I don't really know how to devlog lol

I should talk about what I did to get where I am, but it doesn't really feel that special. hbomberguy recently said something along the lines of "creatives often undervalue what they can do because they're doing it all the time" but if I'm being honest, I haven't been doing it all the time. All I felt like I've done here is do compsci 101 stuff I've learned and forgotten and relearned and forgotten and relearned and forgotten and relearned again. 

feel free to skip to the next strikethrough if you like continuing to respect people lmao. this is just me sabotaging my "documentation" mark. hi professors this is absolutely not for you to read i promise

That's a fault of my own brain I suppose, but it doesn't really help that for the past few years my mainline university courses have mostly been an unfocused mishmash of "watch youtube tutorials for this thing that is almost entirely unrelated to the last thing you just did" and "literally a requirement of the assignment is to tell AI to do at least 40% of the work." You know, the thing you have to socially engineer into not hallucinating? That drank more water than NFTs this year? That's going to crumble in a few months because the excuses for the technology existing range from "It's literally impossible for AI to work without stealing from tens of thousands of people without permission" to "Go fuck yourself we're stealing anyways"? I thought my mainline class was supposed to give me a foundation, but instead I feel like I've placed a bunch of buoys in a vast sea. You can't build on a buoy. They're supposed to help you navigate: here, they're leading me towards the conclusion that I need to learn how to swim. I don't know if that analogy makes any sense. I hear gamedevs say that "gamedev programs should teach principles of design rather than just the specific engines and tools" and jeez, imagine if I spent 5 fucking years in a Unity curriculum just for that to happen, but it would have been nice to have any training in some core concepts and technologies, like for example when you taught us how to use mocap and photogrammetry? But the other half the time I almost feel poisoned by lack of focus, we get very small timelines to work on totally new/different things, and - again, I have to reiterate - pushing AI assignments onto us, the thing that is massively unpopular in basically every breadth of art because silicon valley tech bros and business majors instantly and exclusively started building in the direction of "cutting every possible person in the industry out of the payroll" (like they always do) when we should be building tools, actual tools, like "Magic Marker but Good" and "Computer, Figure Out How I'm Supposed To UV Map This Mess," so we can have more fun making instead of having the making done for us. What happens when I get a template to tweak instead of just building the damn thing myself is that I don't know how tweaking it even works, I play it safe and break things anyways, I throw out things I wanted to do until it works, and I end up with something that is, at best, good enough. Judging by what I've seen from most AI ""artists,"" they set their standards pretty low - "whoa! this thing that isn't even alive stole its way into something that kinda looks like the work of the person I wanted to hire!" and they're perfectly okay with that. 

Personally, I like to make things I'm actually proud of, and I like to come back to my things over and over to admire them, so unless I'm seriously behind on something, "good enough" is never good enough.

genuinely i almost hope you skipped that despite what i say in the literal next paragraph. there is no world in which this was a good idea

Ugh, I got sidetracked again. Sorry everyone. I should probably delete most of that but these are going to be the most personally important writings of my degree, this isn't a formal deliverable, and I clearly care about that, so I think I'll leave it. I am so far behind on handing this in it's happening at the start of the semester after it was supposed to appear, so let's talk about what I'm proud of.

First, we came from a node.js prototype, just to see how it would work. I hadn't started with Godot yet, and I did this specifically because I'd already made a little node.js turn based game for another class that I thought I could retrofit into a prototype for this game, I ended up with some pretty drastic changes as expected. In it I started plotting out the buffs I wanted to add, the damage model, a few important mechanics like your AP/TP resources and the combo system. You can play both of these if you want, they should be downloadable on this page. 

Afterwards I began to learn Godot using that Godot Humble Bundle package from Zenva. I followed along with their intro to Godot development and Micro Turn Based RPG courses to gain a foundation of how the engine and GDScript is structured, then began the process of translating much of my node.js code to GDScript under the node-based structure of the engine. 

Because the game didn't live on a command prompt anymore I spent a bunch of time visualizing a lot of information. Now I wasn't missing things anymore in the word salad, at the time I was most excited about that in the move to Godot. In particular, making the enemy's stagger bar an actual bar made it much easier to parse. Adding tooltips to each button helped a lot with lowering the amount of menuing as well.

Then I heard Godot also has some shady business practices going on! Something about how most of the heads of the Godot Foundation are also selling the ability to port to consoles for a price way worse than Unity under this really weird rationale of, like... "its open source so if you can do it yourself that would be great or even better you could find us a guy who can also do that for everyone :)" or something like that. I'm not sure how credible this is at all, but... at least they're not working with Lockheed Martin yet? 

Cool so anyways, combat works, but it always felt wrong... because the whole time it was always a one-on-one fight, even in the other project I'd based this on from long before this project really started. At least in my node.js prototypes I'd generate a new robot for you to fight, but the Godot tutorial only had a single one-on-one fight in mind. Mobius was never defined by its one-on-one fights. Enemies almost always faced you in numbers. Final waves would be a hit squad of mid-threat enemies or a single boss with a few goons. Pretty much the only one-on-one fights were scenarios where you're meeting a mid-high level threat for the first time, and against real rough customers, chapter end bosses like Demon Wall, Adamanterrapin, or your prophesied battle against Chaos; and in endgame superboss challenges - but in half of those they also just say "fuck it" and throw three Adamanterrapins at you. Have fun with the stampede, fucker. 

So, because this was so integral to Mobius - and also because it just seemed easier with my new mindset/canvas in Godot, and also also because one-on-one fights were really easy and boring, I spent some time making multi-combat work first, with some pretty drastic changes to the old turn manager infrastructure and targeting. The way I handle area attacks feels really dumb but area attacks in these kinda games are probably just dumb like thats anyways.

Because I was rapidly reaching the point of no longer "just making sure it works" I also at the same time set up a system where character stats are stored in their own resource files, turning the nodes that were previously the characters themselves into pretty much like slots for character files to be plugged into, so I wouldn't have to make like 2 million duplicate scenes of the Battlescene itself. This kind of was a thing I was taught to do in the Micro RPG tutorial (for the characters' combat actions) and I feel like I probably would've been taught to just do that with a structure made by someone who has made these before in a non-Micro RPG tutorial, but honestly catching COVID set me back enough that I was just learning on the fly.

I put this same philosophy to work in designing the structure of missions and the waves within the missions. Waves store the character files for each enemy slot plus possibly a few messages that could appear in the message log, and Missions store the waves you fight as well as any unique win/loss messages.

And... here we are! Everything is in place, this is pretty much the vertical slice of the game. I've made a little demo containing a 5-wave introductory mission that is a preview and prequel to the story of the final game. It should also be attached here: The Vertical Slice! It's kinda shit right now because I have no idea which exact part of turn changing is causing my game to get confused and there's also a few other bugs that I'm pretty sure I actually know the cause of but I started writing a presentation in a delirious haze instead. But it mostly works!

From now on I hope I should be able to work on more interesting things to talk about and keep you guys updated. At least I hope I remember to keep you updated and don't just keep adding onto one gigantic monthly blogpost. I was gonna say something like jeez imagine if hakita ultrakill had the first dozen weeks of his devlog be small kinda obvious interactions like 'i figured out how to shoot gun in unty' but looking at it again i guess he did kinda do that so that kinda makes me feel better. and from there his devlogs get more and more interesting too!

Right now a lot of my assets are placeholder - I'm using icons from game-icons.net by Lorc and Delapouite for buff/debuff icons and the targeting arrow. The character visuals are from Oryx Design Lab's 16-bit Sci-fi Tileset. Also there is just no sound right now. I don't intend for this game to stay 2D, I want it to be a 3D experience, but I don't actually see this happening right now as I think I still have a few things to learn before I feel comfortable starting. I've had many experiences where bad practices when I was a beginner have made it rough to continue later on. I'll definitely try within the ~4 month window I have for this game, though in the meantime I think a lot of my in-engine work will still be in the 2D. Most of the graphics are just set dressing anyways.