The Wrong Book - an idler-style demonic bonanza (Compo)!
For this LD compo, I had a very distinct mindset going into it... it'll likely be my last for a long time as I'm going to become a father for the first time in just a few weeks, so I really wanted to sum up everything I've learned and done over the years making Ludum Dare entries. What I ended up with was a game I am immensely happy with, and based on the reactions so far, I'm not alone. This post acts as a post-mortem and also semi-promotion (even though results are a day and a half away).
There are big spoilers below - if you haven't played the game through to the end, I highly recommend doing so. It takes around 15 minutes.
What is the game?
The Wrong Book is a 2D strategy/idle game based around hiring cultists, summoning demons, and putting them all to work to contribute towards the summoning of The Great Old Things.
What did I focus on?
Knowing that humour and audio have been huge winners for me in the past (achieving 3rd place in each of those categories over two consecutive compos), I figured I would lean into my strengths. I did more voice acting than ever before, and spent a good chunk of time on audio design and music composition. In addition to this, a few days before the compo started I was having a conversation with a friend about how games (specifically Civ and Factorio) manage to achieve the "just one more turn" feeling and thought I should implement some of those philosophies into the game design.
What worked?
To be quite frank, a lot worked. The reaction from everyone (fellow entrants, family, friends) has been overwhelmingly positive. The humour landed very well with people, and the audio has impressed many, especially the wide variety of voices and voice lines I managed to put in the game.
The UI design is probably the best it's ever been for any of my games, which I can tell purely by the fact that nobody has commented on it at all. It's my first game with extensive tooltips and context menus, and I think it's worked in my favour to an incredible degree. Everyone I've seen play the game managed to figure out very quickly how to interact with everything and what the gameplay loop was, with no confusion or ambiguity (a first for me). Speaking of...
One of my favourite things about the game is how automation is not available right away. Everyone, by the time automation is researched, has become very much used to manually clicking on units to queue up resources. Taking some inspiration from Factorio, I decided early on that in order to make automation feel useful & fun, the player needs to spend at least some time doing things manually first. The feeling you get when automation is finally unlocked is one of relief, and it might be the best gameplay moment of the game. The whole game was designed with this feeling in mind. Give the player an obstacle first and then give them the tools to overcome it.
But, in my view, the crowning achievement moment comes right at the end (again, massive spoilers). I had so much fun putting that silly little polka song together, and my friends and family haven't stopped singing it at me since the game came out. Very rarely have I seen any of my entries cause people to laugh out loud, but this one managed it.
What didn't work, or needed more work?
In all honesty I feel this section is much more "nitpicky" than previous post-mortems because the game has been so well received there isn't anything huge here; certainly nothing on the game design or balance side.
There was an annoying bug with the Unity engine that caused UI text to disappear midway through the game in the WebGL version only, but I managed to put a fix in a few days later. I'm hoping this bug didn't affect anyone's enjoyment of the game too much - it was infuriating for me to see something so key to the game being ruined by something in the engine itself, unrelated to anything I'd done.
The other thing was also kind of a bug - UI scaling for high resolution monitors. The maths I put together to determine where the tooltip shows completely falls apart if the UI has any dynamic scaling on it. I only learned this right at the end of dev time, just before submission, so I had to turn off scaling and make the UI a fixed size instead. Unfortunately this meant that the text was tiny on very high-res screens.
Final thoughts
I'm so happy with The Wrong Book. I think it's my best Ludum Dare game yet, and if it's my last one (due to impending fatherhood), I'm very happy to go out on such a high. Thanks to everyone who's played it and left such wonderful comments.
Thank you to everyone who tested, rated and commented on my mouche(s)? game. You are great people. Thank you also for all this creativity and all these wonderful games... it was simply fantastic.
Hi, I'm Bo, and I'm an artist on our team!
In this post, I explain how we optimized our pipeline, the challenges we encountered and the decisions we made to overcome them. For game 
We decided that the game will be in 3D, however creating characters in 3D takes much time. For instance, my model of a duck includes: sculpting, retopology, UV-layout, texturing, rigging, and finally animation. But I needed to figure out how to create several characters within a short timeframe.
I decided to create 2.5D characters, which simplified my work. You'll understand why as we progress.
I'm skilled in 3D rather than drawing, starting with _ZBrush where I can rapidly refine the shape of my characters and add cost-effective details. I created a model for 2D rendering, so the most important part is the front. I don't focus on the other sides. Also, I used Polypaint(vertex paint) for texturing characters—an easy but effective method. I didn't need retopology and UVs, which are often the most frustrating parts of 3D work
For rendering, I used _Marmoset Toolbag 4. I separated the model into several layers in ZBrush and exported it to Marmoset. Then, I set up the scene, adjusted the camera angle, and lighting. Additionally, I added an extra render pass for the Normal map. Furthermore, I used this scene for other characters.
I organized and refine the render layers in Photoshop and then exported the PSB file to Unity.
The "2D PSD Importer" automatically separated my layers into sprites
Before this moment, I hadn't tried working with 2.5D. I realized that sprites don't interact well with a 3D environment and lights. This confused me, but our programmer figured out how to fix it and wrote the shader. Sprites began to behave as I wanted; I was able to use the Normal map, and the sprites responded to light.
I rigged the characters using the standard _'2D Animator' and 'IK Components' in Unity. During animation, I focused on my favorite principle of animation: 'Follow through and overlapping action'. We realized that we didn't need triggers for animation, as the animation starts automatically when we call the prefab. The programmer prepared prefabs before for characters that we planned, and I began to fill them with other characters.
I hope you found this post useful and interesting. It's my first time writing something like this. I've always wanted to share about my work on the Jam. Thank you for reading, and _don't forget to play and rate our game 