normal roguelike game should naturally transition to DD. Keyboard controls are one of them: the fact that a good numpad / YKU control set is required for a comfortable and effective experience in something like Crawl or DoomRL is undeniable. Those games were built from the ground up to support keyboard controls, largely because the original Roguelike (er, Rogue) came about at a time when mouse-driven gameplay was still novel and clumsy at best (if it existed at all). Game and UI design were largely informed by this limitation and it’s actually quite hideous trying to play just about any of these titles without a keyboard handy.

Desktop Dungeons, on the other hand, was built solidly mouse-first. This ties in vitally with the game’s core structure: combat prediction in the alpha is impossible without a mouse, players are regularly expected to move across half a dungeon screen at the click of a button (with absolutely no in-game penalty for doing so) and everything from inventory to status effect information is built around (and geared towards) adequate access to mouseover information. We don’t just use the mouse: we construct our systems around it. We depend on it. The game is entirely playable without laying a finger on the keyboard (as I’ve mentioned before, I honestly never do), and when one further considers that our next viable platform is the touchscreen, this makes even more sense.

Keyboard controls are now more feasible with our recent interface additions: in particular, the right-click selection panel actually enables us to write a system where players can finally “bump into” and interact with a monster in a meaningful and information-filled way, and that’s pretty exciting. But even though this brings us closer to the speed and convenience of mouse-driven gameplay, it’s still hard to say that this method is strictly better. In fact, the main argument for a keyboard’s niche value is the speedup of the early dungeon exploration phase – which, while valid, is still not the meat of an average player’s DD experience and can be fixed with access to any “real” mouse (as opposed to the aforementioned coding-on-a-bus trackpad approach).

A few other distinctions and challenges that separate Desktop Dungeons from classic roguelikes:

–          Enemy movement. While enemy relocation is possible in specific situations, they cannot chase, retreat or otherwise move in normal ways that are integral to classic roguelikes. This remains one of the hardest lessons to teach new players, requiring some careful tutorial layout and a great deal of unlearning. In short, the dungeon arena is far too small and dense for meaningful spatial combat due to its single-screen nature.

–          Regeneration through exploration. This is arguably DD’s greatest claim to fame and also happens to be a bloody difficult lesson to teach because it’s about as unintuitive as it is vital. But interestingly enough, this mechanic actually was directly inspired by roguelike regeneration, just abstracted and exaggerated with a single adjustment to make sure that the game didn’t succumb to its own equivalent of Pillar Dancing exploits.

–          The god system. The gods in Desktop Dungeons have gone through a heavy number of iterations, mainly because they started off trying a little *too* hard to base themselves off classic roguelike diety interactions – with little idea of how to meaningfully reduce them to smaller playfields. As it turns out, the long-term worship of a higher power over several hours of play and many, many dungeon situations is pretty damn different to a quick in-and-out session of piety gathering. I still occasionally get the crazy impulse to tear down the entire system and start from scratch on something radically different, but the current iteration is definitely a much more natural – and vital! – puzzle piece in DD sessions.

–          The metagame. Most roguelike game sessions are big and involved enough to never require a metasystem for overall cohesion and progress (though some very light elements would certainly be interesting). In the DD beta, this is our lifeblood. A complicated economic system like the Kingdom would be unnecessary and painful in something like NetHack, but the metagame has always fit DD like a glove and has arguably been a part of the game’s success – in some form or another – from the very early days of the alpha.

But remember, keyboard users: those controls are coming for you! Don’t take this essay too sorely because we’ve always meant to bring this stuff to you, and the day fast approaches when your dreams will be realised. Just sit tight, this beta’s not yet over!

Hugs and kisses.