Alright, it's been a hot minute since the last time I was able to give an update. Life got a bit hectic so I had to take a bit of a hiatus from my project work in order to deal with everything, but now I should be more or less back on my normal work schedule. 

So this update is probably roughly equivalent to 2 weeks of actual dev work spread out across a little over a month, and what have we done during that time? Well, I introduced the concept of "Rooms" to the codebase, which store data for the tile sets and trigger volumes of a small chunk of environment as well as map coordinates (which will come in handy later). They also indirectly hold information on their neighboring rooms via the newly added transition triggers I made, which in turn allows us to move from room to room. 

And that's really what I spent the bulk of the time working on; room transitions. In Metroid, when the player walks through a door, they don't just simply appear in the next room, the game pauses, everything fades to black except for the door itself, the camera pans in the direction of the new room, and then everything fades back in before gameplay resumes. This process hides any load jitters while also being pretty visually appealing at the same time, but each step of that process involves a new system that I had to make.

Pausing the world is relatively simple thanks to Godot's timescale and Processing type options, just need to select the top-level node and tell it to pause, and set a few choice objects to be able to continue to run their update loops even when the timescale is paused. The fade involves attaching a giant black rectangle to the camera and adding scripts to allow it to transition from one color to another over time, and the actual camera movement only needed some slight adjustments to the camera manager system I set up as part of my last update. In order to get the whole sequence put together, I also had to figure out how callback functions work in Godot, which luckily ended up not being that much different from what I'm used to in Unity.  Technically I could have just used signals to get functions to fire when I need them, but having to connect and disconnect to signals repeatedly every time I ran through the process seemed like a lot more work than just passing in a callback function.

There's a little more magic happening behind the scenes when it comes to placing the rooms in the proper position in the world and making sure the player's movement and height are carried over between room transitions, but that's just some simple math at the end of the day. 

Now, as for what's next.... 14 weeks is actually a pretty short amount of time when it comes to game development, but the fact of the matter is that this project was supposed to be a relatively short one. Theoretically I could keep chipping away until I more or less recreate the whole game, but the point of this exercise was to familiarize myself with the Godot engine before I move on to working on something of my own. To that end, I have made a list of topics/features that I want to cover before I can declare this project over. Here's what we have left:

- Mini-map
- Pause Menu
- Save/Load
- Sound and Music

I've determined that basically everything else on the feature list I originally wrote up is either something I have already figured out how to do during my time with the engine, or something that the engine doesn't meaningfully effect the implementation of. As such, those won't be much help as a learning tool, and I'd be better off spending my time elsewhere. So all things considered, we're actually pretty close to the end here. I've been doing some thinking about the project I'll be starting up once I'm done with the Metroid stuff, and I'll talk some more about that when we get there.  For now, next thing on the docket is the map functionality.

​See you next time.

Okay technically we missed a progress check-in last week, but I've been busy with job search matters so I had to put this project on the back-burner for a little bit. This past milestone has been all about our game's camera: getting it to follow the player, controlling where it can and can't go, and handling transition behaviors.

The good news here is that I learned that Godot has a lot of really nice built in camera functionality. It's got smooth follow, tracking bounds, and even built-in deadzone options.  The bad news is that it turns out I couldn't use most of these features for my purposes and had to write my own camera manager instead. So why is that? The short answer is that the default functionality of the godot camera was at odds with the camera design of Super Metroid. This was mainly evident in two areas: tracking and bounds.

For Camera Tracking, Godot has set up its cameras to use a very common design for side-scrollers known as a deadzone.  The idea is that if you are in the middle of the screen, you can move slightly to the left or right before the camera actually starts moving to follow you. This makes the camera feel more dynamic and more smooth than if it simply always tried to keep the player in the exact center. Super Metroid doesn't do that though, rather the camera moves slightly ahead of the player as they move in a particular direction, moving them to the "back" half of the screen to allow them to see farther forward. This makes sense for a game where your primary attacks are projectiles, as it gives you warning for things ahead of you so that you can better engage at range. For this purpose, I built the camera manager to have the camera track an invisible point that moves in whatever direction the player is going, up to a defined limit in each direction.

For the camera bounds, Godot has some built-in variables that you can set to tell the camera not to pan past a certain point. This works fantastically, the only problem is that if you ever try to change the bounds of the camera, say when going through a screen transition, the camera instantly snaps to be inside of the new bounds rather than transitioning to it smoothly. Technically Metroid's camera deals with this in two different ways depending on the specific trigger. Either the camera will simply do a slide transition to its new desire bounds, or the camera will have a ratchet effect applied to it, where moving closer to the bounds shifts the camera as normal, but trying to move away will have it remain in place. 

As you can see, I got it all working, it simply took a bit longer than expected. Anyway, hope you all enjoy the little primer on camera design. Next up I'll be moving on to real room transitions with all the added fun of scene/asset loading. 

See you next time.