Really happy with how my game came out and it's lovely to see that reflected in the ratings, thank you to everyone who played Puppeteer!
If you want to try out some perspective based platforming puzzles you can play it here: https://benskca.itch.io/puppeteer
Hi guys !
I just wanted to thank all of you for this great time on LD57, played great games with fantastic design, had great comments and feedback on my first LD game + on our first game (for the team composed of my brother and I as we used to make games with a team of 5 members, essentially on GGJ every year).
Eleven Studio is just born on this LD57 !
The community here is so kind. I just can't wait for the next one in October ! See you there !
EXCLUSIVE CONTENT :rofl: here one of the future evolution of UI, what d oyou think about ?
Do not hesitate to follow us on itch : https://elevenstudio.itch.io/
First of all — thank you to everyone who played, rated, commented, left feedback, recorded videos, laughed, cried, ran out of ammo, and punched a bee in desperation.
Thanks to you, this was our strongest Ludum Dare performance ever as a team — not just in terms of score, but in how we felt about what we made.
This was the Ludum Dare where we finally clicked.
🛠️ What Went Right
We actually planned first. Instead of charging in with vibes only (RIP Paper Cat Farms), we spent about an hour deciding what we really wanted: a short, punchy game you could beat in about 10 minutes, dense with action, and fun enough to come back to.
We balanced from experience. We followed an old-school design principle (popularized at id Software, creators of Doom and Quake):
"If you can beat your game blindfolded, that's Hard. Normal should come before that."
This helped us keep the main route satisfying but not soul-crushing. Endless Mode is there if you want to hurt yourself. 😉
Our small early maps were intentional. We learned during testing that huge mazes with three enemies hiding at the end were just... boring. Early levels are small and fast on purpose to keep pacing tight and fun. Later levels grow larger and more challenging.
🐝 The Soul of the Game: Ammo Management
We received tons of thoughtful feedback (thank you again!) about ammo scarcity. Many players noted that running out of ammo could feel brutal — but also strategic.
We want to be very clear: this was deliberate. Ammo management wasn’t a flaw — it was the soul of Dive Into the Hive Mind.
We didn’t want you to be Rambo. We wanted you to think like Agent 47:
-Make every shot count.
-Decide when it’s worth the risk to punch.
-Adapt when things don't go to plan.
Some players loved this tension. Some struggled. But we believe it gave the game its identity, and we’re proud we stuck to that vision.
🎨 Behind the Scenes
Bizzybee – Art, Music, and SFX Biz went full mad scientist, doodling like a whirlwind and composing electric, neuron-inspired music centered around the note B ("bee," get it?). Despite a recent chronic fatigue diagnosis, she powered through and made more art for this LD than ever before. Constraint truly made creativity flourish this time.
Fgeva – Programming With up-front design this time, there were no last-minute panics. Fgeva hit all our main goals and even some stretch goals — though he may or may not have been hallucinating from sleep deprivation by the end. (LD tradition!)
Balimaar – Programming Terrified about implementing procedural generation — and then utterly euphoric when it worked. This jam had way fewer "we're doomed" moments thanks to early planning. Also: much less bee-serk panic compared to LD56.
🎯 Moving Forward
We are soon to release a Post-Jam Version fixing many rough edges:
-Polished visuals and SFX balancing
-Smoothed out web build issues
-Controller input fixes
-Small tweaks based on feedback
But importantly: the ammo balance remains true to the original vision. We made a game that demands careful thought and adaptation — and we’re sticking to that.
❤️ Final Thank You
Thank you for diving into the hive mind with us. This was hands-down the best Ludum Dare we’ve had — and it’s thanks to all of you who played, commented, encouraged, critiqued, and took a chance on our weird little bee-brain experiment.
We hope you had fun. We sure did.
🐝💚 — The Hive Mind Team (Bizzybee, Fgeva, Balimaar)
Hey all! It's been a hot minute since I did one of these. Why is that? Well, 3 years ago I graduated college and got a game industry job of my own, woo! But that also means I didn't exactly have the creative energy to do a lot of game dev in my free time. But now I'm back, baby! And with all the experience I've gotten from my job to boot. Did these lessons guide me to game jam greatness? Let's find out!
While the game starts out fun, it really shines once you get into the cool movement options provided by the whip later in the game. The grapple points are just super fun to launch off of, making for some daring setups. But the wall cling is what really turns the game on its head. I especially love the room with the moving blocks, since the grapple interacts with the collision in so many cool ways. It also lets you take some creative shortcuts, if you can spot them. I'd actually be kinda interested to see what a speed run of this game would look like 👀
I've definitely improved on my pixel art ever since I started Ludum Dare back in high school. I'm really proud of just how expressive and bouncy the characters are in this. I mean, just look at my boy move:
Even a simple enemy like this arrow trap got a nice little arm pump:
One aspect I'm particularly proud of is all the little details in the HUD. My favorite has to be the effects on the health bar when taking damage:
One of the major skills I've learned from my job is how to polish an enemy, and I think I did a fairly good job here. Even a simple enemy like the snake has this nice ease in/out to its movement. This is not only visually appealing, but helps test the player on their knowledge of the whip's range:
This bat enemy has to be my favorite though, the way it moves around is just so fun (even if I did phone the art in a little, lol)
One big thing I learned from this job is that my old custom engine... sucked! Like, really sucked. So one of my goals this jam was to use a new custom engine created with all the lessons I've learned over the years. So, in preparation for this jam I spent some time laying the ground work for this engine, getting sprites animating, physics working. But I wound up having to spend a lot of time during the jam both adding new features to the engine, like hitboxes and easing, while also fixing major issues with the physics system. My original intention was to participate in the Compo, but I was nowhere near done and wound up having to take an extra day to finish as part of the Jam, making this my first Jam entry. I actually didn't take Monday off work since I foolishly believed I wouldn't need it, so I had to pull an all nighter on Sunday to finish, which was... not great 😬
As you might expect, new engine means new bugs, and I wasn't able to fix all of them in time. Notably, there were some bugs with the grapple near the end of the game which let the player clip through walls in some places. This didn't make the game unplayable, but was annoying enough that I decided to make a post Jam version to fix it Monday night.
However, some players reported that the game wouldn't start at all, or the sprites would all be missing. Eventually I got in touch with a player who had this issue and was able to get an error log from them. Turns out it was a system language setting issues. It was easy to fix at that point, but a frustrating issue to run into. Let this be a lesson not to rely on system string functions working the same in all languages!
You may be surprised to learn that the grappling mechanics, which are the best part of the game, were actually added on the last day! I always intended this to be a metroidvania, but now I'm not sure if that was such a bright idea. By the time you get the grapple, the game is almost over. There're barely any setups that combine the grapple points with the wall cling before the game just abruptly ends. This also means there's not much time to really learn the grapple before putting it through its paces, so some players had a lot of trouble on the final section. Maybe it would have been better if you got the grapple really early and the whole game was built around it.
Some players reported frustration with the controls. Some of this is normal for a castlevania-esque game like this, but some I think could be improved. In particular, a lot of players were frustrated you couldn't change facing direction in mid air. This was done so you couldn't re-grab the same wall with the grapple (maybe this would have been more apparent if you got it earlier, lol). But I could have just made this a limitation only when jumping from a wall cling, and let you change dir when jumping from the ground. Or, at the very least, I could give you like 10 frames or so at the start of a jump where you can change facing dir to make it a bit more lenient.
I'll be honest, I'm a bit disappointed with my final results. My usual goal is to get top 100 in at least 1 category, which I missed out on this time (although I came close with Fun). This was somewhat expected since it was my first time doing the Jam, which is a more competitive category. But my raw scores were also about on par with my past entries, which felt a bit discouraging. I really felt this game was an improvement.
Part of me wonders if some people who couldn't start the game due to that crash just rated it poorly without playing; one player did comment that they couldn't play but left a review "based on the gifs". But I doubt many people did that, and I'm probably just coping here, lol.
So, was Depth Delver a success? Well, if the point of this game was to take all the experience I've gained over the years and pour it into something that would blow my past entries out of the water, then no, by the numbers it didn't achieve that. BUT if my goal was to gain new experience and improve my skills even further, then I do think it was a success! I was able to make significant progress on my engine, and I'm confident now that I could make a game of a similar scope within the Compo deadline next time (and without the bugs).
So, thanks to everyone who played and rated my game, I value each and every one of your comments more than you know. And, until next time, see yah!
PS: here's the game if you wanna try it yourself: https://ldjam.com/events/ludum-dare/57/depth-delver
I'm an overthinker and usually overprepared for events; after multiple attempts I was burnout and couldn't participate on previous LDJams.
I was laying in bed when noticed a LDJAM stream, I was not aware that it was going to happen that weekend. I challenge myself to a single mechanic, a bite-size web experience, where I could polish my stronger traits. Discuss the game with my wife and proceeded to setup the game models in blender.
I had to kill my darlings, since I only had 1 day to make it: a smaller level, single camera angle, no dialog or humor and shooting mechanic. Submit it at 7 am CST on a Sunday and went to my planned commitments.
Previously; I was accused of flowing the board promoting my games so I kept it to 3 posts, and focus on playing games.
I woke up to my best rated game so far. I was not expecting this! My goal was to have at least one stat within the first one hundred of compo.
https://www.youtube.com/watch?v=EUHZBGWL5T8
...and that might be just why it worked.
(A CLOCK IN post-mortem and a story on telling stories or is it deep to tell a deep story?)
How it started:
How it ended:
It ended up ranking #3 in Mood. (Bragging, I know, but I'm proud that for once I succeeded on what I tried to succeed)
For most Jams, I like to take a few hours (or a whole night) just thinking about the theme, no rush, no pressure, just letting ideas brew while I walk around, talk to family or even sleep on it.
It's honestly one of my favorite parts of the jam: wasting time on weird, probably (likely) bad ideas that only sound good in my head 😅.
This time, the idea came a little faster (for reasons I will discuss a bit downwards) and I spent half a day building it out.
It became one of my favorite memories from this jam: brainstorming a slow, invisible way to break the player.
At some point, the perfectly cruel mechanic clicked:
Spoilers ahead:
If you plan to play the game, go now, it's pretty short, runs in the browser, can even be played from your mobile, one of the feedbacks I got from my wife is that the text was small on the phone, so if you click the post-its, the text enlarges.
Come back here later.
Continuing:
What if the moment the player finally got how the system works, right at that satisfying peak where you're just starting to master it, the game would start taking it away?
Without warning.
One day, the first click happens for you, with an odd subtle sound, minimal feedback. The first one, so you are not sure what happens. It does not happen again, until...
Next day, two clicks.
Until you're just... there.
Present, but no longer necessary.
I wanted players to feel a tiny sting of unfairness.
And then realize:
that’s the point.
There are plenty of small little things that I think add to it:
Hidden visual cues: crosses growing on buildings... the city itself quietly turning into a graveyard for obedient drones.
Post-it notes that come back (depending on the answer you gave, hinting at something darker).
Your base salary just sort of continues...
And even the way the game ends is cryptic: wtf happens?
Well, only you know.
It is possible to end the game on the 3 screens, it's pretty hard to trigger on some (and time-consuming).
I was kind of hoping someone would eventually stumble on them accidentally (some hints on the comments made me think some did).
Well, as usual, drawing takes me FOREVER, does not matter if I try to rush it, I always take way too long to make one button look right (Not a perfectionist, I just suck at it, there is always something weird I many times fix with some lucky pixel I forget to erase).
And well, compo timing being short, and I had even less time with family visiting, so, minimalist art style was not a design choice, it was basically survival.
But it ended up helping the game a lot...
The static screens, muted colors, black screen transitions, and limited interactions made the world feel colder, flatter... drained.
Which is exactly what the story needed.
Instead of showing a bustling city full of life, what we got was that 2 shades of gray 6 shapes of squares kind of city.
Instead of celebrating progress, everything around you just sat there, grey.
The way the city faded into grayscale as you commuted, the old computer screen (with PERFECT buttons), the dead, boring apartment...
Almost like I found a cheat code for making bad art work (I amazingly got 82 in graphics).
Instead of explaining how the game works, I was thinking on foreshadowing things, and how to add them.. etc.. I don't have much to say on that, except that I had a blast doing it (even if only thinking about it).
Recommend.
Also pretty nice to try to just tell enough of something, and later hope someone gets it.. and when some one does? Oh boy, that feels coooooooool. Also Recommend.
One thing that helped a lot was thinking about the world first, before even worrying about the game.
"What kind of world would this exist in?"
"What would someone do there?"
Judging emotions "evolved" from Judging reputation in the ideas realm. Then it became a work game, and rest just came.
You are just categorizing human feelings on acceptable or not acceptable, sort of reducing people 0s and 1s (the reports also sort of devolve in that way). It felt fun to me, but I think that if the game had been longer, it probably would have gotten boring. (Maybe I also found a cheat code for bad mechanic)
This ideas could only come from building the world first. I will definitely try that again, even if I make a dumb flying-armored-tree-bird-killing game once more.
It was so much faster to make everything connect... I usually take forever to be satisfied with an idea..
.
Well, I got BORED testing the game.
By the end of the compo, I wasn’t proud anymore, I genuinely felt bad... all that gray, hidding crosses, heavy messaging and all that thinking on "how to make this darker" eventually DOES GET INTO YOU.
I even started cutting things out: the commute home gets shorter after the first day because I literally couldn’t stand playing through it anymore.
One commenter suggested clicking to stop the alarm clock.. well.. I had a starting thought on that (alarm sound was probably the last thing I added).. but that thought was interrupted by another th.... NO
The game actually has an internal apathy system: some stuff you do (or don't) slowly build it up, and if it reaches a limit, the game ends.
It was extra challenging fiddling with it, because... how do you measure "enough" boredom for someone that will play it only once when you played the DARN THING 200 TIMES?
Another thing: how do you make a boring game on purpose without just making... a boring game?
One of the commenters very cleverly pointed it out that If you succeed at making a boring game on purpose, well... you end up with a boring game:
I think the trouble with making a game exploring a “bleak” or “boring” situation is that if your work succeeds (and I think this does) then it just feels bleak and depressing
And... yeah.
The Fun score kind of reflects that. 😅
(Heh. Maybe I found yet another cheat code.)
I wish I had found (and had time for) a way to let players beat CALM, or at least reach a hidden "happy" ending.
But honestly... I just didn’t have the energy left to do it.
Even though I stopped working on the game with 9 hours to spare, a miracle for me, since I always finish in a mad cursing rush fixing last minute deployment issues (does not matter the engine... amazing, I always f... something up)
GOOD ENOUGH.
I couldn’t even think of a way to beat it.
Now, some ideas are floating around... maybe for a post-compo version.
I think that even if I added, and ended up cutting the happy ending later, just knowing it existed might have made the whole experience lighter for me while making the game.
Instead, it was like I sorta believed the thing I was trying to build. Goddamm weak minded fool, game is not even that good.
(funny thing: vacation ended today and I go back to work tomorrow)
Building the world first helps a lot.
Even if you don't have a full gameplay loop yet, understanding the vibe and rules of the world makes every design decision faster and more natural later.
Emotional games hit you too.
Making a bleak game will bleed into your mood after a while. Plan for that. Hide an happy ending? Take breaks? I did take a few walks outside during development, it was sunny.. helped a bit.
It's FUN telling stories"
Even though it has an expiration date on some cases, it can be fun to tell a story as it is to build a fun mechanic. Will definitely try that again.
This was such a fun experience and my friend and I had a blast. Thanks to everyone who left reviews and rated us. We had a lot of fun playing everyone's games!
I think if it wasn't for us actually enjoying playing the games and engaging with the overall process we might have left less ratings/comments.
Mostly, since it seems like being on the Smart page for a long time and getting more ratings actually hurts you in the long run?
Not sure, but we saw some amazing games that had tons of engagement and thought for sure they were going to win first place but it seems like the more ratings you get the more you actually regress to the mean. Which makes sense unless you have some standout art, or standout concepts like certain games do.
Which actually makes the whole thing feel way more random than we had initially believed it to be.
That being said, it was cool that we know at least 78 people played our game, and coming out of it with our first ever game being rated consistently above 3 stars in every category is an amazing feeling!
Thanks to everyone who played and rated!
|?|???|
|:-|-:|
|
|
|
My first game jam 1 month afer picking up Godot and game dev in general.
I actually thought I wouldn't be able to finish and submit but somehow managed.
If anyone reads this and is on the fence to try - go for it, the learning experience alone is worth far more than building a game in isolation!
Massive thank you to everyone that played my game, it means a lot!
Next, a post-mortem is in the works and I am planning a post-jam update to address the feedback. Not really expecting anyone to come and play the game again but it is important for me to get it done.
What a ride, see you in October! <3
Those are some fantastic scores. Thanks to everyone who played Down Golf! ⛳
I've been trying to sit down and write a post-mortem for way too long now. I also want to make a long-form YouTube video with the post-mortem, which will be a first for me. Stay tuned! Hopefully I'll get it done in the next 2 weeks. ⏳
Postmortem? Eh, not feeling like a full postmortem. Here's some yapping:
...Personal best set! Finally top 100! Hard to point anything out when it's literally the best thing we made yet.
But also...
The game is doing well on Newgrounds! Still receiving plays after ratings on LDJAM closed up.
Also need to work on teamplay. I've teamed up with @byzilio for the first (technically second) time, and it went mostly okayish. We've decided to continue teaming up, he does most of the gameplay-related code and I do graphics, audio and presentation-related code. Always wanted to step away from the low level stuff and up towards what makes the game look well. We still need to figure out how to make our code click together better, because frankly, it's a glorious mess... 🤣
NEXT GOALS: - Reach top 10 on LDJAM! - Reach 4.5 on Newgrounds! - Have more fun! - Make more fun!
Team Cardboard Box, signing out. See you all in October!
My friend @drazil100 and I have been maintaining a spreadsheet of our Ludum Dare scores. Our data also includes scores for my son, @jason-higerd, since he's participated a couple times. Drazil started participating in LD35 and I started participating in the LD36 mini-jam, so we have 20 games in our combined stats now. (Though only 18 have scores, because LD36 was a mini-jam that didn't do ratings, and my LD50 submission was too late to get any feedback.)
While looking over these historical scores, I noticed that the games where I participated solo tended to have pretty poor Fun ratings. Our first collaborative game in a scored jam was Diminishing Lands in LD38, and its Fun score of 3.551 was the best Fun score either of us had achieved up until Drazil knocked it out of the park with Loading Zone, his LD53 entry, with a Fun score of 3.684. But by contrast, my only solo-developed game to do better than 3.1 was Lily vs the Swarm in LD46 with a score of 3.429 -- respectable, but also an outlier compared to all of my other games between LD41 and LD55 ranging between 2.52 and 3.079.
Even my son's first game that he made when he was only 11 years old, Running Out Of Space (LD42), got a 3.444, beating all of my solo games. (This stung particularly badly because my submission for LD42 got my worst Fun score ever.)
So I set a goal for myself for LD57: I wanted to get at least a 3.5 Fun rating.
I knew I was going to have to really focus on what makes a game fun if I was going to break away from my usual ratings. I couldn't get distracted by flashy effects or deep game systems. So I looked at classic arcade games that needed to be fun the moment you dropped a quarter in the machine and thought about what they had in common. I saw that it would need to be easy to pick up but with a moderate challenge, relatively fast-paced, with tight controls. I knew I had to boil the mechanics down to a solid core and make sure that the basic interactions felt fun.
When the time limit ran out, I felt pretty good about it. I felt like I had done the best I could. I hoped that the best that I could do was actually good. And I waited for the rating period to end.
3.761!
I didn't just meet my goal -- I blew it away.
It's the game people are calling The 349th Best Game Of Ludum Dare 57 - and it's just got a new update!
Ludum Dare is a great source of feedback on a game so I've made a few small tweaks to respond to some of the comments I received during the rating period. So let's dive in to what's new in v1.1.0
Chill Mode
Deep Space Delivery Office I think succeeds in somehow being a slow game and quite stressful at the same time. For folks that want a slightly less (based on comments from @tetrapteryx and @christian-zommerfelds) constrained playthrough I've added a new option to the game to make it a bit more relaxed. In Chill Mode, when the reward for delivering a piece of cargo reaches zero, that cargo will simply disappear, rather than clogging up valuable cargo slots on planets and in space stations. This should make it much harder to hit penalties from maxing out cargo space, and much easier to have a longer, more relaxed session. Chill Mode high scores will be recorded separately to non-Chill (stressful mode?) runs, which leads on to the next v1.1.0 change...
More Info On High Scores
High scores will now show the number of deliveries completed on your high score run, along with the number of days you kept the Post Office running.
Quick Cargo Transfers
Rather than having to select a piece of cargo, then right click the freighter / planet / space station you want to transfer it to, you can now right click on a piece of cargo to transfer it instantly. This will only work if there is only 1 place the cargo could reasonably be transferred to. For example, if 2 freighters are docked at a space station, right clicking cargo in the space station won't transfer it - you'll have to use the old style left click to select, right click on target method for that. At least until I get my mind reading engine up and running. This was a bit of an attempt to deal with some more of the feedback from @christian-zommerfelds.
Bug Fixes / Balance Tweaks
Fuel Tankers can now be sent to refuel Shuttles, and Shuttle fuel usage has been decreased Shuttles aren't really meant to run out of fuel, but I've occasionally run into a bug where they really seem to guzzle the stuff up. So this change should turn this from a minor nuisance into a minor-er nuisance. Fixed a bug where the "Show More" button in the upgrades shop could be clicked when it "Show More" button wasn't being shown. Fixed a number of bugs introduced by changing how the high scores worked You know how it is with these things Not really a bug fix or a balance tweak, but the title screen now handily (for me at least) shows the version number
Please go and have a play here
Firstly i'd like to thank everyone for playing, rating and commenting on our game. The feedback has been super valuable.
This is another set of consistently high results, similar to our last entry Checkmite. As we've grown and gotten more professional team members we too have tried to grow as programmers to make the most of their talent. Mood is an especially strong standout. The coming together of audio, graphics, mechanics and level design to really sell the effect of the depths of your subconscious.
We came into this jam with the same goal as always. Make something awesome. And I think we achieved that. Somnium is one of our most ambitious games yet, being fully 3D and free roam. There were moments where we thought we might not finish in time but the finished game is very close to our vision. However there are some areas we weren't able to implement and things to improve.
The field are standouts for going beyond my expectations and being truly beautiful, big shoutout to @Ruslana here for her art and animation work on the horses. She also did a great job with environemental story telling and I hope those who played it were able to find some of the easter eggs she placed in the world. 
The music is also another area that we got good feedback on and really helps sell the uncanny dream like nature of the game.
There are several areas where we could've done better and most of them come down to running out of time. Running out of time is inevitable in a game jam but it's how we use the limited time we have that can be improved. While the feedback has been very positive there are a few consistent weak points people have highlighted.
Firstly is that the audio did not have fading. That was actually something we intended to implement but ran out of time. I have since implemented it while working on a post dare version and this has been a useful learning experience to learn more about Unity's Coroutine system and the intricacies of AudioSources and AudioClips.
The second is that the ending is a bit abrupt and doesn't make sense. Now to some extent that is intentional. Dreams are often nonsensical. But there was one aspect of the final area we didn't get time to implement. That is a voice with 3D audio calling out to you to wake up that would've come from the correct door. This would've then reinforced the theme of depth with your depth perception of audio and also made the ending more natural as you waking up from the dream. Unfortunately due to time we scrapped this and just ended with a fade to black instead. This is something I am also working on for a post dare version.
Another area is the amount of bugs. Most have been fixed but this too is partly due to running out of time. We prioritised new areas and puzzles and while I don't regret that decision it definitely did lead to a few players getting stuck and missing out on experiencing the entire game, which always hurts a lot to see. Part of this is down to the game being ambitious and 3D being fairly new to me and @Pegasys. We've learnt a lot and hopefully next time there won't be as many bugs. Alas as a developer I know that is wishful thinking. Colliders especially were an area that is far more fiddly to get right in 3D than in 2D. 
Lastly is the topic of getting lost and the game being hard to figure out what to do next. This is an interesting area because it is a fundamental aspect of the genre with similar games like Superliminal not holding your hand. Our philosphy here was to let players figure out for themselves what is going on and piece together the world and puzzles however certain areas such as the Infinite Hallway could've been choreographed better. I don't think having a tutorial or hints would be a good idea as it would detract from the mood and dream like nature however perhaps audio cues or other less explicit environmental hints could've been used and that is something we want to explore in future.
So from this feedback how can we improve things for next time? There are 4 ways we plan to learn from our mistakes and improve
*1) Implement audio fading into audio manager *
This one is already done as I mentioned above and should allow us to have improved audio transitions in future games.
*2) Focus on building a MVP by end of day 2. The game should be playable from start to finish. Doesn't have to be final art, music or level design but it should be playable *
This is something we have tried before to middling success. Back when it was us two the last day was often spent with @Pegasys creating art and me on features. However now we can focus purely on programming I think this is something that we can do without worrying about the other elements of the game.
3) Start each day with a Prioritisation session
Using something like MoSCoW to prioritise the various features and assets is something we have done in the past but haven't been as on top of in recent dares. This would greatly help us to decide which aspects are vital to the game and which are nice to have. So this will be a bigger focus next time.
4) Having friends and family playtest our game
Partly due to the time constraints we didn't have time to give a start to finish version of the game to friends till right near the end. This meant we missed out on bugs and other potential flaws that we could've caught. As such doing more regular playtesting, at a minimum once per day, will be something we do going forward.
We really enjoyed making Somnium and exploring new techniques and art styles and it was a pleasure working with @Ruslana and @theoconfidor. We will be back in October and looking forward to what we can make.
If you're interested in trying it out you can play it here: https://pegasys.itch.io/somnium
I can’t shake off a very unpleasant feeling after the contest results, and I can’t stay silent about it. All I want to understand is what formula was used to determine the winner?
The contest rules stated that we needed to rate, play, and comment on other people’s games to boost our own rating. That’s exactly what our team did until the results were announced, but it turned out that this harmed our score, as the number of ratings our game received led to a lower average score.
For example: Participant James got 50 ratings, all 5 points, while participant John got 51 ratings, all 5 points except one, which was 4 points. So, James has a higher average score, even though John has one more rating, and that single 4 ruins everything—this feels unfair.
I just want to understand for myself: what was the equation, how was it calculated? Please help me get rid of this unpleasant feeling, give a "transparent" answer so I can clearly understand.
Why it feels unfair: - Punishment for greater activity: Actively participating (rating, commenting, etc.) led to more ratings, but it also increased the chance of getting a lower score (like John’s 4), which lowered the average. - No consideration for the number of votes: The system doesn’t reward participants for having more votes, which contradicts the rules encouraging engagement. - Impact of a single rating: A single low rating has a big impact when the number of votes is high, making the system sensitive to outliers.
What exact calculation method was used? @pov
And I want to add, we didn’t artificially inflate votes and acted strictly within the rules with respect for all contest participants!
This was my first Ludum Dare game jam. I made a simple game and it was fun getting to just be a part of ludum dare finally after years of lurking and not making an entry. I think it gave me a better idea of what to consider next jam and how to better approach it, so hopefully I improve as I keep on entering! :D
For this jam, I made Fathom Pond, a classic text adventure about finding your glasses in an increasingly surreal underwater world. And hiding in that description is perhaps the biggest error I've made in all of my 20+ jams: deciding to make a text adventure. However, before we get into that, let's start from the beginning.
Before the Jam
For this Ludum Dare, I knew I would be short on time before I even began. I've recently started a new job in which I work weekend mornings, and didn't bother taking time off for LD. This led to me considering some alternative options which would require a little less time than normal while still maintaining the level of scope I like to have in my jam games. Eventually, I landed on interactive fiction. I've been meaning to make a text adventure for quite some time, and this seemed like the perfect time to finally bite the bullet. I figured it would probably be a little more inaccessible than the average entry, but that I would have fun making it, and that it was worth the tradeoff. So off we go to Ludum Dare weekend.
A Brand New Engine
Fathom Pond was made in Inform 7, a natural language-based tool designed around the traditional style of text adventures (think Colossal Cave Adventure, Zork, etc.). It really is a great and deeply powerful tool, but it ended up being a bit more complex than I was prepared for. I had dabbled in the basics here and there, but had never actually made a full game this way, and ended up doing a lot of learning on the fly. This is nothing new for me, back in LD48 I opened Godot for the first time and ended up making depths of dreams, all things considered one of my better games. However, it does still present some struggles considering the time pressure I was under. Lots of functions that would be second nature to someone experienced in the tool took more time to sort out than I would have preferred, and I did start hitting some time snags near the end of Sunday. I finished everything I intended in time, but had a few major issues.
Immediate Reflection
As soon as I had submitted the game, I took a second to consider how the process had gone and what I wish I would have done differently. The big thing was the presence of a few pretty major bugs that hadn't come up in my testing but were found pretty immediately by non-LD playtesters. I would have liked to have gotten that feedback prior to the end of the jam, but the time ultimately didn't allow that to happen. The most major of these (one of which affected literally the final room in the game) were patched the next day, but a few minor-but-annoying ones still persisted. I was also a little nervous about the potential obscurity of many of the puzzles. Coming from an IF background they all seemed relatively doable to me, but I knew the majority of the players of the game wouldn't be as experienced in the genre as I am. This turned out to be a massive understatement, as we'll soon get to. However, ultimately, I was pretty happy with how it turned out! I thought all but the very final puzzles were really fun and interesting, and even now I'm shocked by just how much actual content I managed to squeeze into this given all of the constraints I've mentioned. Unfortunately, most of these puzzles were never cracked by the overwhelming majority of people.
The First Ratings
When I woke up the morning after the Jam had ended and people began rating games, I had 2 comments on the game page. Both of them were extremely lost and unable to progress beyond the first few rooms. Immediately I went into damage control mode, adding an additional disclaimer to the game page and additional helpful information (namely a common commands document and a full walkthrough) as attachments on Itch. I do think these were ultimately good additions, and mitigated the worst of the confusion, but it was still a rude awakening. Frankly, at this point I had done all I could do and just had to let the comments come in.
Complete Confusion
Most of the comments I received were ultimately pretty uninformative (good writing, but I got stuck). A couple people who explicitly mentioned they were text adventure fans mentioned they really enjoyed it and made some more substantial progress. However, there was an overwhelming opinion: pretty much everyone wanted in-game help. Whether this was just a list of possible commands, more specific hints in particularly challenging rooms, or more in-universe guidance from the NPCs, it would have gone a long way. Something like the Inform postcard is useful, but ultimately doesn't help with custom commands for this game like "lure" or "attach". I suppose in my mind simply trying things that could work is much of the point of playing a text adventure, and being given a list of commands would ruin a lot of the immersion and discovery of the experience. But this is also a Ludum Dare entry, which requires some level of mainstream appeal, and so it perhaps would have been more useful to include more explicit help.
One real missed opportunity is a lack of dialogue options from the Eel, an NPC accessible as soon as you enter the pond. He has dialogue relating to a few objects, but it's mostly funny and doesn't lead you in any particular direction. This would have probably struck the best balance between extra help and maintaining the total atmosphere. There was also a pretty big problem with some intended actions not being recognized by the parser. The most egregious example is in the altar, where "place" doesn't work but "put" does. While experimentation with phrases is at the core of IF, intending the right thing but the specific phrasing not being recognized is a pretty major issue.
The Good
Not everything was bad! Those who did make more significant progress said they found the puzzles really engaging and well designed, and the sheer amount of content was very impressive given the compo restrictions. This is one of the things I'm most proud of with the game as well, I have 8 major puzzles in the game and none of them are repeats. Given the relatively limited number of puzzle options in the genre, coming up with novel ideas was tricky but super satisfying.
There was also pretty much universal praise for the writing. I don't tend to get much space to showcase my writing within my jam entries given the nature of compo scope, so having an opportunity to make a game that fully showcased the way I could create mood was really exciting for me. I loved designing this weird, surreal underwater world, and was thrilled by the number of narrative twists and turns I could sneak in without really having an explicit story. Ending the game inside a giant kraken in pitch darkness is a particular highlight.
Results
There ultimately isn't much to say about the final results. They're some of my lowest scores in a while, and I fully expected that going in. This was never going to be one of my best performing games, and I had to accept that before I even started working on it. I am still very proud of my Mood score, considering that was probably the aspect I was targeting the hardest. A complete lack of graphics and audio probably also skewed my scores, but I didn't have much choice in the matter. There are ways to add them to Inform projects, but it's not supported by every interpreter and was just generally more trouble than it was worth. Basically, I did just about exactly as well as I was expecting.
What Next?
I've begun to notice a pretty distinct dichotomy within my Ludum Dare entries. Either I make a game that scores well, or I make a game that I find more enjoyable to create. I've had a few jams where this overlapped, but for the most part it's one or the other. Stork Serenity is still my best scoring LD entry, but it was a pretty mechanically plain game to work on. On the other hand, my most recent games, Gull and Ray and Fathom Pond, were much more mechanically interesting and fun to make, but were inaccessible and didn't score particularly well. There definitely is a balance to find here, but I still appreciate making something like Fathom Pond every once in a while. If I just made the kinds of games I think would score well, I would have burnt out of LD a long time ago. We'll see if I strike a better balance for my 20th(!) Ludum Dare this October.
Is there a list somewhere that would have the top 20 games.
It would be good to know what genre of games received the top scores. Next jam perhaps make a game in the genre that the audience here prefers. Something more enjoyable for the players.
Secondly, of the top 20, how many were either originally web games or converted to web games and how many had to be downloaded and played. That would also be something to look at for scoring as well as how many times the game was played.
Finally, on a game by game basis how each player scored the game. It is not necessary to know the players handle, but just a list of the scoring. Were there a lot who gave it a 5 star rating and a lot that gave it a 1 star rating? Were the ratings about the same? We have an average sure, but that does not speak to any type of min/max differentiation which can be helpful in making the game better.
All of the data is there, it would just be a matter of releasing it and allowing the participants access to it.
It would be interesting also to allow for an in depth look at how each of our games were scored. I don't want any formula, I just want to see the raw data and then come up with my own conclusions.
Ultimately we want the best game we can make in just a few days. However, we are still human and want the best score. Everyone likes to win. These numbers could help us achieve that.
Hi everyone, right after releasing my AI based point and click adventure generator for LD57, I already started on a new project: a shooter authoring tool with AI generated enemy behaviour. The first playable prototype is here!
You can generate enemies by describing their behaviour in natural language. The AI then codes Javascript files for the enemies, and generates sprites as well. The result can be immediately viewed in the browser. You can enable and configure the enemies through the web-based UI, so you can create suitable levels with the available material. The AI supplies additional behaviour parameters with each enemy (these are the options that start with "extra_"), so you can tweak the enemies further.
It can handle prompts that describe multiple enemies and shooting behaviour. Example prompts:
zig zags randomly, periodically spawning minions that circle around it. The minions shoot bullets that explodes into a spray of shrapnel when they are close to the player.
dashes randomly, either in random direction or towards the player. Fires a single arc of bullets when it dashes towards the player.
alternatively stands still and moves quickly in a random direction. Is invisible when standing still. Shoots a bullet when it starts moving. Bullets explode after some time into a spray of shrapnel.
It's still a prototype, you can define only one level that starts playing immediately, and load/save options are limited. You can create an URL to demonstrate your playable level. No unit testing is done on the AI generated code yet, so you will find some malfunctioning enemies. Some planned features are:
