we spent tonight going over concepts, scoping, and setting up a framework, and tomorrow we're going to execute. i also spent a few minutes validating that I know how to deploy (lol). using GMS2, photoshop, ableton, bfxr. this was probably our least favorite theme but we're making it work. i set up a github repo both for tracking what needs to be done and for source control. planning to make it public once we submit.
but for now, we sleep.
Friday night - Don't Settle Never Compromise Do Everything
Saturday - Jesus Christ there's no way I'll ever finish all of these open issues in time
Sunday - hey this is starting to look pretty good
Monday afternoon - 10 bugs are revealed two hours before posting, including a fatal crash
On a five person team, there's no shortage of ideas during the ideation stage. We narrowed our top choices down but the notion to create a tamagotchi within a larger game just kept gnawing at me, so I kept pushing for it. When Rai brought it all together with the marriage you're trying to keep alive, we knew we had a top concept.
Wireframe from the ideation call:
Final:
We'd picked the most ambitious project that had been proposed, but we had clearly delineated our roles and knew what to do, so I started filing issues out, and we were off. We're all getting old (no offense, team) so we didn't pull any all-nighters, but there was definitely a crunch time atmosphere the entire (long) weekend.
Ultimately, I'm extremely proud of what we were able to achieve, and I hope you all enjoyed playing as much as we enjoyed building it.
https://ldjam.com/events/ludum-dare/46/happy-life
I thought it might be fun to show off our earliest "playable" version of Happy Life. Check it out at www.memorydumpgames.com and if you haven't played it yet, here's a link to the submission version. Thanks for playing!
Don't Kill Your Darlings was a challenging one for our third Ludum Dare. We like to play with the theme as much as possible, but it took us forever to find something that was in scope and felt like it spoke for itself (our ideation call was like 3 hours, when they're usually more like 1 or so). But, our artists really pushed themselves, and I had fun exploring GMS2 and GML again after being away for a year or so.
After spending most of the last year learning C++, going back to GML was really eye-opening. A lot of what makes GML so easy to just sit down and start writing (great for a game jam) also makes it extremely unwieldy as projects grow without some cleverness, very careful consideration of architecture, and self-control, all of which are lacking for me during Ludum Dare. I've always known this line about GMS, but I didn't really understand it until now.
We made a pretty critical mistake early, which was to put off writing the script until the end. This led to a ton of problems in the JSON parser that weren't apparent until we had an actual narrative in place, and caused me multiple hours of scrambling. It also made coordinating the art a bit more confusing.
In any case, I think we walked away with some learnings about where we can improve our asset pipeline for our main project and to better plan the scaffolding at the start of a feature, and we published a game in 72 hours that, while simple, feels pretty successful (despite the crunch time bug fixing).
Check out our entry here.
p.s. Thanks to Juju Adams for all the amazing GMS libraries; in particular Scribble, from which I used a default example for our typewriter text :stuckouttonguewinkingeye:
