How to Disappear Completely: The Tale of Azrael’s Tear

[ Editor’s Note: This feature was commissioned as part of the RetroPitch 2017 event. ]


Some games fail. This bears repeating, given Cuphead’s explosive commercial success last month.

Originality, style and strong execution don’t ensure profit. In the games industry, this observation is at least old enough to drive: imaginative triumphs like Black Isle Studios’ Planescape: Torment and Double Fine’s Psychonauts are famous — arguably most famous — for their failure at retail. For every Cuphead, there is at least one Dujanah.

Most of us accept this painful reality. The harder truth is that some games, even incredible games, disappear.

Today, everyone knows that Psychonauts is the best game no one played. Yet this is common knowledge only because we all ended up playing it: word of mouth slowly brought the game’s sales to 1.7 million copies, and climbing. In the years after reissued Planescape: Torment, it was consistently a top seller on the site. Its spiritual sequel, Tides of Numenara, set funding records on Kickstarter.

But not every fantastic failure becomes a poster child. No great game is guaranteed a legacy by default.

What follows is not the story of the best game no one played, but of the best game no one remembered. The story of a game cherished for 21 years by only the tiniest, most disparate group of true believers, even as it disappeared from store shelves and down the cultural memory hole.

This is the tale of Azrael’s Tear.

– – – – – –

Azrael’s Tear (1996), Mindscape/Intelligent Games

I guess when you combine a thief with an ancient subterranean temple, grailstone, weird time-locked creatures, the undead, and a group of insane knights wanting world domination… well, interesting things are bound to happen!

—Ken Haywood, designer

Azrael’s Tear is a game that traffics in mystery, primarily. Like Gone Home, another title based in exploration and discovery, it’s hurt by writers overeager to share their experiences of it. This is a strange game, a dark game, and no description can replace the first-hand act of playing it.

With that in mind, this overeager writer will attempt an outline.

After the introductory cutscene, compelling and opaque, Azrael’s Tear begins. You emerge into a dripping, drafty stone hallway. Lizards resembling miniature dinosaurs wander harmlessly at your feet, and your helmet’s readout declares:

                                                                        Model 7 Head Up Display


                                                                        H U D  S t a r t u p . . .

                                                                        Env Systems…. OK
                                                                        Life Support…. OK
                                                                       Sensor Fix…. OK


                                                                       Redundancy….         0x0aa3
                                                                       ROM Checksum…. 0x453f

                                                                       S y s t e m  I n t e g r i t y :
                                                                        1 0 0 %

                                                                       MISSION · Aeternis
                                                                       penetrated · 12/2/2012 · all
                                                                        systems functional ·
                                                                        OBJECTIVE · holy grail

This is a quest to steal the Holy Grail from the Gothic, underground city of Aeternis, in other words. It is far-flung 2012; you are a kind of cybernetic Indiana Jones; and many of Aeternis’s inhabitants seem to be centuries old. The Knights Templar, a medieval order thought annihilated in the 14th century, are still alive here. And they behave and believe like people from the Middle Ages.

It’s a short walk to the first clue: a torn scrap of parchment, taken from a 12th-century medieval manuscript. These parchment scraps, and the splintered narrative they convey, will soon become familiar.

Azrael’s Tear (1996), Mindscape/Intelligent Games

Despite its protagonist’s-eye view, Azrael’s Tear is not a first-person shooter. It quickly reveals itself as a genre mutt, akin to Metroid Prime with less shooting and jumping.

Like in Prime, the goal is to explore an open, atmospheric world while solving puzzles and collecting backstory fragments. Your helmet scans and describes the locals (and locales). Unlike Prime, though, there’s much talking to do — with medieval knights, fellow thieves and the dead. Many of them are, apparently, insane.

Between you and the Grail lies a world of surprises, and nightmares. Yet the game’s imaginative flights are anchored by an attention to historical detail. Aeternis contains both beautiful illuminated manuscripts and still-living men executed centuries ago; both chthonic beasts and serious, near-scholarly consideration of medieval politics, art, culture and religion.

At this point, certain readers may be wondering why they haven’t heard of Azrael’s Tear. The reasons are complicated.

Soon after its release for DOS in 1996, poor sales drove the game from store shelves. It was an era before commercial bombs could linger, as they do now on and Steam. Physical shelf space was precious.

And so no more copies were printed. Companies went bankrupt; licenses fell into limbo. Azrael’s Tear joined many other inventive computer games of its time, like Terra Nova: Strike Force Centauri, in the bargain bin and then oblivion.

What’s unique is that Azrael’s Tear stayed there.

– – –

In late 1993, Rand and Robyn Miller released Myst for Macintosh computers. This oddball adventure title — composed of mysterious clues and arcane puzzles, live actors and luscious pre-rendered backgrounds — quickly sucked millions into its world.

Although Myst required expensive, uncommon hardware (read: a CD-ROM drive), it sold 200,000 copies by April 1994 and a million by mid-1995. It remained the all-time biggest computer game until The Sims.

But just before Myst mania, around 1992, a self-employed graphic designer named Ken Haywood had an idea. He wanted to write a novel. An ambitious science-fiction novel — a story about the Holy Grail, and the futuristic crypt robbers trying to steal it from an ancient temple.

This story seed, Haywood tells me, “quickly morphed into something more.” It hit him that the idea would work as a computer game.

Myst (1993), Brøderbund/Cyan

 From Australia, but living in England’s city of Winchester, Haywood had “hardly played any games” and had never made one. “I just saw it working well within that space,” he says. Yet the resources for an independent, first-time developer were limited. In that era there was no Unity or GameMaker, no GIMP or Krita or Blender, no Audacity, no GitHub, no Kickstarter or Indiegogo.

Even Britain’s so-called “bedroom programmers” of the ‘80s — the indie upstarts who gave us Elite and Metagalactic Llamas Battle at the Edge of Time — couldn’t have pulled off Haywood’s concept. This project needed a publisher, a budget and a professional team.

So, in the spare moments his graphic design job afforded, Haywood studied the game market and developed a pitch document. The story was bold, unique: a tale that stretched from the Mesozoic Era to the Middle Ages to modern day, inspired partly by conspiracy theories from Holy Blood, Holy Grail. Haywood named his project Raptor, after the high-tech thieves, Raptors, hired to steal the Holy Grail from the temple-city Aeternis.

Despite his newness to the games industry, Haywood nevertheless found an agency — Bad Management — willing to pitch Raptor to a development house.

The developer was Intelligent Games. A six-man team located in nearby London, founded in 1988 by Matthew Stibbe, Intelligent would make its name with SimIsle and Lego Loco. But these came later. When Raptor arrived, many at the company were hardly more experienced than Haywood himself.

Luckily, the Intelligent team liked the idea, and Haywood was hired. However, it wasn’t yet decided what form the boundary-pushing design outlined in Haywood’s pitch — procedural character interaction; a “living story” that developed “algorithmically” — would ultimately take.

Raptor concept art, Ken Haywood

By this point, it was 1993. The Myst volcano was erupting, alongside similar stills-and-actors hits Return to Zork and The 7th Guest. Intelligent Games chose to join the burgeoning Myst-clone gold rush: Raptor would be a first-person adventure game, with live actors superimposed over pre-rendered backgrounds.

Intelligent’s goals were bigger than this choice suggested, though. A blunt, three-word declaration in the team’s pitch to publishers put it simply: “Raptor is different.”

– – –

Dream games by inexperienced developers rarely work. Some readers will recall Battlecruiser 3000AD (1996), a space sim whose ambition and broken promises dwarfed even those of No Man’s Sky.

At the same time, newcomers can succeed with inadvisable projects that wiser minds wouldn’t touch. Improvising amateurs built Ultima Underworld (1992), the landmark RPG that inspired Deus Ex and Dishonored. “We didn’t have any idea that we were doing anything that wasn’t just obvious in some sense,” Underworld’s Doug Church once said, “because we had no context.”

Intelligent Games’ final Raptor pitch in 1993 was thick with unattainable dreams. Non-player characters interacted procedurally with each other and the player, while also competing with the player to solve puzzles and locate the Grail. The first-person view — pre-rendered graphics, live actors and all — was rotatable 360°, a feat of technical magic the Myst games didn’t manage until 2001. And its central “living story” concept still defeats today’s best designers.

Raptor gameplay mockup, Ken Haywood


At the time, Intelligent had shipped only two titles, both of them comparatively-simple strategy projects. For a small team of newcomers to attempt Raptor was unthinkable: most industry veterans would have quailed at its scope.

But Intelligent, like Underworld’s developers, was too new to be discouraged. Raptor’s art lead Richard Evans tells me that “most of us were new to games full stop — so we were just playing around with what we could do.” His sentiment is echoed by Phil Veale, Intelligent’s former lead programmer: “I think we pushed the envelope somewhat because we didn’t really pay much attention to what we couldn’t really do!”

According to Ken Haywood, the pitch went to “various games companies” without a bite.

Eventually, just as Underworld had found a publisher — Origin Systems — willing to gamble on the impossible, Intelligent Games found Mindscape. Headquartered in the States, Mindscape had a history of publishing groundbreakers like 1985’s Déjà Vu, the first true point-and-click adventure game. Raptor would be Mindscape’s Myst-clone, impossible or not.

– – –

More than any other factor, what separated Ultima Underworld from Battlecruiser 3000AD was its team’s capacity to roll with the punches. Two naïve dreams met reality and crumbled — but one team riffed its way, jazzlike, into something that worked. In a way, this is what Intelligent Games did with Raptor.

With the project greenlit, Haywood began working full-time at Intelligent. His chief accomplice was designer Richard Guy, another relative newcomer to the games industry, whom Haywood describes as an intellectual with “a wealth of knowledge on most things.”

Among other points, Guy had a BA in fine art and a fascination with pen-and-paper role-playing. He works today as an academic, with a Ph.D. in the history of the Dutch East India Company’s ship architecture.

“And together,” Haywood tells me, “we completely re-worked and polished the concept into the final game design.”

Raptor concept art, Intelligent Games


Beginning in 1994, their collaboration lasted around nine months. The combination of Guy’s knowledge and Haywood’s own deep interest in history — and heavy research into the Knights Templar — gave Raptor‘s design a sophistication rare in newer games. These were outsiders, not insiders, and their interests stretched far beyond their chosen medium.

The game evolved from Intelligent’s short pitch into an almost unrecognizable design bible. Haywood describes it as a “100-odd page” behemoth — “a step-by-step walk-through” of everything in the game’s roughly 90 planned rooms. He continues:

It included all the puzzle and dialogue details, scene requirements (for interaction, atmosphere and puzzle functions) and character interactions. The game was broken into a series of sealed nodes in which the player had to solve various puzzles and tests to pass through and move deeper inside the temple. The bible was comprehensive enough that anyone working on any part of the game could see immediately what was needed in that part, and see too how that part fitted into the game world.

In recent years, a pre-bible design document for Raptor has surfaced online. It retains many of the pitch’s biggest ambitions — multiple protagonists; real-time competition between Raptors; an outrageously complex dialogue system — that met the cutting-room floor as Haywood and Guy revised the game.

What the two of them concocted instead was more doable, on a technical level, but more creatively inspired. And it closely mirrored the game Raptor would become.

In the end, the pair devised “a very deliberately non-linear plot,” as Guy said in 2012. But not the simulated story-world of the pitch. “There are 4 mandatory ‘gates’ in the design that you have to work through in sequence,” Guy explained, “so it’s a limited kind of non-linearity, but still a huge headache to design, despite its limitations.”

Haywood and Guy added massive depth to Raptor’s setting, but removed — or left willfully, poetically vague — much that the earlier document had overexplained. This included the nature of the mysterious Holy Grail itself. The pair kept the best characters, creatures and locations conceived before (like Jack Sinclair, the Sanctum Atrium and Eugenes), and polished them. And added more.

The early design had focused on mechanics and systems and automation, almost like a competitive board game; but the bible proffered a cinematic experience. It felt authored, intentional. Although the final name would come later, Azrael’s Tear had been born.

Raptor concept art, Intelligent Games


Not all of this revised design derived from a two-man cabal, though. The operation was freewheeling; no one at Intelligent Games was left out. As Guy put it:

My sense is that during that process you really couldn’t have said who designed what – it all came out of conversation and back-and-forth creation, which extended to the whole team of artists, programmers and dialogue writers.

And one of the team’s biggest decisions was to ditch the Myst format.

– – –

It was a choice dictated, in part, by necessity. “We realized that we could just about do it in 3D,” Matthew Stibbe said in 2015, “and that would liberate us massively from the problems of doing it Myst-style.”

Instead of Myst’s beautiful but ultimately 2D world, the team reimagined Raptor as a 3D adventure with total freedom of movement. Pre-rendered visuals might sound easier on paper, but achieving the player mobility required by Raptor’s design “would become quite complex and that involved having lots of imagery,” Stibbe said.

Pre-rendering so many images could get expensive fast. Myst’s 1997 sequel, Riven, cost over $10 million to develop.

And so Intelligent Games created Xanadu, a proprietary graphics engine that pushed the technological boundaries of the day. This move was ambitious, as usual; but it wasn’t beyond the realm of possibility for a small team in the mid-’90s. As Guy later said, “3D graphics were then still at the stage where you could learn to do cutting-edge stuff quite quickly.”

Azrael’s Tear (1996), Mindscape/Intelligent Games


Xanadu’s creation was spearheaded by Phil Veale, a man described as a “technical genius” by Intelligent’s former programmer Martin Fermor.

At a time when 3D games like BioForge and Duke Nukem 3D hedged their bets, using pre-rendered backgrounds or 2D sprite characters, Raptor tried it all. Fully 3D environments and character models; intricately detailed textures and geometry; and a preference for variety over incessant tiling.

“I think it mainly pushed hard because we were focused on the result,” Veale says of the engine, “not any pre-existing limitations.”

“We were trying to figure out how to do a lot of this stuff from scratch,” recalled Stibbe. It wasn’t easy, and it didn’t help that Raptor was a DOS game, with all of the technical limitations that implied. In the era before Windows became standard, memory management was a constant headache. As Veale remembers, “We had to add in some virtual memory handling at the very last moment because we kept blowing up memory with textures that were too large.”

Environments described in the early design proved too large to be rendered, so Intelligent redesigned Aeternis’s layout in the bible. “We didn’t have the processing power to do deep vistas,” said Stibbe, “you couldn’t look a long way or else you use up all your polygons.” Instead, regions were broken up à la Metroid Prime, with load times at each entrance.

Even with these precautions, Raptor‘s system requirements remained enormous, and its frame rate low. “Despite having really smart engine and tools programmers,” Guy lamented, “we just didn’t have the resources to improve the engine significantly.” Still, the results spoke — and speak — for themselves.

– – –

Raptor’s technical achievements were matched by an unmistakable visual flair, eye-catching even now. The title exemplifies the importance, even the supremacy, of art direction in games.

Whenever an area was completed in the design bible, one of Intelligent Games’ 3D artists would begin working on it alone, using a combination of 3D Studio (now Autodesk 3ds Max) and Adobe Photoshop. Richard Guy believed that this setup “had the advantage of making each room quite individual and [internally] cohesive, rather than enforcing a consistent art style too strongly.”

The result was that Raptor‘s environments surprised at every turn, in an era when many 3D games (like Quake) skewed samey.

Azrael’s Tear (1996), Mindscape/Intelligent Games


According to art director Richard Evans, the team drew visual inspiration from Jurassic Park and the films of Terry Gilliam. On close inspection, the influence of Gilliam’s Brazil (1985) appears subtly throughout the game.

Regardless, Raptor’s art wasn’t a matter of careful planning. Evans says that “most of it was just having a good time drawing mutant dinosaurs, ghosts and weird caves.” Neil Crosbourne, another artist on the game, gives Evans’ “vivid imagination” the lion’s share of the credit for Raptor’s unique look.

“His ideas are always pretty ‘out there’,” says Crosbourne. “A brilliant artist who is heavily influenced by manga.”

Even the game’s signature use of darkness arose organically, Evans says, once the artists learned that they could “render the lighting from the 3D models into the textures.” He points out that the careful observer “can almost tell the order that the locations were made by the degree of sophistication of the lighting and colour schemes employed.”

However it happened, Raptor ended up resembling Gothic art combined with German Expressionist film. In many ways, it’s reminiscent of Jiří Barta’s puppet-animated masterpiece The Pied Piper (1986). The game’s visual style embraced the necessities of angular, grainy, low-polygon graphics — and ran with them.

This art direction, run through Raptor‘s cutting-edge technology, stood out in the ‘90s; and it’s aged gracefully since. It doesn’t hurt that 1998’s Thief: The Dark Project, a title still revered today, has carried a very Raptor-esque look into the cultural memory. (Asked about the similarity, Evans is uncertain, but he says he’d be “very happy” if Raptor’s visual direction inspired other artists.)

– – –

In April 1995, just after Raptor’s bible was finished, Ken Haywood left Intelligent Games and returned to Australia. His main work was complete. In his place, Richard Guy assumed the role of creative lead, and shepherded Raptor to completion with the design bible as his blueprint.

Back in Australia, Haywood laid plans for a Raptor novelization — and for two sequels to the game proper. At the same time, he remained a key force in Raptor. “I kept in close contact with the team afterwards and was regularly updated on developments,” he tells me. Nothing major happened, or was changed, without Haywood’s input.

Azrael’s Tear (1996), Mindscape/Intelligent Games


Yet Raptor had a bumpy final stretch. This would be the first link in a chain of troubles that, ultimately, buried the game for over two decades.

Even as Mindscape’s deadline for Raptor loomed, the game was lagging behind. To catch up, Intelligent had no choice but to cut “big chunks [of content] late in development,” according to Guy. Around 90 rooms were reduced to 70-ish. A detailed introductory cutscene, meant to explain the game to new players, was swapped out for a simpler text-based opening.

To complete the salvage job, the team scrambled to “close up some plot plumbing,” Guy said. An unfortunate loss, certainly, but Intelligent plastered the cracks well: most of the signs of crunch-and-rush development were hidden. Anyone familiar with the Planescape: Torment Unfinished Business mod will wonder what was meant to be, though.

Another point of contention was the new name. Instead of Raptor, the game became Azrael’s Tear. Haywood criticizes this change even today, and speculates that it hurt the game at retail. Matthew Stibbe, meanwhile, called the switch “catastrophic” in 2015. “It’s hard to figure out what it means, what the game is about, what you’re playing,” he said.

However, Mindscape’s marketing effort posed an even bigger problem. Haywood calls it “inadequate,” but Guy went further.

“I was never happy with the marketing Mindscape gave it,” Guy said, “but that was entirely outside our control.” In particular, he pointed to Raptor’s lack of exposure via magazine previews, unlike the “much smarter marketing presence” of Tomb Raider. And then there was the box art that Mindscape concocted, which Guy said “made a lot of folks on the team angry”:

Azrael’s Tear box art, UK (left) and US


Adding to the pile, Azrael’s Tear had a near-miss with disaster just before launch. As Intelligent’s Iain McNeil, whose job as production coordinator mostly involved “doing the stuff nobody else wanted to,” remembers:

We received the final disks in their boxes ahead of the launch. We were very excited so opened them up straight away to see how it all looked. Only to find the boxes contained German disks. That resulted in a few frantic conversations with the publisher and the factory, but it somehow all got resolved!

And finally, fatally, there was Quake.

– – –

When Azrael’s Tear launched in mid-1996, the response was strong. Critics ate it up.

In the United States: an 82% from PC Gamer, 4/5 from Computer Games Strategy Plus and 3.5/5 from Computer Gaming World. It fared even better in its home country. Awarding it a rare and coveted 90%, Britain’s PC Zone declared Azrael’s Tear “a yardstick title, a game against which other adventures should be measured.”

As is often the case, though, critical approval didn’t mean public acceptance.

Putting it bluntly, Phil Veale told me that “the sales were not awesome.” Likely the biggest factor was that Azrael’s Tear dropped the same week as Quake, the proverbial ultra-hyped blockbuster, whose graphics were flashier and whose engine was (significantly) faster. This shooter and its technology, Richard Guy said, “became the de facto industry standard overnight.”

Azrael’s Tear (1996), Mindscape/Intelligent Games


Azrael’s Tear struck an unlucky balance. Its huge system requirements barred it from the mainstream audience that Myst had captured. At the same time, it didn’t provide the action-packed gameplay that most gamers with powerful PCs devoured.

In this latter respect, Quake beat all comers — and kept the enthusiast market occupied when Azrael’s Tear needed its attention most.

The bad news didn’t hit all at once. “In those days we had very little information,” Iain McNeil tells me. The gears turned slowly. “There were no online reports,” he explains, “retail sales were reported quarterly.” All of which meant that Intelligent Games long remained unaware of the situation.

“By the time we knew how the game had sold, we’d probably be 3-6 months into development of the next game,” says McNeil.

And so, two oceans away, Haywood powered forward with Azrael’s Tear 2. At the same time, Intelligent Games jumped into another Xanadu-based game, built with an unrelated IP. Eventually the Azrael’s Tear sales numbers arrived.

Eventually, everything was canceled.

– – –

After selling Intelligent Games in 2000, Matthew Stibbe left the games industry. The company folded two years later.

“The job changed on me when I wasn’t looking,” he later said. The computer games business wasn’t the one he’d joined in the ‘80s. Budgets and teams and risk had grown, and the bureaucracy with it; and innovation had withered. By the time Stibbe left, Intelligent

was 70-something people and I sat in my office kind of dealing with money, lawyers, contracts, HR, managing a team that managed a team that managed development, [and] nobody would let me make games anymore.

Richard Guy soon abandoned games, too, for similar reasons. “I got out of the industry partly because I saw that it was pretty unlikely that I’d ever get to work on something as creatively unfettered as Azrael’s Tear ever again,” he said in 2012. And it makes sense: Azrael’s Tear was the kind of game only possible before — when the frontier was open and outsiders had an in.

One team member who did find a path forward in games was Iain McNeil, who’s successfully run an independent wargame house, Slitherine Software, since 2000. Superb titles like Battle Academy 2 continue to be designed by his hand today. Meanwhile, Richard Evans leads the Slitherine art department; and Phil Veale is the company’s technical director.

Battle Academy 2: Eastern Front (2014), Slitherine Software


As for Ken Haywood, he left the games industry after writing one last non-Azrael’s Tear design pitch, which failed to gain traction in Australia. He explains that “the industry here was too small at that time” to support his plans.

As a result, Haywood returned to graphic design. He now owns and manages a design firm, Clik Creative, in Sydney. Azrael’s Tear remains his only game.

– – –

It is a shame that this game disappeared under the sands of time because this is without a doubt, one of the greatest games I have ever played.

—Letter to Matthew Stibbe from an Azrael’s Tear fan, many years later

Game preservation experienced its dark night of the soul during the 2000s.

Visualize a time before digital distribution was mainstream, or even extant. Before Humble Bundle,, Steam. Before DOSBox — or, in the years after DOSBox’s creation in 2002, before it worked. Before 15 years of thread build-up at VOGONS made technical problems solvable.

Back to time when ReplacementDocs maintained an inventory of around 400 documents.

Now, imagine a period before companies like Retroism and Night Dive Studios made a business out of resurrecting old games. Before game publishers saw any value in back catalogues. When company bankruptcies could throw rights into limbo, forever. When a set number of copies was printed for each game, and no more.

There was a viable second-hand market for popular games, certainly; but what of catastrophes like Superhero League of Hoboken, which sold fewer than 30,000 copies worldwide? They were often expensive, if available at all. And they were finicky on new hardware. Older floppy-disc titles, like Ultima Underworld, faced the same and worse: new computers shipped without floppy drives.

Azrael’s Tear (1996), Mindscape/Intelligent Games

The feeling might be impossible to understand for those spared this moment in history — but there existed an extremely real, unbearably present sense that old games were disappearing. Operating systems and hardware became less and less compatible with the classics, just as the classics became less and less available.

It was into this quicksand that Azrael’s Tear fell, alongside thousands of other games, both famous and unknown. During a presentation on game history at GDC Europe 2001, Matthew Stibbe stopped in the middle of the usual timeline — Pong, Zork, SimCity, etc. — to discuss Azrael’s Tear. “Hands up everyone who played it?” he asked. “Who even heard of it?”

Although Stibbe tried to provide a playable version of each game in his history, Azrael’s Tear wouldn’t cooperate:

Ironically, it was also the hardest game to find a playable version. My PC will emulate a BBC, a Spectrum, an Atari 800, even a PDP-1 but I can’t get it to run a DOS game any more!

It was 2001, five short years since Azrael’s Tear launched, and the game was completely forgotten and impossible to play. From the standpoint of that time, Azrael’s Tear had no future, and had barely managed a past.

– – –

The situation improved, somewhat, as the 2000s continued. DOSBox started to work, most of the time. And websites like Abandonia and Home of the Underdogs, originally launched in the late ‘90s, grew larger and more mainstream. Their purpose was simple: making old, lost computer games available for download.

Downloading such games was piracy, technically, even though the original companies were often long dead. Yet the ethics were gray: when no one was around to sue, the crime seemed victimless. Products passed around this scene were therefore termed “abandonware” — illegal, but fair game.

This abandonware scene is where Azrael’s Tear found itself. Intelligent Games was dead, Mindscape didn’t care and no stores still sold the game.

Abandonia’s homepage, circa 2007

Even here, though, Azrael’s Tear was unlucky.

A game like Ultima VII was a handful of megabytes. This was small enough for a shoestring site like Abandonia, the most reliable and polished abandonware source, to host comfortably. Azrael’s Tear was an enormous CD-ROM title. As such, Abandonia’s main site never hosted it. Even the staff’s well-hidden “ISO Cellar” didn’t include Azrael’s Tear until 2009.

Home of the Underdogs, a rickety website smacked regularly with legal threats and takedowns, was the only decent place to download Azrael’s Tear during most of the 2000s. But Underdogs ripped filesize-expanding features out of its CD-ROM games, including cutscenes and voice acting.

Azrael’s Tear was originally hundreds of megabytes; on Home of the Underdogs, it was around 70. It was a shell of a game.

For those unable to procure a second-hand copy, then, the true Azrael’s Tear experience was exclusively available via BitTorrent ISO piracy. That meant finding someone to seed a torrent for a game no one knew about.

– – –

Even still, Azrael’s Tear faced a yet more fatal problem: the average computer couldn’t run it in DOSBox.

DOSBox emulates a second computer inside one’s own — a taxing process, especially in the program’s (much slower) early years. Most DOS games were low-definition enough to be playable anyway. Not so with Azrael’s Tear, perhaps the most hardware-intensive DOS game ever produced. The problem that helped to sink its launch, enormous system requirements, refused to go away.

This situation was eventually put to rest by DOSBox’s final versions — especially v0.74, released in 2010 — and hardware advancement in the industry. Until then, the curious had to play Azrael’s Tear the old-fashioned way: Command Prompt; hacked sound emulation; and MoSlo, to drag the computer’s speed back to 1996, manually.

This was unstable, to say the least. Even when it (sort of) worked, the music was ruined by modern computers’ poor MIDI libraries, scrambling the excellent soundtrack.

Home of the Underdogs’ homepage, circa 2005

Azrael’s Tear was, therefore, almost impossible to find and even more impossible to play. At a time when games like Darklands (1992) were building huge new fanbases as abandonware, Azrael’s Tear was shut out again.

Despite it all, the game’s obvious greatness managed to draw fans, particularly in the 2010s.

The underground forum RPG Codex ran a popular “Let’s Play” of Azrael’s Tear in 2012. Its host called it “quite possibly the most underrated adventure game ever,” and pointed out that it “wins the fandom of everyone that plays it to the end.” In the last few years, 17 dedicated users have generated a 900+ page Wikia about Azrael’s Tear — an impressive display of fandom.

Even today, Matthew Stibbe still receives fan letters about the game.

Yet it hasn’t been enough. Online services like and Steam, and companies like Night Dive, have rescued DOS classics for almost 10 years; but not Azrael’s Tear. As Iain McNeil said in response to my interview request, “I had no idea there was any interest in the game until you sent your mail!”

On, the community wishlist entry for Azrael’s Tear rests today at a respectable 225 votes. A user on that page wrote earlier this year:

Not fair. Nobody knows the game, so nobody knows they should want it.:-( What a pity.

Too obscure to be desired, too forgotten to be remembered: that is the tale of Azrael’s Tear. And there it stays.

– – – – – –

addendum: the story of this story

Or there it stayed until, I guess, right now.

The story you’ve just read is over, and thanks for reading, but there’s still another story to get through before you’re done. That’s the story of what happened because I wrote this article.

I started this process back in July, because I’m a fan of Eric’s, and because versions of this story have lingered in the back of my mind for five or six years, but the timing never felt right — and I assumed that Azrael’s Tear would be re-released eventually, anyway, so why bother?

It didn’t happen. That’s why we’re here today.

All along, my goal’s been to get Azrael’s Tear back on the market. A big feature story that attracted the attention of retrogaming enthusiasts, got them interested in the game, would maybe also attract the attention of Night Dive or Retroism or GOG and so on. For good measure, I thought, I’d get quotes from that assemblage about why Azrael’s Tear of all games was still submerged in the abandonware peat bog.

During our interview, when we talked about Night Dive, Ken Haywood told me, “I understand they’re already open to Azrael’s Tear.” I didn’t know what that meant, but it sounded promising, so I emailed them.

It turns out that Night Dive has wanted Azrael’s Tear, but hasn’t been able to get it. Almost two weeks after I sent in, by which point I figured a reply was ruled out, Stephen (Kick, company founder) found my email and was kind enough to message me back.

Night Dive first tried in November 2015, apparently. They emailed Matthew Stibbe, and he sent this:

I don’t own the rights, which are divided up between all kinds of different people, including the legatees of the old Mindscape company, whoever they are now. I just think it’s an impossible task.

Their attempts to reach Ken ended in failure. I had a hard time finding Ken myself — sending a general message through Clik Creative’s company form, and getting lucky he saw it, ended up being the only option.

Ken, as it happens, does own the rights to Azrael’s Tear, at least in part. And he’s very interested in seeing it brought back. “My wish would be to see a modern remake done, and perhaps some recognition at last for the excellent work the whole team put into it,” he tells me.

So, Ken asked me to put him in touch with Stephen, and Stephen asked me to put him in touch with Ken. I did. They’ve done the digging, they know where the rights landed and they’re in talks now. Night Dive adjudicated the System Shock 2 rights debacle — once considered impossible — and brought that game back out. They’re legends at this.

Ken said that “it seems likely we’ll be going ahead with a revived version of Azrael’s Tear” back in September. That was incredible to hear — but, when I followed up with Stephen about whether the deal was guaranteed, he said, “I’m not quite sure yet!”

That wasn’t the end.

On November 13, Ken said he was “pretty hopeful it will come off soon.” He seemed more-or-less convinced it was going to happen. Stephen tweeted something even more certain: “it’ll be out in the near future.” Ken asked that I write about the revival in this article.

So, it’s happening. After all these years, it’s happening. Set your watches: the crevice into Aeternis is reopening. Azrael’s Tear is coming back from the abyss.

– – – – – –


Thanks to Mark J. P. Wolf’s book Myst and Riven: The World of D’ni for the details about Myst’s success. The Doug Church quotation about Underworld comes from Richard Rouse III’s book Game Design Theory & Practice Second Edition. That book also provides the sales figures for Superhero League of Hoboken. Richard Guy’s remarks, along with lots of other details about the history of Azrael’s Tear, are taken from an informal comment-section interview that occurred in 2012. Connor Eck’s wide-ranging talk with Matthew Stibbe in Abandoned Times #3 (2015) was another invaluable resource. Other material courtesy of personal interviews, Raptor‘s pitch and design documents, MobyGames, Google and more.

Special thanks to Ken Haywood, Neil Crosbourne, Stephen Kick, Phil Veale, Iain McNeil, Richard Evans and Martin Fermor for working with me to make this article a reality. And major thanks to Eric for putting up with everything, especially the delays, that happened along the road. This article would’ve been impossible with any other editor, for any other publication.


Bob Thornbrook is a retrogaming fan from way back. This is his first piece for Skirmish Frogs.

What do you think of this post?
  • Hop! 

About NintendoLegend

Eric Bailey is Top Frog of He also blogs at You can follow him on Twitter, @Nintendo_Legend.


  1. Bob, this is a superb piece of writing – and a bittersweet story.
    I hope there is a chance we will see the game revived and back on player’s screens soon.

    • Thanks so much! That’s my hope as well. After this article was completed, I was informed by Ken Haywood that he’d officially signed a contract with Night Dive Studios, so I think it’s going to happen soon.

      • I’m so glad I found this article. I played this when it came out and loved it. At the time I had received a brand new Acer with a Pentium 166mhz chip as a graduation gift and it played great. I remember I kept waiting for a patch that never came that would support my 3dfx. I never finished it because of this, I thought, this game is great but how awesome will it be accelerated!?! I am 100% certain I still have the disk and instructions in storage. I kept all my games from that time you also mentioned Super Hero League of Hoboken….I have that on 3.5 floppy….memories.

  2. What a really interesting article! I’m glad it’s coming back for another go. This is the sort of game I would have absolutely devoured at the time, but i’d never seen it before clicking on this article. It’s a real shame it didn’t get the sort of marketing push it deserved.

    • Thank you! It really is an underappreciated gem. Here’s hoping that Night Dive has success bringing it to a new generation.

  3. Wow! That’s a very thorough article and it brought back lots of memories. I wish Ken and the guys every success in rebooting Azrael’s Tear.

  4. I’ve loved AT since it first came out…can’t wait for it to show up in a new version, and a possible re-release of the original game. It’s a travesty the game was lost for so long.

  5. Wonderful! As a young mom I loved this game. The music and the scenery use to really kind of creep me out, especially when I would play it at night. This game deserves a rebirth. I too have googled AT over the years hoping for another version. Thank you for the update and your role in getting the right people back in touch!

  6. I found AT on shelf at Bookman’s in Tuscon, AZ, around 97-98 and have cherished it since. It was easy play at first(System requirements are primitive by today’s standards), but today it very hard to play it. I use Dosbox, but are there other emulators I could use?

  7. I have thought about this game maybe twice a month for the last 21 years. It is truly a eerie breathtaking game. I read Charlie brooker PC gamer reviews on it. Outside of that review and the fact that I played it it seemed like it never existed. I could never talk about the game to anyone. I did have it on an emulator and looking back at the game I’m astounded by the voice acting. Creeped by the howling wind which I assume to be at the mouth of the exit that took me several years to get to. This game had no walkthrough the internet was dial up. After learning about computers and doing wireframe meshes and understanding polygon counts and sticking textures to a surface model I’m blown away by what I saw and the whole thing was in DOS. The enormity of the task to be that cutting edge in 1996 really set this game apart technically as well as the amazing storyline.
    Upon replaying it years later I was spoiled by graphics and smooth movements of games today with much more understood and known game engines and wished it could respond as such before reminding myself this game was way ahead of its time when it came out. Seeing people talking about this game is heartwarming good games with originality and effort should be canonised not forgotten.

Leave a Reply

Your email address will not be published. Required fields are marked *