RSS

Category Archives: Digital Antiquaria

New Tricks for an Old Z-Machine, Part 2: Hacking Deeper (or, Follies of Graham Nelson’s Youth)

Earlier this year, I reached out to Graham Nelson, the most important single technical architect of interactive fiction’s last three decades, to open a dialog about his early life and work. I was rewarded with a rich and enjoyable correspondence. But when the time came to write this article based on it, I found myself on the horns of a dilemma. The problem was not, as it too often is, that I lacked for material to flesh out his personal story. It was rather that Graham had told his own story so well that I didn’t know what I could possibly add to it. I saw little point in paraphrasing what Graham wrote in my own words, trampling all over his spry English irony with my clumsy Americanisms. In the end, I decided not to try.

So, today I present to you Graham Nelson’s story, told as only he can tell it. It’s a rare treat given that Graham is, like so many people of real accomplishment, usually reluctant to speak at any length about himself. I’ll just offer a couple of contextual notes before he begins. The “Inform” to which Graham eventually refers is a specialized text-adventure programming language by that name targeting the Z-Machine (and much later a newer virtual machine known as Glulx which has finally come to supersede Infocom’s venerable creation); Inform has been the most popular tool of its type through the last quarter-century. And Curses is the first full-fledged game ever written with Inform, a puzzly yet eminently literary time-traveling epic which took the huddled, beleaguered text-adventure diehards by storm upon its release in 1993, giving them new hope for their beloved form’s future and inspiring many of them to think of making their own games — using Inform more often than not. In the third and final article of this series on the roots of the Interactive Fiction Renaissance, I’ll examine both of these seminal artifacts in depth with the detachment of a third party, trying to place them in their proper historical context for you. For today, though, I give you Graham Nelson unfiltered to tell you his story of how they — and he — came to be…


Great Baddow, the quiet Essex village where Graham Nelson grew up.

I was born in 1968, so I’m coeval with The White Album and Apollo 8. I was born in Chelmsford, in Essex, and grew up mainly in Great Baddow, a quiet suburban village. There were arable farms on one side, where in those days the stubble of the wheat would still be burned off once a year. (In fact, I see that the Wikipedia page for “stubble burning” features a photo from the flat countryside of Essex, taken in 1986. The practice is banned now.) My street, Hollywood Close, had been built in the early 1960s on what used to be Rothman’s Farm. The last trees were still being cut down when I was young, though that was mainly because of Dutch Elm Disease. The houses having been sold all at once, to young families of a similar age, my street was full of seven-year olds when I was seven, and full of fifteen-year olds when I was fifteen. I went to local schools, never more than walking distance away. My primary school, Rothman’s Junior, was built on another field of the same farm, in fact.

My father Peter was an electronics engineer at English Electric Valve. My mother Christine — always “Chris” — was a clerical civil servant before she had me, at the National Assistance Board, which we would call social security today. In those days, women left work when they had a child, which is exactly what she did when she had me and my brother. But later on she trained as a personal assistant, learning Pitman shorthand, which I never picked up, and also typing, which I sort of did: I am a two-fingered typist to this day, but unusually fast at it. I did try the proper technique, but on our home typewriter, my little finger just wasn’t strong enough to strike an “A”. Or perhaps I saw no reason to learn how other people did things.

My parents had met in school in Gosport, a naval village opposite Portsmouth, on the south coast of England. As a result, both sides of my family were in the same town; indeed, we were the eccentric ones, having moved away to Essex. My many aunts, uncles, second cousins, and so on were almost all still in Portsmouth, and we would stay there for every holiday or school break. In effect, it was a second home. Though I didn’t know him for long, a formative influence was my mother’s father Albert, a navy regular who became a postman in civilian life. He was ship’s cook on HMS Belfast during the Second World War; my one successful poem (in the sense of being reprinted, which is the acid test for poems) is in his memory.

None of these people had any higher education at all. I would be the first to go to a university, though my father did the correspondence-course Open University degree in the 1970s, and my mother went to any number of evening classes. (She ended up with a ridiculous number of O-levels, rather the way that some Scouts go on collecting badges until their arms are completely covered.) They both came from genuinely poor backgrounds, where you grew a lot of your own food, and had to make and mend. You didn’t buy books, you borrowed them from the library — though my grandmother did have the Pears Cyclopaedia for 1938 and a dictionary for crosswords. But I didn’t grow up in any way that could be called deprived. My father made a solid middle-class income at a time when that could keep a family of four in a house of their own and run a car. He wasn’t a top-bracket professional, able to sign passport applications as a character reference, like a doctor or a lawyer, but he was definitely white-collar staff, not blue-collar. Yes, he worked in a factory, but in the R&D lab at one end. This is not a Bruce Springsteen song. He would not have known what to do with a six pack of beer.

My brother Toby, who later became a professional computer programmer working at Electronic Arts and other places, was two years younger than me, which meant he passed through school with teachers expecting him to be like me, which he both is and isn’t. He’s my only sibling, though I now also have a brother-in-law and sister-in-law. “Graham” and “Toby” are both definitely unusual names in England in our generation, which is the sort of thing that annoys you as a child, but is then usefully distinctive in later life. At least “Graham” is unabbreviable, for which I have always been grateful.

The local education authority would have expected me to pass the eleven-plus exam, and move up the social ladder to King Edward VI Grammar School, the best in the area by far. But my parents, who believed in universal education, chose not to enter me. So at eleven and a half, I began at Great Baddow Comprehensive School. I didn’t regret this then, and don’t now. I had some fine teachers, and though I was an oddity there, I would have been an oddity anywhere. Besides, I had plenty of friends; it wasn’t the social snake-pit which American high schools always seem to be on television.

Until around 1980, there were no commercial home computers in the UK, which was consistently a couple of years behind the United States in that respect. But my father Peter was also an electronics hobbyist. Practical Electronics magazine tended to be around the house, and even American magazines like Byte, on occasion; I had a copy of the legendary Smalltalk number of Byte, with its famous hot-air-balloon cover. But the gap between these magazines — and the book in my school library about Unix — and reality was enormous. All we had in the house was a breadboard and some TTL chips. Remarkably, my father nevertheless built a computer the size of a typewriter. It had no persistent storage; you had to key in opcodes in hex with a numeric keypad. But it worked. It was a mechanism with no moving parts. It’s hard to explain now how almost alchemical that seemed. He would give a little my-team-has-won-again cheer from his armchair whenever the BBC show Tomorrow’s World used the words “integrated circuits”. (I think this was a little before the term “microchips” came into common usage, or possibly the BBC simply thought it a vulgar colloquialism. They were more old-school back then.)

Until I was twelve years old, then, computing was something done on mainframes – or at any rate “minis” like the DEC VAX, running payroll for medium-sized companies. Schools never had these, or anything else for that matter. In the ordinary way of things, I would never have seen or touched a real computer. But I did, on just a few tantalising occasions.

Great Baddow was not really a tech town, but it was where Marconi had set up, and so there were avionics businesses, such as the one my father worked for, English Electric Valve. Because of that, a rising industry figure named Ian Young lived in our street. His two boys were just about the same age as me and my brother, and he and his wife Gill were good friends of my parents — I caught up with them at my parents’ sixtieth wedding anniversary only a few weeks ago. Ian soon relocated to Reading as an executive climbing the ranks of Digital Equipment Corporation, then the world’s number two computer company after IBM, but our families kept in touch. A couple of times each year my brother and I would go off to spend a week with the Youngs during the school holidays. This is beginning to sound like a Narnia book, and in a way it was a little like that. Ian would sportingly take us four boys to DEC’s headquarters — in particular, to the darkened rooms where the programmers worked, in an industrial space shared with a biscuit factory. (Another fun thing about the Youngs was that they always had plenty of chocolate-coated Club biscuits from factory surplus.) We would sit at a VT-220 terminal with a fluorescent green screen and play the DECUS user group’s collection of games for the VAX. These were entirely textual, though a few, like chess or Star Trek, rendered a board using ASCII art. Most of these games were flimsy nothings: a boxing simulator, I remember, a Towers of Hanoi demo, and so on. But the exception was Crowther and Woods’s Adventure, which I played less than a year after Don Woods’s canonical first version was circulated by DECUS. Adventure was like nothing else, and had a depth and an ability to entrance which is hard to overstate. There was no such thing as saving the game — or if there was, we didn’t know about it. We simply remembered that you had to unlock the grating, and that the rusty iron rod would… and so on. Our sessions almost invariably ended in one of the two unforgiving mazes. But that was somehow not an unsatisfying thing. It seemed like something you were exploring, not something you were trying to win.

It was, of course, maddening to be hooked on a game you could play perhaps once every six months. I got my first actual computer in 1980, for my twelfth birthday: an Acorn Atom. I had the circuit diagram on my wall; it was the first and last computer I’ve ever owned which I understood the physical workings of. My father assembled it from the kit form. This was £50 cheaper — not a trivial sum in those days — and was also rather satisfying for him, both because it was a lovely bit of craftsmanship to put together (involving two weekends of non-stop soldering), and also because he was never such a hero to his son as when we finally plugged it in and it worked flawlessly. Curious how much of this story appears to be about fathers and sons…

At any rate, I began thinking about implementing “adventures” very early on. This was close to impossible on a computer with 12 K of RAM (and even that only after I slowly expanded it, buying 0.5 K memory chips one at a time from a local hardware store). And yet… I can still remember the epiphany when I realised that you could model the location of an object by storing this in a byte which was either a room number or a special value to mean “being carried”. I think the most feasible creation I came up with was a procedurally-generated game on a squared grid, ten rooms wide by infinity rooms long, where certain rooms were overridden with names and puzzles. It had no title, but was known in my family as “the adventure of Igneous the Dwarf”, after its only real character. My first published game was an imitation of the arcade game Frogger for the Acorn Atom. I made something like £70 in royalties from it, but it really had no interactive-fiction content of any kind.

My first experience of commercial interactive fiction came for the BBC Micro, the big brother of the Acorn Atom; my father being my big brother in this instance, since he bought one in 1981. The Scott Adams line made it onto the BBC Micro, and so did ports of the Cambridge mainframe games, marketed first by Acornsoft and then by Topologika. I thus played some of the canonical Cambridge games quite a while before going to Cambridge. (Cambridge was then the lodestone of the UK computing industry; things like the BBC Micro and the ARM chip are easily overlooked in Cambridge’s history, given the university’s work with gravity, evolution, the electron, etc., but this was not a small deal at the time.) In particular, the most ambitious of the Cambridge games, Acheton, came out from Acornsoft on a disk release, and I played it. This was an extraordinary thing; in the United Kingdom, few computer owners had disk drives, and no more than a handful of BBC Micro games were ever released in that format.

I made something fractionally like a graphical adventure, called Crystal Castle, for the BBC Micro. (In 2000, Toby helpfully, if that’s the word, found the last existing cassette tape of this, digitised it to a WAV file, signal-processed the result, and ended up with about 22 K of program and data. To our astonishment, it ran.) It was written in binary machine code, which thus had no source code. Crystal Castle was nearly published, but the deal ultimately fell through. Superior Software, then the best marque for BBC Micro stuff, exchanged friendly letters with me, and for a while it really did look like it would happen. But I really needed an artist, and a bit more design skill. So, they passed. I imagine they had quite a large slush pile of games on cassette sent in by aspiring coders back then. You should not think of me as a teenage entrepreneur; I was mostly unsuccessful.

I did get two BBC Micro games published in 1984 by a cottage-industry sort of software house somewhere in Essex, run by a local teacher. Anybody who could arrange to duplicate cassette tapes and print inlay cards could be a “software house” in those days, and quite a lot of firms with improvised names (“Aardvark Software”, etc.) were actually people running a mail-order business out of their front rooms. They sold my two games as one, in that they were side A and side B of the same cassette. The games had the somewhat Asimovian names Galaxy’s Edge and Escape from Solaris. I honestly remember little about them, except that Escape from Solaris was a two-handed game. To play, you had to connect two BBC Micros back-to-back with an RS-423 cable, and then you had to type alternate commands. One program would stall while the other was active, but the thing worked. I cannot imagine that these games were any good, but the milieu was that of alien science being indistinguishable from magic. The role-playing game Traveller may have been an influence, I suppose, but my local library had also stocked a great deal of golden-age science fiction, and I had read every last dreg of it. (I hadn’t, at that time, played Starcross, though I’d probably seen Level 9’s Snowball.) I do not still have copies, and I am therefore spared the moral dilemma of whether I should make them publicly available. I did get a piece of fan mail, I remember, by someone who asked if I was a chemist. From this memory, I infer that there were some science-based puzzles.

The Quill-written games weren’t any influence on me, nor really the Magnetic Scrolls ones. The Quill was a ZX Spectrum phenomenon — and the Spectrum came from Acorn’s arch-enemy Sinclair. I think my father regarded it as unsound. It certainly did not have a keyboard designed to the requirement that it survive having a cup of coffee poured through it, as the BBC Micro did. But it did have an enormous amount of RAM — or rather, it didn’t consume all of that precious RAM on screen memory. The way that it avoided this was a distasteful hack, but also a stroke of genius, making the Spectrum a perfect games machine. As a result, those of my friends whose fathers knew anything about computers had BBC Micros, and the rest had Spectrums. It is somehow very English of us to have invented a new class distinction in the 1980s, but I rather think we did. Magnetic Scrolls were a different case, since they were adopting an Infocom-like strategy of releasing for multiple platforms, but they came along later, and always seemed to me to be more style than substance. The Pawn was heavily promoted, but I didn’t care for it.

I really must mention Level 9, though. They wrote 200-room cave adventures – albeit sometimes the cave was a starship – and by dint of some ingenious compression were able to get them out on tape. In particular, I played through to completion all three of the original Level 9 fantasy trilogy: the first being an extended version of the Crowther and Woods Adventure, the second and third being new but in the same style. I still think these good, in some relative sense. Level 9’s version of the Crowther and Woods Adventure, Colossal Adventure, was the first version which I fully explored, so that it still half seems to me like the definitive version. Ironically, none of Level 9’s games had levels in the normal gaming sense.

I didn’t play any of Infocom’s games until, I think, 1987. I bought a handful, one at a time, from Harrod’s in Knightsbridge — a department store for the rich and, it would like to imagine, the socially elite. I was neither of those things, but I knew what I wanted. Infocom’s wares were luxury goods, and luxury goods tend to stay on the shelves until they sell. Harrod’s had a modest stock, which almost nobody else in the UK did, though you could find a handful of early Infocom titles such as Suspended for the Commodore 64 if you trawled the more plebeian electronics shops of Tottenham Court Road. The ones I bought were CP/M editions of some of the classic titles of 1983 to 1985: Enchanter, I remember, being the first. These we were able to run on my brother’s computer, which was an Amstrad, a British machine built for word processing, but which — thanks to the cheapness of Alan Sugar, Amstrad’s proprietor, a sort of British version of Commodore’s Jack Tramiel — ran CP/M rather than MS-DOS.

That was just after I had begun as an undergraduate at Cambridge and joined the mainframe there, Phoenix, as a user. Each user had an allocation of “shares”, which governed how much computing time you could have. As the newest kid to arrive, I had ten shares. There were legends of a man in computational chemistry, modelling the Schrödinger equation for polythene, who had something like 10,000. At any rate, ten shares was only just enough to read your email in daytime. To run anything like Dungeon, the IBM port of Zork, you had to sit up at night — which we did, a little. I think Dungeon was the only externally-written game playable on Phoenix; the others were all homegrown, using TSAL, the game assembler written by David Seal and Jonathan Thackray. As I wrote long ago, to me and others who played them them those games “are as redolent of late nights in the User Area as the soapy taste of Nestlé’s vending-machine chocolate or floppy, rapidly-yellowing line printer paper.” As I noted earlier, most of them ultimately migrated to Acornsoft and Topologika releases.

But there were other social aspects to Phoenix as well. There was a rudimentary bulletin board called GROGGS (the “General Reverse-Ordered Gossip-Gathering System”) and it was tacitly encouraged by the Phoenix administrators because it stopped people abusing the Suggest program as a noticeboard. (We did not then have access to Usenet.) GROGGS was unusually egalitarian — students and faculty somewhat mingled, which was not typical of Cambridge then. Its undoubted king was Jonathan Partington (JRP1), a young professor who had a generous, playful wit. The Phoenix administrators dreaded his parodies of their official announcements. In his presence, GROGGS was a little like the salon in which the hangers-on of Oscar Wilde would attempt to keep up. Numerous people had a schtick; mine was to mutate my user-name to some version of the Prufrockian “I am not Prince Hamlet”. Commenting on the new Dire Straits album, I would post as “I am not Mark Knopfler”. That sort of thing. Jonathan wrote some of the Cambridge mainframe games. He taught me for a few second-year options.

There was also a form of direct messaging, the “notify” command, and you had the ability to link your filespace to somebody else’s, in effect giving them shared access. At some point Mark Owen and Matthew Richards, inseparable friends at Trinity College, observed that these links turned the users of Phoenix into a directed graph — what we would now call a social network. Mark and Matthew converted the whole mainframe into a sort of adventure game on this basis, in which user filespaces were the rooms, and links were map connections between them. You could store a little text file in your filespace as your own room description. Mark and Matthew’s system was called MEGA, a name chosen as an anagram of GAME. Mark went on to take a PhD in neural networks, back in the days when they didn’t work and were considered a dead end; he eventually wrote a book on signal processing. Matthew, a gifted algebraist and one of the nicest people I have ever known, died of Hodgkin’s disease only a couple of years into his own PhD — the first shock of death close up that most of us had known. The doctors tried everything to keep him alive. There’s no length they won’t go to with a young, strong patient, however cruel.

At any rate, back in the days of MEGA, it occurred to me that more could be done. Rather than storing just a single room description, each user could store a larger blob of content, and we would then have a form of MUD. This system, jointly coded by myself and a CS student called John Croft, was called TERA (I forget why we didn’t go up from MEGA to GIGA — perhaps there already was one?) and its compiler was “teraform”. This is the origin of the “-form” suffix in Inform’s name.

Cambridge mathematics degrees were in four parts: IA, IB, II, and III. Part III was an optional fourth year, which now earns you a master’s, but which for arcane funding reasons didn’t in my day. The Part III people were the aspiring professionals, hoping for a PhD grant at the end of it. Only seven or eight were available, which lent a competitive edge to a social group which was all too competitive already. I was thoroughly settled in Cambridge, living in an old Victorian house off Trumpington Street with four close friends, down by the river meadows. It was a very happy time in my life, and I had absolutely no intention of giving it up. As a geometer, I was hoping to be a research student of Frank Adams, a legendary topologist but a man with an awkward, stand-offish character. I’m now rather glad that this didn’t happen, though I’m sorry about the reason, which was that he died in a car crash. The only possible alternative, the affable Ray Lickorish, was just going on sabbatical. And so I found myself obliged to apply to Oxford instead. I was very fortunate to become the student of Simon Donaldson, only the fifth British mathematician to win the Fields Medal. (He is warmly remembered at St Anne’s College, where I now am, not for the Fields, or the Crafoord Prize, or for being knighted, or winning a $3 million award — not for any of that, but for having been a good Nursery Fellow, looking after the college crèche.) Having opened up a new and, almost at once, a rapidly-moving field of study, Simon was over-extended with collaborators, and I wasn’t often a good use of his time. Picture me as one of those plodding Viennese students Beethoven was obliged to give piano lessons to. But it was a privilege even to be present at an important moment in the history of modern geometry, and in his quietly kind way, Simon was an inspirational leader.

So, although I did find myself a doctoral perch, I had time on my hands — not work time, as I had plenty to do on that front, but social time, since everyone I knew was back in Cambridge. I read a great many books, buying up remaindered Faber literary paperbacks from the Henry Pordes bookshop in Charing Cross Road, London, whenever I was passing through. The plays of Tom Stoppard, Alan Bennett, David Hare; the poems of Philip Larkin, Seamus Heaney, Auden, Eliot, and so forth. I wrote a novel, which had to do with two people who worked in a research lab doing unethical things attempting to control chimpanzees. He took the work at face value, she didn’t, or perhaps it was the other way around. By the time I finished, I knew enough to know that it wasn’t any good, but in so far as you become a writer simply by writing, I had become a writer. I then wrote four short stories, and a one-act play called A Church by Daylight (a title which is a tag borrowed from Much Ado About Nothing). This play was thin on plot but had to do with loss. I wasn’t much good at dialogue, and in some way I boiled the play down to its essence, which was eventually published as a twelve-line poem called “Requiem”.

It was during my second year as a DPhil student that The Lost Treasures of Infocom came out. At this time my computer was an Acorn Archimedes with a 20 MB hard drive. I bought the MS-DOS box because I could read the story files from the MS-DOS disks, even if I couldn’t run the MS-DOS interpreter. I had no modem or network access from my house, and could only get files on or off by taking a floppy disk to the computing-service building right across town. I used the InfoTaskforce interpreter to actually play the games on my Archimedes.

So, I would say that the existence of a community-written interpreter was an essential precondition for Inform. In the period from 1990 to 1992, there were two significant Infocom-archaeology projects going on independently, though they were certainly aware of each other: the InfoTaskforce interpreter, and a disassembler called “txd” by Mark Howell. The InfoTaskforce people were based in Australia, and I had no contact with them, but I saw their code. Mark, however, I did exchange emails with. I remember emailing him to ask if anyone had written an assembler to make new games for the Z-Machine, and he replied with some wording close to: “Many people have had many dreams”. I set myself the task of faking a story file just well enough to allow it to execute on the InfoTaskforce interpreter.

I recall that my first self-made story file computed a prime factorisation and then printed the result. Except that it didn’t. I would double-click on the story file, and nothing would happen. I would assume that this was because there was some further table in the story file which I needed to fake: that the interpreter was refusing my file because it lacked this table, let’s say. As a result, I got into a cycle of making more and more elaborate fakes, always with negative results. Eventually I found that these faux story files had been correct all along; it was just that the user interface for the Acorn Archimedes port of the InfoTaskforce interpreter displayed nothing onscreen until the first moment when a game’s output hit the bottom of its virtual display and caused a scroll event. My story files, uniquely in the history of the Z-Machine, simply printed a few lines and then quit. They didn’t produce enough output to scroll, so nothing ever showed up onscreen. (This is why, for several years, the first thing that an Inform-written game did was to print a run of newlines.) So, when I finally managed to make a story file which factorised the numbers 2 to 100, and found that it worked correctly, I had a fairly elaborate assembler. This was called “zass”, and eventually became Inform 1.

The project might have gone no further except for the arrival of Usenet and the rec.arts.int-fiction newsgroup. Suddenly my email address was one which people could contact, and my posts were replied to. I was no longer on GROGGS, talking to a handful of people I knew in real life; I was on Usenet, talking to those I would likely never meet. People didn’t really use Inform much until around Inform 3, but still, there was feedback. An appetite seemed to exist.

A curious echo of the fascination the Z-machine held is that a couple of tiny story files produced by me in the course of these experiments — I remember one with two rooms in it and a few sample objects, one of them a football — themselves started to be collected by people. Of course there were soon to be lots of story files, an unending supply of them. But for just a brief period, even the output of Inform had a sort of second-hand glory reflected onto it.

Inform 1 was the result of my experiments to synthesise a story file, so it preceded Curses; it’s not that I set out to create both. Still, I did once write that Inform and Curses were Siamese twins, though the expression makes me flinch now. It’s not a comedic thing to be born conjoined. That aside, was it true, or did it simply sound clever? It’s true in part. I steadily improved Inform as I was building up Curses in size, and Curses undeniably played a role as a proof of concept. Numerous half-finished interactive-fiction systems had been abandoned with no notable games to their credit, but TADS, especially, shone by having been used for full-scale works. Yet this linkage is only part of the story.

In retrospect, the decision to write Curses fits with the pattern of imitation which you tend to find in the juvenilia of writers. I had read some novels, I wrote a novel; I had read some plays, I wrote a play; and so on. Lost Treasures may have played the same role for me, in computer-game terms, that those 1980s Faber & Faber paperbacks played in literary terms. But I also wrote Curses as an entertainment for my friends back in Cambridge, who attacked it without mercy. A very early version caused hilarity not so much for its intrinsic qualities as because the command “unlock fish” crashed it right out.

The title alludes to the recurring ancestral curses of the Meldrew family, each generation doomed never quite to achieve anything. (Read into that what you will, but it caused my father to raise an amused eyebrow.) The name was actually a hindrance for a while. In the days of Archie and Veronica and other pre-Web systems for searching FTP sites, “curses” was a name already taken by the software library for text windows on Unix.

What is Curses about? A few years ago Emily Short and I were interviewed, one after another, at the Seattle Museum of Pop Culture. Emily described Curses as being about the richness of culture and the excitement of discovering it. This may be an overly generous verdict, but I see what she means. Curses has a kind of exuberance to it. The ferment of what I was reading infuses the game, and although most people saw it as a faithful homage to Infocom, it was also a work of Modernism, assembled from the juxtaposed fragments of other texts. At Meldrew Hall, I could connect everything with everything.

There were four main strands here. Most apparent is the many-volume Oxford History of England, an old-school reference work, which lined up on my shelf in pale blue dust jackets. I had collected them by scouring second-hand book shops with the same assiduity as a kid completing an album of football stickers. Something of each went into Curses, from Roman England (Vol. I) through to society paintings by Sir Joshua Reynolds, and so on. The second strand was Eliot and The Waste Land, not solely for its content but also for its permissive style, as if it had authorised me to throw everything together. The third strand was classics: I was reading a lot of those “Cambridge Companion to Ancient Greek Philosophy” type of books, and I liked to grab the picturesque parts. Lastly, of course, the fourth strand is Infocom. Some of the puzzle design is lovingly imitative of Lebling, especially. The hieroglyphics from Infidel make a direct appearance. I also took affectionate swipes at the conventions, as with the infamous “You have missed the point entirely” death incurred simply by going down from the opening room, or the part where the narrator awards some points and then, a few turns later, takes them back again. Or the devil, who gives hints, all of which are lies. People actually filed bug reports over that. But really, I don’t think I did anything so transgressive that Infocom might not have done the same itself.

Those four strands are the main ingredients, but I should also acknowledge the indirect influence of the 1980s turn towards magical realism in fantasy novels, where it became possible to marry the fantastical with the merely historical. I had certainly read John Crowley’s Little, Big, for example. You could, at a stretch, say that Curses lies in the same genre.

The art of the Modernist collage is to somehow provide some cement which will hold the whole thing together. In the case of Curses, that cement is provided by the continuity of the Meldrew family and of the house – to which, and this is crucial, the player is always returning, and which ramifies with endless secret rooms. Moreover, you always experience the house through its behind-the-scenes places, joined in a skeletal way around the public areas which you never get to visit. The game is at its best when this cement is strongest, with the puzzles directly related to family members or to the house’s nooks and crannies. It loses coherence when it goes further afield, and this is why a final proposed addition, to do with the subway systems of various world cities all being joined up, was dropped. It didn’t feel like Curses any more. The weakest parts of Curses are the last parts added, and I suspect that the penultimate release is probably a better experience than the final one.

I am sometimes asked if Curses was autobiographical. As the above makes clear, in one sense yes, in that it’s a logbook of my reading. And in another obvious sense, no: I never actually teleported to ancient Alexandria. Nor have I ever lived in a grand house. My family home was built around 1960. It had seven rooms, none of them secret, and its map was an acyclic graph. There were early players who imagined that I might really be from some cadet branch of the landed gentry, with spacious grounds out of my window. This was not the case. Our estate consisted of one apple tree and two gooseberry bushes. All the same, England is not like America in this respect. Because of the Second World War, and because of inheritance tax, the great stately homes of England had essentially all become public places by the time I was a child. A routine way to entertain visiting grandparents was to take them around, say, the Jacobean manor house at Hatfield, where the Cecils had lived since the reign of James I. You didn’t have to be at all rich to do this.

The Attic area of Curses, where the game begins, does also contain just a little of my real family. The most intriguing place in my childhood home was, for sure, the attic, because it was so seldom accessible to me: a windowless but large space, properly floored, but never converted into a living area. My father would develop photographs up there, pouring chemicals into a tray, under a red lamp with a pull-cord switch. He would allow me to pull this cord. The house also had an airing cupboard — that is, a space around the hot-water boiler where towels could be dried. In this cupboard, my mother at one time made home-brew wine, in a sort of slow chemistry experiment with evil-looking demijohns. My brother doesn’t really make an appearance in Curses, which I’m sad about now, but it’s essential that the protagonist has ancestors rather than contemporaries. Though the protagonist has a spouse and children, mentioned right up front, they never appear, which I think is worth noting in a game where almost everything else that is foreshadowed eventually comes to pass.

Curses is by any reasonable standard too hard. In its first releases, I would update it with new material each time I made bug fixes, so that the game evolved and grew. Some players would play each version as it came out, and this enabled them to get further in, because they had prior experience from earlier builds. A dedicated fan base sent in bug reports, my favourite being that the brass key could not be picked up by the robot mouse, because brass is non-magnetic. The reward for any bug reported was that the reporter could nominate a new song to be added to the radio’s playlist, provided that it was both catchy and objectively dreadful. It would be interesting to extract that playlist now and put it on Spotify.

Feedback from players gave Curses a certain polish, but it wasn’t the only thing. I think it’s noteworthy that, just as Infocom had an editor as well as play-testers, so too I had an editor for at least part of the process: Gareth Rees, a Cambridge friend, author of the very wonderful Christminster. Richard Tucker also weighed in. I have the impression that before 1992 works of interactive fiction didn’t have much quality control, not so much because people didn’t want it, but because networking conditions didn’t allow for it.

To my great regret, the source code for Curses is now lost. It was for a while on a disk promisingly labelled “Curses source code”, but that disk is unreadable, and not for want of trying. Somewhere in my many changes of address and computer, I lost the necessary tech, or damaged it. (And Jigsaw too, alas.) It wouldn’t be hard to resurrect something, by working from a disassembly of the story file: there’s actually a tool to turn story files into Inform 6 out there somewhere. I occasionally think of asking if anyone would like to do that, and perhaps produce a faithful Inform 7 implementation.

Today, people play Curses with a walkthrough by their sides. But the game never quite goes away. Mike Spivey told me recently that he introduced himself to modern interactive fiction – “modern” interactive fiction – by playing Curses in 2017. A few people, at least, still tread Meldrew Hall. I remain fond of the place, as you can probably gather from the length of this reminiscence. Once in a blue moon I am tempted to write a sequel, Curses Foiled. But no. Sometimes you really can’t go back.

 
 

Tags: , , ,

Summer Daze


I want to bring a new project by this blog’s friends Corey and Lori Ann Cole to your attention today. Summer Daze at Hero-U builds upon their earlier Hero-U: From Rogue to Redemption, which was a fine revival in spirit of their much-loved Quest for Glory games of yore. This sequel — or actually prequel — goes in a somewhat different direction, with something of a visual-novel aesthetic and markedly fewer CRPG trappings. The Coles are fine game designers as well as fine people, so I’m sure they’ll pull it off with the same aplomb that made the first Hero-U such a pleasure to play.

The economics of game-making being what they are, however, they could use your support to bring the project to fruition. As of this writing, they’re about $30,000 short of their $100,000 goal on Kickstarter, with six days left to go. Please head on over to the Kickstarter site and give the project a look, and think about pitching in if you like what you see. If I know the Coles at all, I know that they’ll undoubtedly give you your money’s worth in the finished product.

 

New Tricks for an Old Z-Machine, Part 1: Digging the Trenches

One of the most oddly inspiring stories I know of in all computing history is that of the resurrection and re-purposing of the Z-Machine, Infocom’s virtual machine of the 1980s, to serve a whole new community of interactive-fiction enthusiasts in the 1990s and well beyond. Even as the simple 8-bit computers for which it had originally been designed became obsolete, and then became veritable antiques, the Z-Machine just kept soldiering on, continuing to act as the delivery system for hundreds of brand new games that post-dated the company that had created it by years and eventually decades. The community of hobbyist practitioners who spawned the Interactive Fiction Renaissance of the mid-1990s made the Z-Machine one of their technological bedrocks for reasons more sentimental than practical: most of them worshiped Infocom, and loved the way that distributing their games via Infocom’s venerable virtual machine made them feel like the anointed heirs to that legacy. The Z-Machine was reborn, in other words, largely out of nostalgia. Very soon, though, the hobbyists’ restless creativity pushed and twisted the Z-Machine, and the genre of games it hosted, in all sorts of ways of which even Infocom at their most experimental could never have dreamed. Thus a regressive became a progressive impulse.

In the end, then, a design which Joel Berez and Marc Blank first sketched out hurriedly at their kitchen tables in 1979, in response to the urgently immediate problem of how to move their DEC PDP-10 game of Zork out of the MIT computer lab and onto microcomputers, didn’t fall out of general use as a delivery medium for new games until after 2010. And even today it still remains in active use as a legacy technology, the delivery medium for half or more of the best text adventures in the historical canon. In terms of the sheer number of platforms on which it runs, it must have a strong claim to being the most successful virtual machine in history; it runs on everything from e-readers to game consoles, from mobile phones to mainframes, from personal computers to electronic personal assistants. (To paraphrase an old joke, it really wouldn’t surprise me to learn that someone is running it on her toaster…) Its longevity is both a tribute to the fundamental soundness of its original design and to the enduring hold which Infocom’s pioneering interactive fiction of the 1980s has had upon more recent practitioners of the form. Like so many technology stories, in other words, the story of the Z-Machine is really about people.

One of the more ironic aspects of the Z-Machine story is the fact that it was never designed to be promulgated in this way. It was never intended to be a community software project; it was no Linux, no Mozilla, no Java. The ideological framework that would lead to such projects didn’t even exist apart from a handful of closeted university campuses at the time Berez and Blank were drawing it up. The Z-Machine was a closed, proprietary technology, closely guarded by Infocom during their heyday as one of their greatest competitive advantages over their rivals.

The first order of business for anyone outside of Infocom who wished to do anything with it, then, was to figure it out — because Infocom certainly wasn’t telling. This first article in a series of three is the story of those first intrepid Z-Machine archaeologists, who came to it knowing nothing and began, bit by bit, to puzzle it out. Little did they know that they were laying the foundation of an artistic movement. Graham Nelson, the most important single technical and creative architect of the Interactive Fiction Renaissance of the 1990s (and thus the eventual subject of my second and third articles), said it most cogently: “If I have hacked deeper than them, it is because I stand in their trenches.”



Although the Z-Machine was decidedly not intended as a community project, Infocom in their heyday made no particular attempt to hide the abstract fact that they were the proud possessors of some unusual technology. The early- and mid-1980s, Infocom’s commercial peak, was still the Wild West era of personal computing in the United States, with dozens of incompatible models jockeying for space on store shelves. Almost every published profile of Infocom — and there were many of them — made mention of the unique technology which somehow allowed them to write a game on a big DEC PDP-10 of the sort usually found only in universities and research laboratories, then move it onto as many as 25 normally incompatible microcomputers all at once. This was, perhaps even more so than their superb parser and general commitment to good writing and design, their secret weapon, allowing them to make games for the whole of the market, including parts of it that were served by virtually no other publishers.

So, even if highfalutin phrases like “virtual machine” weren’t yet tripping off the tongue of the average bedroom hacker, it wasn’t hard to divine what Infocom must be doing in the broad strokes. The specifics, however, were another matter. For, while Infocom didn’t hide the existence of a Z-Machine in the abstract, they had no vested interest in advertising how it worked.

The very first outsiders to begin to explore the vagaries of the Z-Machine actually had no real awareness of doing so. They were simply trying to devise ways of copying Infocom’s games — most charitably, so that they could make personal backups of them; most likely, so that they could trade them with their friends. They published their findings in organs like The Computist, an underground magazine for Apple II owners which focused mainly on defeating copy protection, hacking games, and otherwise doing things that the software publishers would prefer you didn’t. By 1984, you could learn how Infocom’s (unimpressive) copy-protection scheme worked from the magazine; by 1986, you could type in a program listing from it that would dump most of the text in a game for cheating purposes.

But plumbing the depths of a virtual machine whose very existence was only implicit was hard work, especially when one was forced to carry it out on such a basic computer as the Apple II. People tended to really dive in only when they had some compelling, practical reason. Thus users of the Apple II and other popular, well-supported platforms mostly contented themselves with fairly shallow explorations such as those just described. Users of some other platforms, however, weren’t fortunate enough to enjoy the ongoing support of the company that had made their computer and a large quantity of software on the shelves at their local computer store; they had a stronger motivation for going deeper.

Over the course of the 1980s, the American computing scene became steadily more monolithic, as an industry that had once boasted dozens of incompatible systems collapsed toward the uniformity that would mark most of the 1990s, when MS-DOS, Microsoft Windows, and (to rather a lesser extent) the Apple Macintosh would be the only viable options for anyone wishing to run the latest shrink-wrapped commercial software. This gradual change was reflected in Infocom’s product catalog. After peaking at 25 or so machines in 1984, they released their final few games in 1988 and 1989 on just four of them. The realities of the market by then were such that it just didn’t make sense to support more platforms than that.

But technical transitions like these always come with their fair share of friction. In this case, plenty of people who had been unlucky or unwise enough to purchase one of the orphaned machines were left to consider their options. Some of them gave up on computing altogether, while others sucked it up and bought another model. But some of these folks either couldn’t afford to buy something else, or had fallen hopelessly in love with their first computer, or were just too stubborn to give it up. This state of affairs led directly to the world’s first full-fledged Z-Machine interpreter to be born outside of Infocom.

The orphaned machine at the heart of this story is the Texas Instruments 99/4A, a sturdy, thoughtfully designed little computer in many respects which enjoyed a spectacular Christmas of 1982, only to be buried by Jack Tramiel under an avalanche of Commodore VIC/20s and 64s the following year. On October 28, 1983, Texas Instruments announced they were pulling out of the home-computer market entirely, thus marking the end of one of the more frantic boom-and-bust cycles in computing history. It left in its wake hundreds of thousands of people with 99/4As on their desks or in their closets — both those who had bought the machine when it was still a going proposition and many more who snatched up some of the unsold inventory which Texas Instruments dumped onto the market afterward, at street prices of $50 or less. The number of active 99/4A users would inevitably decrease sharply as time went on, but some clung to their machines like the first loves they often were, for all of the reasons cited above.

This little 99/4A fraternity would prove sufficiently loyal to the platform to support an under-the-radar commercial- software ecosystem of their own into the 1990s. For many users, the platform was appealing not least in that it never lost the homegrown charm of the very earliest days of personal computing, when every user was a programmer to one degree or another, when the magazines were full of do-it-yourself hardware projects and type-in program listings, and when one kid working from his bedroom could change the accepted best practices of everyone else almost overnight. The Z-Machine interpreter that interests us today was a reflection of this can-do spirit.

Infocom’s first taste of major success had corresponded with the 99/4A’s one great Christmas. Naturally, they had made sure their games were available on one of the hottest computers in the country. Even after Texas Instruments officially abandoned the 99/4A, there was no immediate reason to ignore its many owners. Thus Infocom continued to make versions of their games for the machine through The Hitchhiker’s Guide to the Galaxy in September of 1984. In all, they released fourteen 99/4A games.

But continuing to support any given machine eventually tended to become a more complicated proposition than simply continuing to use an already-extant interpreter. The Z-Machine in reality was more of a moving target than the abstract idea behind it might suggest. Infocom’s games got steadily bigger and richer as time went on, with more text, better parsers, and more ambitious world models. The original Z-Machine, as designed in 1979, had a theoretical maximum story-file size of 128 K, but the practical limitations of the machines running the interpreters kept the early games from reaching anything close to this size. (The original Zork, for example, Infocom’s very first game, was just 77 K.) As story files pushed ever closer to their theoretical maximum size in the years that followed, they began to exceed the practical limitations of some existing interpreters. When that happened, Infocom had to decide whether reworking the interpreter to support a larger story file was possible at all, and, if so, whether it was worth the effort in light of a platform’s sales figures. Following Hitchhiker’s (story-file size: 110 K), their fourteenth game, but before Suspect (story-file size: 120 K), their fifteenth, Infocom judged the answer to one or both of those questions to be no in the case of the 99/4A.

Barry Boone, the first person outside of Infocom to create a full-fledged Z-Machine interpreter.

As one might expect, this decision left a number of 99/4A users sorely disappointed. Among them was Barry Boone, a clever young man just out of high school who was already one of the leading lights of 99/4A hackery. Having read enough about Infocom to understand that their game format must be in some sense portable, he started doggedly digging into the details of its implementation. Soon he was able to make a clear delineation between the interpreter running natively on his machine and the story file it executed — a delineation the Apple II crowd writing for The Computist had yet to manage. And then he uncovered the big secret: that the interpreter packaged with one game could actually run the story file from another — even if said story file originated on a platform other than the 99/4A! Boone:

Having worked out the file format, I wrote a program to crunch the non-TI files and build the TI files. The resulting files appeared to work, but I quickly discovered [a] problem. If I converted an older game that already existed in TI format, everything worked perfectly. But with the newer games, there was a big problem.

The problem was that the interpreter software written for the TI had a number of bugs, many of which did not show up with the original set of games, but became all too apparent with the newer ones and made them unplayable. So I began a process of reverse-engineering the Z-Code interpreter for the TI. Once I reached a point of having recreated the source code, I began working on making the code more efficient, and fixing numerous bugs in the implementation. The largest bug I encountered was a vocabulary-table bug. Basically, the original TI interpreter would hit an overflow bug if the vocabulary table was too large, and the binary-search algorithm would start searching the wrong area of memory to look up words. This had the effect of making the last portion of the vocabulary inaccessible, and made the game impossible to play.

I also added a number of enhancements that allowed the games to load about twenty times faster, and modifications to play the games on TI systems equipped with 80-column displays. Finally, I had to make a second variation of the interpreter so that persons who had an extra 8 K of RAM (known as a Super Cart, or Super Space module) could play some of the games that required a larger memory footprint than 24 K of memory buffer. These games included Leather Goddesses.

Boone estimates that he finished his interpreter around 1986, whereupon he promptly began sharing it with his network of friends and fellow 99/4A enthusiasts, who used it to play many of the newer Infocom story files, transferring them from disks for other platforms. Boone was stymied only by the games from Infocom’s Interactive Fiction Plus line, such as A Mind Forever Voyaging and Trinity. Those games used an expanded version of the Z-Machine, known internally as version 4 — the mature version of the original virtual machine was version 3 — which expanded the available memory space to 256 K, far beyond what the 99/4A could possibly manage. Even without them, however, Boone gave himself and his mates ten new Infocom games to play — i.e., all of those released for the 128 K Z-Machine between October of 1984 and July of 1987, when this original incarnation of the virtual machine made its last bow in Infocom packaging.

But even that wasn’t quite the end of the story. An obscure footnote to Infocom’s history took shape in late 1988 or early 1989, when Chris Bobbitt, founder of a company called Asgard, the 99/4A software publisher that most resembled a real business as opposed to a hobbyist project, had the idea of contacting Infocom themselves to ask permission to market the newer games, running under Boone’s interpreter, as legitimate commercial products. Although Bobbitt doubtless didn’t realize it at the time, Infocom was by then on the verge of being shut down, and Mediagenic, their less-than-doting parent company, were also beginning to feel the financial stresses that would force them into bankruptcy in 1990. They saw Bobbitt’s proposal as a handy way to clear their warehouse of old stock and make some desperately needed cash. Jim Fetzner, who worked with Asgard at the time, remembers how the deal went down:

[Bobbitt] contacted Infocom to ask for permission to release the later Infocom releases, and was given permission to do so on one condition: that the packaging and disks had to be originals for other systems, relabeled (the packaging) and reformatted (the disks) for use with the TI. Infocom scoured their warehouse and sent Chris two very large boxes of the titles he was asking to reproduce—and noted on the invoice that these boxes included every single copy of the relevant titles that Infocom still had in their possession. Some of the titles were relatively plentiful, but others were included in much lower numbers. The boxes only contained four copies of Leather Goddesses of Phobos, for example. All other titles had at least ten copies each, and some had a lot more. He was permitted to buy more copies from remainders in the retail channels, though, so it is possible there are more properly badged Asgard copies of the titles that were harder to find. All of the stock he received from Infocom was gone in a matter of months.

These games, which Bobbitt bought for $5 apiece and sold on for several times that, thereby became the last new Infocom games ever sold in their original packaging — out-of-print games from a dead company sold to owners of an orphaned computer.

Asgard prepared their own platform-specific reference card after the Infocom example and inserted it into the box.

Well before Asgard entered the scene, however, another, more structured and sustainable project had led to a Z-Machine interpreter much more amenable to being ported and built upon than Boone’s incarnation of same for an idiosyncratic, bare-bones, orphaned platform. Not long after Boone first started sharing his 99/4A interpreter with friends, a few students at the University of Sydney in far-off Australia started disassembling another of Infocom’s own interpreters — in this case one for Zilog Z80-based computers running the operating system CP/M. The group included in their ranks David Beazley, George Janczuk, Peter Lisle, Russell Hoare, and Chris Tham. They gave themselves the rather grandiose name of the InfoTaskforce, but they initially regarded the project, said Janczuk to me recently, strictly as “a form of mental calisthenics”: “This was never meant to be a public exercise.”

Still, the group had several advantages which Boone had lacked — in addition, that is, to the advantage of sheer numbers. Boone had been a bedroom hacker working on fairly primitive hardware, where cryptic assembly language, highly specific to the computer on which it was running, was the only viable option. The InfoTaskforce, on the other hand, had more advanced hardware at their disposal, and were steeped in the culture of institutional hacking, where portable C was the most popular programming language and software was typically distributed as source code, ready to be analyzed, ported, and expanded upon by people other than its creators, quite possibly working on platforms of which said creators had never dreamed. And then, thanks to their university, the InfoTaskforce was connected to the Internet, long before most people had even heard of such a thing; this gave them a way to share their work quickly with others across a wide, international swath of computing. The contrast with the segregated ghetto that was the world of the 99/4A is telling.

David Beazley, who did almost all of the actual coding for the InfoTaskforce interpreter — the others had their hands full enough with reverse-engineering the Z-Machine architecture — did so in C on a first-generation Apple Macintosh. On May 25, 1987, he used this machine to compile the first truly portable Z-Machine interpreter in history. Within a week, he and his mates had also gotten it compiled and running on an MS-DOS machine and a big DEC VAX. (Ironically, the latter was the successor to the PDP-10 line so famously employed by Infocom themselves; thus one might say that the Z-Machine had already come full-circle.)

As Janczuk remembers it, the first version of the interpreter to reach the Internet actually did so accidentally. He gave it to a friend of his at university, who, as so many friends have done over the years, uploaded it without permission on June 2, 1987. There followed an immediate outpouring of interest from all over the world, which greatly surprised the interpreter’s own creators. It prompted them to release an official version 1.0, capable of playing any story file for the standard — i.e., 128 K — Z-Machine on August 1, 1987. Already by this time, the Commodore Amiga personal computer and several more big machines had been added to the list of confirmed-compatible host platforms. It was a milestone day in the history of interactive fiction; Infocom’s games had been freed from the tyranny of the hardware for which they’d originally shipped. And they could remain free of the vicissitudes and fashions of hardware forevermore, as long as there was an enterprising hacker ready to tweak an existing interpreter’s source code to suit the latest gadget to come down the pipe. (So far, there has been no shortage of such hackers…)

With their university days coming to an end, the InfoTaskforce boys worked on their interpreter only in fits and starts over the years that followed. Not until 1990 did they finish adding support for the Interactive Fiction Plus line; not until 1992, in a final burst of activity, did they add support for Infocom’s last few text-only games, which ran under what was known internally as the version 5 Z-Machine. This last release of the InfoTaskforce interpreter actually attracted a bit of scoffing for its inefficiency, and for generally lagging behind what other hackers had done by that point in other interpreters.

In reality, information and inspiration rather than the software itself were the most important legacies of the InfoTaskForce interpreter. Beazley’s C source told you almost all of what you really needed to know about the Z-Machine, so long as you were sufficiently motivated to dig out the information you needed; doing so was certainly a fair sight more pleasant than poring over eye-watering printouts of cryptic disassembled Z80 machine language, as Beazley and his pals had been forced to do before coming up with it. The InfoTaskforce interpreter thus became the gateway through which the Z-Machine burst into the public domain, even as Infocom was soon to collapse and abandon their virtual machine. This was a role which Boone’s interpreter, for all its naïve brilliance, just wasn’t equipped to play, for all of the reasons we’ve already explored.

An enterprising American hacker named Mark Howell did perhaps the most to build upon the foundation of the InfoTaskforce interpreter during the half-decade after its initial appearance. His own interpreter bore the name of ZIP (for “Z-Machine Implementation Program”), a name it shared with the popular compression format, to enormous confusion all the way around — although, to be fair, this was also the name by which Infocom knew their own interpreters. ZIP was faster and less buggy than the InfoTaskforce interpreter, and for this reason it soon surpassed its older sibling in popularity. But Howell also delved further into the architecture of the Z-Machine than anyone before him, analyzing its design like a computer scientist might rather than as a hacker simply trying to write a quick-and-dirty clone of Infocom’s existing interpreters. When he came up for air, he uploaded his set of “ZTools” — programs for probing story files in all sorts of ways, including a disassembler for the actual code they contained. These tools did much to set the stage for the next phase of the Z-Machine’s resurrection and liberation.

In 1992, another building block fell into place when Activision shipped their Lost Treasures of Infocom collection to unexpected success. It and its sequel collected all of the Infocom games together in one place at a reasonable price, stored as neatly discrete story files ready to be fed into either the original Infocom interpreters included on the disks or an alternative of one’s choice. Lost Treasures shipped only in versions for MS-DOS, the Apple Macintosh, and the Commodore Amiga — the last three commercially viable personal-computing platforms left in North America by that time (and the Amiga wouldn’t enjoy that status much longer). But users of orphaned and non-North American platforms were soon passing around the tip that, if you could just get the story files off of the original Lost Treasures disks, they could be run on their own platforms as well with one of the interpreters that had by now spread far and wide. For example, our old friends at The Computist, still carrying the 8-bit torch in these twilight days of the Apple II, published instructions on how to do just that — a fitting end point to their earliest explorations of the Infocom format.

Across the Atlantic, meanwhile, the magazine Acorn User published a similar article for users of the Acorn Archimedes, a machine that was virtually unknown outside of Britain, a few parts of mainland Europe, and Australasia. (“It’s hard to conceive of videogame nostalgia,” they wrote of the Lost Treasures collections, “but this is as close as it gets.” Little did they know…) It so happened that an Oxford doctoral candidate in mathematics named Graham Nelson was a stalwart Acorn loyalist and a regular reader of that magazine. By the time the article in question appeared, the window opened by the InfoTaskforce interpreter and all the software that had followed it, combined with the Lost Treasures collections, had already led him to begin sliding the next couple of building blocks of the Interactive Fiction Renaissance into place.

Infocom’s The Hitchhiker’s Guide to the Galaxy running on an Acorn Archimedes — a platform for which it was never officially released — under a third-party Z-Machine interpreter by Edouard Poor.

(Sources: The Computist 5, 7, 34, 41, 47, 57, 58, 63, and 86; Acorn User of July 1993; Asgard Software’s newsletters from 1989 and 1990. Online sources include Barry Boone’s memories of writing his Z-Machine interpreter at The Museum of Computer Adventure Game History and his bio for the TI99ers Hall of Fame. The original source for the InfoTaskforce interpreter can be found in various file archives. My huge thanks go to Barry Boone, Jim Fetzner, and George Janczuk for talking to me about their pioneering early work in Z-Machine archaeology.)

 
 

Tags:

Betrayal at Krondor

During the 1960s and 1970s, a new type of game began to appear in increasing numbers on American tabletops: the experiential game. These differed from the purely abstract board and card games of yore in that they purported to simulate a virtual world of sorts which lived behind their surface systems. The paradigm shift this entailed was such that for many players these games ceased to be games at all in the zero-sum sense. When a group came together to play Squad Leader or Dungeons & Dragons, there hung over the plebeian kitchen or basement in which they played a shared vision of the beaches of Normandy or the dungeons of Greyhawk. The games became vehicles for exploring the vagaries of history or the limits of the imagination — vehicles, in other words, for living out shared stories.

In retrospect, it was perhaps inevitable that some of the stories generated in this way would make their way out of the gaming sessions which had spawned them and find a home in more traditional, linear forms of media. And, indeed, just such things were happening by the 1980s, as the first novels born from games arrived.

Needless to say, basing your book on a game you’ve played isn’t much of a path to literary respectability. But for a certain kind of plot-focused genre novel — the kind focusing strictly on what people do rather than why they do it — prototyping the whole thing as a game makes a degree of sense. It can keep you honest by forcing your story to conform to a simulated reality that transcends the mere expediency of what might be cool and exciting to write into the next scene. By pushing against authorial fiat and the deus ex machina, it can give the whole work an internal coherency — an honesty, one might even say — that’s too often missing from novels of this stripe.

The most widely publicized early example of the phenomenon was undoubtedly the one which involved a humble insurance salesman named Tom Clancy, who came out of nowhere with a techno-thriller novel called The Hunt for Red October in 1984. The perfect book for a time of resurgent patriotism and military pride in the United States, it found a fan in no less elevated a personage than President Ronald Reagan, who declared it “my kind of yarn.” As the book topped the bestseller charts and the press rushed to draft their human-interest stories on the man who had written it, they learned that Clancy had gamed out its entire scenario, involving a rogue Soviet submarine captain who wishes to defect along with his vessel to the United States, with a friend of his named Larry Bond, using Harpoon, a tabletop wargame of modern naval combat designed by the latter. Clancy’s follow-up novel, a story of open warfare between East and West called Red Storm Rising, was a product of the same gestation process. To the literary establishment, it all seemed extremely strange and vaguely unsettling; to many a wargamer, it seemed perfectly natural.

Another line of ludic adaptations from the same period didn’t attract as much attention from the New York Times Book Review, much less the president, but nevertheless became almost as successful on its own terms. In 1983, TSR, the publisher of Dungeons & Dragons, decided to make a new series of adventure modules for the game, each of which would feature a different kind of dragon — because, as some of their customers were writing in their letters, the existing Dungeons & Dragons modules “had plenty of dungeons, but not many dragons.” The marketing exercise soon grew into Dragonlance, an elaborately plotted Tolkienesque epic set in a brand new fantasy world — one which, yes, featured plenty of dragons. TSR asked employees Margaret Weis and Tracy Hickman to write a trilogy of novels based on the fourteen Dragonlance adventure modules and source books they planned to publish. Thus Dragons of Autumn Twilight, the first volume of The Dragonlance Chronicles, was published in the same year as The Hunt for Red October. It promptly became a nerdy sensation, the biggest fantasy novel of the year, spawning a whole new business for TSR as a publisher of paperback novels. In time, said novels would become as big a part of their business as the games on which they were based.

A third, only slightly less heralded example of the games-into-books trend actually predates the two I’ve just mentioned by a couple of years. In the late 1970s, a group of students at the University of California San Diego took up the recently published Dungeons & Dragons. Growing dissatisfied with TSR’s rules, they scrapped them one by one, replacing them with their own home-grown versions. Meanwhile they evolved a world in which to play called Midkemia, complete with its own detailed history, bestiary, sociology, and geography. Forming a little company of their own, as so many Dungeons & Dragons fanatics were doing at the time, they published some of their innovations to modest sales.

Raymond E. Feist

But one of their number named Raymond E. Feist had bigger ambitions. He wrote a novel based on some of the group’s exploits in Midkemia. Calling it simply Magician, he got it published through Doubleday in 1982 as the first volume of The Riftwar Saga. It sold very well, and he’s been writing Midkemia novels ever since.

Unlike the later cases of Tom Clancy and Dragonlance, Magician wasn’t widely publicized or advertised as being the product of a game. It was seen instead as merely the latest entry in an exploding branch of genre fiction: lengthy high-fantasy series inspired by J.R.R. Tolkien, often to the point of one-to-one correspondences between characters and plot events, but written in a manner more immediately accessible to the average Middle American reader, with more action, more narrative thrust, less elevated diction, and markedly less digressive songs and poetry. Dragonlance, of course, is an example of the same breed.

I must admit that I’ve personally read only the first book of Feist’s series, and not even to completion at that. This sort of derivative high fantasy doesn’t do much for me as a rule, so I’m not the best person to judge Feist’s output under any circumstances. Anything positive I do say about it runs the risk of damning with faint praise.

To wit: my wife and I used the book as our light bedtime reading, and we made it about two-thirds of the way through before terminal ennui set in and we decided we’d had enough. If that seems like less than a ringing endorsement, know that it’s farther than I generally get with most fantasy novels, including ones with considerably more literary credibility. I thus feel comfortable in saying that at least the early Raymond E. Feist novels are well-crafted examples of their breed, if you happen to like that sort of thing. (I do understand from others that the quality of his work, and particularly of his plotting, began to decline after his first handful of Midkemia novels. Perhaps because he was no longer basing them on his gaming experiences?)

The world of Midkemia is most interesting for our purposes, however, for the computer game it spawned. Yes, a series of novels based on a game got turned back into a very different sort of game. And then, just for good measure, that game got turned into another novel. It’s a crazy old transmedia world.


The more direct origin of Betrayal at Krondor, the game in question, can be traced back to June of 1991 and a chance meeting between John Cutter and Jeff Tunnell at the Summer Consumer Electronics Show. Both names may be familiar to regular readers of these histories.

John Cutter

Cutter had spent several years with Cinemaware, helping to craft many of their most innovative creations, which blended strong narrative elements with play styles that were unorthodox in story-heavy computer games at the time. In late 1990, with Cinemaware in the process of collapsing, he and several colleagues had jumped ship to New World Computing, best known for their Might & Magic series of CRPGs. But he was trapped in a purely administrative role there, without the freedom to create which he had enjoyed at Cinemaware, and was already feeling dissatisfied by the time he met Tunnell at that Summer CES.

Jeff Tunnell

Tunnell, for his part, was the founder of the studio known as Dynamix, now a subsidiary of Sierra Online. They were best known for their 3D graphics technology and the line of vehicular simulators it enabled, but they had fingers in several other pies as well, from adventure games to a burgeoning interest in casual puzzle games.

Recognizing talent when he saw it, Tunnell asked Cutter to leave Southern California, the home of the erstwhile Cinemaware and the current New World, and come to Eugene, Oregon, the home of Dynamix. Not only would he be able to have a creative role there once again, Tunnell promised, but he would be allowed to make whatever game he wanted to. Cutter jumped at the chance.

Once in Eugene, however, he struggled to identify just the right project. His first instinct was to make a point-and-click adventure game in the Sierra mold, but Tunnell, having made three of them in the last couple of years to less than satisfying effect, was feeling burned out on the genre and its limitations, and gently steered him away from it. (Absolute creative freedom, Cutter was learning, is seldom really absolute.)

At last, Tunnell came to Cutter with an idea of his own. He’d been reading a very popular series of fantasy novels by this fellow named Raymond E. Feist, and he thought they’d make a fine CRPG. Dynamix had never dabbled in the genre before, but when had that ever stopped them from trying something new? He suggested that Cutter give the first few of the books a read. If it turned out that he liked them as well and agreed that they’d make a good game, well, perhaps he should ring Feist up and have a chat about just that possibility.

Glad to finally have a clear sense of direction, Cutter did the one thing and then did the other. Feist was very busy, but was himself a long-time computer gamer, having sat down in front of his first Apple II some twelve or thirteen years before. He liked the idea of seeing Midkemia come to life on a computer screen. Although he didn’t have much time for working personally on such a project, he told his agent to make the deal happen if at all possible. So, a contract was signed that gave Dynamix the right to make Midkemia games until January 1, 1995, with Feist given the right of final approval or rejection of each title prior to its release. By one account at least, it was the most expensive literary license yet granted to a game developer, a sign of Feist’s ongoing popularity among readers of fantasy literature.

Another, slightly less welcome sign of same followed immediately after: upon being asked whether he was interested in authoring the game himself, Feist said that his time was money, so he’d need to be paid something beyond the terms of the licensing agreement itself — and, he noted flatly, “you couldn’t afford me.” This posed a dilemma. Cutter believed himself to be a better designer of game systems than a writer, and thus certainly wasn’t going to take on the job personally. Casting about for a likely candidate, his thoughts turned to one Neal Hallford, an enthusiastic young fellow with a way with words whom he’d befriended back at New World Computing.

Neal Hallford

A fresh-out-of-university Hallford had joined New World in the role of writer some months before Cutter himself had arrived. His first assignment there had been to make sense of the poorly translated English text of Tunnels & Trolls: Crusaders of Khazan, a project New World had chosen to outsource to a Japanese developer, with underwhelming results all the way around. After that truly thankless task, he’d worked for a while on Might and Magic III before playing a pivotal role on Planet’s Edge, an ambitious science-fiction CRPG that had tried to do just a little bit too much for its own good. He was just finishing that project when his old friend John Cutter called.

Like Cutter before him, Hallford found Dynamix’s offer difficult to refuse. Eugene struck him as idyllic by contrast with the crowded, smoggy streets of Los Angeles; meanwhile Dynamix’s offices enjoyed the well-deserved reputation of being just about the most stylish and comfortable in the entire industry, vastly outdistancing even the parent company of Sierra in that respect. Certainly they compared favorably with the chaotic jumble of tightly packed cubicles that was the domain of New World. Thus on Halloween Day, 1991, Hallford shook hands with his old colleagues there for the last time and hopped into his Geo Metro for the drive north.

Upon Hallford’s arrival in Eugene, Cutter pulled him into his office and kept him there for a week, while the two hashed out exactly what game they wanted to make and wrote the outline of a script. Hallford still remembers that week of frenzied creativity as “one of the best weeks of my life.” These two friends, different in talents and personality but unified in their vision for the game, would do the vast majority of the creative heavy lifting that would go into it. Broadly stated, Cutter would be the systems guy while Hallford would be the story guy, yet their visions would prove so simpatico that they’d seldom disagree on much of anything at all.

Jeff Tunnell had initially fallen in love with a Midkemia novel called Silverthorn, and the original plan he’d pitched to Cutter had been to make the game a fairly straightforward adaptation of that book’s plot. But such a thing is inherently problematic, for reasons I’ve had ample cause to discuss in earlier articles. Players who buy the game because they read and liked the novel — who are, after all, the whole reason for making a licensed game at all from a business perspective — won’t be excited about stepping through a plot they already know. At the same time, it’s all too easy from the design side to make a game where victory hinges on taking all of the same idiosyncratic, possibly irrational actions as the protagonists of the novel. And so you end up with a game that bores one group of players to tears, even as it frustrates another group who don’t happen to know what Character A needs to do in Situation B in order to replicate the novel’s story.

The biggest appeal of the Midkemia novels, Hallford believed, was indeed the world itself, with its detailed culture and geography and its cast of dozens of well-established characters. It would be better, he thought, to set a brand new story there, one that would let Feist’s many fans meet up with old friends in familiar locales, but that wouldn’t force them to step by rote through a plot they already knew. During the crash course on Midkemia which he’d given himself in the few weeks before starting at Dynamix — like Cutter, he’d come to Feist fandom cold — Hallford had identified a twenty-year “hole” in the chronology where he and Cutter could set a new story: just after A Darkness at Sethanon, the concluding volume in the original Riftwar Cycle that had started the ball rolling. Somewhat to everyone’s surprise, Feist was willing to entrust this young, unproven writer with creating something really new in his world. Betrayal at Krondor was off and running.

Hallford may have come to Midkemia late, but his dogged determination to capture the world exactly as it existed in the novels would come to a large degree to define the project. He calls himself a “born fanboy” by nature. Thus, even though he wasn’t quite of Feist’s hardcore fandom, he had enormous empathy for them. He points back to an experience from his youth: when, as a dedicated Star Trek fan, he started to read the paperback novels based on the television series which Pocket Books published in the 1980s. I read them as well, and can remember that some of them were surprisingly good as novels, at least according to my adolescent sensibilities, while also managing to capture the spirit of the series I saw on television. Others, however… not so much. Hallford points to one disillusioning book in particular, which constantly referred to phasers as “ray guns.” It inculcated in him a sense that any writer who works in a beloved universe owes it to the fans of said universe — even if he’s not really one of them — to be as true to it as is humanly possible.

So, Hallford wrote Betrayal at Krondor with Feist’s fans constantly in mind. He immersed himself in Feist’s works to the point of that he was almost able to become the novelist. The prose he crafted, vivid and effective within its domain, really is virtually indistinguishable from that of its inspiration, whose own involvement was limited to an early in-person meeting and regular phone conversations thereafter. Yet the latter became more rather than less frequent as the project wore on; Feist found his enthusiasm for the game increasing in tandem with his surprise at how earnestly Hallford tried to capture his novels and the extent to which he was managing to succeed with only the most limited coaching. The fan verdict would prove even more telling. To this day, many of them believe that it was Feist himself who scripted Betrayal at Krondor.

But Betrayal of Krondor is notable for more than Neal Hallford’s dedicated fan service. It’s filled to bursting with genuinely original ideas, many of which flew in the face of contemporary fashions in games. Not all of the ideas work — some of them rather pull against one another — but the game’s boldness makes it a bracing study in design.

Following the lead of GUI advocates working with other sorts of software, game designers in the early 1990s were increasingly embracing the gospel of the “mode-less” interface: a single master screen on which everything takes place, as opposed to different displays and interfaces for different play states. (For an excellent example of how a mode-less interface could be implemented in the context of a CRPG, see Origin Systems’s Ultima VII.) Cutter and Hallford, however, pitched this gospel straight into the trash can without a second thought. Betrayal at Krondor has a separate mode for everything.

The closest thing it has to a “home” screen must be the first-person exploration view, which uses 3D graphics technology poached from Dynamix’s flight simulators. But then, you can and probably often will move around from an overhead map view as well. When interesting encounters happen, the screen is given over to text with clickable menus, or to storybook-style illustrated dialog scenes. When you get in a fight, that’s also displayed on a screen of its own; combat is a turn-based affair played on a grid that ends up vaguely resembling the Battle Chess games by Interplay. (Thankfully, it’s also tactically interesting and satisfying.) And then when you come upon a locked chest, you’re dumped into yet another new mode, where you have to work out a word puzzle in order to open it, because why not? All of these modes are accompanied by different styles of graphics: 3D graphics on the main exploration screen, a no-frills Rogue-like display for the overhead movement view, pixel art with the story scenes, digitized real-world actors with the dialog scenes, the sprite-based isometric view that accompanies combat, etc.

The first-person exploration view.

The overhead view.

A bit of exposition. Could this be a side quest before us?

The combat view.

A puzzle chest. The answer to this one, for the record, is “die.” Later riddles get much more complicated, but the mechanics of the puzzles ingeniously prevent them from ever becoming completely insoluble. Many a player has had a significant other who couldn’t care less about the rest of the game, but loves these puzzle chests…

This mishmash of approaches can make the game feel like a throwback to the 1980s, when genres and their established sets of best practices were not yet set in stone, and when many games that may strike us as rather odd mashups today were being produced. We can certainly see John Cutter’s roots in Cinemaware here; that company made a career out of ignoring the rules of ludic genre in favor of whatever systems best conveyed the fictional genre they were attempting to capture. By all rights, Betrayal at Krondor ought not to work, as so many of Cinemaware’s games tended not quite to work. All of these different modes and play styles — the puzzle chests in particular seem beamed in from a different game entirely — ought to add up to a hopelessly confusing muddle. Somehow, though, it does work; Betrayal at Krondor actually isn’t terribly hard to come to grips with initially, and navigating its many modes soon becomes second nature.

One reason for this is doubtless also the reason for much else that’s good about the game: its unusually extended testing period. When development was reaching what everyone thought to be its final stages, Dynamix sent the game to outside testers for what was expected to be a three-month evaluation period. Even this much usability testing would have been more than most studios were doing at this time. But the project, as so many game-development projects tend to do, ran way longer than expected, and three months turned into nine months of constant player feedback. While our universe isn’t entirely bereft of games that seem to have sprung into being fully-formed, by far the most good games attain that status only gradually, through repeated iterations of testing and feedback. Betrayal at Krondor came by its goodness in exactly this hard, honest way. Unlike a dismaying number of games from its time, this game feels like one that’s actually been played — played extensively — before it got released. The niggling problems that dog even many good games from the early 1990s (such as the infuriating inventory management and rudderless combat of Ultima VII) are almost completely absent here. Instead the game is full of thoughtful little touches to head off annoyance, the sort of touches that can only come from real player feedback.

The final verdict on its mishmash of graphical approaches, on the other hand, must be less positive. Betrayal at Krondor wasn’t a notably attractive game even by the standards of its day, and time has done it no favors; the project desperately needed a strong art director able to impose a unified aesthetic vision. The parts of it that have aged the worst by far are those employing digitized actors, who look almost unbelievably ludicrous, cutting violently against any sense of Tolkienesque grandeur Hallford’s prose might be straining to evoke. Most store-bought Halloween costumes look higher rent than this bunch of survivors of an explosion at the Loony Tunes prop department. John Cutter acknowledges the problems:

We digitized a lot of the actors, and we assumed they were going to be so pixelated that the makeup and costumes didn’t have to look that great. They just kind of had to be… close. But by the time we launched the game the technology had improved… yeah. You could see the elastic bands on the fake beards. It was pretty bad. I wasn’t crazy about a lot of the graphics in the game.

Tellingly, the use of digitized actors was the one place where Betrayal at Krondor didn’t blaze its own trail, bowing instead to contemporary trends.



For all of Betrayal at Krondor‘s welcome willingness just to try lots of stuff, its approach to story remains its most memorable and interesting quality of all. This aspect of the game was so front and center in the mind of John Cutter that, when he wrote a brief few paragraphs of “Designer Notes” for the manual, it came to occupy more than half the space:

We decided the game should be an interactive story. Characters would be multidimensional and capable of stirring the player’s emotions. The story would be carefully plotted with lots of surprises, a good mix of humor and pathos, and abundant amounts of mystery and foreshadowing to keep the player intrigued.

Balancing play against plot is the most confounding job any game designer can face on a fantasy role-playing game. In Betrayal at Krondor, we have integrated our plot so that it provides ample gaming opportunities, while also giving the player a sense of time, place, and purpose. This is achieved by making an onscreen map available to the player at all times, and by creating short-term goals — the nine chapters in the game — which give us a unique opportunity to tell a progressive story that still gives the player plenty of freedom to explore and adventure without being confined to a scripted plot.

In thus “balancing play against plot,” Cutter and Hallford were attempting to square a circle that had been bedeviling game designers for a long time. All of the things that mark a rich story — characters with agendas of their own; big reveals and shocking turns; the classic narrative structure of rising action, climax, and denouement; dramatic confrontations with expressive dialog — cut against the player’s freedom to go wherever and do whatever she wants. As a designer, says the conventional wisdom, you can’t have it all: you must rather stake out your spot on a continuum where at one end the player does little more than click her way through a railroaded plot line, and at the other she does absolutely anything she wants, but does it in a world bereft of any larger meaning or purpose. Adventure games tend to lean toward the set-piece-storytelling end of the continuum, CRPGs toward open-ended interactivity.

Even CRPGs from around the time of Betrayal at Krondor which are written expansively and well, such as Ultima VII, generally send you wandering through other people’s stories rather than your own. Each city you explore in that game is full of little story stubs revolving around the inhabitants thereof rather than yourself; your role is merely to nudge these dramas of others along to some sort of resolution before you disappear again. Your larger agenda, meanwhile, boils down to the usual real or metaphorical collecting of pieces to assemble the big whatsit at the end — a series of actions which can be done in any order precisely because they’re so simplistic in terms of plot. You’re in the world, but never really feel yourself to be of it.

Cutter and Hallford, however, refused to accept the conventional wisdom embodied by even so markedly innovative a CRPG as Ultima VII. They were determined to deliver the best of both worlds — an adventure-game-like plot and CRPG-like freedom — in the same game. Unsurprisingly, it doesn’t quite work as a whole. Nevertheless, the attempt is well worth discussing.

Betrayal at Krondor positively trumpets its intentions via the metaphors which its user interface employs. Once again ignoring all of the fashions of its time, which emphasized the definitively non-textual aesthetic of the interactive movie, this game presents itself as an interactive book with an enthusiasm worthy of the 1980s heyday of bookware. The overriding look of the game, to the extent it has one amidst all its clashing graphical styles, is of an illuminated manuscript, ink on yellowing parchment. The story is told in a literary past tense, save points become “bookmarks,” and, as Cutter himself noted in the extract above, the whole experience is divided into nine neat “chapters.”

The game is relentless about describing every single event using full sentences worthy of one of Feist’s novels. Sometimes the end result can verge on the ridiculous. For example, every single time you search the body of an opponent you’ve just killed — something you’ll be doing an awful lot of, what with this being a CRPG and all — you’re greeted with a verbose missive:

Owyn looked for supplies. Feeling like a vulture, he turned the body this way and that as he searched for anything that might be of value to them on their journey. All in all, he supposed that if he were the dead man, it wouldn’t matter to him any longer what happened to his belongings.

Every character has the exact same feeling when searching a dead body, despite very different personalities. This is one of many places where Betrayal at Krondor‘s verbosity winds up undercutting rather than strengthening its sense of mimesis.

Of course, you can and quickly will learn to click right through this message and its one or two random variations each time you search a corpse. But it remains an amusing sign of just how committed Cutter and Halford were to their “interactive storybook” concept in even the most repetitive, mechanical areas of their creation. (Imagine what Pac-Man would be like if the title character stopped to muse about his actions every time he swallowed a power pill and killed another ghost…)

All of this past-tense verbosity has an oddly distancing effect. You don’t feel like you’re having an adventure so much as reading one — or possibly writing one. You’re held at a remove even from the characters in your party, normally the primary locus of player identification in a game like this one. You don’t get to make your own characters; instead you’re assigned three of them who fulfill the needs of the plot. And, while you can guide their development by earning experience points, improving their skills, and buying them new spells and equipment, you don’t even get to hang onto the same bunch through the whole game. Characters are moved in and out of your party from chapter to chapter — again, as the needs of each chapter’s plot requires. The final effect almost smacks of a literary hypertext, as you explore the possibility space of a story rather than actually feeling yourself to be embodying a role or roles in that story. This is certainly unique, and not necessarily a bad thing. It’s just… a little strange in relation to what we tend to think of CRPGs as being. These are, after all, role-playing games.

As I’ve described it so far, Betrayal at Krondor sounds more akin to the typical Japanese than the Western CRPG. The former tend to lie much closer to the set-piece-story end of our continuum of design; they provide a set, fairly linear plot to walk through, generally complete with predefined characters, rather than the degree of world simulation and open-ended exploration that marks the Western tradition. (A Japanese CRPG is, many a critic has scoffed, just a linear story in which you have to fight a battle to see each successive scene.) Yet Betrayal at Krondor actually doesn’t fit comfortably with that bunch either. For, as Cutter also notes above, he and his design partner were determined to “give the player plenty of freedom to explore and adventure without being bound to a scripted plot.”

Their means of accomplishing that relies once again on the chapter system. Each chapter begins and ends with a big helping of set-piece plot and exposition. In between, though, you’re free to go your own way and take your time in satisfying the conditions that will lead to the end of the chapter. In the first chapter, for example, your assignment is to escort a prisoner across much of the map to the capital city of Krondor. How and when you do so is up to you. The map is filled with encounters and quests, most of which have nothing to do with your central mission. And when you eventually do finish the chapter and continue on with the next, the same map gets repopulated with new things to do. This is the origin of a claim from Dynamix’s marketing department that Betrayal at Krondor is really nine CRPGs in one. In truth, it doesn’t quite live up to that billing. Only a subsection of the map is actually available to you in most chapters, much of it being walled off by impenetrable obstacles or monsters you can’t possibly kill. Even the repopulation that happens between chapters is far from comprehensive. Still, it’s an impressively earnest attempt to combine the pleasures of set-piece plotting with those of an emergent, persistent virtual world.

And yet the combination between set-piece storytelling and emergent exploration always feels like just that: a combination rather than a seamless whole. Cutter and Hallford didn’t, in other words, truly square this particular circle. There’s one massive block of cognitive dissonance standing at the center of it all.

Consider: you’re told at the beginning of the first chapter that your mission of escorting your prisoner to the capital is urgent. Political crisis is in the air, war clouds on the horizon. The situation demands that you hurry to Krondor by the shortest, most direct path. And yet what do you do, if you want to get the most out of the game? You head off in the opposite direction at a relaxed doddle, poking your nose into every cranny you come across. There’s a tacit agreement between game and player that the “urgent” sense of crisis in the air won’t actually evolve into anything until you decide to make it do so by hitting the next plot trigger. Thus the fundamental artificiality of the story is recognized at some level by both game and player, in a way that cuts against everything Betrayal at Krondor claims to want to be. This isn’t really an interactive storybook; it’s still at bottom a collection of gameplay elements wired together with chunks of story that don’t really need to be taken all that seriously at the end of the day.

The same sense of separation shows itself in those lengthy chapter-beginning and -ending expository scenes. A lot of stuff happens in these, including fights involving the characters ostensibly under your control, that you have no control over whatsoever — that are external to the world simulation. And then the demands of plot are satisfied for a while, and the simulation engine kicks back in. This is no better or worse than the vast majority of games with stories, but it certainly isn’t the revolution some of the designers’ claims might seem to imply.

Of course, one might say that all of these observations are rather more philosophical than practical, of more interest to game designers and scholars than the average player; you can suspend your disbelief easily enough and enjoy the game just as it is. There are places in Betrayal at Krondor, however, where some of the knock-on effects of the designers’ priorities really do impact your enjoyment in more tangible ways. For this is a game which can leave you marooned halfway through, unable to move forward and unable to go back.

Dead ends where the only option is to restore are normally less associated with CRPGs than adventure games; they played a big role in all but killing that genre as a commercial proposition by the end of the 1990s. CRPGs are usually more forgiving thanks to their more simulation-oriented nature — but, sadly, Betrayal at Krondor is an exception, due to a confluence of design decisions that all seem perfectly reasonable and were all made with the best of intentions. It thus provides a lesson in unexpected, unintended consequences — a lesson which any game designer would be wise to study.

The blogger Chet Bolingbroke, better known as The CRPG Addict, made these comments recently in the context of another game:

One of the notable features of CRPGs in contrast to some other genres is that they almost always support a Plan B. When one way of playing doesn’t work out, you can almost always resort to a more boring, more banal, grindier method of getting something done. I tend to mentally preface these fallback plans with “I can always…” Having a tough time with the final battle? “I can always reload again and again until the initiative rolls go my way.” Can’t overcome the evil wizard at your current level? “I can always grind.” Running out of resources? “I can always retreat from the dungeon, head back to town and buy a ton of healing potions.”

The most frustrating moments in CRPGs are when you suddenly find yourself with no way to finish “I can always” — when there is no Plan B, when luck alone will never save you, when there isn’t even a long way around.

This is precisely the problem which the player of Betrayal at Krondor can all too easily run into. Not only does the game allow you to ignore the urgent call of its plot, but it actually forces you to do so in order to be successful. If you take the impetus of the story seriously and rush to fulfill your tasks in the early chapters, you won’t build up your characters sufficiently to survive the later ones. Even if you do take your time and explore, trying to accrue experience, focusing on the wrong skills and spells can leave you in the same boat. By the time you realize your predicament, your “Plan B” is nonexistent. You can’t get back to those encounters you skipped in the earlier, easier chapters, and thus can’t grind your characters out of their difficulties. There actually are no random encounters whatsoever in the game, only the fixed ones placed on the map at the beginning of each chapter. I’m no fan of grinding, so I’d normally be all in favor of such a choice, which Cutter and Hallford doubtless made in order to make the game less tedious and increase its sense of narrative verisimilitude. In practice, though, it means that the pool of available money and experience is finite, meaning you need not only to forget the plot and explore everywhere in the earlier chapters but make the right choices in terms of character development there if you hope to succeed in the later ones.

On the whole, then, Betrayal at Krondor acquits itself better in its earlier chapters than in its later ones. It can be a very immersive experience indeed when you first start out with a huge map to roam, full of monsters to battle and quests to discover. By the time said map has been repopulated three or four times, however, roaming across its familiar landmarks yet again, looking for whatever might be new, has begun to lose some of its appeal.

And then, as Neal Hallford would be the first to admit, Betrayal at Krondor is written above all for Raymond E. Feist fans, which can be a bit problematic if you don’t happen to be among them. This was my experience, at any rate. As an outsider to Feist’s universe, watching characters I didn’t know talk about things I’d never heard of eventually got old. When an “iconic” character like Jimmy the Hand shows up, I’m supposed to be all aflutter with excitement, but instead I’m just wondering who this latest jerk in a terrible costume is and why I should care. In my view, the game peaks in Chapter 3, which takes the form of a surprisingly complex self-contained murder mystery; this is a place where the game does succeed in integrating its set-piece and emergent sides to a greater extent than elsewhere. If you elect to stop playing after that chapter, you really won’t miss that much.


As I noted already, Betrayal at Krondor ran dramatically over time and over budget. To their credit, Dynamix’s management didn’t push it out the door in an unfinished state, as was happening with so many other games during this period of transition to larger and more complex productions. Yet everyone, especially poor Neal Hallford, felt the pressure of getting it done. Not only did he write almost every word of the considerable amount of text in the game, but he also wrote much of the manual, and somehow even wound up on the hook for the puff pieces about it in Sierra’s customer newsletter. After weeks of virtually living at the office, he collapsed there one day, clutching at his chest. His colleagues rushed him to the hospital, believing he must be having a heart attack even though he was still in his twenties. It turned out that he wasn’t, but the doctor’s orders were clear: “You’re not going back to work for a week. Get some rest and eat something proper. No pizza. No soft drinks. It’s either this or next time you leave work it’ll be in a hearse.” Such are the perils of commercial game development.

Betrayal at Krondor finally shipped on June 15, 1993, an inauspicious time in the history of CRPGs. Origin Systems was about to take the Ultima series in a radically different direction after a less than overwhelming response to Ultima VII; Sir-Tech was about to put their equally long-running Wizardry series on ice for similar reasons; SSI was facing dwindling sales of their Dungeons & Dragons games and was on the verge of losing the once-coveted license; other publishers were quietly dropping less prominent franchises and would-be franchises. The several years to come would be remembered by CRPG fans as the Dark Age of their favored genre; relatively few of games of this stripe would be released at all, and those that were would be greeted by the marketplace with little enthusiasm.

Initially, Dynamix’s first CRPG performed about as well as you might expect in this environment. Despite some strong reviews, and despite whatever commercial advantages the Feist license brought with it, sales were slow. Cutter and Hallford had gone into Betrayal at Krondor imagining it to be only the first entry in a new series, but it soon appeared unlikely that a sequel would come to pass. Sierra, Dynamix’s parent company, was having an ugly year financially and wasn’t in the mood to make another expensive game in a passé genre, while Jeff Tunnell, the man who had had the original idea for Betrayal at Krondor, had stepped down from day-to-day management at Dynamix in favor of running a smaller subsidiary studio. Cutter and Hallford begged their new bosses to give the game time before making any final decisions, noting that good reviews and positive word of mouth among fans of the novels could yet pay dividends. The leadership team responded by laying Cutter off.

But over time, Betrayal at Krondor continued to sell steadily if not spectacularly. Then a genuine surge in sales came in early 1994, when a CD-ROM-based version featuring a lovely soundtrack and enhanced if still less than lovely graphics was released, just as the influential magazine Computer Gaming World was crowning the game the best CRPG of the previous year. Dynamix now made a belated attempt to start work on a sequel, asking Neal Hallford to helm it. But he considered the budget they were proposing to be inadequate, the time frame for development far too compressed. He turned it down, and left the company shortly thereafter. Dynamix would never make a second CRPG, whether set in Midkemia or anywhere else.

Nevertheless, that wasn’t quite the end of the story. Feist had been profoundly impressed by Betrayal at Krondor, and now took the ludic possibilities of his series of novels much more seriously than he had before seeing it. As soon as the Dynamix license expired at the beginning of 1995, he began to shop the property around once again. Initially, however, he found no one willing to pay his price, what with the current state of the CRPG market. While interactive Midkemia was thus in limbo, Sierra came up with another, cheaper idea for capitalizing on the first game’s slow-burning success. Lacking the Midkemia license, they decided to leverage the first half of the Betrayal at Krondor name instead, releasing the in-house-developed Betrayal in Antara in 1997. It copied some of the interface elements and gameplay approaches of its predecessor, but moved the action to a generic fantasy world, to less satisfying effect.

And yet the story still wasn’t over. Feist had finally found a buyer for the Midkemia rights in 1996 in the form of a publisher known as 7th Level, who signed a studio known as PyroTechnix to make a direct sequel to Betrayal at Krondor at last. But when 7th Level ran into financial difficulties, Sierra of all publishers bought back the rights, along with PyroTechnix’s development contract. The latter completed the game and saw it released under the Sierra imprint in 1998. Feist played a much more active role on Return to Krondor, the game in question, than he had on Betrayal at Krondor, yet the result once again pales in comparison to the first Midkemia game, perhaps because Cutter and Hallford once again played no role. Its mixed reception marks the last official implementation of Midkemia on a computer to date, excepting only a brief-lived MMORPG.

Two of Feist’s later books, 1998’s Krondor: The Betrayal and 2000’s Krondor: Tear of the Gods, were based upon the first and second Midkemia computer game respectively. Thus Midkemia completed its long, strange transmedia journey from game to book to game to book again. Feist continued to churn out Midkemia books until 2013, when he announced that that year’s appropriately named Magician’s End, the 30th (!) entry in the series, would be the last. The later books, however, didn’t sell in the same quantities as the earlier ones, bearing as they did the stale odor of a series long past its sell-by date.

For many of us, Betrayal at Krondor will always remain the most memorable entry in the exercise in competent derivation that is Midkemia as a whole; the game is ironically much more innovative in its medium than the novels which spawned it are in theirs. Indeed, it’s thoroughly unique, a welcome breath of bold originality in a genre usually content to rely on the tried and true, a game which doesn’t work perfectly but perhaps works better than it has any right to. As a writer, I can only applaud a game which takes its writing this seriously. If it’s not quite the revolutionary amalgamation of narrative and interactivity that its creators wanted it to be, it’s still a heck of a lot more interesting than your average dungeon crawl.

(Sources: the book Designers and Dragons by Shannon Appelcline; Sierra’s newsletter InterAction of Winter 1992 and June 1993; Compute! of December 1993; Computer Gaming World of February 1993, April 1994, June 1994, August 1996, and October 1998; Electronic Games of October 1992 and June 1993; Questbusters of November 1991, August 1992, April 1993, and August 1993; Retro Gamer 84; Dragon of January 2004; the CD-ROM Today bundled CD-ROM of August/September 1994. Online sources include Matt Barton’s interviews with Neal Hallford, Jeff Tunnell, and John Cutter in Matt Chat episodes 191, 192, 201, 291, 292, and 293; Neal Hallford’s blog series Krondor Confidential; the “History of Midkemia Press” on the same publisher’s website.

Betrayal at Krondor and Betrayal in Antara are available as a package purchase at GOG.com.)

 
 

Tags: , ,

Opening the Gold Box, Part 6: A Troubled Marriage

This pie chart prepared by the investment firm Piper Jaffray Research provides a snapshot of the American computer-game industry as of 1993. Sierra leads the pack with a market share of 11.8 percent, trailed closely by Spectrum Holobyte, who have just increased their profile dramatically by acquiring MicroProse Software. Electronic Arts comes in at only third place here, even following their recent acquisition of Origin Systems, but this chart reflects only their computer-game sales; their total sales including computers and consoles are vastly higher than those of Sierra by this point. SSI manages to come in at a respectable fifth place, thanks not least to the two aforementioned acquisitions of comparably sized competitors, but their trend lines are all moving in the wrong direction; their last new release to cross the magical threshold of 100,000 copies sold was Eye of the Beholder from February of 1991, while their biggest game of 1993 will sell just over 70,000 copies.

Two individually unhappy spouses aren’t the recipe for a happy marriage. By 1992, the computer-game publisher SSI and the tabletop-game publisher TSR, whose announcement of a partnership had so shocked both of their industries back in 1987, were learning this reality the hard way. Dungeons & Dragons, both on the computer and on the tabletop, was in trouble, and the marketing synergy which the two companies had so successfully created just a few years before had now turned into a deadly embrace that threatened to pull them both under.

In many ways, SSI’s problems were typical of any small publisher in their changing industry. Players’ audiovisual expectations of the games they purchased were growing rapidly, and it just wasn’t clear where the money to meet them was to come from. SSI had ridden their Gold Box engine for Dungeons & Dragons CRPGs way too hard, churning out nine games using it — eleven if you count two reskinned science-fictional Buck Rogers games — in a span of less than four years. The engine had received some modest improvements over the course of that time, in the form of rudimentary mouse support and VGA- rather than EGA-standard graphics, but at bottom it still played like what it was: an artifact from an entirely different epoch of gaming, designed around the affordances of the 8-bit Commodore 64 rather than the latest 32-bit Intel wonders. It was so outdated as to seem almost laughable beside a boundary-pushing wunderkind like Origin Systems’s Ultima VII.

Just as distressingly, SSI hardly seemed to be trying anymore even when it came to their Gold Box designs. No later Gold Box game had possessed anything like the creative flair of Pool of Radiance and Curse of the Azure Bonds, the first two games of the line and by far the best.

And so, as the technology had aged and design standards had fallen, gamers had reacted appropriately: sales had dropped almost linearly from title to title. Pool of Radiance had sold 264,536 copies upon its release in August of 1988; Dark Queen of Krynn, the anticlimactic end of the Gold Box line, sold 40,640 copies after its release in May of 1992. SSI was still profitable that year, but only by a whisker: during the fiscal year which ended on September 30, 1992, profits amounted to just $168,000 on sales of a little under $13 million, the latter of which fell short of expectations by $1 million. What would the next year bring?

TSR was a larger and more diversified company, but they were facing the same essential problem: sales of their own Dungeons & Dragons line for the tabletop had been going in the wrong direction as well for the past couple of years, and it wasn’t immediately clear how to reverse that trend. A flood of new rules supplements and settings — by 1993, TSR would offer an extraordinary eight separate boxed “worlds” in which to play the game, ranging from traditional high fantasy to the Arabian Nights to the depths of outer space — certainly wasn’t doing the trick. In fact, by making Dungeons & Dragons ever more impenetrable to newcomers, the torrent of product was arguably hurting TSR more than it was helping them.

Thanks to these trends, Dungeons & Dragons was in danger of seeing its position as the commercial ne plus ultra of tabletop RPGs usurped for the first and only time in the history of the hobby. The biggest threat to its status came from a new RPG called Vampire: The Masquerade, whose rules-lite, storytelling-oriented approach was the antithesis of the baggy monstrosity which Dungeons & Dragons had become. By catching a wave of “goth” inspiration that was sweeping pop culture more generally, Vampire had even accrued a degree of street cred the likes of which TSR’s nerdier, more pedantic offerings couldn’t have hoped to match even in their early 1980s heyday. TSR’s entire Dungeons & Dragons gaming line was in danger of becoming the world’s most elaborate loss leader, fueling sales of the one part of their empire that was still consistently earning money: their vast and ever-growing lineup of fantasy novels based on their gaming properties.

Human nature being what it is, it was perhaps inevitable that SSI and TSR, these two partners with good reason to be profoundly worried about their futures, would each come to blame the other for at least some of their difficulties. SSI noted pointedly that the Gold Box line was supposed to have been a creative as well as financial partnership between the two companies, with TSR’s staff contributing much of the content for the computer games and TSR themselves publishing tie-in products for the tabletop. All of those synergies, however, had dried up after Pool of Radiance and Curse of the Azure Bonds. (The status of these two very first games as the very best of their line begins to seem like less of a puzzle in light of TSR’s active involvement with them.) SSI had been left to their own devices from 1990 on, albeit still subject to the frequently exercised veto power which TSR enjoyed over their ideas.

Meanwhile, even as SSI complained about their creative abandonment, it was hardly lost on TSR that the Gold Box engine had fallen badly behind the state of the art. As they judged it, its antiquity had become extreme enough to actively hurt their brand, not only on the computer but on the tabletop as well; when struggling against their tabletop game’s popular image as a kitschy relic of the 1980s, TSR’s marketers weren’t excited to be confronted with computers games that themselves looked like products of the previous decade. TSR was also unhappy with SSI’s failure to port the Gold Box games from computers to consoles; out of all of them, only Pool of Radiance had been ported to the Nintendo Entertainment System, by a Japanese developer rather than SSI themselves. SSI tried to point out that this port, which played badly and sold worse, only served to illustrate all the ways in which this style of game just wasn’t suitable for consoles, but TSR was having none of it. In their view, the porting issue was a problem for SSI to solve rather than to explain away.

Behind all the bickering loomed a daunting reality: SSI’s exclusive license to Dungeons & Dragons was due to expire on January 1, 1993. One of the partners had far more cause for concern about this fact than the other. For those in the boardroom at TSR, the question of the contract’s renewal was just another business debate to be hashed out, but for SSI it was quite possibly of existential importance. After signing his first contract with TSR, Joel Billings, SSI’s founder and president, had rejiggered the public and private face of his company, from that of a maker of hardcore wargames inspired by the tabletop grognard tradition of Avalon Hill and SPI into the Computerized Home of Dungeons & Dragons. While SSI still published some wargames in the early 1990s, they generally sold even worse than the final stragglers from the Gold Box line, and were made strictly by outside developers; almost the entirety of SSI’s internal development efforts had been devoted to Dungeons & Dragons for the past five years. SSI’s identity had become so bound up with TSR’s flagship property that it wasn’t clear what they could or should be without the Dungeons & Dragons license.

The uncertainty surrounding the future of the contract left SSI paralyzed. It was obvious that they needed a better, more modern engine if they were to continue to make Dungeons & Dragons CRPGs, but it would be foolhardy to embark on that expensive project before they were sure of retaining the license that would let them use it for its intended purpose.

Thus Billings must have breathed a sigh of relief in early 1992, when TSR, despite all their recent misgivings about SSI’s handling of the license, agreed to an eighteen-month contract extension. It would take the license out to July 1, 1994, giving SSI enough time to make a new engine and at least one new game with it. Still, the short length of extension served notice that they were on probation; if the marriage was to continue, SSI would have to deliver a hit of Pool of Radiance proportions.

Billings put his people to work on an engine that would build upon the best ideas of SSI’s competitors, not least Origin’s much-admired Ultima VII engine. Like that one, this one would be designed with a mouse in mind from the start; would offer free-scrolling real-time movement over a large world; would go almost entirely mode-less in terms of interface, integrating combat into the same view where conversation and exploration took place. Gone would be the fussy paragraph books, graph-paper maps, and code wheels of the Gold Box games, which could make the experience of playing them feel almost like a hybrid between a computer and tabletop game. SSI had a very different experience in mind this time out. They planned make the engine effortless enough for the player that it could be ported to the Super Nintendo for play on living-room couches. And if that version did well, other console ports would follow.

TSR, eager to give a boost to one of their sales-challenged alternate settings, convinced SSI to set the first game made with the new engine in the land of Dark Sun, a desert world with a vaguely post-apocalyptic feel. Billings, aware that he was on shaky ground with TSR, also initiated development of an original science-fiction game that was to use the engine as well, just in case the Dungeons & Dragons license went away.

Creating such a complex engine alongside the first two games to use it was a truly enormous task — by far the biggest thing SSI had ever attempted, dwarfing even the initial software engineering that had gone into the Gold Box engine. Development dragged on and on after the Gold Box line had petered out with Dark Queen of Krynn. SSI attempted to plug the Gold Box-sized gap in their product line with such second-string releases as Prophecy of the Shadow, an internally developed, non-licensed CRPG-lite (25,875 copies sold); Pirates of Realmspace, a buggy computerized take on TSR’s Spelljammer setting from an outside developer known as Cybertech (23,280 copies sold); The Summoning, a simple action-RPG from Event Horizon Software (25,273 copies sold); Veil of Darkness, a game of a similar stripe to the previous from the same developer (9866 copies sold); Legends of Valor, a poor man’s Ultima Underworld from Synthetic Dimensions (12,588 copies sold); and Unlimited Adventures, a final hurrah for the Gold Box in the form of a public release of many of SSI’s internal development tools, thereby to let the diehards make more games of their own of the old type (32,362 copies sold).

As the sales figures above attest, none of these games set the world on fire. Indeed, their sales managed to make even the latter days of the Gold Box line look pretty good by comparison. In all, SSI released just three games between the summers of 1992 and 1993 that managed to top 40,000 units: Great Naval Battles in September (43,774 copies sold), Tony La Russa Baseball II in March (70,902 copies sold), and Eye of the Beholder III in May (50,664 copies sold). Of this trio, only the last was a Dungeons & Dragons title, and only the last was developed internally. Needless to say, the bottom line suffered. During the fiscal year which ended on September 30, 1993, revenues fell to $10.5 million, and the company lost $500,000 — the first annual loss SSI had posted in more than a decade.

Joel Billings wrote in that year’s annual report that it had been “the most difficult year in SSI’s 14-year history.” He spoke his personal truth not least. Throughout this period, over the course of which development of the Dark Sun game and its engine kept dragging on far longer than expected, Billings was scrambling madly to stem the bleeding. He put an organization that had always had the atmosphere of a family company through the trauma of its first-ever layoff, slashing the employee rolls from 115 to 75 employees; the memory of doing so still haunts Billings, a gentle soul at heart, to this day. Having been forced to cut the staff needed to create the science-fiction game earmarked for the new engine, he cut that as well, putting all his eggs into the single basket that was the Dark Sun game. Even the Super Nintendo version of that game, which his programmers had been struggling mightily to realize, would have to be set aside as well, at least for now. Much to TSR’s chagrin, this latest Dungeons & Dragons game too would have to live or die on computers.

Yet all of Billings’s scrambling constituted no more than financial triage. The existential obstacle which SSI faced was that of being a small, boutique publisher in an industry whose economies of scale were making it harder and harder for such an entity to survive. It was getting ever harder to win shelf space at retail, harder to pay for advertising in the glossy magazines — and, most of all, harder to foot the ever-increasing bill of developing modern games that met all of the expectations of the 1990s. Billings reluctantly concluded that he had but one choice: he had to sell out, had to find a buyer for the family business he had spent almost a decade and a half building from the ground up.

Accordingly, he spent much of his time in 1993 beating the bushes for just such a buyer. Yet here he was stymied once again by the realities of the marketplace. SSI was far from the only small publisher looking for a port in the storm, and many of the others had — or at least were judged to have — more attractive portfolios of extant and forthcoming games. Thus Billings faced a dispiriting, borderline-humiliating series of near misses, of seeing SSI cast aside in favor of alternative acquisitions in the fast-consolidating industry.

At the beginning of June, he thought he had made a deal with Spectrum HoloByte, an oddly bifurcated publisher that was almost entirely dependent on two wildly divergent games: the ultra-hardcore flight simulator Falcon, whose manual was roughly the size of a Tom Clancy novel, and the casual phenomenon Tetris, a game so brilliantly simple that it took only about 30 seconds of experimentation at the keyboard to spawn a lifetime’s addiction. Both of these games, radically different though they were in personality, were equally successful with their own demographics. Just as importantly, Spectrum HoloByte was absurdly well-connected with the movers and shakers of international finance, and was awash in venture capital as a result.

Due diligence between SSI and Spectrum HoloByte was completed, and a plan was made to meet again and sign a letter of intent as soon as that year’s Summer Consumer Electronics Show was behind both of them. At that show, however, Spectrum HoloByte met with Microprose, whose financial circumstances were even more desperate than those of SSI but who had a much more impressive array of upcoming titles to show to potential suitors. To make a long story short, Spectrum HoloByte bought MicroProse instead, leaving SSI stranded at the altar.

A few months later, the same scenario repeated itself. This time the would-be acquirer was Electronic Arts, a company with which SSI already had a longstanding relationship: Trip Hawkins had been a member of SSI’s board since before he founded EA, SSI had been piggybacking on EA’s distribution network as an “affiliated label” since 1987, and EA in fact already owned 20 percent of SSI thanks to an investment made in 1987, when the smaller company was first scaling up to take on Dungeons & Dragons. For all these reasons, the deal at first seemed a natural one. But Hawkins, the biggest proponent of the acquisition on EA’s side, was busy with a new semi-subsidiary known as 3DO and no longer had the day-to-day involvement necessary with the parent company to push it through. After kicking the tires a bit, the rest of EA’s management decided that SSI just wasn’t worth the asking price — especially given that EA already owned Origin Systems, one of SSI’s biggest rivals in CRPGs. Contrary to Joel Billings’s best intentions, SSI would thus be forced to exit 1993 as they had entered it: still an independent company, facing a future that looked more perilous than ever.

SSI’s struggle to find a buyer was a sign not only of their own weakness but of the diminished commercial profile of Dungeons & Dragons. Five years earlier, three quarters of the industry would have rushed to scoop up SSI, if only to acquire the enviable licensing deal they had recently signed. Now, though, the tabletop game was at a low ebb of its own, even as it seemed hopelessly antithetical to all of the winds of change in digital gaming. Where did this nerdy game played in parents’ basements, all tables and charts and numbers, fit in an industry rushing to make slick, kinetic interactive movies featuring real Hollywood actors? Dungeons & Dragons just wasn’t cool. It had never really been cool, of course, but that hadn’t been a problem when the computer-game industry as well was thoroughly uncool. But now, as computer-game moguls were busily penning paeans to themselves as the next wave in mainstream entertainment, its uncoolness was extremely problematic.

Amidst all of this — in September of 1993, to be specific — Dark Sun: Shattered Lands finally got completed and released. It was the most important game SSI had published since Pool of Radiance; the future of the TSR partnership, and thus their own future as a company, rode on its success or lack thereof.

When viewed separately from all of these external pressures, as just a game to be played and hopefully enjoyed, it revealed itself to be a nobly earnest attempt to improve on SSI’s most recent efforts in the realm of CRPGs, even if it wasn’t an entirely unblemished one. On the technological side, SSI’s next-generation engine largely delivered where it needed to: it was indeed vastly slicker, prettier, easier, and more modern than the Gold Box engine, feeling like a true product of the 1990s rather than a holdover from the last decade. It was an engine that could even stand next to the likes of an Ultima VII without undue embarrassment. Indeed, SSI seemed to have learned from their rival’s mistakes and done Origin one better in some places. For example, in place of the real-time, well-nigh uncontrollable frenzy that was combat in Ultima VII, SSI’s engine lapsed seamlessly into a turn-based mode as soon as a fight began; this allowed combat in Shattered Lands to retain most of the tactical complexity and interest that had marked its implementation in the Gold Box games, with the additional advantages of increased audiovisual interest and a less cryptic interface.

At the same time that they endeavored to keep combat interesting, however, SSI’s design team had clearly made a concerted effort to move beyond the exercises in incessant combat and very little else which the Gold Box games had become by the end. Shattered Lands offered much better-developed characters to talk to, along with heaps of real choices to make and alternative pathways to discover. The new approach was enough to impress even so committed an SSI skeptic as Scorpia, Computer Gaming World magazine’s longtime adventure columnist, who had been roundly criticizing the Gold Box games in print for their “incessant, fight-after-fight” nature for half a decade by this point. Now, she could write that “SSI is taking their role-playing line in a new direction, which is good to see”: “the solution to every problem is not kill, kill, kill.” Shay Addams, another prominent adventure pundit, had a similar take: “It’s no secret that I never liked the Gold Box games. Dark Sun, however, kept me coming back to the dungeon for more: more combat, more exploring, more story.”

Still, the game had its fair share of niggles — more than enough of them, in fact, to prevent its achieving a classic status to rival Pool of Radiance and Curse of the Azure Bonds. While SSI was to be commended for attempting to give the setting and plot more nuance and texture, that just wasn’t the sort of thing they did best, and they were still receiving little to no help from TSR on that front. The writing and plotting were derivative in several different directions at once, hackneyed even by the usual standards of the genre. Mind you, the writing wasn’t actually worse than most of that which had accompanied the Gold Box games — but here, moved as it was from a paragraph book onto the screen and expected as it was to do a lot more heavy lifting, its weaknesses were magnified.

Shattered Lands was also damaged as a computer game by its need to conform to TSR’s tabletop rules. The boxed set which presented the Dark Sun setting for the tabletop included a whole range of new rules complications and variations to distinguish it from the already convoluted Dungeon & Dragons base game, and most of these SSI was expected to implement faithfully as part of their licensing agreement. And so Shattered Lands came complete with a bunch of races and classes unfamiliar even to most Gold Box and tabletop Dungeons & Dragons veterans, along with a veritable baseline expectation that every character would be double- or triple-classed. Clerics suddenly had to choose an “element” to worship, which limited their selection of spells — and now everyone had access to a whole parallel sphere of magic known as psionics, and had to choose a specialty there as well. No game designer starting a CRPG from scratch would ever have inserted so much cruft of such marginal utility to the ultimate goal of fun; it was the sort of thing that could only arise from a company like TSR throwing rule after rule at the wall over the course of years in order to sell more supplements. Certainly none of it made much sense in a game explicitly envisioned as a new beginning for Dungeons & Dragons on computers, a place for fresh players to jump aboard. Nor, for that matter, did the choice of the oddball world of Dark Sun as a setting; for all that critics like me have long railed against the tendency, gamers for time immemorial have been demonstrating their preference for CRPGs set in generic high-fantasy worlds — such as that of TSR’s own Forgotten Realms, home of the most commercially successful of the Gold Box games — over more unique settings like this one.

But whatever its intrinsic strengths and weaknesses, Shattered Lands suffered most of all from one undeniable external failing: it was deeply, thoroughly unfashionable in the context of 1993. At a time when the whole industry was moving toward multimedia “talkies,” its many conversations and descriptions were still implemented via screenful after screenful of boring old text. And in addition to the old-fashioned implementation, there also remained the fact that the Dungeons & Dragons name just wasn’t the force it once had been. A measure of the industry’s attitude toward the game and its commercial prospects can be gleaned from its placement in the magazines. Even as they were giving it reasonably positive reviews, Computer Gaming World buried it on page 124 of 276, Shay Addams’s Questbusters newsletter on page 8 of 16. (The lead review of that issue, evidently judged to be more immediately interesting to the newsletter’s readers than a review of Shattered Lands, was of Legend Entertainment’s Gateway 2, a fine game in its own right but one which still had a parser, for God’s sake.)

So, you’ve probably guessed where this is going: Dark Sun: Shattered Lands proved a devastating disappointment to TSR and especially to SSI. After costing more than $1 million and eighteen months to make, with the additional opportunity cost of preventing SSI’s internal developers from doing much of anything else over the course of that period, it sold just 45,917 copies. To put this figure into perspective, consider that it’s barely 5000 more copies than the last tired release of the old Gold Box line, or that it’s about one-sixth of the sales of Pool of Radiance — this in spite of an expanded marketplace in which the number of copies which a hit game could hope to sell was actually far greater than it had been five years before.

When SSI and TSR met again early in 1994, after it had become all too clear that Shattered Lands wasn’t to be the next Pool of Radiance, TSR stated matter-of-factly that they no longer wished to remain in the marriage. Some tense negotiation followed, during which TSR did make some concessions to a frantic SSI, who were facing down the apocalyptic prospect of a license due to expire in less than six months while they still had a lot more Dungeons & Dragons product from third-party developers in the pipeline. TSR agreed to extend the exclusive license for six more months, to January 1, 1995, and to allow SSI to continue to release new games under a non-exclusive license until July 1, 1995. After that, though, the marriage was through. TSR emphasized that there would be no further settlement agreements.

Thus SSI’s final string of Dungeons & Dragons releases, of which there would still be a considerable number, would have something of the feel of a lame-duck session of government. DreamForge Intertainment provided two real-time CRPGs set in TSR’s Gothic world of Ravenloft and a third set in the Forgotten Realms; Cyberlore Studios provided a similar game set in the Arabian Nights World of Al-Qadim; Lion Entertainment provided a Doom-influenced hack-and-slasher set nowhere in particular. An overoptimistic SSI had launched into Dark Sun: Wake of the Ravager, a sequel to Shattered Lands, before the commercial verdict on the first game and TSR’s final judgment on the whole partnership that had led to it had come in. They finished that game up too, after a fashion anyway, and released it, still full of bugs, unimplemented features, and placeholder writing. It became their final in-house-developed Dungeons & Dragons title. It made for a slightly pathetic way to bow out, but at this point they just couldn’t be bothered to do better; they were now a long way indeed from those enthusiastic early days of Pool of Radiance. None of these games sold more than a few tens of thousands of copies. But then, no one, least of all SSI, had much expected them to.

The news that TSR and SSI were parting ways reached the magazines almost immediately. The two newly minted divorcees couldn’t resist a bit of veiled sniping in the press. SSI, for instance, told Computer Gaming World that they were “unhappy with the rules and restrictions imposed with the license that limited their creativity,” and said they could be perfectly happy and very successful making original CRPGs instead. TSR, for their part, said they’d learned a lesson about binding themselves too inextricably to others, and thus wouldn’t be entering into any more exclusive arrangements at all. Instead they’d play the field, signing deals with publishers on a title-by-title basis, and might just learn how to make computer games of their own.

Yet behind all these brave words lurked a difficult reality for both companies; it was by no means clear that either or both of them would really be better off apart than they’d been together. As if it hadn’t had problems enough already, tabletop Dungeons & Dragons was now getting pummeled by a new arrival with huge appeal to the same demographic: Magic: The Gathering, a fast-playing, accessible “collectible card game” of fantasy combat psychologically engineered to sell an endless amount of content to gamers looking for that one perfect card which could give them an edge over their chums. Magic decks were soon eating up much of the shelf space in hobby stores that had once gone to Dungeons & Dragons, and pushing it out of their display windows entirely. TSR’s only solution was the same as it had always been: to churn out yet more source books. And so the spiral of diminishing returns continued.

The contrast between TSR and Wizards of the Coast, the upstart makers of Magic, was a telling one. The latter engaged with their customers directly at every opportunity, skillfully goosing the grass-roots excitement around their products to yet further extremes. But TSR, still led by the widely disliked non-gamer Lorraine Williams, seemed out of touch, utterly disinterested in their fans and their opinions. Ryan Dancey, who has done a lot of research into TSR’s history, sums up the company’s attitude in damning fashion:

In all my research into TSR’s business, across all the ledgers, notebooks, computer files, and other sources of data, there was one thing I never found — one gaping hole in the mass of data we had available. No customer profiling information. No feedback. No surveys. No “voice of the customer.” TSR, it seems, knew nothing about the people who kept it alive.

The brainy kids who used to fall into the Dungeons & Dragons rabbit hole around the time they entered junior high were now getting their first Magic decks at that age instead. With the red ink beginning to pile up to a truly alarming extent — even the novels were no longer selling like they used to — TSR looked to be headed for an ugly reckoning.

And yet, if TSR was in dire straits, SSI’s position was if anything even worse. Without Dungeons & Dragons, they had almost literally nothing; the strongest remaining item in their portfolio was the Tony La Russa Baseball franchise developed by Stormfront Studios. But a baseball simulation alone wouldn’t be enough to sustain the company, and the sales picture of their other recent products wasn’t pretty. They were still in desperate need of a savior, but now lacked even the TSR connection to offer to potential buyers. Who in the age of multimedia would want to buy a failing publisher of stats-heavy wargames and traditionalist CRPGs? Joel Billings didn’t know, but he had no choice but to keep looking for someone crazy enough to take the plunge.

(Sources: As with all of my SSI articles, much of this one is drawn from the SSI archive at the Strong Museum of Play. Other sources include the book Designers and Dragons by Shannon Appelcline; Computer Gaming World of September 1993, December 1993, April 1994, and December 1994; Questbusters of October 1993. Online sources include Matt Barton’s video interviews with Joel Billings and David Shelley and Laura Bowen.

The two Dark Sun games are available as digital purchases at GOG.com.)

 
21 Comments

Posted by on September 20, 2019 in Digital Antiquaria, Interactive Fiction

 

Tags: ,