Many versions of Angband, such as the MS-DOS port shown above, offer graphical interfaces. (Image: Mobygames.)
Arcane Games
Andi Sidwell was not sure what to make of the information displayed on the computer their big brothers shared. [Author's Note: Sidwell prefers they/their/them pronouns.] Antony was twenty, and Peter was twenty-four. The brothers, who had gone in together on a house, were always introducing ten-year-old Andi to hobbies and ideas. This one, a computer game called Moria, left them bewildered... and intrigued.
Sidwell leaned in closer to the monitor. Moria was a game, yet there were no colorful graphics decorating the monitor. Text characters formed the corridors and chambers of a vast dungeon. There were other symbols strewn around the screen: colorful letters that their brothers said represented fearsome monsters, as well as characters that stood for traps and treasure.
Sidwell was no stranger to esoteric computer programs. At the age of eight, they had come across a programming textbook for the BBC BASIC language used to write software on the BBC Micro, a PC manufactured by Acorn Computers. Later that year, older brother Antony had come home for a visit and brought his work with him; he was writing an application in a language called C. One afternoon, he got up and left the computer. Sidwell crept over to take a look. Antony's compiler, the program in which C programmers wrote and tested code, filled the screen. Slipping into the chair, Sidwell studied the syntax for a few moments, then began to tinker. "It was pretty unusual for someone my age to be doing that. It seems like that experience is more typical for people five to 10 years older than me—but even then, I suppose, still only for the geekiest part of the population."1
Moria stumped Sidwell. Determined to understand the game, Sidwell borrowed a copy of Moria from Peter and Antony. Weeks later, Sidwell had made little progress. The arrow keys moved the character, a little "@" symbol, but the deluge of other in-game actions left them stymied. Every key on the keyboard made something happen, and there were so many commands to remember. Something different happened depending on whether Sidwell pressed a lowercase "d" or an uppercase "D." Frustrated, they gave up.
In 1997, the Sidwell family connected their RiscPC—a successor to the BBC Micro—to the Internet. Once online, Sidwell's world shrank, but in the best possible way. With a click, Sidwell could visit the online homes of people who lived on the other sides of oceans—and download games for free. "I downloaded them systematically, tried the ones I liked, and junked the ones I didn't. There was a lot of crap."
Moria, it seemed, was not crap. People online seemed to love Moria. But for Sidwell, the game remained inscrutable. Sidwell had even tried UMoria, but it, too, proved confusing and frustrating. While on the hunt for new games to try, Sidwell came across a dungeon hack that had been baked in a similar mold. "At some point I must have come across Angband, and it was like an easier to use version of UMoria."
Angband was easier to play, though it was not the most entertaining game. Like Moria, all the action took place in dungeons. Sidwell did some digging and found that Angband was like ice cream: there were many flavors, and they had sampled vanilla first. The other flavors were called variants, derivatives of vanilla Angband that featured different settings, monsters, and magical items. "Soon after, I discovered Zangband [...] in 1999. I loved that game and played it loads. I was curious so I downloaded the source code and read it."
Fueled by curiosity, Sidwell began to code.
"You Have a Superb Feeling..."
With each cassette tape containing Moria that Robert Koeneke had mailed out, the virtual mines grew, ensnaring players around the world eager for a taste of randomly-generated treasure. James Wilson, a student at University of California Berkeley, got addicted to the game and decided to port it to UNIX. He called it UMoria, and released the first version in 1989. A year later, Alex Cutler and Andy Astrand, students attending University of Warwick, played the game and harbored the same ambitions toward UMoria that Koeneke had harbored toward Rogue seven years previous. They wanted more. More items, more monsters—more everything.
Moria lasted fifty levels, the last of which contained the dreaded Balrog. Cutler and Astrand doubled the count. The new master of the dungeon was Morgoth, known in Tolkien lore as the Black Foe of the World. Cutler and Astrand kept the Balrog, but demoted it. One Balrog made the bravest heroes quake in their boots, but the Black Foe had spawned legions of Balrogs to serve him. Even Sauron, the primary antagonist in J.R.R. Tolkien's The Lord of the Rings novels, was merely a lieutenant under Morgoth.
Tolkien's writings loosely guided Cutler and Astrand's design. To change things up, they scrapped the Mines of Moria and changed the setting to Angband, a fortress built by Morgoth beneath a mountain chain. Trolls, dragons, Ringwraiths, and other horrors from Tolkien's stories haunted Angband's halls, and were joined by unique monsters, powerful foes bearing the names and abilities of Morgoth's most powerful servants. Durin's Bane, the Balrog from Moria, was one. Sauron was another.
Players began their adventure through Angband by choosing from dozens of races and classes, then materialized in the town level, which reflected the settlement in Moria. A staircase near one of the shops led down to the first level of the dungeon. To heighten the sense of foreboding as players fought their way down, Cutler and Astrand marked their progress by depth instead of levels, as Robert Koeneke had done in Moria. Every level was fifty feet deeper than the last, which meant players had to descend through 5,000 feet of tricks, traps, and battles to stand a chance of confronting Morgoth, a gray "P" that had the power to smash through walls and summon dragons to his aid.
Upon graduating, Cutler and Astrand left Angband behind and moved on to other interests. Sean Marsh and Geoff Hill, two students who had pitched in on Angband's development, had no desire to see the game wither. They spent months adding artifacts and tightening algorithms. As Warwick students fought their way deeper into the game, Marsh and Hill generated a list of the most accomplished questers and posted it to the rec.games.moria newsgroup. Across the newsgroup and popular hacker channels, whispers of a public release of the game spread. That was exactly what Marsh and Hill had wanted. They had made the list of winners public in order to generate buzz: a successor to Moria, bigger, better, and tougher, was coming.2
On April 11, 1993, Marsh and Hill released Angband version "2.4.frog-knows" (2.4.fk) for UNIX, marking the first time the game had set foot outside University of Warwick. The public release included dozens of character races and classes, special rooms known as vaults where powerful foes guarded the best magical artifacts, and level feelings—printed messages intended to convey the level of danger and quality of items at the current depth. When players entered a new dungeon, the game analyzed the level and calculated a rating based on the monsters, vaults, and artifacts within. The higher the rating, the better the level feeling, which ranged from What a boring place… all the way up to You have a superb feeling about this level.
But Angband was not done yet.
The Years of the Lamps
Unbeknownst to them, Marsh and Hill had started a chain reaction. Shortly after they made Angband public, another programmer named Charles Teague pulled 2.4.fk from the newsgroup and converted it to MS-DOS so PC gamers could meet their fate in Morgoth's fortress. Teague continued to update the game until March 1994 when he rolled out Angband v1.4 and disappeared from the newsgroup, likely because real-life obligations grew too overwhelming.3
Concurrently to Teague's work on the PC adaptation, Charles Swiger began to examine the game's workings. His goal was to clean up the code and filter suggestions from players on the newsgroup, as well as authors of Angband flavors, variants to the game's source code. One of the most popular features of PC Angband 1.4 was colorized text, used by authors to distinguish monsters. For example, the Balrog of Moria appeared as a violet "U." Swiger, Angband's self-appointed maintainer, rolled the change and other suitable modifications into the next version of the game. Years later, official releases of Angband would be known as "vanilla Angband," the codebase and starting point from which any developer was welcome to develop other flavors for free distribution.
In late 1994, Swiger dropped a bombshell on the Angband faithful. He had found a job and no longer had time to maintain the vanilla game's code. Like the Tolkien world from which the game drew inspiration, factions arose and nominated contenders to vie for the position of god of the code. They knew that to be the maintainer of vanilla Angband was to become as powerful as Tolkien's Eru Ilúvatar, divine creator of the world. Each contender argued his or her merits, yet arguing was all that filled the empty void. The throne lay vacant.
On New Year's Day 1995, Ben Harrison made the decision for them.4 He had been playing the Apple Macintosh version of the game when a debilitating bug crippled his progress. Occasionally, an M would appear on-screen, signaling the appearance of a manticore—a monster with the body of a lion, the head of a human, and a mouth full of fangs. The manticore attacked by spitting fangs, triggered the message The manticore shoots spikes -more-, signaling that the player should press the spacebar to advance the fight. But because of the bug, the message repeated endlessly, forcing players to abort the game and lose their progress.
"The thrill. The immersion. The randomness of the levels. The excitement of realizing that you had found a magic item, or better yet, an artifact. The heart wrenching despair of falling through a trap door while running towards the glowing yellow tilde [representing a magical phial] in the corner of the room, and realizing that you had missed your one and only chance—back then—of acquiring the Phial of Galadriel," waxed Harrison, recalling the thrill of adventure and the close shaves he had survived (or not) that had given rise to his Angband addiction. "Some of my fondest memories involve drinking a random potion with bated breath, knowing it might be deadly, but unable to resist the lure of knowledge."5
While the rest of the Angband newsgroup community squabbled, Harrison quietly set about patching and organizing the game's cluttered code. His updates were like corks that plugged the holes left by Swiger's sudden departure. Unaware that so much debate had raged around the maintainership, Harrison was surprised when the code he had submitted on January 1 was declared the official 2.7.0 version, and that he, Harrison, had been appointed the new maintainer by proxy.
Harrison's efforts to organize Angband's underpinnings resulted in code that was easy to read and edit. Retroactively, Angband maintainers refer to his work as the great code clean-up. "Keep it clean, make it elegant," Harrison told me, summarizing his goals during his time as maintainer.
The great code clean-up benefitted other developers, too. Easier-to-read code allowed more programmers than ever to try their hand at downloading the source files and writing modules that added new mechanics, artifacts, monsters, and spells, as well spin off their own flavors of Angband that featured unique themes and settings. Other programmers concentrated on porting the game to a wider range of operating systems including Windows, Linux, and permutations of IBM workstations. Modules written by developers on the newsgroup often found their way into the vanilla codebase, provided the maintainer deemed the changes they introduced appropriate or exemplary.
"Actually, I [cleaned up the codebase] to simplify compilation on the Mac, for which large static code files were difficult to compile," Harrison admitted. "The customization benefit was actually an unintentional side effect."
For five years, each subsequent update from Harrison optimized Angband further, as well as folded in prime suggestions from the game's growing legion of fans. Like Smaug's horde of treasures, the game's trove of artifacts grew more and more resplendent while the monsters set to guard them grew greater in number and strength.
By the time he released 2.8.3, Harrison had succumbed to real-life responsibilities. Robert Rühlmann was next in the line of succession.
The Years of the Trees
In 2000, Rühlmann—known by the handle rr9 on the newsgroup—released an unofficial update. His version was a hodgepodge of bug fixes and features submitted by other players: the easy patch, an update created by Tim Baker that simplified the process of springing traps and opening doors; advanced artificial intelligence for monsters written by Keldon Jones; and randomized properties on artifacts, an enhancement created by Greg Wooledge. As Harrison had done, the simple act of taking charge was enough to anoint Rühlmann the keeper of the code. A few years of patches and updates followed until, inevitably, he announced that other obligations would force him to abdicate the throne.
Once again, diehard supporters who had been writing their own flavors or submitting fixes to the vanilla codebase threw their names into the hat. Among them was Andi Sidwell, known among other Angband players by the online handle takkaria. "The big change happened when rr9 [Rühlmann] said he didn't have time to continue," Sidwell recalled. "Unusually, he actually wrote to the newsgroup saying that he was going to stop working on the game and announced the search for a successor. There was a lot of private discussion amongst people who wanted to make their case for taking over."
Competition was fierce. Leon Marrick, one of the frontrunners, was the creator of Oangband, a flavor renowned for superb item and monster balance; and Sangband, short for Skills Angband, one of the oldest Angband flavors as well as one of the first to introduce unique character skills such as crafting armor, greater proficiency with certain types of weapons, and modes of unarmed combat such as wrestling and karate.6 Another candidate, Bahman Rabii, was Marrick's successor on Oangband. Timo Pietilä was a longtime player who had an eye for refining gameplay.
Rühlmann may or may not have been aware that there was any discussion around who would take his place; in March 2006, he unwittingly ended the debate by selecting Julian Lighton—perhaps, Sidwell posited, due to Lighton's conservatism toward the game, which appealed to long-time players. Lighton proved a divisive choice. He pushed out a few updates and then went quiet in 2007. Discontent rumbled up from the community. Sensing that Lighton might be stepping away from maintainership, Sidwell took stock of features players wanted to see. "I had been producing patches for some time, a lot of them just importing the interface features I missed from variants into Angband proper. Vanilla is something like the tree of life for variants, or at least was once upon a time, and I thought that improving vanilla would mean that future variants would have a better base to build on."
In an announcement email made to variant authors and players, Sidwell made it clear that they were not attempting to usurp Lighton's position as maintainer; their patches were merely an attempt to prevent Angband from rusting. Sidwell's changes were embraced by the community.
However, Sidwell was not looking to take over. They merely hoped that pushing out updates would light a fire under Lighton to return to work. Another variant author, Nick McConnell, paid a visit to the Angband newsgroup and lit a fire of his own.
From: Nick McConnell
Subject: Angband maintainer
Date: 2007-03-23 13:19:35
OK, someone needs to say it.
Andrew Sidwell should be the Angband maintainer.
Comments from anyone, particularly Julian [Lighton], encouraged.
Nick.6
Sidwell discovered the thread and was amazed to see a string of replies echoing McConnell's sentiment. Pleased, Sidwell wasted little time getting to work. "I remember when I was first taking over, I was mostly interested in implementing what you might call the consensus position. I made exhaustive lists of ideas and suggestions that others mentioned on the newsgroup; I had several different text files and over a period of months, read many arguments and copied and pasted into my text files some things I liked, and some I didn't. I came to a decent understanding of some issues this way, and eventually condensed my text files down into bullet points for implementation."
Sifting through topic after topic on the game's official forums, Sidwell paid special attention to arguments that broke out over polarizing features. One debate revolved around item degradation and how players used gold. Items degraded as players attacked and took damage, causing them to pour all their gold into costly repairs, which was not as fun or interesting as saving up to buy new items. After weighing community feedback, Sidwell introduced repair-item scrolls that restored items to good-as-new status. When people complained that the scrolls were overpowered—presumably because they allowed players to cling to tried-and-true items rather than branch out and try new ones—Sidwell removed the scrolls.
"The arguments that happened on the newsgroup, and later on the forums, really shaped my view of what my maintainership should be about—which really was following the lead of the community, neither going too slow or too fast, and trying to bring people along with each advancement," Sidwell explained.
Angband's community responded well to Sidwell's transparency. Previous maintainers had been wizards behind curtains, shaping the game in the way they thought would be best. Sidwell emerged from behind the curtain to take stock in what the players wanted, while also incorporating the cream of the crop from the many flavors floating around online.
After a year or two, Sidwell took the next logical step and converted Angband's monarchy into a meritocracy. Chris Carr, who had taken an interest in submitting ideas for fixes and features after Harrison's great code clean-up, was one of the first to join up. "[Sidwell] said, 'I'm not going to do the dictatorial maintainer-type thing where people send me patches and I'll decide whether or not to include it, and control everything, and release the game as and when I feel like it. I'm going to create a development team.'"7 Carr had been a member of Angband's community for years, migrating from the newsgroup to the forums. He had added his voice to Nick McConnell's when McConnell had proposed that Andi Sidwell assume the throne. [SQ1]
Carr had already been active role in Angband's development. "I had submitted patches to previous maintainers to fix and improve the random-artifact generator, but they had never actually been incorporated [into the official version]. Around 2009, I got in touch with takkaria [Sidwell] and said, 'I'd be really interested in joining your devteam. Here's my first contribution; tell me what you think.' I was made to feel very welcome. My patch was incorporated pretty much straightaway, and I received lots of testing and feedback."
Around the time Carr expressed interest in joining the devteam, several others were welcomed into the fold. Like Sidwell, Erik Osheim was at first bemused and then entranced by the cryptic nature of roguelike games. While in college in 2004, Osheim downloaded an update to the game that added a chart that listed the name of every monster and the colored letter that represented it—Baby Green Dragon and a green lowercase d, for example. Being able to reference the chart when a monster appeared enabled Osheim to recognize beasts that had given him trouble in the past, and determine whether it would be best to confront or sneak past them.
He found the monster chart so helpful that he decided to create a chart for items. "Items are the same way: you see a brown comma or gray slash, but you don't know what it is. I wanted [an item chart] to exist, but it didn't. So I said, 'I know how to code C. I could probably figure this out.'"8 After downloading the code and learning how everything fit together, he submitted his patch and kept his fingers crossed that Sidwell would merge it into the main game. "I earned the trust of Andi Sidwell and became part of the devteam. I picked up bugs from the bug tracker and started fixing them, and becoming familiar with the codebase."
Aaron Bader was another who earned Sidwell's respect by taking initiative. "I was posting on the forums a lot, and I think I made a comment like, 'Yeah, I'd like to modify the code, but I'm not sure how to get started.' They said, 'Just come on the dev channel. We'll help you get set up.' I guess a lot of people are bluffing: 'Oh, yeah, I would totally help, but it's too hard to get started.' But I took them up on that offer, and they helped me get set up. I started playing around with the code."9
Simulating Virtual Reality
To encourage more community members to get involved in development, Sidwell moved vanilla Angband's code to Github (pronounced "get-hub"), a repository where any number of people can submit changes to the code. Previously, Angband had been stored on Subversion (SVN), a repository that made it difficult to keep track of which sections of the code had been changed. Users who modified the same block of code often received errors if their changes did not jive with modifications another user had submitted before them. Github merged code automatically, facilitating organization.
Meanwhile, the devteam tightened loose screws. To expunge tedium and keep the game challenging yet fair, the devteam focused on balancing the game's procedurally-generated elements. A physicist by day, Aaron Bader enjoyed writing simulations to solve problems at work and in Angband. Years of playing the game had left him with burning questions, such as how often certain monsters tended to spawn and whether or not particular items dropped too often. "When you're playing, you really have such a small sample size. But with statistics-gathering, you get actual data. Like, okay, this is a 1-in-1,000 event; this is a 1-in-100 event; this is a 1-in-10 event. You can really answer some of the questions that are impossible to answer from the player's standpoint."
Searching for a reliable method able to produce definitive answers, Bader employed the Monte Carlo method. Often used to solve complex physics problems, the Monte Carlo method applies a series of algorithms that obtain numerical information from repeated samplings of randomized data. Angband is programmed to only generate certain monsters and items once the player reaches certain depths. By feeding his Monte Carlo algorithms data such as which monsters and items are allowed to generate, Bader can observe how often they appear on certain levels.
"For me, the aspect that's interesting is the game never plays the same way. Your capabilities are always different. Every once in a while, you'll use items you've never used before, just because it happens that they're the best items you have in that particular play-through."
Likewise, Erik Osheim runs simulations for the purpose of studying which monsters are able to appear at certain depths. Out-of-depth monsters, those that would normally appear on deeper, tougher levels, are allowed to pop up earlier so long as their presence does not become cloying. "We want [randomized experiences] within parameters. Either there should be no chance of an encounter with, say, a great wyrm [dragon] on level 1 at all, or [the odds] should be so low that they're basically zero. Most players will never see it. And those [odds] basically end up being the same thing. If a person dies to a super-powerful enemy once, they might complain, but as long as it doesn't happen a lot, it's not really a big deal."
Running simulations is critical to maintaining balance in Angband. Unlike NetHack and Rogue, Angband generates new dungeons when players return to levels they cleared. Players tend to fall into two camps. Those who like to clear the same level over and over in the hopes of finding top-tier items known to appear there; and those who prefer to clear a level once and then forge ahead. Ultimately, the devteam views their job as straightforward. "Rule one of Angband is play how you want," Carr explained. "If you want to spend millions of turns hanging around at 2,000 feet because you don't have resistance to poison, it's fine to do that. But we're realistically trying to balance the game according to a certain expectation of pace. That means that if you choose to play really, really slowly, then the game will be a little easier." [SQ2]
Priority number-one for the devteam is eliminating tedium and needless complexity. "Behind Angband's apparent complexity, it's actually a really basic game that hasn't moved on much since its inception," Sidwell explained. Complexity has become the poster child for the belief that roguelikes are too difficult for laypeople to attempt without lots of preparation. "When I took over, loads of crap got generated in the dungeon, and its only role was to use up the player's time without giving any kind of reward. So I think a lot of the difficulty of the game really was in avoiding tedium."
The devteam walks a fine line. Removing aspects that they view as tedious or overly complex may cause veteran players to have an easier time finishing the game. "Without pretty graphics or a compelling storyline, the risk of death is the only real draw—especially in the early versions, which made it [nearly] impossible to cheat death. Note that I never came close to actually winning," Harrison admitted.
Permadeath should remain a learning tool, and should be avoidable in most cases. Harrison believes that "Unlike NetHack, a normal person can actually expect to pick up everything they need to know about Angband via (several characters' worth of) experimentation. Granted, some lessons, like running away from any multi-colored lowercase 'd,' come at a terrible cost. Some levels are intended to be unwinnable, like jelly pits encountered by weak characters, or a level with an out-of-depth dragon wandering around."
Osheim agreed. "A lot of times, I think becoming a better Angband player is all about coming up with [short-term] goals. You can say, 'Okay, I died this way this time. I'm not going to let that happen again.' As long as you're dying a new way each time, you're getting better." He spoke of one player who had become famous for entering the dungeon with nothing—no weapons, no torch. The player delighted in devising his own path through the game, presenting a new and interesting challenge each time. [SQ3]
The Years of the Sun
"I think I had one of the longest innings of all the maintainers," Andi Sidwell reminisced. "When I started with 3.0.7s, I think I was 16, still in compulsory schooling. I've changed a lot since then, and honestly, Angband doesn't hold as much appeal as it once did." Other interests such as radical politics had stolen Sidwell's attention. The time had come to move on.
Ready to hand over the crown, Sidwell visited the forums—the Angband community's new home after Usenet dried up—and turned the tables on Nick McConnell.
October 22, 2013, 12:47
takkaria
OK, someone needs to say it.
Nick McConnell should be the Angband maintainer.
Comments from anyone, particularly myself, encouraged.10
At first, McConnell viewed the comment as nothing more than a pithy joke, Sidwell's way of acknowledging how McConnell had volunteered them for the job years ago. He had been maintaining his own well-received variant, FAAngband (First Age Angband), which hued closely to Tolkien's lore, ever since discovering Angband via a peculiar screensaver running on a colleague's computer at work. The screensaver was the borg, a bot programmed to navigate Angband's dungeons and defeat monsters. After watching it for a while, McConnell sought the game out and began to play. "I'm a big Tolkien fan, so I liked the theme straight away. Then, as I started to actually watch what was going on, I found the game messages tremendously entertaining: You hit the singing, happy drunk. The singing, happy drunk flees in terror! and so on."11
To McConnell's astonishment, other community members echoed Sidwell. Sidwell wrote a reply explaining that the nomination for McConnell was indeed a tongue-in-cheek reciprocation for what McConnell had done years before. But the more users read the thread of posts, the more votes McConnell found in his favor. On October 24th, he wrote a quick reply confirming that he was giving the matter consideration. More replies flooded in, asking when the coronation would take place. After mulling it over, McConnell donned the crown.
Still active today, McConnell has the same priorities as those who came before him: to keep the game challenging for as wide a demographic as possible. "I see roguelikes as very stripped-back games—the game experience is more like an intellectual puzzle. As such, they really need to be challenging to engage the player; as a consequence they tend to get a reputation for being difficult, which just adds to the sense of challenge—and the sense of achievement from winning."
As of the summer of 2014, less than a year into the McConnell regime, the devteam and its new head maintainer were still feeling their way through how they would collaborate. McConnell's first job was to restructure the code, an ongoing process. Until he finishes the restructure, he has asked that the codebase remain frozen, as revisions and new features would only complicate the clean-up. Even so, McConnell wants to assure long-time fans that he has no radical changes in mind. "I think what I would regard as core traditions are the dependence on character progression by items, and the capacity to play at one's own pace. There are also, of course, many monsters and items which are beloved—or deeply hated—and which I would not want to mess with too much. Also: permadeath." [SQ4]
Test of Time
For over twenty years, the titanic doors of Morgoth's fortress have creaked open, swallowing hapless and hearty adventurers alike. Thousands have ventured into the labyrinth, and the devteam—McConnell, Bader, Carr, Osheim, and Robert Au—believe thousands more will brave its depths so long as one tradition remains sacrosanct: "It's very clear that what has kept Angband going is that maintainers have graciously stepped down and handed over the reins. No one has ever said, 'This is my game. I get to decide what it is, and I don't want to work on it anymore, but no one else can either.' Every time a maintainer has stepped down, they've passed the torch amicably to the next person," Bader explained.
Sidwell speculates that interest will continue indefinitely due to cheap, widely available Internet access, which facilitated rapid growth of the game and its community beginning in the 1990s. "There was an outpouring of innovation when variants started being made and they all provide quite a different gaming experience. In some sense it makes a lot of sense to have so many variants; if everyone tried to work on one game, you would have an unintelligible and broken mess. This way you get cross-pollination of ideas, copying of code, and so on. I think it's probably a good reason why Angband has stayed going."
Most importantly, a text-only game that originated on a dumb terminal did not seem outdated—not then, and not now, when the imagination still remains a more powerful graphics processor than the most lavish video card. "Every few years, somebody tries to make a multi-player version [of Angband], or a version with 'better' graphics, but that is, in my opinion, missing the point," Harrison stated.
New players find their way onto the forums, ask questions, and begin their quest. Osheim believes it should be the goal of all devteam members to make the descent into darkness as welcoming as possible.
"I think that despite the fact that it's a somewhat challenging game, I think the experience could be much more welcoming for new players. Some games have a tutorial mode. I would love it if Angband had more of that stuff. We have some things like that, but we have a long way to go."