Reviewing some IFComp games, part 5 of ???
Games reviewed (or marked DNP) so far: 20
Games DNP'd: 3
Current best-in-show: The Witch Girls, by Amy Stevens
Favorite game in this batch: The Tempest of Baraqiel, by Nathan Leigh
Moon Logic, by Lancelot
This is a Twine game that's essentially a parody of Zork – the joke here is that you're playing an 'AI enhanced' version of Zork, in which you select verbs and the game's 'smart' parser then selects the nouns. This is then used to set up puzzles that are solved, more or less, by selecting verbs in the correct order to essentially trick and manipulate the parser into doing what you want.
This is complemented by a sort of split-screen UI; on the left is the actual text of the game-within-a-game, but on the right is running commentary by two friends playing the game, which provides clues (really, more or less explicit instruction). The whole thing is built around figuring out the obscure logic of how the interface works, and then inputting the right sequence of verbs to progress the game.
Mostly I found this puzzle design pretty tedious. It has a big problem with both the number of inputs needed to solve puzzles and the time it takes to input them – by default, the game slowly writes out game text with each step. The most frustrating aspect of the game is the verb 'go'; the whole game map is essentially one big line; you 'go' in one direction and then, at the end when you can make no more progress, turn around and start going back the other way. This means that if you input an errant 'go' command, which is easy to do when trying to solve a certain puzzle, you then have to backtrack through the entire game so far and back before you're allowed to resume whatever you were doing.
This and the overall low-agency way the game operates – you're kind of just hitting verbs like 'drop' and 'take' hoping the game will drop or take the right things, etcetera – just makes this a real slog to crawl through. The jokes are occasionally funny, but I wouldn't say enough to that I was excited to play through the rather clunky puzzles.
The underlying idea here seems sound to me: a parser game 'enhanced' by a bad UI that tries to guess your intent, and puzzles centered around tricking that bad UI, with the underlying foundation being a bunch of standard puzzles from a classic game everyone knows. But I don't think this game's design really solves the problem of how to construct that without it becoming actually frustrating rather than a joke about frustrating games.
Rain Check-In by Zeno Pillan
Did not play; author used an LLM to write it.
The Tempest of Baraqiel, by Nathan Leigh
This is a fairly long, mostly linear choice-based story built in a bespoke system akin to something like ChoiceScript. It's billed a homage to 70's science fiction, which it absolutely is – this is military sci-fi in the vein of something like Robert A Heinlein; humanity is losing the bug war, and you're a naval officer – and scientist – recruited to a project that might just turn it all around.
I played this through to a 'bad' ending; since the game didn't really signal that this was possible, I didn't bother saving. As it also doesn't have an 'undo', and is actually fairly long, I didn't have time to replay it through to a different ending, so I didn't really get a full impression of the story. This seems structured like a classic SF short story, with the plot hinging around an 'aha' moment at the end, but I never got that 'aha' moment. Oops!
There are some other technical issues here. I wasn't able to play this with audio; the game is supposed to feature a procedural music score, but my attempts at playing it with sound resulted in the long load time for all of the audio files stalling. I think IFComp's own fairly slow servers don't enjoy me trying to download 200mb of music stems spread across 300 individual files. So I feel like I didn't get the full experience in this sense either.
Beyond that, the presentation is enriched with illustrations made very much in the style of 90s cd-rom game art – retro 3D renders with hard surfaces and looping animations, presented as pretty crunchy indexed-palette gifs. It's charming, although I think some more polish to the overall interface (as in, layout, typeface choices, colors) could have elevated that more.
My synopsis above makes the writing here sound really schlocky, but that's not really the case. It does a pretty good impression of this style of sci-fi, both in prose style and tone, and it takes its characters and world more or less seriously. There's some good character work here, and some interesting worldbuilding that makes this more interesting than an off-the-shelf retrofuture you could easily imagine this as.
Formally, this is a dialogue-heavy story in which all choices are lines of dialogue. While it has a solid player character protagonist, it also opts to have dialogue choices sometimes belong to other characters, occasionally allowing the player to branch the behavior of various characters in the story. From my single playthrough I couldn't glean whether there was a lot of hidden state or variability here; I think this story has a strict branch-and-bottleneck structure in which player knowledge is the only thing that can separate distinct paths after they rejoin, but I am not 100% certain.
This gives the entire thing a bit of a light choice puzzle feel – it's trying to reward you for picking up on certain clues and knowing the right things to say – but this is paired with it being a fairly unforgiving choice story with constant forward momentum.
Not having seen the denouement, I can't really speak to the overall plot, but I did get the sense that the final act tended to run towards getting a bit lost in the sauce. I found that the dialogue in later sections was not really written to cope with complicated scenes with multiple characters talking and doing different things; it also represents a fairly abrupt and somewhat out of character escalation relative to the story so far (to avoid spoilers).
I also don't know how much of the sci-fi conceits at the center of the story really land, but again, I didn't see a proper denouement.
I think this is one case where the author has nailed some aspects of writing (the tone and prose style of the genre he's operating under) but is missing the mark a bit on others (there's some clarity and pacing issues). Similarly, I think the narrative design here is a pretty unrefined and seems like a first pass at choice design and building a custom system, although there are some interesting ideas.
This is one where I'd encourage the author to do some exploration of prior art in the medium and get a better sense of what's possible, and maybe try building some games with off-the-shelf systems (most of which are open-source and freely available).
Grove of Bones, by Jacic
This is a short dark fantasy story implemented in ChoiceScript.
Structurally, this piece doesn't have a lot of confidence about how it's using agency and choice; the choices are all over the place in terms of their affect. Mostly I'd describe the agency here as customization – you can choose your character's biographical details. The branching is actually fairly minimal. I did not feel like the interactivity was enhancing the story, but it wasn't grating either.
Writing-wise... On a broadly conceptual level the story just isn't doing anything particularly original. The plot concerns a village that sustains itself through human sacrifice. This is a stock story concept, but that's okay; the problem is that it has no specific angle or hook into that idea that makes it memorable or gives it a sense of identity. The milieu is very thin; the story doesn't really situate itself in any particular time or place. There's very little in the way of a world or even characterization here.
The prose ends up reading as very clunky. It's not really in control of its tone or rhythm; It's verbose without actually conveying much information. It has that problem a lot of beginners have when writing fantasy, where every character speaks in an overly formal register so that your mob of peasants choosing who to sacrifice to the evil trees sounds like a high school debate club.
Mostly this just suffers from an overarching lack of an angle – a unique viewpoint on the events of the story, or a particular way of letting the reader into a scene. Everything is very flatly delivered with very standard imagery. An old woman's face is 'lined by countless years'. A ghost says, in so many words, that 'the tormented remains of [their] soul remain tied to this plane'.
I think we can overvalue originality in writing or take it as a goal in itself, but it's also true that when something feels very well-trodden – which this does – it's hard to latch on to anything or even just to pay attention. What I often want out of writing is not so much original ideas but a real individual voice, and I struggle to characterize the author's voice in this because it's so underdeveloped.