Work Post

12:30am - The mental illness has been riding me hard the last few days. I'm not sure what to do with myself.

I've had a lot of time to try and rest and relax, but I haven't really wanted to do anything. I sort-of want to work on my project, but I've been too out of it most of the day, and now the call is on. I figured I'd try it anyways but time has been slipping by too quickly. It's already...

12:44am - Anyways my next task in the TODO list is to make the Use component, for later. I think I'll lower it on the TODO list because later is too far away; I have nothing to 'use' right now.

The next one is Make a Zombie Enemy. But I think something is missing. I've pulled the GUI off the player and made it follow the active player, but it doesn't do anything anymore.

The problem I've got is, if there's more than one player character, does it show the character it's highlighting? Or is there a 'main' player that it always shows the stats of?

I'm not sure. Maybe I shouldn't even have the stats of the characters taking up a big window of the interface like it does.

Maybe I should do some research.

1:20am - I did some research. Similar games have transparent backgrounds rather than solid windows and lower impact GUIs. I think I'll completely re-do the GUI later. For now I just want to get the text log working again.

1:31am - The more I think about it the more I think I'd have to do all the work again when I redo the GUI anyways.

I could use the print function to output to the console log instead, but then it would also make more work, because I would have to replace the code later that put the text in the log into the other log.

I think it would be better for me to just redo the GUI now. But for that I need to solidify the design, and I'm feeling REALLY off right now, so I'm going to take a break.

2:54am - So I've been trying to think of ways to connect the Log in the GUI to the actual signals. If I have each Actor output a different signal for every action, that would be a heck of a lot of hookups for every Actor that I generate. Alternatively I could use one signal for all the actions an Actor can perform, but give it a note or something saying which one it was along with the other variables. But I'd still have to hook up every actor.

One thing I'm considering is putting the notifications for the GUI Log in the Action Pumps instead. Since the two action pumps are in the Level class, all the GUI has to do is watch for signals from Level. The tricky thing though is that Level only works with the Action parent class. The subclasses are different and they have the information.

Maybe I could put the signals in the Actions themselves? But that would only make it more confusing, if every action had to be linked to the GUI.

I could have the Actions return something to put on the log when their execute() method is run, and the Level class could put that into a signal that the GUI, or anything else, could read. I think that would work. The returned value can be ignored, also, in case the Level and Actions are too tightly coupled.

Hmm... but the Actions don't know much about the actions they do, they only tell the Actors to do them, and carry the info of the targets.

So again I have to go back to the Actor class putting out the notes when its functions are executed. But at least I think using a LogNote class would be easier than having a unique signal for every action method.

Alright. I'll modify the Actor class to emit a signal with a LogNote, and make a LogNote class itself.

4:11am - It took a while due to some strange behavior but I've got the log filling up again.

Next up I make a Zombie, along with all its parts, and make it approach and attack the player.

For now though, I need another break, and maybe a sleep.

Work Post

9:06pm - So, last night I had some heavy feelings and it brought me down to wondering what I was doing with myself, again. I figure it would be a good idea to talk about it and help myself be soothed.

The goal I set for my Roguelike Project was to test the theory of procgen dungeon design I had thought up. I needed a game that would use procgen dungeons to test it, and I had already started a Roguelike game from 7DRL this year, so I used that.

I've made some good progress on it, but now I find myself questioning my choices and what I'll do next. I do enjoy the occasional roguelike, but it's not my favorite genre, and often I think this would be better tested and done with a Zelda-like action adventure real time style instead of a turn based dungeon crawler.

I'm really conflicted because I don't want to pivot so much after coming so far on my Roguelike Project, but I worry that's just the sunk cost fallacy slipping in. What's the point of finishing a roguelike if that's not the sort of game I want to make? And yet, there are other goals that have been slipping in as the project continues, like making the whole thing open source so that others can follow the path I carved. What does that have to do with testing my theory of procgen dungeon design?

There's a ton of Roguelikes out there already, even if there aren't any open source made with Godot. I'm not exactly treading new ground. On the other hand, the whole point of this is to do what I want, and perhaps make some money along the way.

I was thinking of making this Roguelike Project into my next big thing that I wouldn't stop until I was done, but now I'm questioning if I should even keep going or pivot into something else.

I don't think this is something that should be a snap decision anyways. I'll have to think about it. But let's consider it.

A lot of what I've done I can keep from the Roguelike Game to the Zeldalike Game, so I wouldn't be starting from scratch, but there would be a lot of things I'd have to change or make new. It's also early enough in the game's development that big pivots can still happen; it may seem like I've gotten really far but that's just because I'm working alone.

Before I was writing this as a result of my deep dark emotional period last night, I had been thinking about exploring what I would have done differently if I had focused more on making the game in the Godot 'style' using Signals more. I wouldn't want to redo the work I had already done but I could think about what I might have done differently.

I think that's still worthwhile, and it might shed some light into this problem I'm having. I was thinking at the start of this post that I'd figure out that it was better to keep on the Roguelike path, but now I'm not so sure. So I'll do that.

9:55pm - But I'm short on time, so I'll post this for now.