Reviews by Wade Clarke

(I published the first version of this review on 3 October 2012 as part of my blog of IFComp 2012. Guilded Youth was the 3rd of 26 games I reviewed. The review has since been revised to address updates to the game.)

Guilded Youth is a smooth playing and charmingly presented story about the suburban adventures of teenaged Tony. Tony is preoccupied with the cool local manor which is soon to be demolished, and he sneaks in there each night in search of treasure. The game is set in the 1980s and delivered through the prism of the imaginative life shared by Tony and his friends on their local Dungeons & Dragons BBS game, where he's Tony the Thief.

Guilded Youth is linear, but not "I don't feel like I'm doing anything" linear, and its production values, including some graphics and sounds, are impeccable. Its atmosphere works on a couple of levels, that of the world of teenagers in general, and also with a degree of specific nostalgia or period feel for folks who grew up in the 80s or like 80s things, and for the entertainment from that time. The original IFComp version of the game had a disappointing ending (I wasn't the only person to say so) but Jim Munroe revised the tail completely in response, and it's now as good as everything else.

The BBS world is presented in pretty green monochrome which makes way for a contemporary online style when you take Tony out on one of his nightly jaunts. Character portraits of your friends and inventory images surround the screen, and there are cool discrete sound effects if you play with the Chrome web browser. The game is also very simple to control, letting you know that there are only a handful of commands which are actually required for play, though others will work. On each of your trips to the manor at night you are accompanied by different friends from the BBS whom you attract to your party with loot from the previous night. I should make it clear that there is no actual RPG engine in play. Each object only persuades a certain friend to come with you, and only on a particular night, so the linearity extends to all areas of the game.

Guilded Youth recalled for me the atmosphere of many 1980s movies about gangs of adventurous teenagers. The characters in the game are excited that they're getting to have such adventures, and hope to enjoy and prolong this feeling as much as they can before the manor is torn down, symbolically ending the fun and ushering them further towards adulthood. The characters are only as open to us as Tony's point of view allows, and à la their BBS characters (or the cast of The Goonies) each teen demonstrates a knack for a particular skill or way of doing things that fits neatly with the structure of going out with a different one or two of them each night. Tony even gets to (Spoiler - click to show)smooch the cool girl, which was the romance every boy liked to imagine himself in when he watched something like The Goonies. At least I assume the other boys were thinking the same things as me but that we all just never spoke about it. What's funny about the NPCs in the game (or just jealous-making) is that even though they don't say a tremendous amount of stuff or stay onscreen long enough to do a tremendous amount of stuff, they manage to leave a great impression of their liveliness, individual foibles and relationships in a way that's both fun and realistic.

KING OF BEES IN FANTASY LAND is a brisk fantasy/sci-fi CYOA which, in common with the same author's terrific You Will Select a Decision, is presented as an artefact from the recent past written in less than ideal English. Starting up King feels like starting up an '80s coin-op or Nintendo game, albeit one without pictures. The lettering is an all caps 8-bit font and the copyright notice says 1989. The style of the writing is that of mildly enthusiastic Japlish.

The player, addressed as Space Knight during the patriotic opening spiel, is charged with the mission of taking down the "evil King of Bees from Bee Fort" so that humans, who have wrecked their own planet, can colonise the bee planet Garaxas, aka Fantasy Land. The outward silliness of this plot and the game's presentation both put you in in a frame of mind in which you're immediately hungry for fun and success.

The fun is easy to come by. Whatever decisions you make once you hit the planet's surface, the game rolls with them. Even when you face seemingly important binary choices, like whether or not to trust the boardwalk which crosses the alligator swamp, you'll find that all roads tend to lead to ultimate success by their own methods, or that a blocked path will produce a discouraging loop which quickly pushes you onto an unblocked one. Messages will appear proclaiming exciting bonuses you've acquired for non-existent (mechanically speaking) skills, and whatever you do, the occasional exclamation mark is there to suggest that you're doing good, or The Right Thing.

The planet is busy with its bee inhabitants, and they're mostly friendly, chatty folk who offer no opposition to your march across their territory, or even an impression of being aware that opposing you is relevant or necessary. So even though the excitable 8-bit plot and tone of the game will have primed most players savvy to 8-bit conventions for combative action, most players will also find themselves pretty uninterested in vaporising friendly unarmed folk, except in the also 8-bit manner whereby they might just want to see what happens if they act like a jerk. The text gets prejudicial when the bees show up, with terse but aggressive options appearing like ERADICATE, but the delivery remains paradoxically light and encouraging, whether you're acting like Rambo or not.

The first time I played King, the contrary signals being sent simultaneously by different levels of the game about what I was doing as Space Knight started to put me in a nervous and suspicious mood. I was wondering if the game was going to suddenly turn around and tell me (or at least strongly imply) that I was a harmfully suggestible dumb-dumb of the kind who can easily be made to follow any orders. That might sound like a strong reaction to an ostensibly light game, but there seem to be an increasing number of IF games around which impart this lesson through degrees of player deception. It's not that I oppose games deceiving players per se; in fact IF is particularly good at doing this in lots of different ways, and to different ends. But sometimes in the case of games which offered the lesson, "You should have resisted the game's path for moral reasons", I had felt, when I reached the outcome, that I had simply been tricked.

I'm definitely not saying that this lesson or this schtick are the upshots of King, only that these issues do come into play. And I have deliberately not addressed a lot of King's content to avoid spoiling anything. There are some interesting, entertaining and surprising little turns of events tucked into this quite short game, and it's frequently cute, even while it's being serious. To understand all the aspects of what might be going on will take at least a couple of plays, and there's some new fun to be had on the way through each time. The voice of the prose is very authentic in reproducing the earnest and focused tone of Japanese 8-bit games, and the arrangement of the screen, fonts and colours are all attractive. The game is a fine example of how cute and simple aesthetics should not be underestimated in terms of their ability to deliver clever or thoughtful outcomes. Probably the biggest cleverness of King is that the expectations and aesthetics of the 8-bit are used both sincerely and for commentative purposes at the same time. My final advice on this game comes from the attract mode of 1980 coin-op Moon Cresta: "Try it now!! You can get a lot of fun and thrill"

The intent behind this game is clear - to teach some IF basics through a simple scenario involving you getting the newspaper from your mailbox, finding yourself locked out of your house and having to find an alternate way back in. The game fails because it barely works. It doesn't even demonstrate consistent behaviour within the nuances of the Quest interface.

The geography is confusing. The illustrations, while cute, only enhance that confusion. It was a poor choice to create those illustrations in black and white when the text consistently emphasises how sunny the world is. There are no synonyms for anything; "paper" will not do for newspaper. "OPEN MAILBOX" doesn't work; you must "search" or "look inside". The game fails to acknowledge that you have acquired the newspaper until a thunderstorm event happens. Some important and obvious objects do not appear in the "Places and Objects" menu. There is a silly instant death-by-car if you try to cross the road in front of your house, in spite of the game having previously suggested that "the grass is greener on the other side of the road". You can't undo from the instant death if you're using the online Quest player.

I don't think the game is completeable, which is to say that I couldn't complete it, and I made several serious attempts to do so. Amongst the buginess, inconsistent approaches and the general newbie-unfriendliness, the game certainly would not be good for newbies, and isn't suitable for anyone else, yet.

In 1981, you play a man stalking a woman. If "play" sounds light, perhaps it's more accurate to say you are almost chained to the actions of a man in the grip of erotomania (a deluded and obsessive romantic fixation). Due to the shortness of the experience, I will omit further description of either of the main characters, as I can imagine how some players may prefer not to know certain fundamental details of the setup going in. What is certain is the strength of the writing, which thoroughly sinks you into the headspace of the obsessed protagonist, and into his vivid fantasy life.

At the rawest level of game mechanics, 1981 caters to very few player actions; it is what is typically called linear. The obsessed PC imparts clear thoughts in the prose about the next important course of action, and it's generally a waste of time trying anything else. What's interesting is the extent to which this approach could be considered to work well with the subject matter of this game. The psychological disorder at work here is characterised by the subject's complete resistance to all attempts to convince them that the situation is other than they believe it to be. Perhaps this is the best "excuse" for linearity that there can be.

This raises the question of what the player's role is here. "Creepy" and "uncomfortable" have been common review descriptions of the experience of playing 1981. Players tend not to like playing "bad" characters in realistic situations, or even facilitating their actions. I think this remains a difficult or weak point for the prospects of certain kinds of storytelling being done with IF. 1981 may again supply its own solution with its subject matter. With the PC's character being presented so monomaniacally, the player is likely to feel a degree of separation from the PC's actions. If you try to break off the path, the PC either doesn't want to do your thing because it's irrelevant to his plans, or your thing isn't implemented, or both.

It would be difficult and tedious for me to try and describe how 1981 could work as easily as a short story as IF. I think it could, and in that form it would be clear of the "playing a villain" hurdle. But it works in this form with the caveats that you must play the villain and accept mechanical linearity, positions which are unpopular and still querulous to many, respectively. What you will get for this is the sensation that you are shackled to the PC's ruinous path and that there's no getting off. This kind of story trajectory fascinates me because being privy to the amount of effort that a human can devote to going entirely the wrong way in life is strangely illuminating about our capabilities as a species. 1981 is a psychologically strong excursion into this territory – though with little extra implementation – and also an interesting demonstration of one way to traverse a lot of difficult IF terrain to do with unlikeable protagonists and realism.

Spoiling background on the game: (Spoiler - click to show)This is an imagined recreation of the real case of John Hinckley Jr. and his obsession with actress Jodie Foster, resulting in his assassination attempt on Ronald Reagan. I knew about the case before playing the game, but the game doesn't reveal who the parties are until halfway through. I experienced a minor letdown at the moment of revelation, to an extent, only because I then felt I knew what was going to happen. But inevitability is such a strong part of this game anyway - this was really just my idiosyncratic reaction and no reflection on the game.

I had a problem during the 2011 Interactive Fiction Competition. I was supposed to be getting a move on and reviewing all of the other entrants' games, but I kept procrastinating by sneaking away to play Kerkerkruip. By the time the competition was over, I had played it at least 50 times in my quest to complete the game on Normal difficulty. This is testimony to Kerkerkruip's addictiveness, which grows out of the stiff but strategically overcomeable challenge it presents and the relatively infinite pool of circumstantial variations it offers to dungeongoers. The latter quality is what makes the game really memorable and anecdote-worthy once a player has got a handle on its mechanics.

A moment's divergence for the consumer guide part of this review: Kerkerkruip is certainly not a traditional IF game or text adventure in which the player solves unchanging puzzles en route to particular goals while possibly becoming involved in a narrative the author has laid out. This is a high-stakes game of Dungeons & Dragons adventuring in randomly generated dungeons. At the same time, it is delivered by text and controlled by a parser, and uses explorative elements in some typical adventure game-like ways. In all of these capacities, it is obviously from the school of text adventures, and not completely unlike a combat MUD or a modern incarnation of Eamon, though a plotless one. Also note that it is essential to at least read the Beginner's Guide before playing (I found this three page guide to be the easiest way into the game, as opposed to traveling through multiple inline HELP menus) or Kerkerkruip will promptly kick you to the pavement.

Your goal is to find and kill the evil wizard Malygris of the dungeon Kerkerkruip. You begin armed with a rapier; more significant weaponry and equipment must be found in the dungeon. Usually there will be about five other groups of monsters lurking around, and it is only by defeating these monsters and absorbing their powers and health in a wisely chosen order (new powers only accumulate if they are weaker than powers you already possess) that you will have a hope of becoming powerful enough to defeat the wizard. The dungeon contents and layout and the roster of monsters change every time you die or restart. You can't save the game except to take a break, and there is no UNDO. These danger-increasing elements are common to another genre of game Kerkerkruip announces that it belongs to: the roguelike, named, unsurprisingly, after a particular game called Rogue.

Movement is handled with the traditional compass commands, augmented by a "go to" command and a handy "remember" command, but the combat makes use of the ATTACK system originated by the author and is divided into Action and Reaction phases. By working with just a handful of well balanced temporal elements, Kerkerkruip ensures each decision you make about what to do next in battle carries significant weight. Should you Attack now, or build up the strength of your next attack by pausing to Concentrate? You can try to build up to three levels of concentration, but if you're struck in the meantime, your concentration will be broken. On the other hand, if you never concentrate, your attacks won't grow strong enough to finish off the bad guys before they finish you.

This core system is simple enough for anyone to understand, but its application in any moment is modified by a huge number of variables, amongst them: the geography of the room you're in (e.g. it doesn't pay to Dodge while fighting on a narrow bridge over lava), the nature and habits of the enemy you're facing (e.g. animated daggers attack ceaselessly and break your concentration as often – the jumping bomb will never break your concentration, but if it gathers enough concentration itself, it will explode and kill you instantly), the Tension in the air (how long has it been since anyone last struck a blow?), your current status and arsenal of powers, and the interference of a further array of supernatural stuff like fickle dungeon gods or weird summoned entities.

The sum effect of the play amongst all these interrelated elements is that Kerkerkruip is capable of generating the exciting sense that with almost every move you make, the whole game is at stake. The circumstances of danger can rearrange themselves into so many different patterns that a lot of your battles will strike you as uniquely memorable, even when you're dealing with the same small roster of monsters over and over again. You can marvel at a seemingly (or actually) brilliant series of moves you make that succeed in resuscitating your prospects when you're down to 1 hp. Similarly you can laugh at the results of a particularly bold, stupid or unlucky move that backfires spectacularly, or at some confluence of events so extraordinary that you'll feel like telling someone else about it. You will certainly die far more often than you will win, but this is a game where experience, exploration and repeat plays really pay off, and the strategic element is always vivid, the prospect of victory always tantalising.

Ultimately, Kerkerkruip is an essential and massively replayable game for dungeon and combat fans, and also demonstrates the kind of novelty and elegant design that is inspiring in general.

The Surprising Case of Brian Timmons is a Lovecraftian adventure based on a scenario for the Call of Cthulhu tabletop RPG, a scenario in turn based on H.P. Lovecraft's short story 'The Case of Charles Dexter Ward'. In spite of its convoluted sounding provenance, this game is actually one of the most accessible Lovecraft IF games out there. A player doesn't need any prior knowledge of the source material or of Lovecraft's work to be able to get into it, and while it's of moderate size, it's more about linear action than the kind of painstaking puzzling folks often associate with Lovecraftian games ala Anchorhead. A word of caution; it's also a game which gets shootier and bloodier as it goes on.

While Lovecraft's protagonists usually have some kind of personal involvement in the supernatural goings-on they face, the PC in Brian Timmons doesn't. He's a detective from the hardboiled school who gets mixed up in a stranger's supernatural goings-on only because they stand between him and his next paycheck. The novelty of adopting an outsider's viewpoint is a welcome one in this busy IF subgenre, and the detective brings humour, attitude and action to the table – three things you normally don't much associate with Lovecraft. The resulting game is straightforward, episodic in a good way and becomes quite gripping as you move towards its climax, though some elements of the delivery could be improved.

Brian Timmons is divided up into scenes set in different locations. Each car trip you take from one location to the next acts like a chapter break, and you don't have to worry about deciding where to go. The hero chooses the next relevant stop as soon as he's got enough fresh leads from the current one. While the game itself suggests you should use ASK and TELL to communicate with its characters – and at times it's essential to use these methods – the majority of communication actually consists of the NPCs telling you their stories one line at a time. While a lot of games use this method and it gets the job done, the game could be richer if it would allow the player to interject with some relevant ASKing and TELLing (as is, the characters only respond on the most vital of topics), though I acknowledge this is never an easy area to program. The characters do a lot of neat fidgeting of their own accord when not speaking, and the game is also generally strong in the area of random atmospheric detail, throwing in lots of little snippets about passers-by, the weather and other environmental changes.

Where the game has some trouble is in getting all of its content to live in the same place tonally, at least at once. When the hardboiled shtick and language are in evidence, they really dominate. But they vanish too easily when the detective isn't delivering his Chandler-esque wisecracks, allowing the game to be overtaken by more utilitarian descriptive text. The sexy dame character is a bit cringy in this light – she triggers the "poured into her dress" remarks in extremis, but in isolation, and thus comes across more as a reminder of the game's tonal wobbling than an authentic seeming femme fatale character justified by the genre and context.

I have a few other nitpicks. The game suffers a bit from empty porch syndrome. It needs a little more proofreading. The inventory limit can aggravate, though this last point is mitigated by the coolness of having a trench coat with pockets of seemingly infinite depth. And it's just fun to wear a trench coat and Fedora in general. I enjoyed The Surprising Case of Brian Timmons a lot. It's also a game which comes without hints, and I was pleased to be made to solve it off my own back, pausing occasionally to scratch my head.

Voodoo Castle (1979) was the fourth game from Adventure International (AI). It was written by Scott Adams's then wife, Alexis, who had previously assisted on Pirate Adventure, and its opening enthusiastically proclaims that it is "DEDICATED TO MOMS EVERYWHERE!".

The goal of Voodoo Castle is to lift the curse that afflicts Count Cristo, a goal established after the player has opened the coffin in the game's first location and examined the man therein. In the context of the Adams game engine, this is a fairly abstract goal; recall that all of the prose must be extremely minimal (room descriptions generally come in at under 40 characters in length), the parser only accepts two words, and the whole affair has to fit into 16KB of RAM. Doing something like finding treasures and dropping them in a target room, ala Adventureland, is an easy-to-grasp concept in the context of these limitations, but accomplishing a goal as broad as lifting a curse is harder to think about in a vacuum, and potentially a little more intimidating to contemplate when you first fire up this game.

The game's castle isn't actually called "Voodoo Castle", but it is the castle where the action takes place, and Voodoo is clearly afoot. Fascinating paraphernalia can be found lying around in its corridors, including a voodoo doll, a Ju-Ju bag, a witch's brew and a room full of exploding chemicals. With no more to go on than the game's initial exhortation that the player lift a curse, he or she must experiment with these interesting props and advance through the solving of a succession of puzzles, and ultimately of the game. The experience is a lot of fun, and while Voodoo Castle's official difficulty label is Moderate, I find it to be one of the easier AI games. However, I should point out that this was not one of the AI games I had the opportunity to play back in the day. By the time I came to it in the 2000s, I was (a) way older and wiser, (b) had solved a lot of adventure games in general, and (c) had solved a decent number of AI games and acquired a strong sense of their workings.

What is interesting about Voodoo Castle is that there are no antagonists in it. While there are still lots of ways to die or wreck your game, including inescapable rooms and destructible crucial items, there are no people, monsters or other entities that are out to get you. In fact, a theme of Voodoo Castle (if 'theme' isn't too lofty a word in the circumstances) is that people who might seem scary at first are probably not threats, but sources of potential help. Except for the maid, who chases you downstairs if you happen to track soot through the castle. Back in the realm of objects, the cause and effect relationships between a lot of the game's artifacts and things that might happen to you during play are often unintuitive (E.G. "I've recently stopped being blown up by exploding test tubes. Why?") and require much trial and error and game saving to discern.

It would be a struggle to qualify any observations I might be tempted to make about the nature of games Alexis authored or influenced in this series versus the ones her husband authored, but it's certainly fun to speculate. My sense is that when Alexis was involved, the games were a little kinder in tone, though not necessarily in content. The absence of antagonistic characters in Voodoo Castle speaks to this idea, as does its altruistic goal for the player, and the very positive image with which the game ends. Scott of course gave us several games featuring instant death by bear mauling, and he gave us Savage Island Parts I and II, two of the most difficult and masochistic jaunts to ever grace adventuredom. But Adams also opposed the idea of the player having to commit any acts of violence against other creatures to advance in his games. The attitude of the AI games is that violent acts may be visited upon you, usually by nature, if you are stupid or unlucky enough - and we have to take the AI concept of player stupidity with a grain of salt.

Voodoo Castle features a couple of AI's most loveable/hateable guess-the-action and guess-the-verb moments (you won't believe what you have to do with the Ju-Ju bag, and I mean that in a banal way) but fortunately the AI clue sheet cyphers make getting help fun in these games. And I always particularly liked Voodoo Castle's clue sheet. It was the first AI clue sheet I ever encountered, and I encountered it as a kid well before I played the game, back in the Adventurers Corner column of a 1986 issue of Australian Apple Review.

If you haven't tried an AI game before, I wouldn't recommend this one to start with due to the abstract nature of its goal. It's probably best to familiarise yourself with the nature of these very early adventures by first playing a straightforward treasure hunt like Adventureland. But in the scheme of the AI series, Voodoo Castle sports some distinctive features, a castle stocked with lots of interesting objects, and a good dose of that elemental, imminent style of puzzle-solving which is the hallmark of the AI games.

I feel that the best way to describe the aesthetic of this game is as follows: Imagine that the computers used to render the graphics which portray the horrible, gritty world overrun by machines in the film The Matrix were corrupted by a hacker who replaced most of the textures with imagery from Peanuts comics. This is the imagery rendered by the prose of this game. And yes, you do need to have read a moderate amount of Peanuts to recognise the iconography, and knowing The Matrix some will also help. I guess The Simpsons anticipated all of this with that poster outside the Googleplex which says 'You're in the Matrix, Charlie Brown!'

However – I am already struck by the difficulty I feel in describing this as a game. Perhaps it is in the area of pieces like this which the term Interactive Fiction will come into its own in a more literal sense. Mastaba Snoopy is a poetic prose story with junction points which determine what may be read next, but there's a low sense of consequence based on what you click – I confirmed this at least for myself by repeatedly rolling back one move, trying the other option(s) and seeing if my feeling about the whole moved a different way as a result of what I read there. It didn't, except at a handful of major branches; there's a kind of uniform forward velocity into this clever concoction of an alien meets future-internet world based on Peanuts comics, no matter what choice you click on, but I can't say that the different facets of it feel very different to each other. The world is rendered with effective writing, and the immediate effect of the piece is different to that a static piece of writing, but the combination of the piece's overall abstraction and its low consequence of action mean that its emotional effect is still closest to that of a static piece of writing, albeit one which can be rotated to be viewed from a few different angles.

Peanuts has always been and will always be a big part of my life through all of its sense of humour, writing and artwork. I doubt I missed any of the numerous references in Mastaba Snoopy, whose whole world is built out of an alien's interpretation of Peanuts comics. Some of the iterations are darkly amusing, though nobody is likely to guffaw at the bleakness of the whole. Coming into this game as a Peanuts guy, my mental state was along the lines of, "Alright, bring it." I came out disappointed that Mastaba Snoopy was neither specifically as humorous nor as thoughtful enough about ideas from Peanuts as I'd hoped it might be. It's probably hard to be specific when you're also being abstract. I didn't feel that any more meaning emerged from the throbbing of Snoopy's loins – a scene in Mastaba Snoopy – than it would have from the throbbing of, say, Hello Kitty's loins. Or rather, both may be saying the same thing (whatever that is). Mastaba delivers a fair bit on the cyber/veneral imagery front in general.

In spite of the quality of the writing, I was disappointed re: Peanuts and I missed the presence of more game-like consequences which might have made me get more into this world. If the writing alone is enough for you, you may like it a lot more, and the whole idea is very imaginative.

Our Island appears at first, and mostly, to be a lyrical holiday game set in an idyllic seaside community. You can wander along Wharf Street, buy an ice cream from the pretty ice cream girl, look for flotsam and jetsam, observe local wildlife and feel sparkly about life as you note the presence of the local community centre. Admittedly the majority of these things are only implemented at a very rudimentary level, and some only exist in the location description text, but there's still no denying the game's coherent and romantic sense of place. The map is big and demands real mapping with pen and paper, at least if the player isn't to miss anything. This is also assuming the player makes it past the painful intro, which is basically a My Apartment game set in a holiday house where you have to fight with an inventory limit.

The trick with Our Island is that there's a whole other level of play in it, one involving puzzles and unexpected zaniness, but it's so obscurely integrated into an outwardly goal-less holiday game that it's a huge ask of players to (a) even identify that it's there and (b) negotiate it and solve it – especially with Our Island being in as rough a state as it is. Presenting a rich environment and then slowly allowing the player to become aware of some upheaval within it can make for a great dramatic design, but it demands strong execution to be successful. The biggest problem for Our Island is its erratic implementation. If the player starts to sense that more than half the interesting-looking content in the game is just painted on, they're going to stop checking the content. And in the cases where the content is relevant to the puzzles, the game doesn't signal it. There's also just tons of plain old bugs, typos, failures to describe exits, scenes that play at the wrong times, and even the odd location with no description. So for now, the game's island-sized ambitions considerably outstrip its quality of delivery. The groundwork for something great is here, and I already enjoyed exploring the environment, but the whole thing needs lots more work.

Strange Odyssey, released in 1979, was the sixth of Scott Adams's games in the series today referred to as the Scott Adams Classic Adventures. This game was a childhood favourite of mine and remains a favourite in adulthood. In plain mechanical terms, it's a treasure hunt in space, but its use of multiple alien settings gives it a sense of exploratory danger which feels unique in the series. This isn't to say that the perils in the likes of Adventureland or Pyramid of Doom aren't exciting – it is to say that those games are about exploring one dangerous world, while Strange Odyssey involves visiting a series of unrelated dangerous worlds, never knowing what to expect as you step into each one.

This is a dense game even for Adams, whose Classic series entries each had to fit into 16kb of RAM. Many objects have multiple uses and need to be carted back and forth between different worlds. Time pressure comes in the form of the finite air supply in your spacesuit, and working out how and where you can refill it is a significant puzzle. Odyssey also has more locations than most of its siblings, but the reason it feels more expansive than them is because of its intergalactic nature. Its little text strings have to act as seeds to help the player imagine whole environments at a time, rather than just one room or a corridor.

The fundamental puzzle in Strange Odyssey, the one which is most likely to cause players to stand around for awhile going "Hm," is the one involving working out how to move between worlds. It is quite an abstract puzzle (dare I say Zorkian) in a game canon that rarely supported abstract puzzles due to the simplicity of the game engine and the necessary briefness of all the prose. Another interesting element of this puzzle is the way it mobilises split-second glimpses of text. Unfortunately, this special effect only exists in the original Apple II, Atari and TRS-80 versions of the game. I recommend against playing versions of the game which are missing it (C64, Inform, Spectrum) since the game's quality and sense are hurt by its absence.

Dying and dead-ending are frequent occurrences in Odyssey, so it's wise to save frequently. Just stepping through a door can kill you if the gravity or air happen to be unfavourable on the other side. Several objects can run out of gas or power, it's possible to destroy crucial items with your phaser and most of the wildlife is aggressive. When I was a kid, I loved all of this unheralded danger because I always liked stories in which you never knew what bizarre thing might be on the other side of a door or teleporter. This quality of the game still speaks to me today, and while Adams's games have come in for a lot of criticism over the years, Strange Odyssey's alien dangerousness seems to coincide perfectly with the relatively hostile nature of adventure games from this era. A major reason that a lot of old school adventures are disliked today is that players find it too aggravating that they can mess up by taking actions they might reasonably expect to have inoffensive consequences within the world of a particular game – if that game had much logic about itself. In Strange Odyssey, all of the hardships make sense and thus does the form of the whole. Space is dangerous, the worlds you visit aren't explained and alien hardware doesn't come with instructions. In retrospect, I think Strange Odyssey was one of the designs which best fit Adams's minimalist game system.