text
stringlengths
0
1.55k
**Nick Nisi:** It just said, "Surely not everyone was kung fu fighting." \[laughter\]
**Jerod Santo:** Oh, yeah. That is a good one.
**Brian Breiholz:** That's a good one, yeah.
**Jerod Santo:** And this disco warrior game has an epic soundtrack, a very loud soundtrack, so when you pop the tab open and get warrioring, it's in your ears... As well as some nice voiceovers, like "Get off of our disco", or like "You have no moves..." I don't know, I can't remember the exact things that you said, b...
**Brian Breiholz:** I don't know, there were like 20 hours left -- by the way, it was also horrible time management on the first one... So I introduced tons of features on the last night, and it backfired at some point as well... But the voice lines did happen. I spent like two or three hours on some AI voice generatio...
**Jerod Santo:** A lot of iterating to get to that actual...
**Brian Breiholz:** Yeah, yeah...
**Jerod Santo:** That's when you're just done coding, and you're like "I need to do something else for a while. I'm gonna go make some voices, and integrate these."
**Brian Breiholz:** Yeah, I needed a break at that point, so I did like two or three hours of that.
**Jerod Santo:** That's the fun stuff. So was your main problem with time management was just introducing too many things too late in the game?
**Brian Breiholz:** Yeah. Previously for the game engine that we were working on, I've written this huge animation system, character animation system. In modern games there are these insanely complex state machines where you have a character that goes from idle to walk, to run, and potentially those different states ar...
**Jerod Santo:** Do you just stop and move on? Or do you keep polishing your game after it's all over with?
**Brian Breiholz:** After the jam?
**Jerod Santo:** Yeah.
**Brian Breiholz:** No, I actually never touch that codebase again, because it was like 70% --
**Jerod Santo:** \[laughs\] I know that feeling.
**Brian Breiholz:** ...like 70% through the thing. Everything was looking pretty good. I was thinking "Okay, you could actually keep on working on top of that." But it just degenerated completely over the last two days. Lack of sleep, lack of time, so you just end up hacking stuff in the worst way possible... So I didn...
**Break**: \[11:32\]
**Nick Nisi:** You mentioned a couple times that you're working on a game engine, and I'm curious about - like, when it comes to these... I assume that you're using your game engine in this, but is that something that you come pre-prepared with? Do others come with their own game engines, or their shared -- or how does...
**Brian Breiholz:** So to start with, we have the requirement it has to be React, right?
**Nick Nisi:** Yup.
**Brian Breiholz:** And in the case of these games, they are not just, let's say, a Three.js app, and inside a React wrapper, where it's like, okay, there's an app and a full-screen canvas, and that's it. There's no more React. No. This is an actual React game engine, so you will find React components that say like -- ...
**Nick Nisi:** Okay. Yeah, that was kind of gonna be my question, is what's the difference between like Three.js and React Three Fiber, and then the game engine. What does Three provide, and then what does the game engine provide on top of that?
**Brian Breiholz:** Okay. So yeah, let's start with vanilla Three. Vanilla Three.js is a renderer, so there are no game-specific tools. Obviously, there are a lot of tools that you also use for games, but you can also just use it for some product page. And probably the most use cases of Three.js aren't games; they are ...
Then there's React Three Fiber, which is just 3GS, but it gives you a React syntax for it. So it allows you to express your scene as JSX. Meaning instead of having a bunch of scene.addmesh1, scene.addmesh2, mesh2. addchild, you have your JSX hierarchy, where it's, okay, mesh, and then child mash, and maybe another chil...
There's obviously a lot of people who prefer one over the other. I personally like Three Fiber a lot, because JSX just gives you this immediate visibility of what's going on. You open a file and you see what is there. And in turn, it's gives you the ability to abstract a lot.
With Three Fiber we have this collection of components called Drei, which - they're mostly rendering abstractions, but they are insanely helpful. A lot of stuff that you would have to do yourself if you're using vanilla Three.js, that just saves tons of time to make a scene look good. And that's kind of the question, "...
**Nick Nisi:** No, this is super-interesting. It's always nebulous when I hear game engine... To me, that's the really cool part of game development. I've never really done much with game development, but that's where -- so Three and React Three Fiber, it sounds like that's using React to kind of give you the tools to ...
**Brian Breiholz:** Yeah, that hits pretty close. Where three fiber provides a lot of rendering abstractions, our engine is able to provide behavior abstractions. So in React Three Fiber there's no way to attach data to a mesh. And if you say "Okay, your mesh is your player, for example." It's an example. Then, okay, h...
So that's where our engine comes in and it allows you to, at the same time separate the data, have it live outside of React, update it imperatively 60 times a second, but at the same time, compose it inside JSX and have it there together with your mesh, with your player, where you're looking for it.
**Nick Nisi:** With the React wrapper around this - like you were mentioning, being able to see everything from a JSX kind of level, see everything that's loaded into the page, and things like that... That's really cool, and I agree with that. I'm curious if other metaphors from React carry over. Do you end up writing ...
**Brian Breiholz:** Definitely. There are hooks, as I mentioned. There are engine uses, what's called an ECS, Entity Component System; we probably don't have the time to go into that, but it's a way to manage your data and functionality. And for example, we provide hooks, like use component, where it's,okay, you give a...
So there are definitely hooks, which are similar to classical React; they are higher-order components, where in Cooper & Onita for example I had these big, red buttons that you could walk on, that would lift a platform. That was basically the mesh for the button, together with a higher order component for - I think I c...
**Nick Nisi:** Regarding the game engine, you mentioned one thing, and that was state machines. So I had to ask about that. Is that something that the game engine provides? And then also, towards that, does the game engine provide things like -- I'm just thinking of my very vague experience with things like Unity, and ...
**Brian Breiholz:** Okay, so first off, there are basic tools that you will encounter in any game engine that we offer, like an input system, a decent camera system, things like that. Then we are working on the editor as well, or on a editor, that will probably be a later release. So right now if you're asking about le...
**Nick Nisi:** Nice.
**Jerod Santo:** So if you don't know Blender, you're up a creek.
**Brian Breiholz:** At least for now.
**Jerod Santo:** Okay... Sorry, Nick. We're done. We're done here.
**Nick Nisi:** I was done a long time ago, but it's super-interesting... \[laughter\]
**Jerod Santo:** This is why we're flat -- what do you call yourself, a flatlander? That's why we're flatlanders, you know? 3D is just a different breed. Is this game engine going to be open source? Is it going to be for sale? How's it going to be released?
**Brian Breiholz:** It's going to be open source. It's going to be part of the -- I'm always not sure how to pronounce it... \[unintelligible 00:26:58.05\] The collective around Three Fiber, that has a bunch of other libraries under it, like Zustand, and React Spring, and a bunch of others.
**Jerod Santo:** Now your next React Game Jam, you went from a meager third place to first place... Which you can't actually go any higher than that; I wonder if you knew that or not... So you might as well just give up now and not do another one, but... First place, best game overall. This was a Cooper & Onita: Midnig...
**Brian Breiholz:** Yeah. So when the second theme was announced, it was a lot more stressful than the first time around, because I had no idea how to incorporate the theme of cooperation into the game. You can't just fake it and say "Yeah, look, it's a disco. It's on the theme." Right?
**Jerod Santo:** \[laughs\] Right.
**Brian Breiholz:** So it has to be something that actually involves gameplay. So I was contemplating what to do. Eventually I ended up on the Unity store again, browsing for cool-looking models, and I was already working with the Cooper model, the dog, earlier, doing animation work... And the artist has absolutely ama...
Aside from that, time management was a bit better. I didn't scoop in that many features on the last night, but it was also -- it was still stressful. I think for everyone participating in these jams, it's just -- it's a lot. And I had to fix Engine Box for like three days straight, where the scene switching wouldn't wo...
Originally, those were supposed to be different scenes, handled by our scene managing entity. But it turned out that it was horribly buggy, and they were bugs that you couldn't ignore. It was like "Okay, the new scene gets loaded. Suddenly, Onita is 100 meters tall." And you have like the whole model across the screen....
**Jerod Santo:** So you haven't touched the code since, just like the other time.
**Brian Breiholz:** Nope. Nope. I've been thinking about doing a second version of it. I wanted to make some games for one of the online platforms, like Rune, for example, or Pokey... Crazy Games I think is another one... And that's when I thought "Okay, maybe do a remake of Cooper & Onita." But I haven't gotten to tha...