The Saltcast Adventure IFComp 2024 postmortem

The original concept for this bears almost no resemblance to its final form. I watch a lot of indie horror letsplays, and a common trope in that genre is exploration of some secluded and dangerous area with lore delivered via notes the player character finds. The first germ of an idea was to follow the journey of the note-writer more directly, with some player agency on what featured in the messages they left behind, then retrace a changed version of the initial path with a new character who uses those messages to help them progress through the puzzles around them. A sort of temporally-split choice and puzzle game. The only aspects of this idea which survived into Saltcast were the cave system setting and the pov change in the third actā€”although the significance of the latter is again totally different. But it got me started thinking about the world, and the creation myth and magic system gradually materialised, and that got me the beginnings of a plot.

The other plan, once I decided I was actually going for this, was to write something longer than I was used to. Iā€™ve spent the last few years writing short stories. I wanted to build up the stamina to go for a novel. I figured a fairly basic fantasy concept would be fun and would help me avoid getting lost in intricacies of plot, and that the conventional branching paths of IF would allow me to mess about with interesting possibilities while still focusing on a limited number of story beats. It grew a lot in the tellingā€”I exceeded my planned word-count by about 20K. I think it mostly justifies its length, though itā€™s definitely more for people who donā€™t mind something a little ponderous.

Points of tension:

ā€”I struggled a lot with balancing player agency with pacing. I ended up cutting a couple of encounters I was fond of and which provided choice-beats in the narrative because they just made progression feel too slow. There are still fewer meaningful choices than Iā€™d like. Some of the one-click deaths are there just because it felt like there needed to be player engagement at that point, but narratively I could only think of one or two ways to succeedā€”at least without committing to a huge branch in the path which might never rejoin the main flow. It feels a bit like cheating to have them, so Iā€™m not satisfied with that solution, but I still havenā€™t really thought of a better one.

ā€”There were supposed to be puzzles! The second act was supposed to be a gauntlet of puzzle-death-traps you had to pass to get to your goal. When it came to actually writing it, though, I couldnā€™t internally justify why the villain would have left tricky ways of deactivating his protections. Giving up on that, I tried to instead add mini-games based around dodging through the traps, little things where you had to click the correct tile out of a line-up within three seconds, and so on. Here I discovered that coding makes me cry. I was still trying to get them to work a couple days before the deadline, but they went on being buggy nightmares and, much more damningly, were simply not fun to play. I decided getting rid of them was the right call, and Iā€™m very glad I did, though I lost a lot of the functionality of the ā€˜woundsā€™ tracked state with themā€”this would be the main place that stat would come have into play. At the moment, having really high wounds in the first two arcs should block you out of one choice required to have the perfect ending, but the good ending isnā€™t that different, so it does feel like there arenā€™t enough consequences for something which features so prominently. (Wounds in the last arc are actually a different stat, and if theyā€™re high and you get the bad end, they add a scene which softens it a little because the Saltcast are grateful you got hurt helping them out. This is probably also not cued enough.)

ā€”I was maybe more blasĆ© about the third-act pov change than I should have been, even though it mostly worked out! I knew I risked killing momentum with it, but didnā€™t really dare to hope people would be attached enough to Madelaine that the attachment itself might be a factor. I think I underestimated how much role-playing through a defined characterā€™s choices builds a connection with them. I think I got away with it because I personally did care about the character when writing her, and the third act shows that even though I wasnā€™t confident Iā€™d carry the audience with me on the emotional line. Iā€™m delighted people did care, and I think itā€™s a key lesson Iā€™ll take with me.

ā€”Passages are really long. I thought Iā€™d get penalised in the scoring a lot more on that than ended up being the case, so Iā€™m grateful youā€™re all so patient. It was interesting how different the play-times people reported were, presumably just based on reading time.

Despite writing for more than a year with the comp in mind, just before the deadline I was so discouraged with the game that I was seriously considering not entering, mainly because I thought it was so long-winded no-one would want to finish. Iā€™m really, really glad I did. This has been a wonderful experience. I think with future entries Iā€™ll be less worried about the results. Probably nobody is going to chase me out of the community with pitchforks no matter what I produce, and someone out there will enjoy some aspect of it. That said, Iā€™m blown away and honoured to come in the top twenty: it was well beyond my highest ambitions. All the feedback I got was useful, and much of it was delightful. Next year I hope to participate more myself in judging and reviewing.

Final noteā€”I did notice that I was towards the bottom end for number of ratings. Thereā€™s no real way to know why that isā€”length, genre, something in the blurb, something in the cover art? But I do wonder, and please feel free to share any speculation you have on that score. I kind of wonder if it was an advantage: if I just happened to luck into the people who liked the game and never hit the radar of people who wouldnā€™t have. That said, obviously the people who appreciated my masterpiece have immaculate judgement and were entirely correct on the merits, so it all worked out!

16 Likes

Only two choice-based games were listed as ā€˜Longer Than Two Hoursā€™ and both have a low amount of reviews.

Itā€™s pretty normal for parser games that win to be 3-4 hours long (even though the limit is supposed to be 2 hours), but itā€™s more normal for choice-based games to be around 1-1.5 hours long. Nobody says that out loud, itā€™s not a rule, but itā€™s a subconscious thing.

I think itā€™s partly because parser games make it easy to see the progress youā€™ve made; if there are unlocked doors, you know you need to go through them; if thereā€™s a score, you need to reach it. So even if the game is longer, it feels like progress.

But with choice-based games, you could have three hours left or five minutes and it can be hard to tell. A lot of the best games in the past have given people a way to feel the progress of time; Birdland is organized into days, Will Not Let Me Go has a progress bar at the top.

You had Acts, which definitely helped me, but getting to the first act can take some time.

I donā€™t know if that was the main reason or if there are an quick solutions, just some thoughts.

Loved your magic system and your writing! Would love to read more work by you.

5 Likes

Mmm I love hearing how authorā€™s concepts change throughout the development process.

Also interesting to hear your thoughts about the meaningful choice issue, and fascinating that we almost had puzzles or minigames!

OK this does answer some questions I had. :wink:

I donā€™t know, but since speculation is encouraged I would speculate it may have had to do with the length tag? Only a handful of games were entered as ā€œover two hours,ā€ which is a big commitment. Several of the other ā€œover two hourā€ games were by authors with other extant games that may have attracted people.

5 Likes

I am really happy that you submitted this. It was one of my super-favorite pieces (top 2). And as someone who cares about the writing first and interactivity second, it makes me feel good to see that well-plotted longer pieces that have interactivity, but not necessarily as the overwhelming focus, are welcome in this space. So I definitely wouldā€™ve missed not seeing it here.

My speculation (similar to what has been said above) would be that length was the biggest factor. You mention that it is a 70,000 word piece in the blurb. In other contexts, this would be great! When reading work in the wild (a phrase which here means, mainly AO3 I guess :skull:) I much prefer longer works because it feels more worth my time to get invested in that world.

But in this specific context, a competition where judges are arbitrarily required to evaluate a work in only 2 hours rather than having an open-ended reading time, among a sea of 60+ entries, I could see it coming across as an overwhelming amount to try to read quickly.

5 Likes

Yep, that seems like almost definitely the answer. I figured it probably was, but since I am a stack of shivering anxieties concealed in a trenchcoat, I was worried there was some other hugely off-putting element in the blurb I wasnā€™t picking up on.
Iā€™m so glad you liked the story! I loved reading your review.

7 Likes

on scrapping the puzzles (gaining in the Zarfian scale, I guess) in favor of dodging with mini-games, seems that you forget what I consider a main tenet on villains and villaining: no villain wants to literally fell on its own traps ! hence, a villain worth of his title will always care to have a local, fallback, well-hidden mean of deactivation, isnā€™t ?

Best regards from Italy,
dott. Piergiorgio.