Why people still play NetHack

When I tell people that I play NetHack, I usually get one of two responses: “Oh, I haven’t heard of that one,” or “People still play NetHack?” I’ve spent more hours of my life playing NetHack, a game that was first released before I was born, than any other game.

NetHack dates back to 1987 and is still in development today; the newest release, NetHack 3.6.0, just came out in December of last year. The open-source code has even been used in the development of over a dozen variant games. While it may not have the largest following, there is an active community of NetHack players who still enjoy the ASCII-graphic roguelike after all these years.

I myself am an avid player, having won well over 100 games of NetHack, including some notable games in the 2012 and 2013 /dev/null/nethack tournaments. In 2013, I completed a game in which my character did not eat, read or write, fight with a wielded weapon, wish for any objects, polymorph himself or any objects, genocide any monster types, or accept any help from the gods. It was a staggering run which took me several weeks to complete (not to mention months of training) and pushed the limitations of the game in a way that I find exhilarating.

When I first started playing NetHack five years ago, I was amazed by how everything you can do in the game has intricate consequences. A common saying within the community is “The DevTeam Thinks of Everything.” Fall into some water? Your potions can be diluted and your scrolls erased (unless you’ve remembered to grease your bag). Polymorph into a large monster? Your body armor will break apart in the process (unless you’re wearing dragon scales and polymorph into a dragon). Engage in melee on a floor tile that contains written letters? The writing will get rubbed away (unless you’re levitating). Little details like these show just how much thought the developers put into the game. And in situations where I have found myself unable to use a mechanic the exact way that I wanted to, it was a pretty safe bet that the DevTeam designed it that way for a reason.

When you reach the bottom of the dungeon, you need to find a vibrating floor tile before you can move on. There is no in-game shortcut; you have to walk through the whole maze until you find it.

When you reach the bottom of the dungeon, you need to find a vibrating floor tile before you can move on. There is no in-game shortcut; you have to walk through the whole maze until you find it.

Another thing that I noticed pretty quickly: NetHack is hard. Like, really hard. As a new player, you may find yourself asking, “Is NetHack too hard?” Unlike a platformer where the early levels are meant to serve as a tutorial, NetHack (like most other roguelikes) dumps you into the dungeon, vulnerable from the start. Early on, you will die a lot, and you may not understand what killed you. The learning curve is so steep that a new player will need to consult ‘spoilers’ in order to have any shot at surviving for a significant number of turns.

In the NetHack community, ‘spoilers’ refers to any information that you learn outside of playing the game on your own. Monster behavior, optimal item usage, and causes of character death are some examples. Plot spoilers are not much of a concern, as the story line is short and tangential to gameplay. Sources of spoilers include the source code, the wiki, and other players. And when I say you will need to consult spoilers, this is not just a polite suggestion. In the entire history of the game, there is only one recorded case of a player ascending (which means winning the game, as your character is said to “ascend” to the status of demigod/demigoddess) without the use of spoilers. Even with spoilers, it will still take you months to achieve your first ascension.

While NetHack is a difficult game, like most games it becomes more predictable (and thus more winnable) the more familiar you become with the intricacies of gameplay. But even after you’ve ascended your first character, it retains its appeal by offering thirteen different character classes (also known as roles).

Trying to dip a potion into itself will give you a funny error message. Many situations like this will return their own messages.

Trying to dip a potion into itself will give you a funny error message. Many situations like this will return their own messages.

Although some of the fighter roles can feel quite similar to each other, there is still a good amount of variety between fighters, spellcasters, and even a more difficult Tourist role, which is a personal favorite of mine. With a lower win rate than any of the other character classes, the Tourist starts out with many disadvantages–no melee weapon, no known spells, and the weakest armor class of any role. If a Tourist can survive the early game and go on to complete the Quest (a boss battle that differs for each character class), the reward is great: they receive the Platinum Yendorian Express Card, a “charge card” which can be invoked to re-charge critical items, making the late game a breeze.

In the same way that I have a penchant for the Tourist role, many players come to prefer certain roles over others and develop their own playstyles around these preferences. There is no single winning strategy for NetHack–ask a group of veteran players what the “best move” is in a particular situation, and you’re bound to get several different answers. Once you have ascended a few roles, you will start to feel comfortable with your own play style.

At this point you may start to wonder, “Is NetHack too easy?” Luckily, the dev team has still got you covered. The game tracks twelve different voluntary challenges (known as conducts). Conducts are like achievements in which you are rewarded for refraining from certain behaviors like reading, eating, or even killing monsters. Breaking a conduct that you intended to keep can be considered just as much of a failure as a character death would be, and many players (myself included) choose to quit games and start over immediately.

Conducts also have a tendency to become exponentially more difficult the more of them you attempt on a single game, with very few players having recorded ten or eleven. To date, no player has achieved a twelve-conduct ascension, though the feasibility of such a game has been discussed in the community. I’m sure that I’m not alone in wanting to be the first player to pull off a twelve-conduct game.

In the same way that the game’s thirteen different roles allow players to ascend with different play styles, conducts (and other unofficial challenges) allow players to use their own ideas of what constitutes winning. For example, I no longer play just to ascend a character. I play with the intention of keeping as many conducts intact as possible. This means that I rarely ascend anymore, but when I do, it’s that much more memorable. Other players may view winning as having a high ascension rate, taking fewer risks in each individual game with the goal of having as few character deaths as possible. Still others play to ascend in the fewest number of turns, in the fastest clock time, or with the highest (or lowest) score. It would be impossible to say which player is the best at NetHack because NetHack is a different game to each player.

Checking conducts. The game tracks voluntary challenges for players who wish to make the game more difficult.

Checking conducts. The game tracks voluntary challenges for players who wish to make the game more difficult.

Of course, that doesn’t mean that NetHack can’t be a competitive endeavor. The month-long /dev/null/nethack tournament has taken place every year since 1999, making it the longest continuously-running gaming tournament on the internet. This tournament offers players a chance to show off their skills, with several virtual trophies such as “Best of 13” (most ascensions in 13 consecutive games), “Best Behaved Ascension” (most conducts in a single game), and even “Most Unique Deaths” (yes, there is an award devoted to games in which you do not ascend). A legitimate effort to claim any one of these (or one of several other elite trophies) will take enough time that a single player is unlikely to win more than one of them. There is so much indeterminacy in NetHack and so many conditions required to pull off a flashy ascension that a player who excels one year may not even ascend the following year. This is where a difference in play style really becomes apparent–some players see every character as ascendable and consider a high ascension rate to be a good measure of success, while other players see a particular goal as achievable under the right circumstances, no matter how many character deaths it takes to get there.

That philosophical divide, between minimizing risk to avoid death and maximizing risk to increase potential reward, is what I find most compelling about NetHack. For some players, a win is a win and a loss is a loss. They see a high ascension rate as the mark of a successful player, and consider most risks as just shortcuts for players who may be too impatient to play slowly and carefully. Taking the wrong risk can undo hours of good gameplay and kill a promising character. And with so many elements of randomness in the game, a veteran player will have seen several things happen that they might have thought were near impossible. After playing for five years, I have a real sense of what a 1% chance of a negative outcome means (hint: it’s not “basically never”).

Checking conducts. The game tracks voluntary challenges for players who wish to make the game more difficult.

Checking conducts. The game tracks voluntary challenges for players who wish to make the game more difficult.

So, a careful player will spend time avoiding situations in which a bad turn from the RNG could end their game. This style of play mirrors the decisions that most of us make in real life. If, for example, running a particular stop sign would give you a 1% chance of hitting another car, you probably aren’t going to take that risk. The few extra seconds you save by taking such a risk aren’t going to be worth that one time out of 100 that you end up crashing. Indeed, in real life you should take reasonable precautions to minimize your risk of dying.

But I don’t play NetHack for the sake of seeing how well things go when I act cautiously. I follow the “no risk, no reward” philosophy and will gladly lose dozens or even hundreds of ascendable characters all in search of one really impressive one. When I see a set of rules, I instinctively think, “How far can I push these rules? What’s the craziest thing I can do within the limitations of the game?” NetHack remains interesting to me because I still haven’t fully answered those questions. Putting heavy limitations on myself while playing a multi-conduct game forces me to be creative and resourceful, which makes for an exhilarating gaming experience. All of the time spent on characters that ended up dying seems worth it to me when I finally achieve my goal, knowing that I’ve completed a game in a way that very few other players can do.

It’s that unique sense of achievement that pushes me to keep playing toward the mythical twelve-conduct ascension. And while I may never achieve such a game, I’m also not sure that it would be impossible to do so. But just when I thought I was getting close to figuring it out, the release of NetHack 3.6.0 could set me back quite a bit. The developers have made several changes to gameplay, many of which are geared towards making the game harder. After a decade-long hiatus, perhaps this release serves as a tacit response to the question of whether NetHack is too easy.