RSS

Tag Archives: falstein

The Dig

As you would imagine, a lot of the things you can do in a comedy game just don’t work when trying to remain serious. You can’t cover up a bad puzzle with a funny line of self-referential dialog. Er, not that I ever did that. But anyway, it was also a challenge to maintain the tone and some semblance of a dramatic arc. Another challenge was cultural — we were trying to build this game in an environment where everyone else was building funny games, telling jokes, and being pretty outlandish. It was like trying to cram for a physics final during a dorm party. It would have been a lot easier to join the party.

— Sean Clark, fifth (and last) project lead on The Dig

On October 17, 1989, the senior staff of LucasArts[1]LucasArts was known as Lucasfilm Games until the summer of 1992. To avoid confusion, I use the name “LucasArts” throughout this article. assembled in the Main House of Skywalker Ranch for one of their regular planning meetings. In the course of proceedings, Noah Falstein, a designer and programmer who had been with the studio almost from the beginning, learned that he was to be given stewardship of an exciting new project called The Dig, born from an idea for an adventure game that had been presented to LucasArts by none other than Steven Spielberg. Soon after that bit of business was taken care of, remembers Falstein, “we felt the room start to shake — not too unusual, we’d been through many earthquakes in California — but then suddenly it got much stronger, and we started to hear someone scream, and some glass crash to the floor somewhere, and most of us dived under the mahogany conference table to ride it out.” It was the Loma Prieta Earthquake, which would kill 63 people, seriously injure another 400, and do untold amounts of property damage all around Northern California.

Perhaps Falstein and his colleagues should have taken it as an omen. The Dig would turn into a slow-motion fiasco that crushed experienced game developers under its weight with the same assiduity with which the earthquake collapsed Oakland’s Nimitz Freeway. When a finished version of the game finally appeared on store shelves in late 1995, one rather ungenerous question would be hard to avoid asking: it took you six years to make this?



In order to tell the full story of The Dig, the most famously troubled project in the history of LucasArts, we have to wind the clock back yet further: all the way back to the mid-1980s, when Steven Spielberg was flying high on the strength of blockbusters like Raiders of the Lost Ark and E.T.: The Extra-Terrestrial. During this period, many years before the advent of Prestige TV, Spielberg approached NBC with a proposal for a new anthology series named Amazing Stories, after the pulp magazine that had been such an incubator of printed science fiction in the 1930s and 1940s. He would direct the occasional episode himself, he promised, but would mostly just throw out outlines which could be turned into reality by other screenwriters and directors. Among those willing to direct episodes were some of the most respected filmmakers in Hollywood: people like Martin Scorsese, Irvin Kershner, Robert Zemeckis, and Clint Eastwood. Naturally, NBC was all over it; nowhere else on the television of the 1980s could you hope to see a roster of big-screen talent anything like that. The new series debuted with much hype on September 29, 1985.

But somehow it just never came together for Amazing Stories; right from the first episodes, the dominant reaction from both critics and the public was one of vague disappointment. Part of the problem was each episode’s running time of just half an hour, or 22 minutes once commercials and credits were factored in; there wasn’t much scope for story or character development in that paltry span of time. But another, even bigger problem was that what story and characters were there weren’t often all that interesting or original. Spielberg kept his promise to serve as the show’s idea man, personally providing the genesis of some 80 percent of the 45 episodes that were completed, but the outlines he tossed off were too often retreads of things that others had already done better. When he had an idea he really liked — such as the one about a group of miniature aliens who help the residents of an earthbound apartment block with their very earthbound problems — he tended to shop it elsewhere. The aforementioned idea, for example, led to the film Batteries Not Included.

The episode idea that would become the computer game The Dig after many torturous twists and turns was less original than that one. It involved a team of futuristic archaeologists digging in the ruins of what the audience would be led to assume was a lost alien civilization. Until, that is, the final shot set up the big reveal: the strange statue the archaeologists had been uncovering would be shown to be Mickey Mouse, while the enormous building behind it was the Sleeping Beauty Castle. They were digging at Disneyland, right here on Planet Earth!

The problem here was that we had seen all of this before, most notably at the end of Planet of the Apes, whose own climax had come when its own trio of astronauts stranded on its own apparently alien world had discovered the Statue of Liberty half-buried in the sand. Thus it was no great loss to posterity when this particular idea was judged too expensive for Amazing Stories to produce. But the core concept of archaeology in the future got stuck in Spielberg’s craw, to be trotted out again later in a very different context.

In the meantime, the show’s ratings were falling off quickly. As soon as the initial contract for two seasons had been fulfilled, Amazing Stories quietly disappeared from the airwaves. It became an object lesson that nothing is guaranteed in commercial media, not even Steven Spielberg’s Midas touch.

Fast-forward a couple of years, to when Spielberg was in the post-production phase of his latest cinematic blockbuster, Indiana Jones and the Last Crusade, which he was making in partnership with his good friend George Lucas. Noah Falstein of the latter’s very own games studio had been drafted to design an adventure game of the movie. Despite his lack of a games studio of his own, Spielberg was ironically far more personally interested in computer games than Lucas; he followed Falstein’s project quite closely, to the point of serving as a sort of unofficial beta tester. Even after the movie and game were released, Spielberg would ring up LucasArts from time to time to beg for hints for their other adventures, or sometimes just to shoot the breeze; he was clearly intrigued by the rapidly evolving world of interactive media. During one of these conversations, he said he had a concept whose origins dated back to Amazing Stories, one which he believed might work well as a game. And then he asked if he could bring it over to Skywalker Ranch. He didn’t have to ask twice.

The story that Spielberg outlined retained futuristic archaeology as its core motif, but wisely abandoned the clichéd reveal of Mickey Mouse. Instead the archaeologists would be on an actual alien planet, discovering impossibly advanced technology in what Spielberg conceived as an homage to the 1950s science-fiction classic Forbidden Planet. Over time, the individual archaeologists would come to distrust and eventually go to war with one another; this part of the plot hearkened back to another film that Spielberg loved, the classic Western The Treasure of the Sierra Madre. Over to you, Noah Falstein — after the unpleasant business of the earthquake was behind everybody, that is.

Noah Falstein

The offices of LucasArts were filled with young men who had grown up worshiping at the shrines of Star Wars and Indiana Jones, and who now found themselves in the completely unexpected position of going to work every day at Skywalker Ranch, surrounded by the memorabilia of their gods and sometimes by the deities themselves. Their stories of divine contact are always entertaining, not least for the way that they tend to sound more like a plot from one of Spielberg’s films than any plausible reality; surely ordinary middle-class kids in the real world don’t just stumble into a job working for the mastermind of Star Wars, do they? Well, it turns out that in some cases they do. Dave Grossman, an aspiring LucasArts game designer at the time, was present at a follow-up meeting with Spielberg that also included Lucas, Falstein, and game designer Ron Gilbert of Maniac Mansion and Monkey Island fame. His account so magnificently captures what it was like to be a starstruck youngster in those circumstances that I want to quote it in full here.

The Main House at Skywalker is a pretty swanky place, and the meeting is in a boardroom with a table the size of a railroad car, made of oak or mahogany or some other sort of expensive wood. I’m a fidgety young kid with clothes that come pre-wrinkled, and this room makes me feel about as out of place as a cigarette butt in a soufflé. I’m a little on edge just being in here.

Then George and Steven show up and we all say hello. Now, I’ve been playing it cool like it’s no big deal, and I know they’re just people who sneeze and drop forks like everybody else, but… it’s Lucas and Spielberg! These guys are famous and powerful and rich and, although they don’t act like any of those things, I’m totally intimidated. (I should mention that although I’ve been working for George for a year or so at this point, this is only the second time I’ve met him.) I realize I’m really fairly nervous now.

George and Steven chit-chat with each other for a little bit. They’ve been friends a long time and it shows. George seems particularly excited to tell Steven about his new car, an Acura I think – they’re not even available to the public yet, but he’s managed to get the first one off the boat, and it’s parked conspicuously right in front of the building.

Pretty soon they start talking about ideas for The Dig, and they are Rapid-Fire Machine Guns of Creativity. Clearly they do this a lot. It’s all very high-concept and all over the map, and I have no idea how we’re going to make any of it into a game, but that’s kind of what brainstorming sessions are all about. Ron and Noah offer up a few thoughts. I have a few myself, but somehow I don’t feel worthy enough to break in with them. So I sit and listen, and gradually my nervousness is joined by embarrassment that I’m not saying anything.

A snack has been provided for the gathering, some sort of crumbly carbohydrate item, corn bread, if I remember correctly. So I take a piece – I’m kind of hungry, and it gives me something to do with my hands. I take a bite. Normally, the food at Skywalker Ranch is absolutely amazing, but this particular corn bread has been made extra dry. Chalk dry. My mouth is already parched from being nervous, so it takes me a while before I’m able to swallow the bite, and as I chomp and smack at it I’m sure I’m making more noise than a dozen weasels in a paper bag, even though everyone pretends not to notice. There are drinks in the room, but they have been placed out of the way, approximately a quarter-mile from where we’re sitting, and I can’t get up to get one without disrupting everything, and I’m sure by now George and Steven are wondering why I’m in the meeting in the first place.

I want to abandon the corn bread, but it’s begun falling apart, and I can’t put it down on my tiny napkin without making a huge mess. So I eat the whole piece. It takes about twenty minutes. I myself am covered with tiny crumbs, but at least there aren’t any on the gorgeous table.

By now the stakes are quite high. Because I’ve been quiet so long, the mere fact of my speaking up will be a noteworthy event, and anything I say has to measure up to that noteworthiness. You can’t break a long silence with a throwaway comment, it has to be a weighty, breathtaking observation that causes each person in the room to re-examine himself in its light. While I’m waiting for a thought that good, more time goes by and raises the bar even higher. I spend the rest of the meeting in a state of near-total paralysis, trying to figure out how I can get out of the room without anyone noticing, or, better yet, how I can go back in time and arrange not to be there in the first place.

So, yes, I did technically get to meet Steven Spielberg face-to-face once while we were working on The Dig. I actually talked to him later on, when he called to get hints on one of our other games (I think it was Day of the Tentacle), which he was playing with his son. (One of the lesser-known perks of being a famous filmmaker is that you can talk directly to the game designers for hints instead of calling the hint line.) Nice guy.

The broader world of computer gaming’s reaction to Spielberg’s involvement in The Dig would parallel the behavior of Dave Grossman at this meeting. At the same time that some bold industry scribes were beginning to call games a more exciting medium than cinema, destined for even more popularity thanks to the special sauce of interactivity, the press that surrounded The Dig would point out with merciless clarity just how shallow their bravado was, how deep gaming’s inferiority complex really ran: Spielberg’s name was guaranteed to show up in the first paragraph of every advertisement, preview, or, eventually, review. “Steven Spielberg is deigning to show an interest in little old us!” ran the implicit message.

It must be said that the hype was somewhat out of proportion to his actual contribution. After providing the initial idea for the game — an idea that would be transformed beyond all recognition by the time the game was released — Spielberg continued to make himself available for occasional consultations; he met with Falstein and his colleagues for four brainstorming sessions, two of which also included his buddy George Lucas, over the course of about eighteen months. (Thanks no doubt to the prompting of his friend, Lucas’s own involvement with The Dig was as hands-on as he ever got with one of his games studio’s creations.) Yet it’s rather less clear whether these conversations were of much real, practical use to the developers down in the trenches. Neither Spielberg nor Lucas was, to state the obvious, a game designer, and thus they tended to focus on things that might yield watchable movies but were less helpful for making a playable game. Noah Falstein soon discovered that heading a project which involved two such high-profile figures was a less enviable role than he had envisioned it to be; he has since circumspectly described a project where “everyone wanted to put their two cents in, and that can be extremely hard to manage.”

In his quest for a game that could be implemented within the strictures of SCUMM, LucasArts’s in-house point-and-click adventure engine, Falstein whittled away at Spielberg’s idea of two teams of archaeologists who enter into open war with one another. His final design document, last updated on January 30, 1991, takes place in “the future, nearly 80 years since the McKillip Drive made faster-than-light travel a possibility, and only 50 years since the first star colonies were founded.” In another nod back to Spielberg’s old Amazing Stories outline that got the ball rolling, an unmanned probe has recently discovered an immense statue towering amidst other alien ruins on the surface of a heretofore unexplored planet; in a nod to the most famous poem by Percy Shelley, the planet has been named Ozymandias. Three humans have now come to Ozymandias to investigate the probe’s findings — but they’re no longer proper archaeologists, only opportunistic treasure hunters, led by a sketchy character named Major Tom (presumably a nod to David Bowie). The player can choose either of Major Tom’s two subordinates as her avatar.

A series of unfortunate events ensues shortly after the humans make their landing, over the course of which Major Tom is killed and their spaceship damaged beyond any obvious possibility of repair. The two survivors have an argument and go their separate ways, but in this version of the script theirs is a cold rather than a hot war. As the game goes on, the player discovers that a primitive race of aliens living amidst the ruins are in fact the descendants of far more advanced ancestors, who long ago destroyed their civilization and almost wiped out their entire species with internecine germ warfare. But, the player goes on to learn, there are survivors of both factions who fought the apocalyptic final war suspended in cryogenic sleep beneath the surface of the planet. Her ultimate goal becomes to awaken these survivors and negotiate a peace between them, both because it’s simply the right thing to do and because these aliens should have the knowledge and tools she needs to repair her damaged spaceship.

This image by Ken Macklin is one of the few pieces of concept art to have survived from Noah Falstein’s version of The Dig.

For better or for worse, this pared-down but still ambitious vision for The Dig never developed much beyond that final design document and a considerable amount of accompanying concept art. “There was a little bit of SCUMM programming done on one of the more interesting puzzles, but not much [more],” says Falstein. He was pulled off the project very early in 1991, assigned instead to help Hal Barwood with Indiana Jones and the Fate of Atlantis. And when this, his second Indiana Jones game, was finished, he was laid off despite a long and largely exemplary track record.

Meanwhile The Dig spent a year or more in limbo, until it was passed to Brian Moriarty, the writer and designer of three games for the 1980s text-adventure giant Infocom and of LucasArts’s own lovely, lyrical Loom. Of late, he’d been drafting a plan for a game based on The Young Indiana Jones Chronicles, the franchise’s slightly disappointing foray into television, but a lack of personal enthusiasm for the project had led to a frustrating lack of progress. Moriarty was known as one of the most “literary” of game designers by temperament; his old colleagues at Infocom had called him “Professor Moriarty,” more as a nod to his general disposition than to the milieu of Sherlock Holmes. And indeed, his Trinity is as close as Infocom ever got to publishing a work of high literature, while his Loom possesses almost an equally haunting beauty. Seeing himself with some justification as a genuine interactive auteur, he demanded total control of every aspect of The Dig as a condition of taking it on. Bowing to his stellar reputation, LucasArts’s management agreed.

Brian Moriarty

Much of of what went on during the eighteen months that Moriarty spent working on The Dig remains obscure, but it plainly turned into a very troubled, acrimonious project. He got off on the wrong foot with many on his team by summarily binning Falstein’s vision — a vision which they had liked or even in some cases actively contributed to. Instead he devised an entirely new framing plot.

Rather than the far future, The Dig would now take place in 1998; in fact, its beginning would prominently feature the Atlantis, a Space Shuttle that was currently being flown by NASA. A massive asteroid is on a collision course with Earth. Humanity’s only hope is to meet it in space and plant a set of nuclear bombs on its surface. Once exploded, they will hopefully deflect the asteroid just enough to avoid the Earth. (The similarity with not one but two terrible 1998 movies is presumably coincidental.) You play Boston Low, the commander of the mission.

But carrying the mission out successfully and saving the Earth is only a prelude to the real plot. Once you have the leisure to explore the asteroid, you and your crew begin to discover a number of oddities about it, evidence that another form of intelligent being has been here before you. In the midst of your investigations, you set off a booby trap which whisks you and three other crew members light years away to a mysterious world littered with remnants of alien technology but bereft of any living specimens. Yet it’s certainly not bereft of danger: one crew member gets killed in gruesome fashion almost immediately when he bumbles into a rain of acid. Having thus established its bona fides as a serious story, a million light years away from the typical LucasArts cartoon comedy, the game now begins to show a closer resemblance to Falstein’s concept. You must explore this alien world, solve its puzzles, and ferret out the secrets of the civilization that once existed here if you ever hope to see Earth again. In doing so, you’re challenged not only by the environment itself but by bickering dissension in your own ranks.

This last element of the plot corresponded uncomfortably with the mood inside the project. LucasArts had now moved out of the idyllic environs of Skywalker Ranch and into a sprawling, anonymous office complex, where the designers and programmers working on The Dig found themselves in a completely separate building from the artists and administrators. Reading just slightly between the lines here, the root of the project’s troubles seems to have been a marked disconnect between the two buildings. Moriarty, who felt compelled to create meaningful, thematically ambitious games, became every accountant and project planner’s nightmare, piling on element after element, flying without a net (or a definitive design document). He imagined an interface where you would be able to carry ideas around with you like physical inventory items, a maze that would reconfigure itself every time you entered it, a Klein bottle your characters would pass through with strange metaphysical and audiovisual effects. To make all this happen, his programmers would need to create a whole new game engine of their own rather than relying on SCUMM. They named it StoryDroid.

A screenshot from Moriarty’s version of The Dig. Note the menu of verb icons at the bottom of the screen. These would disappear from later versions in favor of the more streamlined style of interface which LucasArts had begun to employ with Sam and Max Hit the Road.

There were some good days on Moriarty’s Dig, especially early on. Bill Tiller, an artist on the project, recalls their one in-person meeting with Steven Spielberg, in his office just behind the Universal Studios Theme Park. Moriarty brought a demo of the work-in-progress, along with a “portable” computer the size of a suitcase to run it. And he also brought a special treat for Spielberg, who continued to genuinely enjoy games in all the ways George Lucas didn’t. Tiller:

Brian brought an expansion disk for one of the aerial battle games Larry Holland was making. Spielberg was a big computer-game geek! He was waiting for this upgrade/mission expansion thing. He called his assistant in and just mentioned what it was. She immediately knew what he meant and said she’d send it home and tell someone to have it installed and running for him when he arrived. I decided at that moment I would have an assistant like that someday.

Anyway, when we were through we told him we had a few hours to kill and wondered what rides we should get on back at the theme park. He said the E.T. ride, since he helped design it. It was brand new at the time. His people said that he was really crazy about it and wanted to show it off to everyone. One of his assistants took us there on a back-lot golf cart. We didn’t have to get another taxi. We didn’t even have to stand in line! They took us straight to the ride and cut us in the line in front of everyone, like real V.I.P.s. Everyone had to stand back and watch, probably trying to figure out who we were. All I remember is Brian with the stupid giant suitcase going through the ride.

But the best part of the whole thing for me was [Spielberg’s] enthusiasm. He really likes games. This wasn’t work to him to have to hear us go on about The Dig.

Brian Moriarty’s version of The Dig was more violent than later versions, a quality which Steven Spielberg reportedly encouraged. Here an astronaut meets a gruesome end after being exposed to an alien acid rain.

But the bonhomie of the Universal Studios visit faded as the months wore on. Moriarty’s highfalutin aspirations began to strike others on the team — especially the artists who were trying to realize his ever-expanding vision — as prima-donna-ish; at the end of the day, after all, it was just a computer game they were making. “I used to tell Brian, when he got all excited about what people would think of our creation, that in ten years no one will even remember The Dig,” recalls Bill Eaken, the first head artist to work under him. He believes that Moriarty may even have imagined Spielberg giving him a screenwriting or directing job if The Dig sufficiently impressed him. Eaken:

I liked Brian. Brian is a smart and creative guy. I still have good memories of sitting in his office and just brainstorming. The sky was the limit. That’s how it should be. Those were good times. But I think as time went on he had stars in his eyes. I think he wanted to show Spielberg what he could do and it became too much pressure on him. After a while he just seemed to bog down under the pressure. When all the politics and Hollywood drama started to impede us, when it wasn’t even a Hollywood gig, I [got] temperamental.

The programming was a complete disaster. I had been working for several years at LucasArts at that time and had a very good feel for the programming. I taught programming in college, and though I wasn’t a programmer on any games, I understood programming enough to know something was amiss on The Dig. I went to one of my friends at the company who was a great programmer and told him my concerns. He went and tried to chat with the programmers about this or that to get a look at their code, but whenever he walked into the room they would shut off their monitors, things like that. What he could see confirmed my worries: the code was way too long, and mostly not working.

The project was “completely out of control and management wouldn’t listen to me about it,” Eaken claims today. So, he quit LucasArts, whereupon his role fell to his erstwhile second-in-command, the aforementioned Bill Tiller. The latter says that he “liked and disliked” Moriarty.

Brian was fun to talk with and was very energetic and was full of good ideas, but he and I started to rub each other the wrong way due to our disagreement over how the art should be done. I wanted the art organized in a tight budget and have it all planned out, just like in a typical animation production, and so did my boss, who mandated I push for an organized art schedule. Brian bristled at being restricted with his creativity. He felt that the creative process was hindered by art schedules and strict budgets. And he was right. But the days of just two or three people making a game were over, and the days of large productions and big budgets were dawning, and I feel Brian had a hard time adjusting to this new age.

Games were going through a transition at that time, from games done by a few programmers with little art, to becoming full-blown animated productions where the artists outnumber the programmers four to one. Add to the mix the enormous pressure of what a Spielberg/Lucas project should be like [and] internal jealousy about the hype, and you have a recipe for disaster.

He wanted to do as much of the game by himself as possible so that it was truly his vision, but I think he felt overwhelmed by the vastness of the game, which required so much graphics programming and asset creation. He was used to low-res graphics and a small intimate team of maybe four people or less. Then there is the pressure of doing the first Spielberg/Lucas game. I mean, come on! That is a tough, tough position for one guy to be in.

One of LucasArt’s longstanding traditions was the “pizza orgy,” in which everyone was invited to drop whatever they were doing, come to the main conference room, eat some pizza, and play a game that had reached a significant milestone in its development. The first Dig pizza orgy, which took place in the fall of 1993, was accompanied by an unusual amount of drama. As folks shuffled in to play the game for the very first time, they were told that Moriarty had quit that very morning.

We’re unlikely ever to know exactly what was going through Moriarty’s head at this juncture; he’s an intensely private individual, as it is of course his right to be, and is not at all given to baring his soul in public. What does seem clear, however, is that The Dig drained from him some fragile reservoir of heedless self-belief which every creative person needs in order to keep creating. Although he’s remained active in the games industry in various roles, those have tended to be managerial rather than creative; Brian Moriarty, one of the best pure writers ever to explore the potential of interactive narratives, never seriously attempted to write another one of them after The Dig. In an interview he did in 2006 for Jason Scott’s film Get Lamp, he mused vaguely during a pensive interlude that “I’m always looking for another Infocom. But sometimes I think we won’t give ourselves permission.” (Who precisely is the “we” here?) This statement may, I would suggest, reveal more than Moriarty intended, about more of his career than just his time at Infocom.

At any rate, Moriarty left LucasArts with one very unwieldy, confused, overambitious project to try to sort out. It struck someone there as wise to give The Dig to Hal Barwood, a former filmmaker himself who had been friends with Steven Spielberg for two decades. But Barwood proved less than enthusiastic about it — which was not terribly surprising in light of how badly The Dig had already derailed the careers of two of LucasArts’s other designers. Following one fluffy interview where he dutifully played up the involvement of Spielberg for all it was worth — “We’re doing our best to capture the essence of the experience he wants to create” — he finagled a way off the project.

At this point, the hot potato was passed to Dave Grossman, who had, as noted above, worked for a time with Noah Falstein on its first incarnation. “I was basically a hedge trimmer,” he says. “There was a general feeling, which I shared, that the design needed more work, and I was asked to fix it up while retaining as much as possible of what had been been done so far — starting over yet again would have been prohibitively expensive. So I went in with my editing scissors, snip snip snip, and held a lot of brainstorming meetings with the team to try to iron out the kinks.” But Grosssman too found something better to do as quickly as possible, whereupon the game lay neglected for the better part of a year while much of Moriarty’s old team went to work on Tim Schafer’s Full Throttle: “a project that the company loved,” says Bill Tiller, drawing an implicit comparison with this other, unloved one.

In late 1994, The Dig was resurrected for the last time, being passed to Sean Clark, a long-serving LucasArts programmer who had moved up to become the producer and co-designer of Sam and Max Hit the Road, and who now saw becoming the man who finally shepherded this infamously vexed project to completion as a good way to continue his ascent. “My plan when I came in on the final incarnation was to take a game that was in production and finish it,” he says. “I didn’t get a lot of pressure or specific objectives from management. I think they were mainly interested in getting the project done so they could have a product plan that didn’t have The Dig listed on it.” Clark has admitted that, when he realized what a sorry state the game was actually in, he went to his bosses and recommended that they simply cancel it once and for all. “I got a lot of resistance, which surprised me,” he says. “It was hard to resist the potential [of having] a game out there with a name like Spielberg’s on it.” In a way, George Lucas was a bigger problem than Spielberg in this context: no one wanted to go to the boss of bosses at LucasArts and tell him they had just cancelled his close friend’s game.

Sean Clark with a hot slice. Pizza was a way of life at LucasArts, as at most games studios. Asked about the negative aspects of his job, one poor tester said that he was “getting really, really tired of pizza. I just can’t look at pizza anymore.”

So, Clark rolled up his sleeves and got to work instead. His first major decision was to ditch the half-finished StoryDroid engine and move the project back to SCUMM. He stuck to Brian Moriarty’s basic plot and characters, but excised without a trace of hesitation or regret anything that was too difficult to implement in SCUMM or too philosophically esoteric. His goal was not to create Art, not to stretch the boundaries of what adventure games could be, but just to get ‘er done. Bill Tiller and many others from the old team returned to the project with the same frame of reference. By now, LucasArts had moved offices yet again, to a chic new space where the programmers and artists could mingle: “Feedback was quick and all-encompassing,” says Tiller. If there still wasn’t a lot of love for the game in the air, there was at least a measure of esprit de corps. LucasArts even sprang for a couple more (reasonably) big names to add to The Dig‘s star-studded marque, hiring the science-fiction author Orson Scott Card, author of the much-admired Ender’s Game among other novels, to write the dialogue, and Robert Patrick, Arnold Schwarzenegger’s principal antagonist from Terminator 2, to head up the cast of voice actors. Remarkably in light of how long the project had gone on and how far it had strayed from his original vision, Steven Spielberg took several more meetings with the team. “He actually called me at home one evening as he was playing through a release candidate,” says Sean Clark. “He was all excited and having fun, but was frustrated because he had gotten stuck on a puzzle and needed a hint.”

Clark’s practicality and pragmatism won the day where the more rarefied visions of Falstein and Moriarty had failed: The Dig finally shipped just in time for the Christmas of 1995. LucasArts gave it the full-court press in terms of promotion, going so far as to call it their “highest-profile product yet.” They arranged for a licensed strategy guide, a novelization by the king of tie-in novelists Alan Dean Foster, an “audio drama” of his book, and even a CD version of Michael Land’s haunting soundtrack to be available within weeks of the game itself. And of course they hyped the Spielberg connection for all it was worth, despite the fact that the finished game betrayed only the slightest similarity to the proposal he had pitched six years before.

Composer Michael Land plays a timpani for The Dig soundtrack. One can make a strong argument that his intensely atmospheric, almost avant-garde score is the best thing about the finished game. Much of it is built from heavily processed, sometimes even backwards-playing samples of Beethoven and Wagner. Sean Clark has described, accurately, how it sounds “strange and yet slightly familiar.”


But the reaction on the street proved somewhat less effusive than LucasArts might have wished. Reviews were surprisingly lukewarm, and gamers were less excited by the involvement of Steven Spielberg than the marketers had so confidently predicted. Bill Tiller feels that the Spielberg connection may have been more of a hindrance than a help in the end: “Spielberg’s name was a tough thing to have attached to this project because people have expectations associated with him. The general public thought this was going to be a live-action [and/or] 3D interactive movie, not an adventure game.” The game wasn’t a commercial disaster, but sold at less than a third the pace of Full Throttle, its immediate predecessor among LucasArts adventures. Within a few months, the marketers had moved on from their “highest-profile product yet” to redouble their focus on the Star Wars games that were accounting for more and more of LucasArts’s profits.

One can certainly chalk up some of the nonplussed reaction to The Dig to its rather comprehensive failure to match the public’s expectations of a LucasArts adventure game. In a catalog that consisted almost exclusively of cartoon comedies, it was a serious, even gloomy game. In a catalog of anarchically social, dialog-driven adventures that were seen by many gamers as the necessary antithesis to the sterile, solitary Myst-style adventure games that were now coming out by the handful, it forced you to spend most of its length all alone, solving mechanical puzzles that struck many as painfully reminiscent of Myst. Additionally, The Dig‘s graphics, although well-composed and well-drawn, reflected the extended saga of its creation; they ran in low-resolution VGA at a time when virtually the whole industry had moved to higher-resolution Super VGA, and they reflected as well the limitations of the paint programs and 3D-rendering software that had been used to create them, in many cases literally years before the game shipped. In the technology-obsessed gaming milieu of the mid-1990s, when flash meant a heck of a lot, such things could be ruinous to a new release’s prospects.

But today, we can presumably look past such concerns to the fundamentals of the game that lives underneath its surface technology. Unfortunately, The Dig proves far from a satisfying experience even on these terms.

An adventure game needs to be, if nothing else, reasonably good company, but The Dig fails this test. In an effort to create “dramatic” characters, it falls into the trap of merely making its leads unlikable. All of them are walking, talking clichés: the unflappable Chuck Yeager-type who’s in charge, the female overachiever with a chip on her shoulder who bickers with his every order, the arrogant German scientist who transforms into the villain of the piece. Orson Scott Card’s dialog is shockingly clunky, full of tired retreads of action-movie one-liners; one would never imagine that it comes from the pen of an award-winning novelist if it didn’t say so in the credits. And, even more unusually for LucasArts, the voice acting is little more inspired. All of which is to say that it comes as something of a relief when everyone else just goes away and leaves Boston Low alone to solve puzzles, although even then you still have to tolerate Robert Patrick’s portrayal of the stoic mission commander; he approaches an unknown alien civilization on the other side of the galaxy with all the enthusiasm of a gourmand with a full belly reading aloud from a McDonald’s menu.

Alas, one soon discovers that the puzzle design isn’t any better than the writing or acting. While the puzzles may have some of the flavor of Myst, they evince none of that game’s rigorous commitment to internal logic and environmental coherence. In contrast to the free exploration offered by Myst, The Dig turns out to be a quite rigidly linear game, with only a single path through its puzzles. Most of these require you just to poke at things rather than to truly enter into the logic of the world, meaning you frequently find yourself “solving” them without knowing how or why.

But this will definitely not happen in at least two grievous cases. At one point, you’re expected to piece together an alien skeleton from stray bones when you have no idea what said alien is even supposed to look like. And another puzzle, involving a cryptic alien control panel, is even more impossible to figure out absent hours of mind-numbing trial and error. “I had no clue that was such a hard puzzle,” says Bill Tiller. “We all thought it was simple. Boy, were we wrong.” And so we learn the ugly truth: despite the six years it spent in development, nobody ever tried to play The Dig cold before it was sent out the door. It was the second LucasArts game in a row of which this was true, indicative of a worrisome decline in quality control from a studio that had made a name for themselves by emphasizing good design.

At the end of The Dig, the resolution of the alien mystery is as banal as it is nonsensical, a 2001: A Space Odyssey with a lobotomy. It most definitely isn’t “an in-depth story in which the exploration of human emotion plays as important a role as the exploration of a game world,” as LucasArts breathlessly promised.

So, The Dig still manages to come across today as simultaneously overstuffed and threadbare. It broaches a lot of Big Ideas (a legacy of Falstein and Moriarty’s expansive visions), but few of them really go anywhere (a legacy of Grossman and Clark’s pragmatic trimming). It winds up just another extended exercise in object manipulation, but it doesn’t do even this particularly well. Although its audiovisuals can create an evocative atmosphere at times, even they come across too often as disjointed, being a hodgepodge of too many different technologies and aesthetics. Long experience has taught many of us to beware of creative expressions of any stripe that take too long to make and pass through too many hands in the process. The Dig only proves this rule: it’s no better than its tortured creation story makes you think it will be. Its neutered final version is put together competently, but not always well, and never with inspiration. And so it winds up being the one thing a game should never be: it’s just kind of… well, boring.

As regular readers of this site are doubtless well aware, I’m a big fan of LucasArts’s earlier adventures of the 1990s. The one complaint I’ve tended to ding them with is a certain failure of ambition — specifically, a failure to leave their designers’ wheelhouse of cartoon comedy. And yet The Dig, LucasArts’s one concerted attempt to break that mold, ironically winds up conveying the opposite message: that sometimes it’s best just to continue to do what you do best. The last of their charmingly pixelated “classic-look” adventure games, The Dig is sadly among the least satisfying of the lot, with a development history far more interesting than either its gameplay or its fiction. A number of people looked at it with stars in their eyes over the six years it remained on LucasArts’s list of ongoing projects, but it proved a stubbornly ill-starred proposition for all of them in the end.

(Sources: the book The Dig: Official Player’s Guide by Jo Ashburn; Computer Gaming World of March 1994, September 1994, September 1995, October 1995, December 1995, and February 1996; Starlog of October 1985; LucasArts’s customer newsletter The Adventurer of Spring 1993, Winter 1994, Summer 1994, Summer 1995, and Winter 1995. Online sources include Noah Falstein’s 2017 interview on Celebrity Interview, Falstein’s presentation on his history with Lucasfilm Games for Øredev 2017, the “secret history” of The Dig at International House of Mojo, the same site’s now-defunct Dig Museum,” ATMachine’s now-defunct pages on the game, Brian Moriarty’s 2006 interview for Adventure Classic Gaming, and Moriarty’s Loom postmortem at the 2015 Game Developers Conference. Finally, thank you to Jason Scott for sharing his full Get Lamp interview archives with me years ago.

The Dig is available for digital purchase on GOG.com.)

Footnotes

Footnotes
1 LucasArts was known as Lucasfilm Games until the summer of 1992. To avoid confusion, I use the name “LucasArts” throughout this article.
 

Tags: , , , ,

Indiana Jones and the Fate of Atlantis (or, Of Movies and Games and Whether the Twain Shall Meet)

You ask why there are movements in movie history. Why all of a sudden there are great Japanese films, or great Italian films, or great Australian films, or whatever. And it’s usually because there are a number of people that cross-pollinated each other.

— Francis Ford Coppola

Over the course of the late 1970s and early 1980s, George Lucas and Steven Spielberg remade the very business of film-making for better or for worse, shifting its focus from message movies and character dramas to the special-effects-heavy, escapist blockbusters that still drive Hollywood’s profits to this day. These events, which we might call the rise of the culture of the blockbuster, have long since been enshrined into the canonical history of film, filed under the names of these two leading lights.

Yet no two personalities could possibly have brought about such a dramatic shift on their own. Orbiting around Lucas and Spielberg was an entire galaxy of less prominent talents whose professional lives were profoundly affected by their association with these two new faces of modern Hollywood. They were the fellow travelers who helped Lucas and Spielberg to change the movie industry, all without ever quite being aware that they were part of any particular movement at all. Among the group were names like John Milius, Walter Murch, Willard Huyck, Randall Kleiser, Matthew Robbins, and Hal Barwood. “I can’t speak for the others,” says Robbins, “but it was my impression that nobody had the foggiest idea that there was any ‘next wave’ coming. Nobody had set their sights — except perhaps for George [Lucas].”

Out of this group of slightly lesser but undeniably accomplished lights, Hal Barwood is our special person of interest for today. He first met George Lucas in the mid-1960s, when the two were students together in the University of Southern California’s film program. Lucas, who had only recently abandoned his original dream of driving race cars in favor of this new one of making movies, was shy almost to the point of complete inarticulateness, and was far more comfortable futzing over a Moviola editing machine than he was trying to cajole live actors into doing his bidding. Nevertheless, there was a drive to this awkward young man that gradually revealed itself over the course of a longer acquaintance, and Barwood — on the surface, a far more assertive, impressive personality — soon joined Lucas’s loose clique in the role of follower rather than leader. Flying in the face of a Hollywood culture which valued realistic character dramas above all else, Lucas, Barwood, and their pals loved science fiction and fantasy, didn’t consider escapism to be a dirty word, and found the visual aesthetics of film to be every bit as interesting as their actors’ performances.

The bond thus forged would remain strong for many years. When Lucas, through the intermediation of his friend and mentor Francis Ford Coppola, got the chance to direct an actual feature film, Barwood added the first professional film credit to his own CV by helping out with the special effects on what became THX-1138, a foreboding, low-budget work of dystopic science fiction that was released to little attention in 1971. When Lucas hit it big two years later with his very different second film, the warmly nostalgic coming-of-age story American Graffiti, he bought a big old ramshackle mansion in San Anselmo, California, with the first of the proceeds, and Barwood became one of several fellow travelers who joined him and his wife in this first headquarters of a nascent corporate entity called Lucasfilm.

George Lucas, standing at far right, discusses Star Wars with his sounding board in the mid-1970s. Steve Spielberg is wearing the orange cap, and Hal Barwood sits on the fence behind him.

Working together with his good friend Matthew Robbins, Barwood wrote a science-fiction script called Star Dancing during the same period. The two commissioned a former Boeing technical illustrator and CBS animator named Ralph McQuarrie to paint some concept art, thereby to help them pitch the project to the studios. In the end, though, Star Dancing never went anywhere — until Lucas, who was toying with ideas for a second, more crowd-pleasing science-fiction movie of his own, saw McQuarrie’s paintings and was greatly inspired by them, hiring him to develop the vision further for what would become Star Wars. McQuarrie’s concept art had much to do with the eventual green-lighting of Star Wars by 20th Century Fox, and he would continue to shape the look of that film and its two sequels over the long course of their production.

Barwood and Robbins, for their part, became two of the eight people entrusted to read the first draft of the film’s script. He and the others in that San Anselmo house then proceeded to slowly shape the Star Wars script we know today over the course of draft after draft.

Even as they helped Lucas with Star Wars, Barwood and Robbins were still trying to make it as screenwriters in their own right. They sold their first script, a chase caper called The Sugarland Express, to George Lucas’s up-and-coming pal Steven Spielberg, a more recently arrived member of the San Anselmo collective; he turned it into his feature-film directorial debut in 1974. More screenwriting followed, including an uncredited rewrite of the 1977 Spielberg blockbuster Close Encounters of the Third Kind, the first film to benefit in a big way commercially from the new interest in science fiction ignited by Star Wars, which had been released about six months prior to it.

Yet the pair found screenwriting to be an inherently frustrating profession in an industry which regarded the director as a movie’s ultimate creative voice. “In writing, you’re always watching directors ruin your stuff,” says Barwood. “As a writer, you have a certain flavor, style, and emphasis in mind when you write the script, and you’re always shocked when the director comes back with something else. There’s a tendency to want to get your hands on the controls and do it yourself.” Accordingly, Matthew Robbins personally directed the duo’s 1978 comedy Corvette Summer, starring Mark Hamill — Luke Skywalker himself — in his first big post-Star Wars role. The film was a commercial success, even if the reviews weren’t great; it turned out that there was only so much you could do with Hamill, the very archetype of an actor who’s good in one role and one role only.

The duo’s next big project that was their most ambitious, time-consuming, and expensive undertaking yet. Dragonslayer was a fantasy epic based loosely on the legend of St. George and the dragon, and was once again directed by Robbins. The special effects were provided by George Lucas’s Industrial Light and Magic. In fact, Dragonslayer became the very first outside, non-Star Wars project the famous effects house took on. Being at least as challenging as anything in any of the Star Wars films, the Dragonslayer effects took them some eighteen months to complete.

Dragonslayer‘s pedigree was such that it was widely heralded in the Hollywood trade press as a “surefire success” prior to its release. But it had the misfortune to arrive in theaters on June 26, 1981, two weeks after Lucas and Spielberg’s new blockbuster collaboration, Raiders of the Lost Ark. The latter film shared with Dragonslayer the distributor Paramount Pictures. “Paramount was quite satisfied to go through the summer with the money they were going to get from Raiders of the Lost Ark,” says Barwood, “and paid no attention to our movie. They just dropped it. They just forgot about it.” Dragonslayer flopped utterly, badly damaging the duo’s reputation inside Hollywood as purveyors of marketable cinema.

Barwood got his one and only chance to direct a feature film in 1985, when he took the reins of Warning Sign, yet another screenplay by himself and Robbins. In a telling sign of the damage Dragonslayer‘s failure had done to their careers, this latest film, a fairly predictable thriller about a genetic-engineering project run amok, had a budget of about one-quarter what the duo had had to work with on their fantasy epic. It garnered mediocre reviews and box-office receipts, and no one seemed eager to entrust its first-time director with more movies. Barwood and Robbins, both frustrated by the directions their careers had taken since Corvette Summer, decided their long creative partnership had run its course.

George Lucas (far left) and some of his old compatriots at Skywalker Ranch in the mid-1980s. Matthew Robbins is third from left, Hal Barwood fourth from left.

So, Hal Barwood found himself at something of a loose end as the 1980s drew to a close. He was still friendly with George Lucas, if perhaps not quite the bosom buddy he once had been, and he still knew many of the most powerful people in the movie industry, starting with Steven Spielberg — who had gradually shown himself to be, even more so than Lucas, the personification of the new, blockbuster-oriented Hollywood, his prolific career cruising along with hit after hit. But, as Spielberg basked in his success, Barwood had parted ways with his partner and seen his directorial debut become a bust. He hadn’t had a hand in a real hit since 1978, and he hadn’t sold a script at all in quite some time. Just to rub salt into the wounds, his old compatriot Matthew Robbins managed to score another modest hit of his own at last just after the breakup, with the distinctly Spielbergian science-fiction comedy Batteries Not Included, which he directed and whose screenplay he had written with others.

Perhaps it was time for Hal Barwood to try something completely different. He had actually been mulling over his future in this cutthroat industry for some time already. During the promotional tour for Warning Sign, he had made a rather odd comment to Starlog magazine, accompanied by what his interviewer described as a “nervous laugh”: “If movies don’t pan out for me, I have a second career lurking around the corner in entertainment software, working on animated computer games, which I’m doing right now. They’re very sophisticated, animated adventure games.”

Barwood had in fact been fascinated by computers for a long time, ever since he had first encountered the hulking number-crunching monstrosities of the 1960s at university. In 1980, while hanging around the set of Dragonslayer, he had programmed his first game on a state-of-the-art HP-41C calculator as a way of passing the time between takes. Soon after, he joined the PC Revolution, buying an Apple II and starting to tinker. He worked for years on a CRPG for that machine in his spare time — worked for so long that it was still in progress when the Apple II games market began to collapse. Undaunted, he moved on to a Macintosh, where he programmed a storyboarding system for movie makers like himself in HyperCard, selling it as shareware under the name of StoryCard.

With all this experience behind him, it was natural for Barwood now to consider a future in games instead of movies — a future in an industry where the budgets were smaller, the competitors were fewer, and it was much easier to come up with an idea and actually see it through from beginning to end in something like its original form. All of this made quite a contrast to the industry where he had cut his teeth. “The movie business is very difficult for most of us,” he says. “We don’t usually get a majority of our projects to completion. Most of our dreams turn into screenplays, but they stall out at that stage.”

Given his long association with George Lucas, Barwood decided to talk to Lucasfilm Games. They were more than happy to have him, and could certainly relate to the reasons that brought him to their doorstep — for, like Barwood, they had had a somewhat complicated life of it to date in the long shadow cast by Lucas.


Before there was Lucasfilm Games, there was the Lucasfilm Computer Division, founded in 1979 to experiment with computer animation and digital effects, technologies with obvious applications for the making of special-effects-heavy films. Lucasfilm Games had been almost literally an afterthought, an outgrowth of the Computer Division that was formed in 1982, a time when George Lucas and Lucasfilm were flying high and throwing money about willy-nilly.

In those days, a hit computer game, one into which Lucasfilm Games had poured their hearts and souls, might be worth about as much to the parent company’s bottom line as a single Jawa action figure — such was the difference in scale between the computer-games industry of the early 1980s and the other markets where Lucasfilm was a player. George Lucas personally had absolutely no interest in or understanding of games, which didn’t do much for the games division’s profile inside his company. And, most frustrating of all for the young developers who came to work for The House That Star Wars Built, they weren’t allowed to make Star Wars games — nor, for that matter, even Indiana Jones games — thanks to Lucas having signed away those rights to others at the height of the Atari VCS fad. Noah Falstein, one of those young developers, would later characterize this situation as “the best thing that could have happened” to them, as it forced them to develop original fictions instead — leading, he believes, to better, more original games. At the time, however, it couldn’t help but frustrate that the only Lucasfilm properties the games division had access to were middling fare like Labyrinth.

Still, somebody inside Lucasfilm apparently believed in the low-profile division’s potential, for it survived the great bloodletting of the mid-1980s. In response to George Lucas’s expensive divorce settlement and the realization that, with the Star Wars trilogy now completed, there would be no more enormous guaranteed paydays in the future, the company’s executives, with Lucas’s tacit blessing, took an axe to many of their more uncertain or idealistic ventures at that time. Among the divisions that were sold off was the rest of the Computer Division that had indirectly spawned Lucasfilm Games; it would go on to worldwide fame and fortune under the name of Pixar. As for the games people: in 1986, they got to move into some of the vacant space all the downsizing had opened up at Skywalker Ranch, Lucasfilm’s sprawling summer camp cum corporate campus in Marin County, California.

The wall separating Lucasfilm Games from the parent company’s most desirable intellectual properties finally began to fall at the end of the 1980s, when the games people were given access to… no, not yet to Star Wars, but to the next best thing: Indiana Jones, George Lucas’s other great cinematic success story. Raiders of the Lost Ark, the first breakneck tale of the adventurous 1930s archaeologist, as conceived by Lucas and passed on to Steven Spielberg to direct, had become the highest-grossing film of 1981 by nearly a factor of two over its nearest competitor; as we’ve seen, it had trampled less fortunate rivals like poor Hal Barwood’s Dragonslayer into the dust during that year’s summer-blockbuster season. A 1984 sequel, Indiana Jones and the Temple of Doom, had done nearly as well. Now a third and presumably final film, to be called Indiana Jones and the Last Crusade, was in the offing. With the earlier licensing deals they had made for the property now expired, the parent company wanted their games division to make an adventure game out of it.

Indiana Jones and the Last Crusade: The Graphic Adventure [1]An Action Game was also published under the auspices of Lucasfilm Games, but its development was outsourced to a British house. was designed by Noah Falstein, David Fox, and Ron Gilbert, all of whom had worked on previous Lucasfilm adventure games, and written using the division’s established SCUMM adventuring engine. This committee approach to the game’s design is typical of the workaday nature of the project as a whole. The designers were given a copy of the movie’s shooting script, and were expected not to deviate too much from it. Ron Gilbert, a comedy writer by disposition and talent, found the need to play it relatively straight particularly frustrating, but it seems safe to say that all of the designers’ creative instincts were somewhat hemmed in by the project’s fixed rules. The end result, while competently executed, hasn’t the same vim and vinegar as Maniac Mansion, the first SCUMM adventure, nor even Zak McKracken and the Alien Mindbenders, the rather less satisfying second SCUMM game.

The boxing scene which opens the game of Indiana Jones and the Last Crusade was a part of the film’s original script which was cut in the editing room. Scenes like these make the game almost of more interest to film historians than game historians, serving as a window into the movie as it was conceived by its screenwriter Jeffrey Boam.

Perhaps the most interesting aspect of the game arises from the fact that its designers were adapting from the shooting script rather than the finished movie, which they got to see in the Skywalker Ranch theater only when their own project was in the final stages of bug-swatting and polishing. They actually implemented parts of Jeffrey Boam’s script for the movie far more faithfully than Steven Spielberg wound up doing, including numerous scenes — like the boxing sequence at the very beginning of the game — that wound up being cut from the movie. Nevertheless, the game suffers from the fundamental problem of all such overly faithful adaptations from other media: if you’ve seen the movie — and it seemed safe to assume that just about everybody who played the game had seen the movie — what’s the point in walking through the same story again in game form? The designers went to considerable lengths to accommodate curious (or cantankerous) players who make different choices from those of Indiana Jones in the movie, turning their choices into viable alternative pathways rather than mere game overs. But there was only so much they could do in even that respect, given the constraints under which they labored.

Of course, licensed games exist first and foremost because licenses sell games, not because they lead to better ones. Indiana Jones and the Last Crusade became the cinematic blockbuster of the year upon its release in May of 1989, and the game also did extremely well when it hit stores a few weeks later. Having taken this first step into the territory of Lucasfilm’s biggest franchises and been so amply rewarded for it, the people at the games division naturally wanted to keep the good times going. There were likely to be no more Indiana Jones movies; Harrison Ford, the series’s famously prickly star, was publicly declaring himself to be through with playing the character. But did that necessarily mean that there couldn’t be more Indiana Jones games? With the license now free and clear for their use, no one at Lucasfilm Games saw any reason to assume so.


It was at this juncture that Hal Barwood entered the picture, interested in trying a new career in games on for size. Just about any developer in the industry would have jumped at the chance to bring someone like him aboard. Talk of a merger of games with cinema to create a whole new genre of mass-media entertainment — the interactive movie, preferably published on CD-ROM complete with voice acting and perhaps even real-world video footage — dominated games-industry conferences and magazine editorials alike as the 1990s began. But for all their grandiose talk, the game developers clustered in Northern California were all too aware that they lacked the artistic respectability necessary to tempt most of those working with traditional film and video in the southern part of the state into working with them on interactive projects. Hal Barwood might have been no more than a mid-tier Hollywood player at best, but suffice to say that there wasn’t exactly a surfeit of other Hollywood veterans of any stripe who were willing to work on games.

In an online CompuServe conference involving many prominent adventure-game designers that was held on August 24, 1990, not long after Barwood’s arrival at Lucasfilm Games, Noah Falstein could hardly keep himself from openly taunting his competitors about what he had and they didn’t:

One way we’re trying to incorporate real stories into games is to use real storytellers. Next year, we have a game coming out by Hal Barwood, who’s been a successful screenwriter, director, and producer for years. His most well-known movies probably are the un-credited work he did on Close Encounters and Dragonslayer, which he co-wrote and produced. He’s also programmed his own Apple II games in 6502 assembly in his spare time. I’ve already learned a great deal about pacing, tension, character, and other “basic” techniques that come naturally — or seem to — to him. I highly recommend such collaborations to you all. I think we’ve got a game with a new level of story on the way.

Falstein was in a position to learn so much from Barwood because he had been assigned to work with him as his design partner on his first project — the idea being that Barwood would take care of all the “basic” cinematic techniques Falstein enthuses about above, while Falstein would keep the project on track and make sure it worked as a playable adventure game, with soluble puzzles and all the rest.

Ironically given what Raiders of the Lost Ark had done to Dragonslayer, Hal Barwood’s one big chance to become a truly major Hollywood player, the game in question was to be another Indiana Jones game — albeit one with an original story, not bound to any movie. The initial plan had been to sift through the pile of rejected scripts for the third Indiana Jones film and select a likely candidate from them for adaptation into a game. But it turned out that the scripts were all pretty bad, or at least not terribly suitable for interactive adaptation.

The Azores is one of the many exotic locations Indy visits in Indiana Jones and the Fate of Atlantis, all of which are brought to life with aplomb by Lucasfilm Games’s accomplished art team.

So, Barwood and Falstein decided to invent their own story, and thus went looking for legends of lost civilizations that might be worthy of an intrepid archeologist who had already found the Lost Ark of the Covenant and the Holy Grail. “George [Lucas] has established a criterion for Indiana Jones adventures,” said Barwood, “and it’s basically that he should only find things that actually existed — or at least could have existed.” The fabled sunken island of Atlantis seemed the right mixture of myth and history. Barwood:

Our eyes fell upon Atlantis because not only is it an ancient myth known by almost everyone, but it also has wonderful credentials, in that it was first mentioned by Plato a couple thousand years ago. In addition to that, in the early part of this century, the idea was taken over by spiritualists and mystics, who attributed to the Atlanteans this fantastic technology, with airships flying 100 milers per hour, powered by vrill and firestone. When we found this out, we thought to ourselves, “Does this sound as interesting as the Holy Grail? Yes, it does.” Even though it’s a myth, the myth is grounded in a wonderful collection of lore.

The legend’s wide-ranging wellsprings would allow them to send Indy traipsing between exotic locations scattered over much of the world: New York City, Iceland, Guatemala, the Azores, Algiers, Monte Carlo, Crete, Santorini, finally ending up under the ocean at the site of Atlantis itself.

The game known as Indiana Jones and the Fate of Atlantis hits all the notes familiar to anyone who has seen an Indiana Jones film. It seems that the mythical Atlanteans were quite a clever lot, having harnessed energies inconceivable to modern scientists. The Nazis have gotten wind of this, and are fast piecing together the clues that will let them find the undersea site of Atlantis, enter it, and take the technology for themselves, thereby to conquer the world. It’s a fine premise for a globetrotting story of thrills and spills — silly but no more silly than any of those that got made into movies. There’s even a female sparring partner/love interest for Indy, just like in the films. This time her name is Sophia, and she’s a former archaeologist who’s become a professional psychic, much to the scientific-minded Indy’s dismay. Let the battle of barbs begin!

Barwood’s first interactive script really is a good one, with deftly drawn plot beats and characters that, if not exactly deep, nobly fulfill their genre-story purposes as engines of action, tension, or comic relief. Other game writers of the early 1990s weren’t always or even usually all that adept at such basic techniques of fiction as “pacing, tension, and character.” To see how painful a game can be that wants to be like the Indiana Jones movies but lacks the writers to pull it off, one need look no further than Dynamix’s Heart of China, with its humor that lands with the most leaden of thuds, its hero who wants to be a charming rogue but misplaced his charm, and its dull supporting characters who are little more than crude ethnic stereotypes. When you play Fate of Atlantis, by contrast, you feel yourself to be in the hands of a writer who knows exactly what he’s doing.

The game is full of callbacks to the movies’ most famous catchphrases.

That said, the degree to which Indiana Jones and the Fate of Atlantis is truly cinematic can be and too often is overstated. Games are not movies — an obvious point that game developers of the early 1990s frequently lost sight of in their rush to make their interactive movies. Hal Barwood, the Hollywood veteran brought into Lucasfilm Games to apply his cinematic expertise, was ironically far more aware of this fact than were many other game designers who lacked his experience in the medium they were all so eager to ape. Speaking to the science-fiction writer and frequent games-industry commentator Orson Scott Card in 1990, Barwood made some telling observations:

“The companies making animated games keep talking as if games resembled movies,” he [Barwood] said. “But they don’t resemble movies all that much.”

He granted some resemblances, of course, especially with animated film. The dependence on artists; the trickle rate of production, where you’re producing the game or film at the rate of only minutes, or even seconds, of usable footage a day; and the dominant role of the editing process.

Still, though, when it comes to the art of composing a game, inventing it, he said, “What it really resembles is theater. Plays.”

Why? Because as with a play, you have only a few settings you can work with, and they can usually be viewed from only a single angle and at the same distance. You can’t do any meaningful work with closeups (to design and program genuine realistic facial expressions just isn’t worth the huge investment in time and disk space). It’s so hard to make actions clear that you must either rely on dialogue, like most plays, or show only the simplest, most obvious actions.

In movies, it’s just the opposite. You control the pace and rhythm of film by cutting and shifting the action from place to place. The camera never gazes at any one thing for long.

The computer games of this era which most clearly did understand the kinetic language of cinema — the language of a roving camera and a keen-eyed editor — weren’t any of the avowed interactive movies that were being presented in the form of plot- and dialog-heavy adventure games, but rather the comparatively minimalist action games Prince of Persia and Another World, both essentially one-man productions that employed few to no words. Both of these games were aesthetically masterful, but somewhat more problematic in terms of providing their players with interesting forms of interactivity, thus inadvertently illustrating some of the drawbacks of fetishizing movies as the ideal aesthetic model for games.

All the other people who thought they were making interactive movies were “filming” their productions the way only the very earliest movie directors had filmed, before a proper language of film had been created: through a single static “camera.” The end results were anything but cinematic in the way a fellow like Hal Barwood, steeped throughout his life in the language of film, understood that term. His long experience in film-making allowed him to see the essential fact that games were not movies. They might borrow the occasional technique from cinema, but games were a medium — or, perhaps better stated, a matrix of mediums, only one of which was the point-and-click adventure game — with their own unique sets of aesthetic affordances. Countless game developers seemed to be using the term “interactive movie” to designate any game that had a lot of story, but the qualities of being cinematic and being narrative-oriented were really orthogonal to one another.

As in the first Indiana Jones movie, a Nazi submarine features prominently in Fate of Atlantis.

In later years, Hal Barwood would describe a narrative-driven game as more akin to a vintage Russian novel than a movie, a “continuous experience in a fictional world”: something the player lives with over a period of days or weeks, working through it at her own pace, mulling it over even when she isn’t actively sitting in front of the computer. The control or lack thereof of pacing is a critical distinction: a game which leaves any reasonable scope of agency to its player must necessarily cede much or all of the control of pacing to her. And yet pacing is absolutely key to the final effect of any movie, so much so that the director may very well spend months in the editing room after all the shooting is done, trying to get the pacing just right. The game designer doesn’t have anything like the same measure of direct control over the player’s experience, and so must deliver a very different sort of fiction.

Indiana Jones and the Fate of Atlantis works because it understands these differences in media. It plays with the settings, characters, and themes of the Indiana Jones movies to fine effect, but never forgets that it’s an adventure game. The driving mechanic of an adventure game — the solving of intellectual puzzles — is quite distinct from that which drives the movies, and the plot points must be adapted to match. Can you imagine the cinematic Indy sticking wads of chewing gum to the soles of his shoes so as to climb up a coal chute? Or using a bathroom plunger as a handy replacement for a missing control lever inside a Nazi submarine? Puzzles of this nature inevitably make Indy himself into a rather different sort of personality — one a little less cool and collected, a little more prone to be the butt of the joke rather than the source of it. It’s clear from the very opening scenes of Fate of Atlantis, an innovative interactive credits sequence in which poor Indy must endure pratfall after pratfall, that this isn’t quite the same character we thrilled to in the movies. Harrison Ford would have walked out if asked to play a series of scenes like these.

This phenomenon, which we might called the Guybrush Threepwoodization of the hero, is very common in adventure games adapted from other media, given that the point-and-click adventure game as a medium wants always to collapse back into comedy as its default setting. (See, for example, the Neuromancer game, which similarly turns the cool cat Case from its source novel into a put-upon loser, and winds up becoming a pretty great game in the process.) Barwood and Falstein as well decided that Indiana Jones must adapt to the adventure-game genre rather than the adventure-game genre adapting to Indiana Jones. This was most definitely the right approach to take, and is the overarching reason why this game succeeds when so many other interactive adaptations fail.

The one place where the otherwise traditionalist Fate of Atlantis clearly does try to do something new has nothing directly to do with its source material. It rather takes the form of three possible narrative through lines, depending on the player’s individual predilections. After playing through the introductory stages of the game, you’re given a choice between a “Team” path, where Indy and Sophia travel together and cooperate to solve problems; a “Wits” path, where Indy travels largely alone and solves problems using his noggin; and a “Fists” path, where Indy travels alone and solves some though by no means all of his problems using more, shall we say, active measures, which translate into little action-oriented minigames for the player. The last is seemingly the closest to the spirit of the films, but is, tellingly, almost universally agreed to be the least interesting way to play the game.

The Team path, like all of them, has its advantages and disadvantages. It’s great to have Sophia around to help out with things — until she falls down a pit and needs your help getting out.

Although the message would get a little muddled once the game reached stores — “three games in one!” was a tagline few marketers could resist — Barwood and Falstein’s primary motivation in making these separate paths wasn’t to create a more replayable game, but rather a more accessible one. Lucasfilm Games always placed great emphasis on giving their players a positive, non-frustrating experience. Different players would prefer to play in different ways, Barwood and Falstein reasoned, and their game should adapt to that. “Socially-oriented” players — possibly including the female players they were always hoping to reach — would enjoy the Team path with its constant banter and pronounced romantic tension between Indy and Sophia; the stereotypically hardcore, cerebral adventure gamers would enjoy the Wits path; those who just wanted to get through the story and check out all the exotic destinations could go for the Fists path.

Falstein liked to call Fate of Atlantis a “self-tuning game.” In this spirit, until very late in development the branching pathways were presented not as an explicit choice but rather as a more subtle in-game personality test. Early on, Indy needs to get into a theater even though he doesn’t have a ticket. There are three ways to accomplish this: talking his way past the guard at the door; puzzling his way through a maze of boxes to find a hidden fire-escape ladder; or simply sucker-punching the guard. Thus would the player’s predilections be determined. In the interest of transparency and as a sop to replayability, however, the personality test wound up being replaced by a simple menu for choosing your pathway.

The first substantial interactive scene in the game, taking place outside and inside a theater in New York City where Indy’s adventuring partner-to-be Sophia is giving a lecture, was intended to function as a personality test of sorts, determining whether the player was sent down the Team, Wits, or Fists path. In the end, though, its finding were softened to a mere recommendation preceding an explicit choice of paths which is offered to the player at its conclusion.

The idea of multiple pathways turned out not to be as compelling in practice as in theory. Most players did take it more as an invitation to play the game three times than an opportunity to play it once their way, and were disappointed to discover that the branching pathways encompass only about 60 percent of the game as a whole; the first 10 percent or so, as well as the lengthy climax, are the same no matter which pathway you choose. Nor are the Team and Wits pathways different enough from one another to give the game all that much of a different flavor; they both ultimately come down to solving a series of logic puzzles. The designers’ time would probably have been better spent making one pathway through the game that combined elements of the Team and Wits pathways. Lucasfilm Games never tried anything similar again. The branching pathways were an experiment, and in that sense at least they served their purpose.

A substantial but by no means enormous game, Indiana Jones and the Fate of Atlantis nevertheless spent some two years in development, a lengthy span of time indeed by the standards of the early 1990s, being at least three times as long as it had taken to make the Last Crusade game. The protracted development cycle wasn’t a symptom of acrimony, lack of focus, or disorganization, as such things so often tend to be. It was rather a byproduct of the three pathways and, most of all, of Lucasfilm Games’s steadfast commitment to getting everything right, prioritizing quality and polish over release dates in that way that always set them apart from the majority of their peers.

Shortly before the belated release of Fate of Atlantis in the summer of 1992, Lucasfilm Games became LucasArts. The slicker, less subservient appellation was a sign of their rising status within the hierarchy of their parent company, as their games sold in bigger quantities and became a substantial revenue stream in their own right, less and less dwarfed by the money that could be made in movies. Those changing circumstances would prove a not-unmixed blessing for them, forcing them to move out of the rustic environs of Skywalker Ranch and shed much of the personality of a quirky artists’ collective for that of a more hard-nosed media enterprise. On the other hand, at least they’d finally get to make Star Wars games…

But that’s an article for another day. I should conclude this one by noting that Indiana Jones and the Fate of Atlantis [2]Once again, there was also a Fate of Atlantis action game, made in Britain with a particular eye to the 8-bit machines in Europe which couldn’t run the adventure game. And once again, it garnered little attention in comparison to its big brother. was greeted with superlative reviews and equally strong sales; even Steven Spielberg, who unlike his friend George Lucas was a big fan of games, played through it and reportedly enjoyed it very much. A year after the original floppy-disk-based release, LucasArts made a “talkie” version for CD-ROM. Getting Harrison Ford to play Indiana Jones was, as you might imagine, out of the question, but they found a credible soundalike, and handled the voice acting as a whole with their usual commitment to quality, recruiting professional voice talent in Hollywood and recording them in the state-of-the-art facilities of Skywalker Sound.

While hard sales numbers for LucasArts’s adventure games have never surfaced to my knowledge, Noah Falstein claims that Indiana Jones and the Fate of Atlantis sold the most of all of them — a claim I can easily imagine to be correct, given its rapturous critical reception and the intrinsic appeal of its license. Today, it tends to be placed just half a step down from the most-loved of the LucasArts adventures, lacking perhaps some of the manic inspiration of the studio’s completely original creations. Nonetheless, it’s a fine, fine game, well worth playing through twice or thrice — at least its middle section, where the pathways diverge — to experience all it has to offer. This game adapted from a movie franchise, which succeeds by not trying to be a movie, marked a fine start for Hal Barwood’s new career.

(Sources: the books The Secret History of Star Wars by Michael Kaminski and Droidmaker: George Lucas and the Digital Revolution by Michael Rubin; LucasArts’s Adventurer magazine of Fall 1991, Spring 1992, and Spring 1993; Starlog of July 1981, September 1981, August 1982, May 1985, September 1985, November 1985, December 1985, and February 1988; Amiga Format of February 1992; Compute! of February 1991; Computer Gaming World of September 1992; CU Amiga of June 1992; Electronic Games of October 1992; MacWorld of June 1989; Next Generation of October 1998; PC Review of September 1992; PC Zone of January 2000; Questbusters of September 1992; Zero of August 1991 and March 1992. Online sources include Arcade Attack‘s interviews with Noah Falstein and Hal Barwood; Noah Falstein’s Reddit AMA; MCV‘s articles on “The Early Days of LucasArts”; Noah Falstein’s presentation on LucasArts at Øredev 2017.

Indiana Jones and the Fate of Atlantis is available for purchase on GOG.com.)

Footnotes

Footnotes
1 An Action Game was also published under the auspices of Lucasfilm Games, but its development was outsourced to a British house.
2 Once again, there was also a Fate of Atlantis action game, made in Britain with a particular eye to the 8-bit machines in Europe which couldn’t run the adventure game. And once again, it garnered little attention in comparison to its big brother.
 
37 Comments

Posted by on September 28, 2018 in Digital Antiquaria, Interactive Fiction

 

Tags: , , , ,

Adventure-Game Rock Stars Live in Conference

On August 24, 1990, CompuServe hosted an online discussion on adventure-game design which included Ron Gilbert, Noah Falstein, Bob Bates, Steve Meretzky, Mike Berlyn, Dave Lebling, Roberta Williams, Al Lowe, Corey and Lori Ann Cole, and Guruka Singh Khalsa. This is, needless to say, an incredible gathering of adventuring star power. In fact, I’m not sure that I’ve ever heard of its like in any other (virtual) place. Bob Bates, who has become a great friend of this blog in many ways, found the conference transcript buried away on some remote corner of his hard drive, and was kind enough to share it with me so that I could share it with you today.

If you’re a regular reader of this blog, you probably recognize all of the names I’ve just listed, with the likely exception only of Khalsa. But, just to anchor this thing in time a bit better, let me take a moment to describe where each of them was and what he or she was working on that August.

Ron Gilbert and Noah Falstein were at Lucasfilm Games (which was soon to be renamed LucasArts). Gilbert had already created the classic Maniac Mansion a few years before, and was about to see published his most beloved creation of all, one that would have as great an impact among his fellow designers as it would among gamers in general: The Secret of Monkey Island. Falstein had created Indiana Jones and the Last Crusade for Lucasfilm in 1989. Their publisher had also recently released Brian Moriarty’s Loom, whose radically simplified interface, short length, and relatively easy puzzles were prompting much contemporaneous debate.

Bob Bates, Steve Meretzky, Mike Berlyn, and Dave Lebling had all written multiple games for the now-defunct Infocom during the previous decade. Bates had recently co-founded Legend Entertainment, where he was working on his own game Timequest and preparing to publish Spellcasting 101: Sorcerers Get All the Girls, Meretzky’s first post-Infocom game and Legend’s first game ever, in a matter of weeks. Berlyn had been kicking around the industry since leaving Infocom in 1985, creating perhaps most notably Tass Times in Tonetown for Interplay; he was just finishing up a science-fiction epic called Altered Destiny for Accolade, and would shortly thereafter embark on the Les Manley games, a pair of Leisure Suit Larry clones, for the same publisher. Lebling was at something of a loose end after the shuttering of Infocom the previous year, unsure whether he even wanted to remain in the games industry; he would eventually decide that the answer to that question was no, and would never design another game.

Roberta Williams, Al Lowe, Corey and Lori Ann Cole, and Guruka Singh Khalsa were all working at Sierra. Williams was in the latter stages of making her latest King’s Quest, the first to use 256-color VGA graphics and a point-and-click interface, and the first to be earmarked for CD-ROM as a “talkie.” Al Lowe was, as usual, hard at work on the latest Leisure Suit Larry game, which also utilized Sierra’s newer, prettier, parser-less engine. The Coles were just finishing up Quest for Glory II: Trial by Fire, which would become the last Sierra game in 16-color EGA and the last with a parser.

Khalsa is the only non-designer here, and, as already noted, the only name here with which longtime readers are unlikely to be familiar. He was another of those unsung heroes to be found behind the scenes at so many developers. At Sierra, he played a role that can perhaps best be compared to that played by the similarly indispensable Jon Palace at Infocom. As the “producer” of Sierra’s adventure games, he made sure the designers had the support they needed, acted as a buffer between them and the more business-oriented people, and gently pushed his charges to make their games just a little bit better in various ways. In keeping with his unsung status, he answers only one question here.

We find all of our participants grappling with the many tensions that marked their field in 1990: the urgent need to attract new players in the face of escalating development budgets; the looming presence of CD-ROM and other disruptive new technologies just over the horizon; the fate of text in this emerging multimedia age; the frustration of not always being able to do truly innovative or meaningful work, thanks to a buying public that largely just seemed to want more of the same old fantasy and comedy. It’s intriguing to see how the individual designers respond to these issues here, just as it is to see how those responses took concrete form in the games themselves. By no means is the group of one mind; there’s a spirited back-and-forth on many questions.

I’ve cleaned up the transcript that follows for readability’s sake, editing out heaps of extraneous comments, correcting spelling and grammar, and rejiggering the flow a bit to make everything more coherent. I’ve also added a few footnotes to clarify things or to insert quick comments of my own. Mostly, though, I’ve managed resist the urge to pontificate on any of what’s said here. You all already know my opinions on many of the topics that are raised. Today, I’m going to let the designers speak for themselves. I hope you’ll find their discussion as interesting and enjoyable as I do.


 

Let’s plunge right into the questions. Before I start, I’d like to thank Eeyore, Flying Gerbil, Steve Horton, Tsunami, Hercules, Mr. Adventure, and Randy Snow for submitting questions… and I apologize for mangling their questions with my editing. And now — drum roll! — on to the first question!

Imagine ourselves five years down the road, with all the technological developments that implies: CD-ROMs, faster machines, etc. Describe what, for you, the “ideal” adventure will look like. How will it be different from current adventures?

Roberta Williams: I think that “five years down the road” is actually just a year or two away. Meaning that a year or two from now, adventure games are going to have a very slick, sophisticated, professional look, feel, and sound to them, and that that’s the way they’re going to stay for a while — standardization, if you will. I mean, how can you improve on realistic images that look like paintings or photographs? How can you improve on CD-quality voices and music? How can you improve on real movement caught with a movie camera, or drawn by a professional animator? That’s the kind of adventure game that the public is going to start seeing within a year or two. Once adventure games reach a certain level of sophistication in look and feel, standardization will set in, which will actually be a boon for all concerned, both buyers and developers alike. After that, the improvements will primarily be in the performance on a particular machine, but the look will stay essentially the same for a while.

Dave Lebling: But if those wonderful pictures and hi-fi sound are driven by a clunky parser or a mythical “parser-less interface,” is this a big improvement? I think not. We can spend $2 million or $5 million developing a prettier version of Colossal Cave. Let’s improve the story and the interface! That doesn’t have to mean text adventures, but there’s more to adventure games than pictures.

Steve Meretzky: I think that in the future the scope of games won’t be limited by hardware but by the marketplace. Unless the market for adventure games expands, it won’t be economical to create super-large environments, even though the hardware is there to support them.

Mike Berlyn: Well, I think that technology can create products which drive the market and create end users — people who need or want to experience something they could experience only on a computer. In the future, I would like to explore “plot” as a structure, something which is currently impossible due to the state of the current technology. Plot cannot be a variable until storage increases and engines get smarter. I can easily see a plot that becomes a network of possibilities.

Corey and Lori Ann Cole: We hope as well that the improvements will be in story and design as well as flash: richer stories, more realistic character interaction, etc. Technology, beyond a certain point which we’ve already reached, really isn’t a big deal. Creativity, and an understanding of the differences between “interactive movies” and games is! The move to professional writers and game designers in the industry is helping.

Ron Gilbert: I think that plot has nothing to do with technology. They are almost unrelated. It’s not CD-ROM or VGA that is going to make the difference, it’s learning how to tell a story. Anyone who is any good can tell a great story in 160 X 200-resolution, 4-color graphics on two disks.

Roberta Williams: It’s not that I don’t think a good plot is important! Obviously it is.

Dave Lebling: I didn’t mean to accuse you of not caring about plot. You of all people know about that! I just think the emphasis on flash is a symptom of the fact that we know how to do flash. Just give us a bigger machine or CD-ROM, and, wham, flash! What we don’t know how to do is plot. I don’t think today’s plots feel more “real” than those of five or eight years ago. Will they be better in five years? I hope so, but I’m not sure. We can’t just blindly duplicate other media without concentrating on the interactivity and control that make ours special. If we work on improving control and the illusion that what we interact with is as rich as reality, then we can do something that none of those other media can touch.

Corey and Lori Ann Cole: We have never really used the computer as a medium in own right.

Steve Meretzky: You haven’t used it to contact the spirit world? [1]One of my favorite things about this transcript is the way that Steve Meretzky and Al Lowe keep making these stupid jokes, and everybody just keeps ignoring them. I fancy I can almost hear the sighs…

Corey and Lorin Ann Cole: There are things that can be done on a computer that can’t be done with other mediums. Unfortunately, the trend seems to be away from the computer and towards scanned images and traditional film and animation techniques. [2]It’s worth noting that the trend the Coles describe as “unfortunate” was exactly the direction in which Sierra, their employer, was moving in very aggressive fashion. The Coles thus found themselves blowing against the political winds in designing their games their way. Perhaps not coincidentally, they were also designing the best games coming out of Sierra during this period. If this trend continues, it may be a long time before we truly discover what can be done uniquely with the computer medium. One small example: the much-chastised saved game is a wonderful time- and mind-travel technique that can be a rich tool instead of an unfortunate necessity.

Bob Bates: I agree. You can’t ask a painter at the Art Institute of Chicago to paint you a different scene. You can’t ask a singer at the Met to sing you a different song. (Well, I guess you could, but they frown on requests.) The essence of a computer game is that the player controls the action. The point is to make beautiful music and art that helps the player’s sense of involvement in the game.

I have noticed that a lot of games coming out now are in 256 colors. Does this mean that 256-color VGA is going to be the standard? Has anyone thought about 256 colors in 640 X 480 yet? And how does anyone know who has what?

Bob Bates: The market research on who has what is abominable. As for us, we are releasing our titles with hi-res EGA, which gives us really good graphics on a relatively popular standard, as well as very nice text letters instead of the big clunky ones.

Steve Meretzky: I often get big clunky letters from my Aunt Matilda.

Guruka Singh Khalsa: We’ve been doing a bit of research on who has what hardware, and an amazing number of Sierra customers have VGA cards. Looks like around 60 percent right now. As for 640 X 480 in 256 colors: there’s no hardware standard for that resolution since it’s not an official VGA mode. You won’t see games in that resolution until the engines are more powerful — got to shove them pixels around! — and until it’s an official mode. All SVGA cards use somewhat different calls.

Dave Lebling: The emerging commercial standard is a 386 with VGA and 2 to 4 megs of memory, with a 40-meg hard drive. The home standard tends to lag the commercial one by a few years. But expect this soon, with Windows as the interface.

Does anyone have any plans to develop strictly for or take advantage of the Windows environment?

Dave Lebling: Windows is on the leading edge of the commercial-adoption wave. The newest Windows is the first one that’s really usable to write serious software. There are about 1 million copies of Windows out there. No one is going to put big bucks into it yet. But in a few years, yes, because porting will be easier, and there is a GUI already built, virtual memory, etc., etc. But not now.

With the coming parser-less interfaces and digitized sound, it seems as if text may eventually disappear completely from adventures. Once, of course, adventures were all text. What was gained and what was lost by this shift? Are adventures still a more “literate” form of computer game?

Bob Bates: Well, of course text has become a dirty word of sorts in the business. But I think the problem has always been the barrier the keyboard presents as an input device for those who can’t type. Plus the problems an inadequate or uncaring game designer can create for the player when he doesn’t consider alternate inputs as solutions to puzzles. I think there will always be words coming across the screen from the game. We hope we have solved this with our new interface, but it’s hard for people to judge that since our first game won’t be out for another month…

Corey and Lori Ann Cole: Text will not disappear. Nor should it. We will see text games, parser-less games, and non-text games. And who cares about being “literate”; fun is what matters! I like words. Lori likes words. But words are no longer enough if one also likes to eat — and we do. We also like graphics and music and those other fun things too, so it’s not too big a loss.

Roberta Williams: It’s true that in books stories can be more developed, involving, and interesting than in movies. I believe that there is still room for interactive books. Hopefully there is a company out there who will forget about all the “video” stuff and just concentrate on good interactive stories in text, and, as such, will have more developed stories than the graphic adventure games. But as we progress adventure games in general are going to become more like interactive movies. The movie industry is a larger and more lucrative business than the book industry. For the most part, the adventure-game business will go along with that trend. Currently adventure games are the most literate of computer games, but that may change as more and more text will be lost in the coming years, to be replaced by speech, sound effects, and animation. But I do predict that some company out there will see a huge opportunity in bringing back well-written, high-quality interactive books. It will be for a smaller audience, but still well worth the effort.

Dave Lebling: I think you’re too  optimistic about “some company” putting out text products. We are moving from interactive books to interactive movies. I’m not optimistic about the commercial survival of text except in very small doses. [3]This was not what many participating in the conference probably wanted to hear, but it wins the prize of being the most prescient single statement of the evening. Note that Lebling not only predicted the complete commercial demise of text adventures, but he also predicted that they would survive as a hobbyist endeavor; the emphasis on the word “commercial” is original. Unlike in science fiction, you don’t have to follow a trend until it goes asymptotic. Text won’t go away, but its role will be reduced in commercial adventures. Graphics and sound are here to stay.

Al Lowe: With the coming of talkies, it seems as if all those wonderful dialog cards disappeared! You know, the ones that make silent movies so literate? It’s a visual medium! No one asks for silent movies; most Americans won’t even watch a black-and-white movie. Yes, text-only games are more “literate.” So?

Mike Berlyn: As far as the future of text is concerned, my money is on it sticking around. But I’m not sure it’s at all necessary in these kinds of games. The adventure I’m just finishing up has a little bit of text that reiterates what is obvious on the screen, and manages to add to the player’s inputs in other ways to a create fuller experience. But I still don’t think it’s necessary. I’ve done two completely text-less designs, though neither made it to the market.

Bob Bates: I don’t think it’s the loss of text as output that creates a problem for the designer; I think it’s text as input. It’s hard to design tough puzzles that can be solved just by pointing and clicking at things. And if there are no puzzles — tough puzzles — you’re just watching a movie on a very small screen. The days of the text-only adventure are over. Graphics are here to stay, and that’s not a bad thing, as long as they supplement the story instead of trying to replace it.

We’ve seen fantasy adventures, science-fiction adventures, mystery adventures, humorous adventures. Are there any new settings or themes for adventures? Is there any subject or theme that you’ve always wanted to put in an adventure but never had the chance?

Al Lowe: I’ve had ideas for a Wall Street setting for a game, but somehow I can’t get out of this Larry rut. I’d also like to do a very serious game — something without one cheap laugh, just to see if I could. Probably couldn’t, though. A serious romance would be good too.

Roberta Williams: There should be as many settings or themes for adventure games as there are for fictionalized books and movies. After all, an adventure game is really just an interactive story with puzzles and exploration woven into it. There are many themes that I personally would like to do, and hopefully will someday: an historical or series of historical adventure games; a horror game; an archaeological game of some sort; possibly a western. In between King’s Quests, of course.

Noah Falstein: I’ve always wanted to do a time-travel game with the following features: no manual save or load, it’s built automatically into the story line as a function of your time-travel device; the opportunity to play through a sequence with yourself in a later — and then earlier — time; and the ability to go back and change your changes, ad infinitum. Of course, the reason I’m mentioning all this is that I — and others here — have fried our brains trying to figure out how this could be accomplished. We’d rather see someone else do it right. Or die trying.

Ad infinitum? Won’t that take a lot of memory?

Noah Falstein: Recursion!

Dave Lebling: Gosh, my fantasy is your fantasy! I’ve always wanted to do a game based on Fritz Leiber’s Change War stories — you know, “tomorrow we go back and nuke ancient Rome!” Funny thing is, I’ve always run up against the same problem you ran up against.

Mike Berlyn: My fantasy is to finish a game that my wife Muffy and I were working on for the — sniff! — dead Infocom. It was a reality-based game that had a main character going through multiple/parallel lives, meeting people he’d met before but who were different this time through. In that way, the relationships would be different, the plot would be different, and their lives would interact differently.

Steve Meretzky: In my fantasy, I answer the door and Goldie Hawn is standing there wearing… oh, we’re talking adventure games now, aren’t we? A lot of the genres I was going to mention have already been mentioned. But one is historical interactive nonfiction. I know that Stu Galley has always wanted to do a game in which you play Paul Revere in April of 1775. And before I die I’m going to do a Titanic game. [4]Steve Meretzky’s perennial Titanic proposal, which he pitched to every publisher he ever worked with, became something of an industry in-joke. There’s just no market for such a game, insisted each of the various publishers. When James Cameron’s 1997 film Titanic became the first ever to top $1 billion at the box office, and a modest little should-have-been-an-obscurity from another design team called Titanic: Adventure Out of Time rode those coattails to sales of 1 million copies, the accusations flew thick and fast from Meretzky’s quarter. But to no avail; he still hasn’t gotten to make his Titanic game. On the other hand, he’s nowhere near death, so there’s still time to fulfill his promise… Also, in my ongoing effort to offend every man, woman, and child in the universe, someday I’d like to write an Interactive Bible, which would be an irreverent comedy, of course. Also, I’d like to see a collection of “short story” adventure games for all those ideas which aren’t big enough to be a whole game. [5]Meretzky had pitched both of these ideas as well to Infocom without success. In the longer term, however, he would get one of his wishes, at least after a fashion. “Short stories” have become the norm in modern interactive fiction, thanks largely to the Interactive Fiction Competition and its guideline that it should be possible to play an entrant to completion within two hours.

Bible Quest: So You Want to Be a God?. I like it, I like it.

Corey and Lori Ann Cole: Ah, but someone will sue over the trademark… [6]Legal threats from the makers of the board game HeroQuest had recently forced the Coles to change the name of their burgeoning series of adventure/CRPG hybrids from the perfect Hero’s Quest to the rather less perfect Quest for Glory. Obviously the fresh wound still smarted.

Bob Bates: The problem of course is marketing. The kinds of games we want to write aren’t always the kinds of games that will sell. This presents something of a quandary for those of us who like to eat.

This question was submitted by Tsunami, and I’ll let him ask in his own words: “Virtually every game I have played on my computer is at least partially tongue-in-cheek. What I am interested in is games with mature themes, or at least a more mature approach to their subjects. Games that, like good movies or plays, really scare a player, really make them feel a tragedy, or even make them angry. What are each of you doing to try to push games to this next level of human interaction?”

Steve Meretzky: Well, I think I already did that with A Mind Forever Voyaging, and it did worse commercially speaking than any other game I’ve ever done. As Bob just said, we have to eat. I’d much rather write a Mind Forever Voyaging than a Leather Goddesses of Phobos, but unless I become independently wealthy, or unless some rich benefactor wants to underwrite such projects, or unless the marketplace changes a lot, I don’t think I’ll be doing a game like A Mind Forever Voyaging in the near future. Sigh.

Corey and Lori Ann Cole: Computers are so stupid that even the smartest game tends to do silly things. So, it’s easier to write a silly game. And the development process on a humorous game tends to be more fun. Quest for Glory II: Trial By Fire is fundamentally a very serious game in terms of story line, but we kept lots of silly stuff in to break up the tension. I call it the “roller-coaster effect.” We want the player to get extremely intense about the game at points, but then have a chance to catch his or her breath with comic relief and plain fun.

Bob Bates: My games are usually fairly “mature,” but when 90 percent of what a player tries to do in a game is wrong, you have to keep him interested when he is not solving a puzzle. The easiest way to do this is with humor; you don’t want him mad at you, after all. But I agree that we all should strive to create emotions in the player like what we all felt when Floyd died in Planetfall.

Roberta Williams: I agree with the sentiment that most adventure games, at least up to now, have been not quite “serious” in their approach to the subject matter at hand. I think the reason for that, for the most part, is that professional writers or storytellers have not had their hands in the design of a game. It’s been mostly programmers who have been behind them. I’m not a professional writer either, but I’m trying to improve myself in that area. With The Colonel’s Bequest, I did attempt a new theme, a murder mystery, and tried to make it more mature in its subject matter — more “plot” oriented. I attempted to put in classic “scare” tactics and suspense. I tried to put in different levels of emotion, from repulsion to sadness to hilarity. Whether I accomplished those goals is up to the player experiencing the game. At least I tried!

Noah Falstein: I venture to predict that we all intend to push games this way, or want to but can’t afford it — or can’t convince a publisher to afford it. But I’ll toot the Lucasfilm horn a bit; imagine the Star Wars fanfare here. One way we’re trying to incorporate real stories into games is to use real storytellers. Next year, we have a game coming out by Hal Barwood, who’s been a successful screenwriter, director, and producer for years. His most well-known movies probably are the un-credited work he did on Close Encounters and Dragonslayer, which he co-wrote and produced. He’s also programmed his own Apple II games in 6502 assembly in his spare time. I’ve already learned a great deal about pacing, tension, character, and other “basic” techniques that come naturally — or seem to — to him. I highly recommend such collaborations to you all. I think we’ve got a game with a new level of story on the way. [7]After some delays, the game Falstein is talking about here would be released in 1992 as Indiana Jones and the Fate of Atlantis. It would prove to be a very good adventure game, if not quite the medium-changer Falstein describes.

Mike Berlyn: I disagree with the idea that hiring professional storytellers from other media will solve our problems for us. Creating emotions is the goal here, if I understood the question. It isn’t whether we write humor or horror, it’s how well we do it. This poses a serious problem. Interactivity is the opposite of the thing that most… well, all storytellers, regardless of medium, require to create emotion. Emotion is created by manipulation. And it is impossible to manipulate emotions when you don’t know where the player has been and you don’t know where the player is going. In linear fiction, where you know what the “player” has just experienced; you can deliberately and continuously set them up. This is the essence of drama, humor, horror, etc. Doing this in games requires a whole different approach. Utilizing an experienced linear writer only tends to make games less game-ish, less interactive, and more linear. In a linear game like Loom, you’re not providing an interactive story or an adventure game. All you’re doing is making the player work to see a movie.

Dave Lebling: Well, emotion also comes from identification with the character in the story. You can’t easily identify in a serious way with a character who looks like a 16 X 16-pixel sprite. [8]It’s interesting to see Lebling still using the rhetoric from Infocom’s iconic early advertising campaigns. If he or she is silly-looking, he or she isn’t much more silly-looking than if he’s serious-looking: for example, Larry Laffer versus Indy in Indiana Jones and the Last Crusade. So, you are at a disadvantage being serious in graphical games. Better graphics will improve that eventually. But even so, I think Bob hit the point perfectly: the player does a lot of silly things, even if there is no parser — running into rocks in the graphic games, for example — and you can’t stay serious. The other thing is that, in my experience, serious games don’t sell. Infocom’s more serious games sold poorly. Few others have tried, and most of those have sold poorly too.

Corey and Lori Ann Cole: A really good game — or story — elicits emotions rather than creating them. A good design opens up the player’s imagination instead of forcing them along a path. A frustrated player is too busy being angry at the computer to experience the wonder and mystery of his or her character and the game’s world. By having fair puzzles and “open” stories, we allow players to emote and imagine.

Okay, now we turn from software to hardware. One of the most striking developments over the last few years has been the growing use of MS-DOS machines for game development. This has led some Amiga and Mac owners to complain that there aren’t any good adventures out for their machines, or that the games that are out for those platforms don’t make good use of their full graphics and sound capabilities. How can this problem be solved?

Corey and Lori Ann Cole: Well, I just about went broke trying to develop Atari ST software a few years ago. This was what made it possible to pull up roots and come to Sierra to do games. But I think the real value of all the alternative platforms has been to force IBM and the clone-makers to play catch-up. Myself, I’m waiting for ubiquitous CD-ROM and telecom. I’d really like to be doing multiplayer games in a few years. In the meantime, the cold hard reality is that IBM clones is where the money is — and money is a good thing.

Roberta Williams: Ha! We at Sierra, probably the most guilty of developing our games on MS-DOS machines, are trying to rectify that problem. This past year, we have put teams of programmers on the more important non-MS-DOS platforms to implement our new game-development system in the best way possible for those machines. Emphasis is on the unique capabilities of each machine, and to truly be of high quality on each of them. Our new Amiga games have been shipping for several months now, and have been favorably received — and our Mac games are nearly ready.

Dave Lebling: Get an installed base of 10 million Macs or Amigas and you’ll see plenty of games for them. Probably even fewer are needed, since programmers have the hots for those platforms. But in reality what you need is companies like Sierra that can leverage their development system to move to different platforms. As Windows and 386-based machines become the IBM standard, the differences among the platforms become less significant, and using an object-oriented development system lets you port relatively easily, just like in the old days. Graphics will still be a problem, as the transforms from one machine to another will still be a pain.

Al Lowe: Money talks. When Mac games outsell MS-DOS games, you’ll see Mac-designed games ported to PCs. When Amiga games are hot, etc. In other words, as long as MS-DOS sales are 80 percent or more of the market, who can afford to do otherwise?

Mike Berlyn: I think we all want our games on as many systems as possible, but in reality the publishers are the ones who make the decisions.

When you design a game, do you decide how hard it’s going to be first, or does the difficulty level just evolve?

Ron Gilbert: I know that I have a general idea of how hard I want the game to be. Almost every game I have done has ended up being a little longer and harder than I would have liked.

Noah Falstein: I agree. I’ve often put in puzzles that I thought were easy, only to find in play-testing that the public disagreed. But since Indiana Jones and the Last Crusade I firmly believe that one good way to go is to put in multiple solutions to any puzzles that are showstoppers, and to make the remaining ones pretty easy. I think that’s the best for the players.

Dave Lebling: I think alternate solution are a red herring because you can’t make them radically different in difficulty or the easier one will always be found first.

Noah Falstein: But if you provide incentives to replay the game, you can make both beginners happy, who will find the easy alternative, and experienced gamers happy, who will want to find every solution…

Dave Lebling: Yes, but what percentage of people replay any game? What percentage even finish?

Steve Meretzky: Games that are intended for beginners — e.g., Wishbringer — are designed to be really easy, and games intended for veterans — e.g., Spellbreaker — are designed to be ball-busters. But since of course you end up getting both types for any game, my own theory is to start out with easy puzzles, have some medium-tough puzzles in the mid-game, and then wrap it up with the real whoppers. (Don’t ask me what the Babel-fish puzzle was doing right near the beginning of Hitchhiker’s Guide to the Galaxy.)

Roberta Williams: Usually the decision of how difficult the game is going to be is made about the time that the design actually begins. And that decision is based on who the main player of the game is going to be. In other words, if it’s an adventure game for children, then obviously the game will be easier. If it’s for families, the game will be harder than for children, but easier than a game strictly for adults. If it’s a game with adults in mind, then the difficulty level lies with the designer as he or she weaves the various puzzles into the plot of the story. I think even then, though, the decision of how difficult it’s going to be is made around the start of the design. Speaking personally, I usually have a good sense of which puzzles are going to be more difficult and which ones are easier to solve. There have been a few times when I miscalculated a puzzle. For instance, in King’s Quest II I thought the bridle-and-snake puzzle was fairly straightforward, but no, it wasn’t. And in The Colonel’s Bequest I didn’t think that discovering the secret passage in the house would be as difficult for some people as it turned out to be.

Corey and Lori Ann Cole: We try to keep the puzzles on the easy side in the sense of being fair; hints are somewhere in the game. But sometimes the best-laid plans of designers and developers go out the window when programming push-time comes, to mix several metaphors. But we definitely plan difficulty level in advance. The Quest for Glory series was intended to be somewhat on the easy side as adventure games go because we were introducing the concept of role-playing at the same time.

Dave Lebling: I think it’s relatively easy to make a game really hard or really easy. What’s tough is the middle-ground game. They tend to slop over to one extreme or the other, sometimes both in different puzzles, and you get a mishmash.

Mike Berlyn: I tend to design games that have various levels of difficulty within themselves, and so can appeal to a broad range of players. Like Steve, I like to open with an easy one and then mix up the middle game, saving the toughest stuff for the endgame.

Corey and Lori Ann Cole: We made a real effort to graduate the puzzles in Quest for Glory I, easier ones in the early phases.

Al Lowe: Does anyone else feel we should lighten up on our difficulty level so as to attract a broader audience and broaden our base of players?

Mike Berlyn: Making games easier isn’t going to attract more players. What will is designing and implementing them better.

Roberta Williams: Perhaps a parser-less interface would help. But I still think that each game should be thought out in advance as to who the target audience is, and then go from there on difficulty level.

Bob Bates: I agree that what is needed is not easier puzzles. I think that players want tough but fair puzzles. Where’s the rush that comes from solving an easy puzzle? What will keep them coming back for more?

Dave Lebling: One person’s easy puzzle is another’s never-solved brain-buster. There need to be a range of games and a range of puzzles in each game. Even Wishbringer, Infocom’s “easiest” game, had huge numbers of people stuck on the “easiest” puzzles.

Adventure designs have recently been criticized for becoming shorter and/or easier. Do you agree with this criticism, and, if so, how do you change a design to make a product longer and/or harder? And are harder games commercially viable?

Dave Lebling: Games are already too easy and not easy enough, and other paradoxes. Meaning that the intentional puzzles are getting too easy, and the unintentional ones — caused by size limitations, laziness, lousy parsers, bugs, etc. — are still too hard. Harder games are commercially viable, but only if the unintentional difficulty is reduced. We aren’t real good at that yet.

Roberta Williams: It may be true, to a certain extent, that adventure games have become shorter and/or easier than in the past. Four to ten years ago, adventure games were primarily text-oriented, and, as such, could be more extensive in scope, size, and complexity. Since the introduction of graphics, animation, and sound — and, coming up, speech — it is much more difficult, if not impossible, to achieve the same sort of scope that the earlier adventure games were able to accomplish. The reason for this is mainly limitations of memory, disk space, time, and cost. We adventure-game developers increasingly have to worry about cramming in beautiful graphics, realistic animation, wonderful sound, and absorbing plots, along with as many places to explore as possible, alternate paths or choices, and interesting puzzles. There is just so much space to put all that in. Something has to give. Even CD technology will not totally solve that problem. Though there is a very large disk capacity with CD, there is still a relatively small memory capacity. Also, the way the adventure-game program needs to be arranged on the CD creates problems. And as usual, with the new CD capabilities, we adventure-game developers are sure to create the most beautiful graphics you’ve ever seen, the most beautiful music you’ve ever heard, etc., etc. And that uses up disk space, even on CD.

Mike Berlyn: Shorter? Yeah, I suppose some of the newer games, whose names will remain untyped, are easier, shorter, etc. But unfortunately, they aren’t cheaper to make. I hate to tell you how much Altered Destiny is going to cost before it’s done. Accolade and myself have over ten man-years in this puppy, and a cast of many is creating it. When I created Oo-Topos or Cyborg or even Suspended, the time and money for development were a fraction of what this baby will cost. In addition, games like King’s Quest IV are larger, give more bang for the buck, and outshine many of the older games.

Steve Meretzky: A few years ago, I totally agreed with the statement that adventure games were getting too short and easy. Then I did Zork Zero, which was massive and ultimately quite hard. A good percentage of the feedback distilled down to “Too big!” It just took too long to play, and it was too hard to keep straight everything you had to do to win the game. Plus, of course, it was a major, major effort to design and implement and debug such a huge game. So, I’ve now come to the conclusion that a nice, average, 50-to-100-room, 20-to-30-hours-of-play-time, medium-level-of-difficulty game is just about right.

Corey and Lori Ann Cole: There is plenty of room left for easier games, especially since most “hard” games are hard only because they are full of unfair outguess-the-designer — or programmer or parser — puzzles. Nobody wants to play a game and feel lost and frustrated. Most of us get enough of that in our daily lives! We want smaller, richer games rather than large, empty ones, and we want to see puzzles that further the story rather than ones that are just thrown in to make the game “hard.”

Al Lowe: I’ve been trying for years to make ’em longer and harder!

Groan…

Al Lowe: But seriously, I have mixed emotions. I work hard on these things, and I hate to think that most people will never see the last half of them because they give up in defeat. On the other hand, gamers want meaty puzzles, and you don’t want to disappoint your proven audience. I think many games will become easier and easier, if only to attract more people to the medium. Of course, hard games will always be needed too, to satisfy the hardcore addicts. Geez, what a cop-out answer!

Bob Bates: You have to give the player his money’s worth, and if you can just waltz through a game, then all you have is an exercise in typing or clicking. The problem is that the definition of who the player is is changing. In trying to reach a mass market, some companies are getting away from our puzzle roots. The quandary here is that this works. The big bucks are in the mass market, and those people don’t want tough puzzles. The designers who stay behind and cater to the puzzle market may well be painting themselves into a niche.

Noah Falstein: Al and Bob have eloquently given the lead-in I was intending. But I’d like to go farther and say that we’re all painting ourselves into a corner if we keep catering to the 500,000 or so people that are regular players — and, more importantly, buyers — of adventure games. It’s like the saber-toothed tiger growing over-specialized. There are over 15 million IBM PC owners out there, and most of them have already given up on us because the games are too… geeky. Sorry, folks! Without mentioning that game that’s looming over this discussion, we’ve found that by making a very easy game, we’ve gotten more vehement, angry letters than ever before — as well as more raves from people who never played or enjoyed such games before. It seems to be financially worthwhile even now, and if more of us cater to this novice crowd, with better stories instead of harder puzzles, there will be a snowball effect. I think this is worth working towards, and I hope some of you will put part of your efforts into this. There’s always still some room for the “standard-audience” games. Interestingly enough, 60 to 100 rooms and 20 to 30 hours is precisely the niche we arrived at too! But let’s put out at least one more accessible game each year.

Dave Lebling: Most of the points I wanted to make have been made, and made well, but I’d like to add one more. What about those 20 million or more Nintendo owners out there? What kinds of games will hook them, if any? Have they written us off? I don’t think our fraction of the IBM market is quite as small as Noah’s figures make it look. Many of those IBM machines are not usable for games by policy, as they are in corporate settings. But all of the Nintendos are in home settings. Sure, they don’t have keyboards, but if there was a demand for our sort of game — a “puzzle” game, for want of a better word — there would be a keyboard-like interface or attachment, like the silly gun or the power glove. There isn’t. Why? Are we too geeky? Are puzzles and even the modicum of text that is left too much? We will have the opportunity to find out when the new game systems with keyboards start appearing in the US.

What do you all think about the idea of labeling difficulty levels and/or estimated playing time on the box, like Infocom used to do at one time?

Steve Meretzky: That was a pretty big failure. As was said earlier about puzzles, one person’s easy is another person’s hard.

Al Lowe: Heh, heh…

Steve Meretzky: For example, I found Suspended to be pretty easy, having a mind nearly as warped as Berlyn’s, but many people consider it one of Infocom’s hardest.

Bob Bates: The other Infocommies here can probably be more accurate, but my recollection is that labeling a game “advanced” scared off people, and labeling a game “easy” or “beginner” turned off lots of people too. So most of the games wound up being released as “standard,” until they dropped the scheme altogether. Still, I think some sort of indication on a very easy game, like the ones Noah was talking about, is in order. The customer has a right to know what he is purchasing.

Corey and Lori Ann Cole: But Loom was rated as an easy game, and people who were stumped on a puzzle felt like this meant they were dumb or something.

Mike Berlyn: Good point! I’m not sure that labeling a product as being easy, medium, or difficult is a real solution. I know some games which were labeled “beginner” level were too tough for me. What we as designers need to do is write better, fairer, more rounded games that don’t stop players from exploring, that don’t close off avenues. It isn’t easy, but it’s sure my goal, and I like to think that others share this goal.

Okay, this is the last question. What is your favorite adventure game and why?

Noah Falstein: This will sound like an ad, but our audience constitutes a mass market. Ron Gilbert’s next game, The Secret of Monkey Island, is the funniest and most enjoyable adventure game I’ve ever played, including the others our company has done. I’ve laughed out loud reading and rereading the best scenes.

Steve Meretzky: Based simply on the games I’ve had the most fun playing, it’s a tie between Starcross — the first ever adventure game in my genre of choice, science fiction — and the vastly ignored and underrated Nord and Bert Couldn’t Make Head or Tail of It.

Roberta Williams: I hate to say it, but I don’t play many adventure games, including our own! I really love adventure games, though. It was this love of adventure gaming that brought me into this business. However, nowadays I’m so busy, what with working on games of my own, helping my husband run the company, taking care of the kids and the house, and doing other extracurricular activities, that I literally don’t have time to play adventure games — and we all know how much time it does take to play them! Of the adventure games that I’ve played and/or seen, I like the games that Lucasfilm produces; I have a lot of respect for them. And I also enjoy the Space Quest and Leisure Suit Larry series that my company, Sierra, produces. Of my own games, I always seem to favor the game I’m currently working on since I’m most attached to it at that given moment. Right now, that would be King’s Quest V. But aside from that, I am particularly proud of The Colonel’s Bequest since it was a departure for me, and very interesting and complicated to do. I am also proud of Mixed-Up Mother Goose, especially the new version coming out. And looking way back, I still have fond memories of Time Zone, for any of you who may remember that one.

Corey and Lori Ann Cole: Of adventure games, we liked the original mainframe Zork and Space Quest III. But our favorite games are Dungeon Master and Rogue, the only games we keep going back to replay. As for our favorite of all two games we’ve done, we’re particularly proud of what we are doing with Quest for Glory II: Trial By Fire. We’re also proud of the first game, but we think Trial by Fire is going to be really great. Okay, end of commercial, at least as soon as I say, “Buy our game!” But seriously, we’re pleased with what we’ve done with the design.

Bob Bates: “You are standing outside a white house. There is a mailbox here.”

Mike Berlyn: This is my least favorite question in the world. (Well, okay, I could think up some I’d like less.) But it’s a toss-up between A Mind Forever Voyaging, Starcross, and the soon-to-be-forgotten masterpiece, Scott Adams’s Pirate Adventure. Yoho.

Dave Lebling: Hitchhiker’s Guide and Trinity. Both well thought-out, with great themes. But beyond those, the original Adventure. I just played it a little bit last night, and I still get a thrill from it. We owe a lot to Will Crowther and Don Woods, and I think that’s an appropriate sentiment to close with.

Footnotes

Footnotes
1 One of my favorite things about this transcript is the way that Steve Meretzky and Al Lowe keep making these stupid jokes, and everybody just keeps ignoring them. I fancy I can almost hear the sighs…
2 It’s worth noting that the trend the Coles describe as “unfortunate” was exactly the direction in which Sierra, their employer, was moving in very aggressive fashion. The Coles thus found themselves blowing against the political winds in designing their games their way. Perhaps not coincidentally, they were also designing the best games coming out of Sierra during this period.
3 This was not what many participating in the conference probably wanted to hear, but it wins the prize of being the most prescient single statement of the evening. Note that Lebling not only predicted the complete commercial demise of text adventures, but he also predicted that they would survive as a hobbyist endeavor; the emphasis on the word “commercial” is original.
4 Steve Meretzky’s perennial Titanic proposal, which he pitched to every publisher he ever worked with, became something of an industry in-joke. There’s just no market for such a game, insisted each of the various publishers. When James Cameron’s 1997 film Titanic became the first ever to top $1 billion at the box office, and a modest little should-have-been-an-obscurity from another design team called Titanic: Adventure Out of Time rode those coattails to sales of 1 million copies, the accusations flew thick and fast from Meretzky’s quarter. But to no avail; he still hasn’t gotten to make his Titanic game. On the other hand, he’s nowhere near death, so there’s still time to fulfill his promise…
5 Meretzky had pitched both of these ideas as well to Infocom without success. In the longer term, however, he would get one of his wishes, at least after a fashion. “Short stories” have become the norm in modern interactive fiction, thanks largely to the Interactive Fiction Competition and its guideline that it should be possible to play an entrant to completion within two hours.
6 Legal threats from the makers of the board game HeroQuest had recently forced the Coles to change the name of their burgeoning series of adventure/CRPG hybrids from the perfect Hero’s Quest to the rather less perfect Quest for Glory. Obviously the fresh wound still smarted.
7 After some delays, the game Falstein is talking about here would be released in 1992 as Indiana Jones and the Fate of Atlantis. It would prove to be a very good adventure game, if not quite the medium-changer Falstein describes.
8 It’s interesting to see Lebling still using the rhetoric from Infocom’s iconic early advertising campaigns.
 
50 Comments

Posted by on February 16, 2018 in Digital Antiquaria, Interactive Fiction

 

Tags: , , , , , , , , , , , ,