Lithier
patreon
Dev Journal: This is Why I Do It
🕑 Added 2023-03-29 20:17:04 +0000 UTCIn my last public journal, I said I'd be trying hard to post a supporter journal sooner than usual, to break out of this cycle of posting just twice per update with the “I finally got over the hump” for supporters and the “coming out soon!” for the public. It had become a habit of mine for some time for various reasons, but I wanted to break that habit and make more frequent, smaller posts.
So, while this post is still a significant length of time after the last one, this is still a much earlier post than I might have made before, and I'm tentatively chalking this up as a successful first step. I'm still in the thick of figuring things out, I'm still struggling and frustrated, as I suspected I might be for quite some time.
But when I resolved to get this post out today, I realized something as I was thinking over what I would talk about. Wouldn't this post be sort of... boring? The news would be mildly interesting, but not very exciting or emotionally engaging, probably. My usual cycle of posts before was getting repetitive, but it was a pretty emotional experience each time.
That's when I realized that maybe there had been another reason I'd been putting off these posts until certain moments: it's easier to make a post with “emotional content” to it, a post that has a story with ups and downs and some resolution, at those key moments. I guess that partly I felt compelled as a storyteller to wait until the “story” was “ready”?
But these posts aren't supposed to be stories. They're to let the world know I'm still working at it, and give the people paying for my food a rough idea of where I am in the whole process.
So I'm going to try and buckle down and just give an honest assessment of where I am on a more regular basis, even if it feels “boring” or anticlimactic at the times I put aside to update you. I don't know if I'll ever be getting them out as regularly as my original goal, but there's certainly a lot of room for improvement from where we've been.
Alright, enough news about news. Where are we with the actual development?
I did do a lot of soul-searching on what I should work on for this update. As I explored in the announcement for v0.05, there are a few options for things that maybe should be done first, but the big question was, should I tackle AI next? The AI overhaul has been my big goal for... well, since the project began, really. I knew the existing system for controlling NPCs was a placeholder from the start, and one I wasn't very happy with.
At this point, just about everything about the game's future development can be spoken of in terms of AI. The majority of feedback and suggestions I get, I couldn't implement until after the AI is overhauled. There are so many things the game itself needs and wants that would just create loads of extra work trying to implement them before I upgrade the AI. I do want PWO to be a game that is emotionally engaging, that has more “meaning” to it than other porn games, but there's no way I could even begin to pull that off with the primitive AI we have now. I might even be tempted to work on other projects, like importing MVOL to Unity or squeezing in another prototype for something fun and different, but I'd feel awful leaving PWO with its current AI while I was working on other stuff.
I eventually realized that I've basically been focusing the majority of my time and effort on getting ready for AI for quite some time now, and if I wasn't working toward the better AI in some way, I'd probably go crazy. So I decided to take the plunge.
My single goal for PWO v0.06 is to implement at least the most barebones version of the overhauled AI system. I don't want to get sucked into the time whirlpool of overdesigning that I did with the Location system (although this might be the single place where no amount of design is too much), I want to figure out how the foundation will work and build it, even if it means cobbling it together into an ugly hybrid with the existing system just as a halfway step for the coming version.
Since then, I have plunged deep into design theory and struggled to find ways to turn my understanding of the psyche into something I can plot out with numbers and spreadsheets. It has been a strange, almost surreal experience and an incredible struggle, but I also feel just a little like this is, in a way, what I have been trying to prepare myself to do my entire life.
From a very young age, I was a storyteller, a game designer, and a student of the mind itself. I played with other children by making up scenarios and asking, “what do you do?” Then spinning their choices into the story. I basically was playing the DM from Dungeons & Dragons before I'd ever heard of the game. Eventually I did become a DM for multiple campaigns, and of course, eventually I got into making games proper.
But one of the biggest things to me in those early days was trying to better understand the nature of people. “What would you do in this unlikely situation?” “What would it take to change that answer?” “When these two priorities of yours are at odds, which will you pick?” I've been trying to study people in this and many other ways all my life, including trying to disassemble and examine all the pieces of my own mind, and even trying to reassemble them other ways to better understand and predict the behavior of others.
All of that helped me in becoming a writer, and a large part of the experience both as a writer and game designer was seeing how people responded and how they engaged with my work. It all fed into my understanding of the mind.
Today, I feel like I am very comfortably familiar with how minds of many different kinds work, and in a much more specific way I can actually describe and talk about than even most people that tend to learn about others in a more instinctual or purely emotional way. I'm very experienced with internally simulating the minds of others and identifying all the little subtleties of behavior, and what drives people to act one way or another. On top of that, I'm gradually becoming more comfortable with programming, and have experience making a person “feel real” through a mixture of programming and storytelling.
It feels like designing this AI is something of a culmination of both the skills I've been accumulating over my life and of all my interests, desires, and goals. It's grueling work, and I keep trying to shake off this sense of pressure, that “I've got to get this perfect no matter what,” given all of this. I have to pace myself very carefully with it, and that's been frustrating, but I already spend a lot of my time tired out and half-fried. I keep taking breaks, then finding myself lunging back into it, hungry for more even when I'm still aching.
But the important thing is, I'm making steady progress, and I'm trying to avoid delving too deep into anything that can be pushed off to the future. It's true that in laying these foundations I need to do a lot of anticipatory design, to make sure I don't make it impossible or too costly to add or change something I'll need in the future. I'm trying to apply every lesson I've learned in the depths of designing this project to make something that I can be proud of for the rest of my life, one piece at a time.
The tough part is that I know, as intense and laborious as this initial stage is, laying out the most basic ideas of how I will even begin to build this system, it is only the first fraction of the work I'll have to do for this to actually manifest as a playable component in the game. It makes me want to hurry, but I know that if I hurry now it will only make the later stages more complicated and troublesome. I need to design this right the first time.
So far, I have laid out the basics of all core motivators that drives a sentient being. I've sketched out many different mechanisms for how those motivators interact with each other, and how the AI keeps track of its various needs and desires in more subtle ways. I've explored mechanisms for how it assesses its situation now and in the future, and I've started delving deeper into how it can naturally form plans to achieve complex goals based on expectations for how the world works: if it can bring about situation X and do action Y, then it will be more likely to achieve goal Z. I've been working to figure out the best way to reduce these very abstract ideas to Objects and numbers the game can run as quickly and efficiently as possible for every NPC in the game, so that even all the characters out of sight can be making rational and emotional decisions that shape them and their relationships before they meet you.
The funniest part in all this is that, like a lot of the work I've done with PWO, the sheer amount of work going into all this will not be very visible to players, especially early on. It's very common for games to try and cheat to make it look a certain way with minimal effort, but those always fall apart under closer examination. The world quickly feels hollow and meaningless. I want to make something better, and I'm well aware that for a lot of people, this would be a foolish thing to even attempt. I'm ready to compromise and pare things down if I have to, but to some extent I feel like if anyone were going to make this work, it might well be me.
I have to try. So much of the work I've done, so much of what I am and why I'm even in this field demands it. It doesn't have to be perfect. I know it won't be. But I'm going to pour my heart into making something special, something good enough that it stands tall not for pandering, not for being well-targeted, not for being the smart business choice, but because it is good and it deserves to exist.
That's what brought MVOL into being, and with your help, I was able to make a living even off something as absurd as an emotional cat simulator. I feel like the stage is finally set for me to make something truly special again, even if it's a long road before all this hard work truly comes to fruition.
Thank you for making this possible. Thank you for giving me the time and the space to make the thing I truly want to make, to try to crystallize all of my experience and passion into something beautiful.
Haha, I guess I ended up making an emotional story out of this post after all. Well... I'll try to accept that it doesn't need all this fanfare next time. I'm still working on it.
Thank you for your support and your patience. Thank you for believing in me. I couldn't have gotten to this moment without your help, and that means the world to me. I'd say I'll try not to disappoint you, but at this point, it's more honest to say I'm trying not to disappoint myself. But... that's how it should be, really.
Maybe above everything else, thank you for that.
