Castle Amnos: The First Legend by John Evans [Comp00]

IFDB page: Castle Amnos
Final placement: 30th place (of 53) in the 2000 Interactive Fiction Competition

Today is November 7th. It’s 11:30pm, Colorado time. I’m writing this review on my creaky but trusty 386 laptop, the television on in front of me (muted) so that I can continue to keep tabs on the US presidential election, which is still in a dead heat after seesawing all night long. It was, to say the least, a bad day to play Castle Amnos. Not that I didn’t have the time — I played my first hour over my lunch break at work, then did another hour after the commute home, dinner, watching election returns, and being at least somewhat present in my marriage (oh yeah, that). Then a quick shower to clear my head, and I’m ready to review.

With most comp games, this time investment would have been sufficient. Not with this one, though. For one thing, Castle Amnos is big. I didn’t expect this, since the game file size is 122K, but that’ll teach me to rely on file sizes — turns out you can squeeze in quite a few rooms, objects, and puzzles when you implement hardly any first-level nouns, code minimal responses for reasonable object actions, and only give your NPCs the bare minimum necessary for them to participate in their puzzles. Consequently, even when I reached the end of my two hours, I had no sense of closure at all — I doubt I’d even seen a third of the game.

What’s more, there are several mazelike sections in the game, and a few others where the geography is quite non-intuitive. I had made maps of these during my first-hour session, but I forgot and left them at work, and I’d be damned if I was going to draw them again. I have a pretty good memory for IF geography, so this didn’t cripple me in my second hour, but it did slow me down. I wonder how many comp judges’ experience is like mine: fragmented, squeezed in at the edges of our lives. I wonder how many others felt frustrated at playing a game that clearly didn’t fit into those small spaces we can create for it.

This is my 42nd game of this comp, and usually by this point I’m ready to name a trend for the year. Last year it was non-interactive games. The year before that, one-room games. This year, though there are some rather surprising similarities — two games prominently featuring peyote (?), several pointless joke games, far too many starvation puzzles — I’m not sure I can put my finger on one overall trend among the games for the entire comp, except that their index of quality has tended to be higher. I have, however, noticed a personal trend: I’m getting more and more impatient with games that, in my opinion, don’t belong in the IF competition.

Mostly, these are games that are so large, it’s really unlikely that most players will see anything like a majority of the game in two hours’ playing time. It’s discouraging to make a sincere effort to play a huge number of games in a six-week timespan, only to discover that several of those games simply cannot be played to any satisfactory conclusion in the two hours allotted. It makes me want to just avoid playing anything over a certain file size, but even that strategy would fail: My Angel is almost twice the size of this game, but it’s easily finishable in two hours. Amnos, on the other hand, is anything but.

I’m not sure what the solution is, but I will say this: Authors, I implore you. Please think carefully about whether your game can be played in two hours. If, realistically, it cannot, I urge you not to enter it in the competition. I know about the feedback problem — people are working on it. In the meantime, isn’t it better to have your game played all the way through by 50 appreciative people than to have its first 25% played, then the whole thing dropped, by 200 people who are terribly pressed for time?

Okay, I realize that I’ve gotten on my comp soapbox and delivered very little useful feedback about the game itself. To remedy that: Castle Amnos appears to have some very interesting ideas at its core. I found sections, and resonances, that intrigued me a lot. I obviously wasn’t able to get far enough in the game to get any sort of resolution on what had been set up, so I don’t know if the payoff fulfills the promise of that setup, but if it does, I think it’ll be an interesting game.

It is, however, hampered by several problematic design decisions which are a bit of a throwback to the earlier days of IF. There’s a more or less pointless inventory limit, which forces you to keep all your objects in one central location and trundle back and forth between it and whatever puzzle you happen to be working on. There is a room somewhat reminiscent of the Round Room in Zork II, only about twice as aggravating because you have to perform an action before the randomizer will run again each time, and it opens on fewer options, making repetition more necessary. It appears to be somewhat circumnavigable, but only somewhat. Then again, who knows whether there isn’t some later section of the game that makes that room behave in a deterministic fashion? Certainly not me.

In addition to these problems, there are (as I mentioned earlier) several mazelike sections of the game. To me, that kind of thing is just no fun. Mileage, I’m sure, varies. All this is not to say that it’s a bad game. It is implemented minimally, but competently. I don’t think I found any major bugs, though the game’s fascination with non-standard geography and randomness sometimes made it difficult to tell what was a bug and what wasn’t. The prose, like the code, is sparse but error-free. Perhaps, if I was able to play it all the way through, I’d even think that Amnos is a really good game, or at least a draft of something on the way to becoming a really good game. With what I was able to see, though, all I was able to tell was that its entry as a competition game impaired my ability to enjoy it.

Rating: 5.7

My Angel by Jon Ingold [Comp00]

IFDB page: My Angel
Final placement: 6th place (of 53) in the 2000 Interactive Fiction Competition

Interactive Fiction is a new enough medium that at least once a year, usually more, a work appears that approaches the form in a way that has never yet been attempted. The IF competition, with its emphasis on short works, has become a haven for this kind of experimentation, and this year is no exception. The most interesting experiment I’ve come across thus far is this game, My Angel. My Angel‘s experiment is to offer the player a “NOVEL” mode, one which actually fulfills both senses of the word: not only does it more closely resemble a prose novel than does any other IF, it is also something that has never quite been seen before.

In NOVEL mode, all input and stock parser responses appear in a two-line status window above the main game text window. In the lower panel, story-related output appears in neatly formatted paragraphs. Because this format can sometimes be hard to follow (it’s difficult to tell where the last output leaves off and the new one begins), the game thoughtfully provides an ALTERNATE command, which prints every other piece of output (more or less) in bold text. I’ve sometimes thought about how IF might be done in this format, but could never quite get past certain mental stumbling blocks: the possibility of repetition, the need for stock responses, and the problem of sequiturs.

My Angel applies a number of ingenious solutions to these problems. First, it adopts the already rather oblique tone of literary fantasy — this particular prose style suffers from a dearth of sequiturs anyway, so the sudden shifts of focus brought about by a flailing player become less obvious. In addition, the game takes stock responses such as “I don’t know the word <whatever>” or “You can’t see that here” and shifts them to the status line, not allowing them to interrupt the flow of the story. It’s also, as far as I can determine, not above simply ignoring the player’s input and continuing with the story, though this doesn’t happen often, and it’s rather difficult to tell for certain when it has happened, because there’s sometimes a significant (but not complete) disconnect between the player’s input and the game’s responses. Finally, My Angel restricts the player’s choice of verbs (though not drastically) in order to reduce the combinatorial explosion of necessary text.

As for repetition, well, My Angel didn’t quite lick that one. I still encountered several instances where the game repeated the same output two or more times in a row. You wouldn’t see this kind of sequence in any but the most avant-garde literary novel, so its appearance detracted from the game’s illusion of an unfolding prose story. For the most part, however, My Angel succeeds admirably in maintaining that illusion. Its writing is mostly quite good, and the intersection between its writing and its coding sometimes verges on the brilliant, not just for NOVEL mode itself but for the way that it deftly picks up on a player’s command and weaves it into the ongoing narrative.

The narrative itself, on the other hand, is a bit confusing, or at least it was to me. Ironically, I think the story’s problem may be one of too much novelty. We are presented with an unfamiliar world, one which appears to be more or less the generic medieval pastoral setting of copious genre fantasy, but which occasionally hints at mysterious magics, the logic of which we are left to figure out. Complicating the situation is the fact that these magics are the sort that are only visible to one person, and therefore we can never be entirely certain that they aren’t simply the delusion of the viewpoint character. That character shares a very unusual relationship with another character, one which (again) isn’t clearly spelled out.

Finally, these two are embroiled in a specific plot, one which begins entirely in medias res, leaving us to piece together the backstory through flashbacks that sometimes obscure more than they reveal. Between trying to decipher the setting, the plot, the characters, the backstory, and the sometimes impenetrable prose style, I frequently found myself feeling very much at sea in the story, lacking adequate purchase in any sort of familiar concept. When the climactic moment came, it didn’t have much of an effect on me, because I never quite understood what exactly underlay the climax’s emotional thrust.

On the other hand, the parts that I did understand, I found quite evocative. There are a number of very arresting images in the game, some wonderful turns of phrase, and the central relationship is a fascinating one. Even better, there are a few lovely puzzles. These puzzles are the best kind — unique and interesting, but of a piece with the story rather than simply tacked on. NOVEL mode makes them feel even more integrated, creating as it does the appearance of an unbroken flow of text.

I encountered one or two grammatical errors while traversing the story, and the occasions when it appeared to be ignoring my input annoyed me somewhat, but overall I came away feeling impressed as hell with My Angel. It’s a daring experiment, executed with grace and care, and it provided me not only with some vivid impressions of setting and story, but with a good deal of food for thought about the possible future directions of IF.

Rating: 8.6