Reviews by MathBrush

This is my third and final Lamppost Projects game to play during this competition, and it is quite different than the rest in some ways.

All three games are set in a D&D-lite world with orcs, half-orcs, halflings, tieflings, magicians, and a setting a little later in European history than most fantasy I've dealt with (this one seems to be around 1600s or later, maybe even 1700s).

All three games also feature watercolor-looking art and a collection of four or more romanceable characters per game, of varying races and genders.

Where this game differs from the rest is that you have skills and animated dice rolls; the others had no randomness at all. The animated dice rolls look really satisfying and seeing the numbers and the target difficulty (and the way the game encourages you to try and fail and keep trying, just like a good GM) makes this a much more pleasant randomized experience for me than most.

You are a private investigator brought into to protect an opera from a threat of robbery. You have to meet the various performers and backstage people and take careful notes, while making use of the background knowledge you chose beforehand. I focused on observation but made myself clumsy, so I did great in conversations but pretty bad when trying to sneak peeks at things covertly.

One outstanding feature of the game is that you can guess the truth of the game at any time starting near act 1, and the game rolls with it if you get it right, which I did right at the end of act 2. You have to pick the right suspect, motive, means, etc. and what's great is that you only have to be mostly right (I had the wrong motive, but otherwise succeeded). If you succeed most of your rolls for your good skills, the villain is fairly obvious, but the target and motives eluded me at first.

I think I like this game best of the three despite a few rough edges (there is romance but it's all packed at the very end of the game and feels separate from the rest), because I have a personal fondness for detective stories, and deduction is very hard to model but this system is one I'll mention in the future when others ask about mystery game advice in the future.

I've played a lot of Stewart Baker's games, and I usually associate him with lighthearted longform narrative-focused choice-based games.

This is a more serious choice-based game with a complex world map and a lot of navigation and some light tracking of objects. To me, it feels like an experiment in shifting tones, and so I'm framing my review with that mindset. That may be an incorrect interpretation, which may render parts of this review less valid.

In this game, you are on a sea, your memories shattered into several (visible) pieces and you have to visit the important locations of your life to remember and revive that part of your life. You learn about your experiences as a youth and as a powerful wizard, and can visit the islands in any order. The world map is navigated with compass directions, as are the individual islands. In each memory, you wander around completing tasks, often tasked with going to specific parts in the map one at a time. Some memories are much shorter. As you complete memories, you have some leeway in how to end them, which raises your score in one of three attributes.

I think recovering your memories as a powerful but defeated creature is a solid trope and works well here. It reminds me of the game Dreamhold, a parser game where you are similarly navigate a space collecting your past memories of a life involving magic and power.

The most effective parts of the game to me were the heavily unusual parts, like what happened to our childhood schoolfriend and what lies hidden below the caves we explore. The author has a talent for describing the truly unusual in an unsettling way.

I was less enthusiastic about the world model and compass navigation. There were large swathes of maps that were essentially 'Hallway D' (but the outdoors equivalent'. I remember something Adam Cadre wrote in a review of Galatea:

In interviews I've been asked to give potential IF authors out there advice, and one of my usual lines is, "The pieces of text you write are the player's reward for thinking of the command that calls them up. So make them rewarding."

While this is choice-based and you don't need to think up commands, it still holds when it comes to discovering new text through exploration. Some of these descriptions could use some more excitement:

This hut is nothing special. Twenty strides by twenty, it holds whatever the village needs holding.

You can reach the veranda of your home to the south-east, while a dirt path leads south-west towards the main road.

This isn't universal advice; Wizard Sniffer has shockingly bare room descriptions and a lot of connecting hallways yet is still well-beloved, but that's mostly due to the large and lively cast that provides flavor in those rooms. (Sorry for the long digression!)

I chose to mostly focus on Despair, one of the three stats you start with. I enjoyed the freedom to choose what to focus on. I think there was one very minor bug near the end where, even though I had used despair the most (both getting and giving), the options I had to choose were ones the walkthrough indicated as applying to coldness. (Specifically, [spoiler]the trials I faced were fire and a rickety bridge, though my stats were 3 1 1 (D C R) and I spent 2 1 1 on the four islands.[/spoiler]

The overall plot is something I liked, and it felt like a replay would definitely be meaningful, since most islands have very different endings to the stories depending on you choice, and there's no back button (there are saves though). So I thought that part of the game was particularly well-constructed.

Lady Thalia and the Case of Clephan

This is the latest entry in a longstanding series of Twine games featuring a lady thief protagonist, a series of stealth and theft puzzles, and conversations with different tacts.

In this game, our lady thief Thalia has reformed, and is now helping to run a detective agency with her will-she-won't-she former police officer Mel. With no more thieving, Thalia is a bit down in the dumps. Things take a turn, though, when a new thief appears:

Lady Thalia. Another lady Thalia, that is. This case of stolen identity is resolved through four different chapters.

The main conversation system worked well for me in this entry. You pick between being Friendly, Direct, and Leading On, depending on the personality of your listener. If you pick the wrong one, it will give you verbal cues showing the problem with your approach, and you can adjust. I do admit I reloaded some saves (not necessary, just wanted to) to 'fix' some mistakes I had made, but it always felt fair.

The exploration portions include some code-finding for locks, some searching of rooms, etc. I did get confused a bit in one area because I didn't map it out, but this fits the game itself, where paying close attention earlier and taking notes can give you big advantages later on.

It's probably recency bias but I like this one more than my memory of liking the others in the series, and since I liked those, this must be pretty good!

(I have the feeling that I forgot to comment on some essential part of the game, but what it is has slipped my mind. If someone who's played it feels I omitted something important, let me know!)

Edit: Maybe it was the fact that your persuasion doesn't work on your love interest and that that is commented on in-game, which was a fun way to handle things and a commentary on the usual formula of 'say five things this person likes and then you can make out'. I do like some games where it's not what you say that someone likes but who you are, so as you make choices that adjust your personality, you fall in and out of the range of likability for the ROs. In this game I think your true nature is pretty fixed, and that's what Mel likes.

Editedit: After some reminders, I remember what it was. With Mel on your side, you can now tag-team conversations, deciding who says what and in what order, which was really a huge amount of complexity (it felt like to me) but not punishing, so very nice.

Also, I felt the game was setting up a very obvious twist in the villain reveal, but it subverted it at the end in a way that made all the earlier clues still kind of make sense.

Before I played this game, someone told me that it had significantly improved on the last two games, which I thought was pretty neat. After playing, I do agree that it's the strongest of the three.

This is part of a series of three games involving hard-bitten detectives, drug rings, exotic dancers and organized crime--except, everything is candy or sweets: the people, the blood, the drugs, the river. Each game has a murder that you investigate, then track down the murderer and accuse them. I don't completely recall everything from the other two games, but all the locations and many people this time seemed familiar. There's a lot of new stuff, though.

The main features of this game are Jimmy Pinata, the strung-up, disemboweled victim (very normal for a pinata, but not normal for a sentient being); and blue rock candy, an ultra-pure drug that's flooded the markets. You have to track these both down.

Gameplay is a mix of classic parser take/drop/lock/unlock and ask/tell/show conversation. The topics available are pretty robust, with most labelled in bold but allowing you to learn of a topic in one conversation and use it in another.

In the past, I associated these gumshoe games with having an incredible setting, a solid story, compelling characters, and kind of shaky implementation. The implementation has gotten substantially better over time, but I found myself fighting over synonyms for a lot of the game. I did try playing without the walkthrough (as opposed to my early ifcomp days where I'd use a walkthrough from the beginning to power through as many games as possible), and got really far, but there were a few times where I was foiled by lack of synonyms or alternative solutions not working (or, just being dumb!). I didn't identify any bugs, though, and I'm sending a transcript to the author, so I suspect if you're reading this from the mildly distant future that you may not have as much trouble as me (which, again, wasn't really that bad. This is among my most enjoyed games this comp so far).

This gimmick is almost infinitely exploitable; you can put any hardboiled old story in here and make it work. You could branch out and do a candy version of the movie Vertigo, or modernize it and do a candy Bourne Ultimatum. This is essentially the IF version of the muppets, where you can do a take on any story and make it funnier, so I hope it continues.

Ben Jackson has had several smooth, polished twine games released in recent years, so I had high expectations for this game, and I feel it delivered. At one point I thought to myself, "I wonder if I would ever be able to make something like this."

It leans hard on classic sci-fi tropes, including the classic 'wake up from a pod on a destroyed spaceship while you have amnesia' and the evergreen 'work with a ship's AI that you can't be sure is working for or against you'. They're classics for a reason, because they can work great in an interactive fiction setting. And here, the author has expanded on them to give them a distinct and unique touch.

You play as a crew member on a ship that has been 'reprinted' as a backup after all crew have, presumably, died. Most of the ship is lacking oxygen, and you have to get out to explore.

The author has used escape-room puzzles in past games like codes and minigames, and this is no exception. We have a lot of doors to open and a few other code-style or 'which item will help here?' puzzles.

But the major change here is recycling and fabricating. Throughout the ship, we can find floating clouds of debris that include things like fabrication recipes and junk. Throwing the junk into the recycling bin, we can get materials in 5 or 6 different categories which are used to make new items.

I enjoyed the progression of the game, especially when I reached a point where I could build equipment that completely solved several long-term problems. It reminded me in a good way of Trigaea, one of my favorite twine games where you gather resources, come back to base, fabricate new stuff and get new recipes.

I hit a wall with two different puzzles. The first turned out to be optional and was hard on purpose, although I feel like completionists would have found it more easily due to trying out everything else in the game. The second was the final puzzle, where I thought I had locked myself out of victory (but turned out not to have).

I felt like both crafting and storytelling had 'real' decisions. My decision on how to handle the AI early on radically changed parts of the game, and the order in which you get upgrades can make a big difference in your play experience.

Like others, I enjoyed the final plot twists, which seemed well-hinted at but still surprising in the exact way it plays out.

Overall, a lot of fun.

This game is written in Dialog and can be played entirely using hyperlinks. It has a minimap at the top that can be used for navigation, persistent links at the bottom, context-sensitive links for the current object being utilized, and green links for trivia and definitions.

You play as a dog in the Hittite empire. Magic is real, and as a dog, you can absorb it and transmit it to other things (Not people though! That's actually against the law. Straight to jail, believe it or not).

You work for a wise-woman, but she has been afflicted by a curse! You need to help her, but you can't even get out of your own home. Once free, your world expands more and more. You can help friends and gain new curses and blessings in your large village, visit the capital and make money, and gain greater power than you thought possible.

Like Daniel Stelzer's other hyperlinked Dialog game about a dog interacting with the supernatural and their afflicted older woman master (Miss Gosling's Last Case), this game has context-sensitive tips and tutorial messages at the beginning, but they've been tuned to be less intrusive, which is nice. There is also a 'think' function that tells you what puzzles you can solve, and the minimap also does that automatically. That's helpful in this sprawling game with many options.

Puzzle difficulty was hit or miss with me. Several times I felt like there were several reasonable options that the author ruled out for what felt like arbitrary reasons to me. A common source of frustration for me was intuiting when a movable object could be affected by fragility or by wind (or the opposites). Rather than making puzzles simpler, the author has instead added a lot of hints (as mentioned above) and made most puzzles optional; for the two largest areas, you only need to reach a certain minimum number of puzzles solved before you can move on or win the game. This reminds me of math tests: is it better to have a test most people can get a high score on, or to make it very hard and then just 'curve' it significantly? This is a 'curved' game.

The background material on the Hittites was fascinating. One common theme was that words had one pronunciation but are written with symbols that have another pronunciation, which reminded me of kanji with Chinese and Japanese readings.

Overall, I found the game substantial and fun. I got stuck several times and used the hints about 4-5 times. This is also the first shady ancient copper merchant I've found in a game that wasn't Ea-Nasir.

This parser takes a bold and innovative direction, and while I think it struggles with the execution, it's nice to see people experimenting and having fun.

In this parser game, you rule a country, but instead of moving from room to room or working with menus, you just give orders. The game itself doesn't give you any real instructions besides 'type what an emperor might do'.

I tried talking to my advisor, who suggested reviewing military deployments. I typed 'REVIEW MILITARY DEPLOYMENTS' and got a list of troops and number of places. I decided to recruit more by typing RECRUIT TROOPS IN , but I ran out of money. So I tried RAISE TAXES, and that worked. Some barbarians attacked, so I tried ATTACK and that worked.

I kind of ran out of steam then. There are some random events that you can respond to, but by that point I couldn't think of anything else to do. I peeked at the walkthrough and saw a list of actions I could try like 'condemn' (although it didn't let me condemn most things I tried).

I then restarted and tried the actual walkthrough. It had a lot of actions I hadn't considered (especially since some were in response to random events I hadn't seen yet), and due to randomization the walkthrough didn't 'work' and I'm not sure there's any ending to the game. Although, as I type this, I decided to try and type z.z.z.z.z.z and copy it over and over again, and was able to get a bad ending as my capital was sacked.

I think the concept (you can type anything!) is exciting, and a lot is implemented, but without stated restrictions or guidance it felt like I was stumbling blindfolded around a large, mostly empty room, trying to find scattered objects placed here and there (here the large room is the state space of all possible parser commands and the objects are the implemented actions).

Every writer writes for different audiences, so I may or may not be the target, but I think I would have had more fun if I had an idea of my long-term goal and about the relative amounts and specificity of things (does raising taxes give lots of money or little? Do I tax everyone, only some people, or only some things?).

Outside of that, the game is smooth, well-polished, and the writing clearly communicated what had occurred.

This was the second-most substantial game I've played this IFComp, and took me around 3.5 hours to play, with dinner in the middle, so about 2.5-3 hours of playtime.

This is a rich and complex game. Instead of focusing on a multitude of small choices, it has a small set of meaningful choices that shape the later narrative. There's a lot of branches here, leading to 500K words, though much of that is due to minor variations of text; with efficient variable use, it would probably be 200-300K, still very large, but that would also probably make it buggy, so I think the author made a smart move here.

The choices themselves manage to be some of the most difficult ones in my recent gameplaying history. Each one is a compromise, giving something you want and something you don't.

But I guess I should describe the story. For those who aren't aware DemonApologist is a talented author whose works almost exclusively focus on what I might describe as monsteryaoi, where a human engages in a romantic relationship with a being typically depicted as evil or malevolent by others. I've found their past works to be engaging and good with dialogue and emotion, so much so that I have tried to study them as I practice writing romance for commercial fiction.

This game is no exception to the monster-loving pattern; in fact, it's the most well-formed example of it I've played. It also, as I describe earlier, provides an enhanced sense of agency.

Our protagonist is a humble initiate who has been prevented from ascending in the ranks of magicdom due to a cruel and callous advisor who won't let him graduate due to his sympathy towards demons. The advisor even summons demons and has them fight each other to the death to demonstrate how unimportant their lives are. He then gives the initiate an impossible task: to light the Pharos Fidelis, an extremely cursed lighthouse at the center of a magical storm.

He is also explicitly asked not to summon demons, and, if he does, to expect them to kill him.

Our initiate, therefore, takes the most logical action in the moment, which is to summon the hottest demon he knows (the one who was defeated in the earlier duel) and to flirt with him awkwardly.

Fortunately, it works out! Or not. That's where the choices come into play. At critical moments, 'you', the viewpoint character (different form the protagonist), get to influence the demon towards one of two options. Each option comes with one benefit and one drawback. These critical moments stack, producing numerous branching timelines and a ton of different endings.

The game looks great. The UI has little gears that pop open side comments and commentary, and I especially loved the background color change when the big event occurred. That event itself was described quite beautifully.

Perhaps surprisingly, the romance in this game peaks in the middle, not the end, allowing us plenty of time to see what a fully formed relationship might look like. I am reticent to play explicit games, but the game is very tasteful in describing our interactions with the demon and I did not feel distressed.

The game left me with a question in my mind about demon love and the concept of demon apologism in general. What is the essential core and appeal of the demon? Is it to be evil, itself? Or is it to be called evil by others? Would a sainted angel who is angelic in nature but hated by a cruel world still feel like a demon, or is it more important that the demon be ruthless and aggressive in nature and only tamed by the touch of man? If a demon turns out to be a good guy, does that erase his demon nature? If to be a demon is to be evil incarnate, can a demon truly make someone happy, a decidedly non-evil act? If it is not evil incarnate, then what makes a demon? It is a paradox, and not just a fruitless one. A lot of romance and even stories in general pivot on the notion of a 'bad boy/girl' that ends up having a heart of gold and doesn't really do bad boy things at all. This isn't directed to the author, it's just something their game made me think about a lot, because I think it's core to a large swathe of storytelling.

Penny Nichols, Troubleshooter

This game consists of instructions to an AI on how to run a game for a human player.

I was excited to see this, because I've often thought when playing an AI-generated game: "I bet the prompt the author used would be so much more interesting than the output I have to read." And here I had the prompt itself!

So I tried playing it twice. In the first, I made myself both player and DM. Instead of AI, I used it as a writing prompt as I explored what might happen with these characters and this setting. I had a lot of fun; I usually struggle to write more than 300 words at a time and 1000 in a day, but I got up to 2270 words in less than an hour. I relied heavily on cliches and tropes but I liked the setting and concept.

I then plugged it into Copilot. Copilot made it look fancier and was much, much faster. Parts of it were interesting and fun to read. It was much more fast paced than my own transcript, and I had been trying to go fast myself. Amusingly, we used a lot of the same cliches and tropes. Where I was most disappointed is that, where I had tried hard to spool out the mysteries and fill in the backstories and characters, the AI just gave away half of the secrets in the first few paragraphs and mostly ignored my companion characters and almost all of the backstory. I felt like it wasn't really making use of the extensive mechanics sheet and was more just giving a series of climactic scenes without real buildup or denouement. It makes me feel like I couldn't personally rely on an LLM to follow my instructions if I were to make such a game myself.

Interesting concept, very glad I just got the prompt instead of having to suffer through pre-generated pages and pages of boring prose and instead got to write my own pages of boring prose.

A Conversation in a Dark Room

This is an author's first game, but is well-polished and has multiple endings and engaging dialogue.

You play as a man hired to be involved in a death. You meet a man in a dark bar late at night, and the two of you have an in-depth conversation. You are a reporter, but it's not clear that you'll be doing any reporting tonight. This seems more intimate. Your counterpart is old and wealthy, very wealthy in fact.

The story is split into three chapters, two of which are in the same location. Mechanically, you have three different stats that you can increase, which the game helpfully clarifies with some about text early on and popups when a stat goes up. Depending on your stats, you can get one of five different endings.

I think this is a promising start and that this author has hurdled over many of the mistakes new authors make.

There were a couple of things that would have enhanced or changed my experience.

While I just recently posted a review praising a different game for slow text, which I usually detest, my experience with this game was a return to form. On chapter titles and a few other select screens, text is spooled out painfully slowly. There are usually two reasons, I suspect, that people use slow text:

One, they want to control the experience of the reader by emphasizing important lines or moments. Many times I believe this is due to a lack of confidence in the power of static organization. Paragraph breaks, fonts, font size, and page breaks naturally provide a pacing for text that has been used effectively for thousands of years (like the elaborate capitals in illustrated manuscripts). At times, slowing down can provide drama by keeping the most interesting tidbits to the end, but in this case it was just regular text that was slowed down.

Second, some people slow down text for cinematic effect, to be like a movie. I think that can be used appropriately (Ryan Veeder uses a nice intro technique in some of his Little Match Girl games), but I personally appreciate it better when it's part of an overall audiovisual strategy and not the only movie-like element used.

So I downloaded the game and edited the code to speed it up. One good rule of thumb is that, if you have trouble sitting through slow text while playing your game (and everyone should play their game while writing it), the player won't enjoy it either.

The second thing is that I found it a little difficult to strategize the different paths of the game. The stats are genius, and they already provide branching and replay value, but I often found it hard to figure out what the effect of each choice would have on stats. I personally would have found it more fun to be more clear.

But these all are just my opinions; there's no one true way to write games, so I offer this as only my own account of my game experience.

As a final side note, I don't drink, but see it a lot in media, and I was pretty surprised by the drink count of my character by the end of the game. I took one drink early on, and after that they ended up pounding down seven drinks in the night, were then invited over for whiskey, and, in one ending, propose going to another bar. I checked and it looks like that amount of drinking can get a lot of people blackout drunk, and is pretty high above the bar for 'binge drinking', so I wonder if our character is a hardcore alcoholic or is going to have a really bad day tomorrow. Both are completely appropriate for this game, so this is not a criticism, it was just fascinating to learn more about drinking culture.