DEVLOG 2023-10-21 - End of Prototyping, Livestreams paused

This week I decided to go forward and simply switch into actual non-prototyping gamedev mode. Yay! Well, the dev of MEGASTRUCTURES will start actually next week or the week after because I have to handle some organisational stuffs, make decisions and next week I’ll be in a conference for dayjob stuffs (Packaging-Con if anyone wants to meet). That also makes me unavailable for the social sharing of Roguelike Celebration, but I noted to try be free for next year. Can’t wait to see the talks.

So first important change: I have to stop the livestreams for some time, because I still didnt decide what I will do on this side of things. There are many ways to share gamedev online but also there is the option to limit sharing to the strict necessary, like these written devlogs. I started livestreaming to give me a specific time in the week where I could progress on the project or share with a few people the progress I made. It kind of worked, I kept having the livestream date in mind and it acted as a soft-deadline. However, for the game proper I’m not sure it’s useful. It could be maybe if I was doing more livestreams like two per weeks, and longer. I could also do livestreams that only focus on showing the state of the game. I could record these in a way easilly posted on youtube. I could edit them, but that owuld take gamedev time. I could skip the livestream and just record each week what I did and show it, many people do it here. However that requires some graphics doesnt it? I dont think I’ll do anything other than cubes and cylinders for a long time, so I’m not really convinced it’ll be interesting to see for the beginning at least. I could also do proper gamedev logs through videos but the perspective of having to edit is daunting. I love editing, but it’s also one of the most time consuming thing ever if you dont want to just make a corporate slides-based talk video. I’m an artistic person and that means I will necessarilly get into that mode where you infuse style into any bland material you have. It takes time. I dont think I have time for that in addition to the gamedev, at least for now.

All these options also only appeal to an audience of other game-making people, it doesnt really create a potential players audience, doenst grow a community around the game etc. I’m not trying to do that as said before, but what I mean is that I cannot justify the time without it being considered “marketing”. I suspect it could be considered marketing as soon as I managed to make proper graphics, but that will not be soon.

Yeah thinking about all this is not easy. If you have an opinion feel free to share with me.

Starting this project proper, I have to setup tons of things: the repo from scratch (because prototypes are not made to be iterated upon to make real projects!), some technical setup to do, some non-technical decisions to make, tools to setup for plannign and making design notes etc. I might end up just putting everything into text files inside the project, but even that needs to be decided. I also have a very large list of things I need to do and the first thing is to setup some priorities. For now I’m prioritizing setting up priorities hahaha!

By the way this made me think that I didnt spend tons of time on thinking about the graphics. I noted to take a week (not sure which though) to decide a general graphic direction. The game is in 3D, but I do not want it to be fancy because 1) I want the animations to be composable (a whole topic I might write about another time), 2) I can’t afford fancy 3) I want to retain the same helpful simplicity that ascciart or simple symbols gives us when making traditional roguelikes. So my general plan so far is just to keep the symbolic aspect but with 3D stuffs, then do animations for now only through code, and think about the other aspects later. This should work but I still need to start thinking about the general aspect of the game once I will start to publish versions. Because I want the style to still be immediately recognizable, unique ideally, and very stylized. I will probably need an artist at some point (in some years?) but I need to think about which kind of art to target. Also, obviously MEGASTRCUTURES needs to be partially inspired graphically by it’s inspirations like “Blame!”, “Ghost In The Shell” and many other references I listed in the presentations of the game. So even if the game will first look symbolic, it needs a general ambiance related to this, but with a unique twist. I have some ideas but I really need to stop coding and organizing to focus on this at some point.

The same question came up for audio, but at least for audio I can experiment myself as I’m a musician and compose sometime for small amateur games. I was thinking maybe when I can focus on that aspect I’ll just livestream making some music for MEGASTRUCTURES. I did that in the past, it was a bit exhausting but it was fun.

As for some other decisions I took the time to ponder, there is the spatial and temporal/turn structure question. Here is what I’m going with:

• the game space is a 3D grid (as planned initially), uniform
• a normal human body will have a height of 2 cubes, which means that things can take less or more than one cube
• this implies a “collision” system that handles some shapes not being able to go through some narrow spaces. This also allows to use size of a body to play in the balance of pros and cons when chosing a body. Say, you find a GITS tank, Tachikoma or Fushikoma style of synthetic body, you tradeof a lot of powerful features, including walking on walls, with size preventing you from going through some doors.
• this also implies that things have an orientation, including towards up or down and rotations. Basically I’ll end up with a 3D tetris collision solver hahaha
• time have a finite granularity, something close to the millisecond
• actions have a time span between their engagement and being done with them. This might mean the action is doing something in that time-span, or that the action’s effect happen at the end of the time-span. Another way to think about it is that effects have a time-span and actions launches effects. So basically, walking from one position to another implies advancing time for like a hundread or thousand clicks/ticks/cycles (not sure how I will call that, but probably not *-seconds).
• actions take resources from the body to be able to start and only release these resources. You need locomotion limbs (aka legs) to walk, in addition to a small amount of attention. You need a bit lot of attention to be hacking. This means you could be hacking while doing melee combat IFF you also have the resources from your body to handle that. Or some other stuffs that boost your mind etc.
• the game plays on 3 parallel level of reality: basis physical world (I’ll have to find a simple name for this), the network (where connections and hacking happens), virtual worlds (where you can escape with your mind and use as transfer spaces). Actions applies to one or several of these levels of reality. For each level, they have a different cost in time and body resources. This means you could be sending emails to your friends in parallel of defending against an adversary, because sending messages would be low-cost and very fast going on in your mind. Virtual worlds also allow downloading yourself into them completely in which case you abandon your physical body so you spend less resources in both worlds.
• the perceived granularity of time will depend on the mind-speed (or something in that direction) so some high-speed actions requires your mind to be fast too, and some also requires your body to be fast enough.
• small object takes a part of spatial cubes. Not sure if I will put a limit of how many such objects would take space in a cube, or if I will simplify to categories with infinite and specific numbers.

One option I would like to try also is to allow bodies to share space (at least partially), usually to handle close-combat (there are other close-body activities possible… but let’s focus on what the game is about first haha). Not sure if it will be interesting or even technically work, but that’s an idea that would make the game very original and interestingly weird.

Note that this is the core structure I’m going with, this is not what the player will be presented with at first. This structure opens potential to implement “naturally” other features that I planned. For example, suddenly high-speed combat in slow-motion makes sense if you have a body and knowledge (which drives the mind) to do high-speed actions. The real difficulty for me here, from a game development point of view, will be to convey the details to the player. I want that the players at first dont see all this, they play very simply because a lot of thigns are simplified by either the lack of resources in their body (they can’t see the real granularity of the time cycles cause they are slow) and mind-software doing things automatically for them. I intend to have “automatic combat” kinds of software that would handle everything so that the player handles things “macro”. They would go into the details only with advanced understanding, advanced tools and bodies that would allow them.

Of course I might end up simplifying or cutting some aspects but these core points are kind of intertwined because, well, it’s space-time XD

Anyway, there is a lot of things for me to prepare, including starting to develop the code architecture to handle all this. I already have a general idea at least for.

One thing I’m wondering is if I will go with the same logic I used in Hard Glitch where there was a set of general rules (objects with functoins) which were providing the code run every turn, the code run every time we need to determine the actions possible for a actor. Technically this is not the most efficient way to do things and I’m not totally sure it will be necessary.

Anyway we’ll see, as you can see I got a lot on my mind and a lot of setup to do so next few weeks will be busy.