I Tried to Make a Video Game. Here's What Happened.

Friedemann: I am making my very first game. This is a tower defense game for musicians where you have to use your ears to defeat enemies. So, if there's one thing that I'm known for, I think as a creator, is that I love systems. I love figuring out how something works. I love seeing beyond the curtain. I mean, The Matrix is my favorite movie of all time. And I love that feeling you get when you see a problem, maybe from a different point of view, and suddenly something clicks and you're like, my God, I figured it out. this is what drove me to make holistic songwriting academy: is figuring out how groove works, how melodies work, how repetition works. It's what drove me to make my own board games and it's what's driving me to make my own video games now. I've always had a fascination with games and really game design. I've always loved the process of putting something together that then works as a functioning unit that you can interact with to have fun. I mean, that's like the most German explanation of it that I could probably have thought of, but that's what I love. I think it's fascinating to create a machine that essentially spills out fun for people who interact with it. That's freaking awesome, right? So I've always been a systems thinker, which also means I am particularly drawn to a type of game called the Euro game. I am really interested in high strategy, low luck. My skill should be the determining factor in whether I win the game or not. Because that forces me to interact with the system in the most meaningful way, I think, where I have to the best player is the player who can understand the system the best. so over the years I've done a lot of game design myself. It it happens like this for me. When I'm fascinated by something, I'll watch a yacht of a lot of YouTube stuff on it. And the more I watch, the more I'm like feeling the urge to create something like this myself. That's how I built this guitar over there, because I saw people Building their own guitars online. I was like, damn, I want to do that. That's awesome. And so that's how I came to board game design as well. I started with this series right here. It's the Cantaloop series, which has consists of three distinct parts. These have been published with Lookoutspiele and have sold over 40,000 copies. So they've been very successful in the board game realm. I, of course, made This deck of cards, the songwriting decks, which isn't really a game, but it's a systematized way of thinking about songwriting that also uses physical components, cards. And I'm currently working on another game. nerds versus gnomes. I'll talk about all of those games in a future video and And how I made them and what kind of how I got the ideas. if this whole connection of me being into b board games is strange to you, I think it's it's not so strange. I think teaching and board games really go hand in hand. And it's not just because I do enjoy teaching games to new players, it's not just because I actually like reading the rules. I remember very distinctly reading the rules to this awesome game by Alexana Fista, Great Western Trail, which is has the rules written by Viktor Kubilke, I believe. that's right. And these rules here was the first time that I read some rules and I thought, I'm actually excited about this. I can't wait to to play this and to try this out. because these were so well written, I had I I could really feel the joy of teaching. I could feel the joy of learning, and I could see myself playing it. reading a lot of board game manuals really has made me a better teacher. because when you read something like that that is so well structured and explains your concept in the most clear way, it really pushes me to s to do the same thing for songwriting, where this is not the norm at all. I feel like a lot of songwriting teachers are really vague and they're not able to put their thoughts into clear words. Board games have pushed me to do better at that. and also I think good players are naturally good learners. because if somebody explains a game to me that I've never played before and I'm able to grok very quickly, aha, that's how this system interacts with that system. So if I do this, this will lead to this, and that means I can score some points off of here. so my goal, my personal goal to myself is I want to win every first game, every first play that I have with a new game. Not as a challenge against the other players, but just for myself to really push myself. And if I lose, it's fine, right? But it's to push myself to really understand everything about the game at a glance. Like that's what I expect myself to do as a designer. So then video games. I've always wanted to make video games. I've always been fascinated by them. I've always wanted to be involved in making one, which is why one of my minors when I studied music was game audio. but that never really there never really came anything off of it. the stars never really aligned. I have some friends who are programmers and game designers, but for some reason our schedules never matched up and we never managed to make something together. I've always been a little scared of programming because I've always felt like, man, it's like a whole new language that I would have to learn. And I don't really know if I'm, you know, still young enough to do that. That's that was kind of my fear. at some point, and this must have been maybe 2020 or so. I commissioned a holistic songwriting ear training app because I had so many ideas for apps and games that I wanted to bring them out into the world. And back then the YouTube channel was booming. I had tons of money. And so I decided to put all of that money into an app. And it was gonna be a quiz game type thing where people could test their ears and improve their own ear training. By answering questions, kind of like a little quiz app that would play you a note and you'd have to figure out, this is the four and the key or whatever, this is the four chord. and the biggest difference between this and other apps like Earmaster and whatever they're called, Tone Gym and whatever is out there, would be that this one would look fun and be fast, right? And it would test your skill to to get faster at doing this thing. Cause the the issue I've always had with Earmaster is that it's so ugly. And so boring. And it does not feel like a game at all. It feels like you're taking a test at a an ancient university, right? With the my god, with the staff and the notes and everything. It's just the font of everything is just so ugly. And ugh. I I mean I hate that. I wanted to make something that looks fun and flashy and that rewards you when you press the right button. It's just does like a little explosion on the screen and it does like a cool sound effect, you know. That's what I wanted to do. And so I sunk. About a car's worth of money into this, and an expensive car, not just into the developing and the programming, which was incredibly expensive, but also into graphics and artwork. The problem with it was, and why it never came to fruition, is because the developers obviously didn't know music theory to the degree that they would have needed to know it. So that was a big issue. it was difficult communicating with them because I was like, I've already tried to make this as clear as I possibly can, but this is just music theory. It is weird and clunky and it's old and it None of it makes sense. You know, that's just how music theory is. And then in 2021, my father died, and I was still doing about seven projects at the same time, like I am doing again now. and I had to put most of that on hold. And the this app really cost me a lot of patience. And so I decided to put that on ice for the time being. When I came back about six months later, after I had recovered a little bit. the company that had made this app was no more or had transitioned to a different company. I tried multiple times to get them to finish the game and that didn't happen. And whenever I think about that story, it it hurts. 'Cause I wanted to do something good with all this money that people gave me. And I messed it up and that sucks. And the unfortunate truth is that sometimes this kind of stuff happens. The lesson that I drew from all of this was that developing games is super expensive and I should not do it because I obviously don't know what I'm doing. Fast forward a couple of years, AI is popping onto the market. And after a few years of being completely overwhelmed by it and being terrified of it, actually my notes here say after a few years of F U C K in capital letters, I decided I was going to, instead of letting it turn me into a passive person, which was the risk, very much the risk for a while, I decided I'm going to use this to Up my game to push myself to do bigger and bolder projects instead. And because I'd heard from a lot of people that coding is now a lot simpler if you have AI, I decided, okay, then I'm gonna do it. I am going to learn how to code. You know, I I don't want to vibe code my game. I at least want to be able to understand what the code is saying, and I want to be able to read it at least, if I can write it all the better. But at least I want to have some basic insight into what it is that my game is doing. I quickly figured out that I'm going to be doing this in a program called Godot, and I found an awesome school online called GDQuest. Very much recommended. I had never done any programming before, and it really taught me the basics so well. And showed me how to program my own first games. It was an awesome experience, highly recommended. long story short, I can read code now and write some of it, right? When AI shows me some of the stuff, I continuously still have to ask, what does that mean? What does this expression mean? but I'm feeling a lot more confident in asking those questions. And about 70 to 80 percent of all the code that I'm giving now makes sense to me. Which has obviously been massively helpful for programming my own game. I'm using AI, but I'm able to read what it's giving me. So I had a couple of ideas for what kind of game I wanted to make. I kind of narrowed it down to four or five out of a huge document that I have with ideas for projects. And then I picked the smallest one off of that list, which is This tower defense ear training idea. for it, I already have an audience, musicians like most of you, I'm guessing. and it felt like a pretty small game that I could probably make in three months because that's the time that I gave myself for it. So, ear training, tower defense. what is the idea behind that? You have a tower that you need to defend, and from the right and left, enemies are coming in that you need to defend against. These enemies will play notes or they'll play chords or maybe melodies. And your job is to press the right degree on your keyboard. So, for example, let's say you're in C major and the enemy plays an E, right? You would have to play three on your keyboard to kill that enemy. Or maybe later on you'd have a more difficult enemy that would play E, and you would actually have to press E on your keyboard to kill that enemy. So that really was the idea to help musicians and really my students at Conservatory who really struggle with this on a regular basis to learn ear training. I personally believe that ear training is the most useful, but also the hardest skill to master in music. But it is so incredibly valuable if you can pick out what chords you're hearing. What the melody is doing. That's the perfect analysis tool. And so much will come out of that if you are able to hear things in real time. So now I'm getting into designing. And I quickly realized that instead of picking the shortest game, I really should have picked the simplest one. And again, I have to find that music theory does not translate well to beeps and bops. I mean, there are so many weird things about music theory if you think about it. For example, why is there G sharp and a G flat? But E sharp is actually F? Like there's a bunch of things that are just really weird in music theory, and it's hard to explain those to a machine. But anyway, during that first month, I was building towards creating an alpha. An alpha prototype is basically a version that is that works, that doesn't look very good, but all the functionality is there. And you can see what the final game might feel like, even though it doesn't look the part, there's the sound effects aren't incorporated yet. It doesn't feel like a finished game yet. So I worked on this all day, but made pretty good progress actually. I was surprised that things got working a lot quicker than I had anticipated. And I think after three or four days I already had something where it had an enemy type that I could fight, which felt pretty amazing. so what does the work of a developer actually look like? I can only speak for myself, of course, but for me it was this process. First, I had to have a a vision. I had to figure out, okay, here's how I want this gaming system, let's say the attacking system to work. That's my vision. Here's what that should look like in the end. Then you figure out the simplest version of that and you make that. Because if you try to make that, it's going to take you forever. You need to play test your game as soon as possible to you know erode any errors that might creep up, because there'll always be loads of those. Then, once you have a simple version, you play test that and you improve on that until it gets fun. And then, as the last step, you make it pretty. I think that's actually a really great lesson for music creation as well. And I've I've been talking about that ever since I made my own board games. Is that we shouldn't spend so much time on making our music sound pretty and so sound and production. We should write the song first, figure out what is going to be my song. Is that fun? Does that make for an enjoyable experience? And then we make it pretty. That should be the last step, not the inverse of that, which is what most people do. Is they start to make something that sounds pretty. But then they don't have a song, they don't have anything that's actually satisfying to listen to as a whole. So once I had that, I would then tell ChatGPT, I know, I know, I should have been using Claude, what I want, and kind of read through what it sent back to me, giving notes about the code and finally implementing that code into my game. So I didn't write most of my code myself. I did let ChatGPT handle that because there's a lot of things that can go wrong, especially for a beginner like me. for example, if I'm going from a one chord enemy to a two chord enemy, that doesn't just mean I add another chord to the enemy. it also means I probably want to figure out the voice lating between those two chords. So it sounds like an actual chord progression you would hear in a song. Because if both are just like, you know, one is down here, the other one is up here, it's gonna sound weird. And so dealing with a problem like that, I'm like, my God, voice leading? How do I explain voice leading to a computer? That's so complicated if you think about it. what I figured out after a while is I can actually tell Chat GPT, here's the vision, here's the s the things that I want to happen eventually. How do we break this down? Where do we start? How do we make this as simply as possible? And then how do we gradually make it more and more complex, testing it every stage of the way? And so ChatGPT would actually help me make a plan for myself and figure out how to tackle these difficult challenges. I don't want it to sound like I let ChatGPT do everything. Again, I read through everything, and when I saw changes that had to be made, I tried to make those changes myself. I did all of the designing for it. So the game design is all mine. The vision obviously is mine. I wrote all of the music and made all of the sound effects myself, obviously. And for the beta, which we're not gonna talk about today, I wrote the entire story by myself, of course. So what is it like to design a game? Man, it's really fun. It's better than I had anticipated. it's fun to see changes immediately take place, right? That's awesome. Of course, I should also say it is incredibly hard. It is a steep learning curve, and making a simple little mistake somewhere can easily haunt you for weeks. sometimes mistakes don't break your game, but they kind of stick around. And then two weeks later you do something and it doesn't work, and you're like, but this should be working, right? And then finding and backtracking until you find that mistake. Hard. Man, really difficult. And it's the little things that sometimes make things so hard. Like one day I thought, okay, let's spend the next half hour to make a pause menu where you hit escape, pause menu comes up, you hit escape again, pause menu closes. Should be simple, right? I thought half an hour would be enough for that. Turns out what ended up happening was I hit escape, nothing happens. Like, what's going on? Very strange, right? So I hit escape, I hit escape, nothing happens. I check all the code, everything looks fine. Until I realized that that escape press immediately triggered the closing of the pause menu as well. So it did actually pop up. It was just so fast that I didn't even see it. So that took me an entire day to figure out. Then there's a thing called keyboard ghosting, which was a huge problem. I had never even heard of that before. sometimes in my game you have to hold a direction key, then press a note, and then press up or down to make it sharp or flat. And that was a problem. For some notes, that didn't work. And I had no idea why. I checked all the code, everything made sense. I didn't get why it worked for G sharp, but not for B sharp or whatever, and it worked on my laptop, it didn't work on the computer. Apparently, it's the problem is the keyboard itself some of them are built cheaply, frankly, and other w ones are built more robust, so they can actually register all of those button presses. Other keyboards cannot. And so that's something you figure out through trial and error. I only figure that out because I have some really good friends, Lisa and Ranjit, who managed to help me with that stuff. So shout out to you guys. Thank you. It was also Lisa and Rangit, by the way, who suggested I install GitHub on my computer, which is a program that saves different states of the game. So you have like different, you can log. Okay, today I did this, save. The next day you today I did this, save. And if you ever Have something, if you ever do something that was stupid and you realize, I should go back to that earlier save, you can, right? So that's what GitHub is. And it's actually really awesome, except it didn't really work in the beginning and there were some weird issues with it, which means I lost a lot of progress in the first week. So a lot of stuff that I programmed that was already finished suddenly was all gone. And that was a big experience for me where I realized, you know what? The These programs like GitHub and Godot, they expect me to really understand programming language. They understand me to be a developer. They understand expect me to be smart and to not make mistakes like this. They the training wheels are off, right? Usually when you're on a Mac, this is what I'm working on. On a Mac, everything is so safe. Nothing can happen. You can't accidentally say, you know, delete something. Not the case when you're working with these programs. So that was a big learning experience for me. Be so much more careful. When you click OK or cancel on a text box. Now, I know a lot of you won't be happy that I am using AI for this. It's the truth of it, is just I wouldn't have been able to make this game without AI because I don't have a six-digit budget for something like this. It's just completely infeasible for me. Of course, I would love to make this with a real artist, but it would just be so much more expensive. It would have taken so much more time, and I wouldn't have the control that I need to make it and to put it into this game. So I don't really feel like I'm putting anybody out of work by doing this because this whole thing wouldn't have happened without AI in the first place. Anyway, besides the point, using AI was also a big learning experience because as I said, it never quite gives you what you want. And I I had found this site online called autosprite.ao, which is pretty cool. and I had generated a bunch of enemy types like orcs and you know, different types of like fantasy looking monsters or whatever. And I was excited to finally feed that into auto sprite, and then it gave me back my first sprite. Sprite is basically an animation sheet. It's the different states your character is in, right? And then the program puts those together into a fluent motion. so I fed that into Auto Sprite, and it was all wonky, and the legs just kind of like moved through each other, and it was just like all the worst AI crap you've ever seen, right? So that didn't work. What I decided to do instead was just change the enemies from orcs to ghosts, because ghosts don't have legs. And so that's not even an issue in the first place. So there were lots of moments during this process where I was really struggling. and I could feel my brain trying to pull me in other directions where it wanted me to go to YouTube or wanted me to go to the computer, and I could feel my body just wanting me to not do this. And I had to consciously force myself to know, you are figuring this out right now. it's it's a challenge, man. I mean, the war of art talks about resistance and a very much felt that, right? If you've ever read that book, That Manifesto, I suggest every musician, every artist needs to read that at least once in their lifetime. anyway, one month in, I finally have a playable alpha, a first playable version of the game. It's the last time I'm gonna head out to the conservatory before the summer break to teach ear training. And I really wanted to get it done before that date because I wanted my students To see it and to be able to practice with the program over the holidays. Which is every student's dream, am I right? so I set up a custom mode, like a screen where you could select which enemies you want to fight, how fast they should be, and how many of them you want to fight and which keys you want to play in, made a web export so they could play it online and just gave that to them over the summer break and said, like, hey, knock yourselves out. I'd love to get all the feedback that you have for me. and feedback I have gotten, and some of it was really brutal and t pushed me much further than I ever anticipated with this game. But we'll talk about that next time. one more thing before I let you go. We are doing another run of next level songwriter, and this will be the last run in quite a while, I believe. So this might be your last chance to get on board if you don't want to wait for a really long time. where next level songwriter, in case you don't know, is a 12 week program, super actionable. Every day you get a video that explains what you're supposed to do. And you have about 10 to 15 minutes every day to finish that challenge. very guided, super fun, lots of learning. This is our best off from everything that I've ever done. condensed into a very short and precise course. if you want to be a part of that, if you're not part of our mailing list yet, you go to holistic songwriting.com and you download our freebie. it's called fill that page. You can find it on the products page. and then you're part of the mailing list and you will be seeing messages about this new program, next level songwriter. So do check that out. And again, it's it's probably gonna be your last chance to join that in quite a while. So I if if you wanted to take it for a while, this would be a good moment to to jump on. All right. That's it for me. Thank you so much for listening. and see you again next week.