(keyboard typing) (dramatic music) - [Danny] On October 1st, 2021 a 19-year-old indie
developer in a remote pocket of Washington State posted a trailer. It was a trailer for a horror video game called Choo Choo Charles. A game that didn't exist
yet aside from a Steam page. The trailer went viral and
wishlists started pouring in. And this young indie dev Gavin Eisenbeisz suddenly had an audience
and a game to make. But this viral trailer
wasn't happenstance. Neither was the game's genre, the look of its antagonist,
or the engine it was using. It was all part of a plan. A strategy born from a developer with a ticking clock
hanging over his head. You see, Gavin wanted
to be a game developer but his parents, like any good parents, wanted him to have an
education to fall back on. He'd released a few
indie games in the past, but Gavin needed a
bonafide hit to convince his parents that he could bypass college and jump straight into
professional game dev. With no formal training
or industry connections, Gavin had a mountain to climb. To create a hit game he was going to have to do something special. Something that grabbed the attention of players around the world. In getting thousands of people to wishlist Choo Choo Charles Gavin had made that crucial first step. Now all he had to do was make a hit game. I interviewed Gavin on the No Clip Podcast a few weeks after that trailer first hit and I was struck by how levelheaded he was despite the tough road
that he had ahead of him. Hype can be a death sentence
if not handled the right way and I was really
interested in how he'd cope in not just making the game, but stopping his audience's expectations
from running wild. This was, after all, a game
made by a single developer. What happened with Choo Choo Charles over the next 15 months was fascinating. And so once the dust had settled, I traveled to the northern coast of Washington's Olympic Peninsula to catch up with Gavin
about how it all went down. The things that went right,
the stuff that went wrong, the four of 10 from IGN,
and what all those fans eventually thought when they got their own hands on Choo Choo Charles. (lofi music) - I liked video games when I was younger, obviously like pretty much everyone. And you know, eventually when I learned that you could just make games on your own for free on your home
computer I just, I don't know, it was just an enthralling concept to me. I say I was a kindergarten dropout. Went to public school until kindergarten, then was homeschooled until
I graduated high school. Originally I started on mobile. I really wanted to make mobile games, so I did a game called Sumo Skies about a skydiving sumo wrestler. Strange concept, but
it sounded really cool as a, I don't know, 14-year-old. I did a game called Behind These Eyes I think when I was 16, 17. That was like my first real like PC game. My first Steam game. Horrible game. But it made it to Steam
and I made a couple hundred bucks off of it,
so I was happy with it. A couple of smaller
Game Jam games as well. Cloud Climber and My Friend is a Raven over the course of a couple years. And then I did My Beautiful Paper Smile which was kind of my first major like sort of serious commercial game. It did okay, but it wasn't quite where I wanted things to be. And then after that I started
doing Choo Choo Charles. - [Danny] So how old
were you when you were releasing My Beautiful Paper Smile? - I started that one I
want to say when I was 16. And finished it, or published it actually, like a month after I
announced Choo Charles. So just like a year ago is when
I finished that at like 20. - [Danny] Well when did you
start learning a language? And what was it? - Never. I just used Blueprint. That's actually why I
started using Unreal Engine is 'cause I specifically, I was like I want to do this visual scripting thing. That sounds fun. And I just haven't had to change at all. It's worked well enough. My parents, they just
had their home office PC just for, you know, whatever. Emails, web surfing, all that. And that was kind of
where I first started, you know, doing development stuff. - [Danny] Nice, were they cool with you using it to make games? - Yeah. Yeah, I think so. I think so. They limited my development time though. They had screen time limits. That made me very angry,
but I forgive you, Dad! - When he started making video games, that was his way to do
this educational thing of learning how to make video games, but he would make video games and spend, you know, hours playing and
testing his video games. So he was kind of getting
around that screen time limit. - [Danny] What was Gavin
like as a kid growing up? - He was really smart. Very creative. He and his brother used to
do a lot of imaginary things. I kind of remember them making
videos about gold mining and moonshining and all
those kind of things. He took it very seriously
from the very beginning. And I'd say probably when
he started working on My Beautiful Paper
Smile, that's when I knew that he was gonna be
doing this professionally. I could see that he was
picking it up really fast. He was doing high quality work and he was getting better and better. It was easy to see that
he knew what he was doing. The worry was knowing that video game development is a creative endeavor. It's an artistic work and not everybody appreciates everybody's art. And he can do a good job,
he can make a a great game, but it doesn't mean that
it'll be well received. You know, there's a lot of
starving artists out there. That was the worry. Is he gonna spend all this time and maybe not have a commercial success? - [Danny] Right. - [Don] Will he need to have
an education to back him up? When he was still working
on My Beautiful Paper Smile, that was when we had the
conversation about college. - So originally I would've been going to college I think about a year before I started working on Choo Choo Charles. And so I kind of managed
to get my way out of it 'cause I signed a deal with a publisher on My Beautiful Paper Smile, so now I was legally obligated to work on that game and finish that game. By the time that that game finished, I'd started making money off
of some programming courses. And so college was starting to
make a little bit less sense 'cause I was already making
money kind of on my own. And then Choo Choo Charles
went viral a few months after that and it was like okay, yeah, so this is something that I can, you know, work on and something that I could bet on. - [Danny] Yeah. By the time Gavin started on this project he had accumulated a
healthy amount of solo dev experience for somebody his age. More importantly, he
seemed to have a very good understanding of where
his weaknesses lied. The areas he lacked experience. But even more importantly
than that perhaps, he knew that without an audience none of that was going to matter. He could spend a year
or two working on a game and have absolutely nobody play it if they never found out about it. So what he decided to do
instead was to pitch a game to the internet and gauge
interest from there. So how did he come up with the pitch? - So the goal was to try
and make it go viral. I had never gone viral
with anything before, so I didn't know if it
would actually work. I just did everything that I could. So that was kind of the goal. And so I had kind of a bit of a unique perspective having done YouTube for a bit. Not super successfully,
but I kind of understood the content creator mindset I guess. It was a big point for me to try to do anything I could to like make the game go just be more inherently viral. I mean obviously a big
one is Spider Train. Spider Train is a pretty big part of that. But there's so much stuff
like just the lighting, art style, the visual
elements to the gameplay. Trains, color schemes, it's just a lot of- - [Danny] The name. - The name, yeah, lots of stuff. You know, horror games, they, all the time they'll take something
that's cute or innocent and they'll make it scary, you know? Oh, evil dolls, evil animatronics,
whatever it might be. And so I was kind of realizing that there weren't many horror games or horror movies that kind of played off
of like kids TV shows. And like those are huge. I mean everybody has watched
a kids TV show as a kid. And so my favorite one growing up was Thomas the Tank Engine. ("Thomas & Friends" theme song) So I was like, well you know, a train-based game would
actually be kind of cool. So that's where the initial concept of a train-based game came from. So I was like well okay, you
could have a train monster. And then I had seen Tom Coben's animations pop up over the years. He's done like these
Cursed Thomas, Thomas.exe type things where it's
like a spider Thomas. And so I was like you
know, I could just kind of sort of take the spider legs and just sort of put them on this new character. And so that's where that came from. So I spent about a month
working on the game. Putting in place enough mechanics that I could make a trailer. 'Cause in the indie world
exposure is everything, traffic is everything, and you want to try and start marketing as soon as possible. So I was like all right, I
need to get my trailer out, I need to get my Steam page out, I'm just gonna get some basic gameplay in. I don't care if there's
bugs, anything like that. Whipped together a trailer
in a month and put that out on Twitter and it took off immediately. - [Danny] So that was from like no dev to trailer amount of dev was one month? - Yes. - [Danny] Wow, that's crazy. And so had you used Unreal before? - Yeah, so I'd been using Unreal for probably six, seven
years at that point. It was my first like big,
like first person game, so it was still a bit of a
learning curve, but yeah. - [Danny] So it goes
viral like straight away? - Yeah, like day one. - [Danny] Okay, so- - Well day two. Day two was when it hit its peak. - [Danny] What was that like? - It was overwhelming. A bit. A good overwhelming though. Yeah, I was kind of just glued to Twitter and YouTube just reloading. It was crazy. It was crazy to see it. It was the first time in as
long as I've been making games that something went to plan and better. Everything had always gone
worse than I hoped for, so this was a nice change of pace. Day one, dropped the trailer. It started picking up some steam. I think it got maybe 20, 30,000 views or something on Twitter
which was great for me. That was way more than I'd ever done. I think on YouTube it
got maybe 10 or 20,000, did a little bit of, you
know, some numbers on Reddit. And that translated to
about a thousand wishlists. And I was like holy crap.
A thousand wishlists. That's awesome. That's a lot. That was like the most I'd
ever gotten in one day. On any of my games. And then day two came
around and it pulled maybe, I think maybe like a a million or close to a million on Twitter, and then a little bit more on Reddit. And that translated to 14,000 wishlists. And that was my new biggest day now. That was when I was like okay, so this is actually, this is something. Like for My Beautiful
Paper Smile it took me three years to get that many wishlists and I just did this in a day. It did like 80,000 wishlists
in the first couple weeks. And 100,000 I guess probably
in the first month or so. - [Danny] What was it
at when you launched? - When I launched it was at 550,000. - [Danny] (laughs) That's crazy. - Yeah, I was pretty happy with that. Good count. (wind howling) (thunder booming) (ominous music) (Charlie snarling) - [Danny] Okay, let's
get off the hype train for a second and focus
on design for a bit. For those of you who haven't played it, Choo Choo Charles is a
game that takes place on an island where the community is being terrorized by
a large spider train. You play as an outsider who has come in to help the islanders. And you do this by
upgrading your own train with new weapons and
collecting a set of eggs which will trigger a climactic battle. The island also has a host
of interesting side quests and piles of scrap that you
can use to upgrade your train. It's a tight loop and the game can be completed in a handful of hours. So how did Gavin scope out this project? What problems did he run into? Where did he spend most
of his development time? And where did he cut corners to meet his one year development deadline? (dramatic music) - I pretty much had the
general gameplay loop honed in before I placed a single node. Yeah, I mean that was a big thing for me 'cause I'm very focused on format. 'Cause like in making
a game inherently viral I feel like there's a
lot of stuff that needs to be really baked into
just the core of the game. So a lot of the gameplay
elements like being able to drive a train around, the
fact that it's open world. That stuff was important to have in place right from the start. So that was just like
in the first few weeks that I was kind of concepting the game, that was when that stuff came in. The story, the mob and Warren and all like that kind of stuff, shrines and all that,
that came in very late. The story went through I
think three major shifts. I probably had four different
scripts that I went through. So that stuff I was playing
around with constantly until about six months before launch is when I finally sort of nailed that in. So the first thing that
I kind of had to do was just get the island itself, getting the track system set up, you know, kind of mapping out
where generally I wanted like points of interest
to be around the map. Just getting the rail system. That was kind of the first
big thing was just making it so you could actually get around the map. After that, I think it was kind of getting into like making the different missions, so I made like the three main missions. Went in and did the
different weapon missions. Went in and did the side missions. And then kind of throughout that time I was also working on, you
know, improving Charles' AI. That was something that
was, that was such a pain. I had to go through a lot of iterations just throughout development. Like if I had tried to make the game in a short period of time
it wouldn't have worked 'cause Charles' AI would've
been just horrible. So that was something that
needed just years to mature. Just making sure he actually is able to navigate effectively. That was a pain. It was really easy for him to get stuck. - [Danny] Right. - Especially with him
being as large as he is. One of the first things
I had to do was add various tests and checks in
order to check if he's stuck. Having different stages of like what he'll do if he does get stuck. So like first step, if he gets stuck, he'll run around in a circle
until he gets unstuck. If that doesn't work and
he's still not moving, then he has to like
teleport to a new location. And if that still doesn't
work then he'll keep teleporting until he gets to a location where he like actually has navigation. And that was another thing is like when you've got like these
steep cliffs and stuff. It's difficult getting the AI to like run effectively up a cliff. You know, it was common
for him to like slide into areas 'cause also
since he's so big and heavy he has to kinda like
drift around a little bit. So that results in him sliding
out of the navigation mesh into areas where he he just gets stuck. So then he has to have
more additional checks in case he slides out of bounds. Let's see, what are some other things? I mean just the difficulty. Just balancing it, you know? That took a lot of time,
just a lot of tuning, just having people try
it out, saying like okay, this was a little
difficult in the beginning, a little hard later on. There's like a slight
dynamic difficulty system. Just kind of a basic one like depending on how leveled up your train
is, he'll be scared off more easily or, you know, or less easily. His patrolling system,
that was another big one. Because if you have a spider
train in a massive island. That's just walking around
randomly, you'll never find him. He'll just be somewhere. He'll just be in a pit,
he'll be in a cave. He'll just be walking
around in circles somewhere and you won't see him the entire game. And people probably wouldn't enjoy buying a game called Choo Choo Charles if they never saw Choo Choo Charles. So yes, I had to design a handful of different systems that
I tried using for that. The first one was where
every like 10 or 20 minutes he would randomly just attack you. But that was kind of a
problem because it felt very unfair to be like
sitting in a building and then all of a sudden you see over the top of a mountain three miles away Charles charging straight
towards you, pinpointing you. That didn't feel very good. People didn't like that. I didn't even like that. It was pissing me off while
I was testing the game. - [Danny] Right.
- So I decided you know what? Maybe I should change that up a bit. Encounters with Charles were most fun when you could kind of see him nearby. And you could effectively avoid him. I ended up going with a
system where he essentially, he always knows where the player is. But he spirals around them and he gets closer and farther depending on, you know, various timers and stuff. So that way he'll be far away, you know, like right after you've fought him or after you've spawned in and started the game so that you're totally safe. And then a little later on
he'll get a little closer and you can kind of see him over the hills for a bit so you get kind of a warning. And then he comes in a little bit closer where you start hearing the music. And so then it's like, you know, maybe you have some close encounters. And then, you know, if you survive that, you know, and you can avoid
him for a few minutes, then he'll start kind
of like going back out and he'll give you a few minutes to get back to your train or whatever. And so tuning that pattern as well, just the scaling of his patrolling zone. That was another thing that just, it took months of just
playing and adjusting and playing and adjusting
until it felt natural. It's weird, in order to
make something feel natural it has to be so unnatural
behind the scenes. Aside from Charles' AI, the
other most difficult thing would've been the mob AI
which was 10 times worse. But honestly I regret having
the mob in the game at all. Because those stealth sections,
people did not like them. - The wrong place, my friend! - [Gavin] That's probably one of the biggest complaints I've had. And their AI was a pain,
so that made it even worse. It's like I spent lots
and lots and lots of time trying to make that AI and
that stealth stuff work. And it just didn't, it still fell flat. The game, like from a technical standpoint it's not that crazy. I'm not really doing anything
that hasn't been done before. So it was mostly pretty simple. One of the biggest things
probably would've just been like the open world
environment design stuff. That was pretty tedious. There
was a lot of stuff to do. That was probably a solid four months of development just day in, day out. I did a lot of live streams of it as well to try and make it a
little bit less tedious. Which did help. The scrap economy took a
little bit of balancing. Just figuring out like, okay, how much should this stuff cost? How much, you know, how
effective should the upgrades be? That stuff just took
some balancing over time. Again, it was one of these things where it's just months of, you know, as more people play it,
you know, you figure out which strategies are really overpowered and which ones are just
horrible that everybody avoids. It was just balancing over time. But the scrap system
and upgrade system and, you know, getting the
weapons from a mission, that that was always there. - Why don't you bring that
box of rocket ammunition back here from inside
the bunker down the rail? - [Danny] So during those 15 months you're obviously doing
so much of this work. Do you have help in any way
or is there people helping with writing or animation
or music or anything? Do you have any other
collaborators on this project? - A little bit, yeah. I mean I use, you know, like some pre-made assets here and there. You know, there's a lot
of, you know, like trees that I didn't make myself,
so help in that way. I hired a musician to
do some original tracks. I had a friend who helped me out with kind of the lore side of
things and the backstory. My family helped me out a lot with like kind of editing the script. My brother, he contributed a lot of top tier jokes to the game. - [Danny] Who's responsible
for Pickle Lady? - Pickle Lady was me. That was me. I take responsibility. (Danny laughs) - I need my pickles! - I did not have a lot
of people play testing until probably the last
three months of development. So I had just like friends and
family testing it early on. Which was part of where
some of the stress stuff was coming from 'cause like I had no idea how good or bad the game was 'cause not enough people had played it. Initially like it didn't
even cross my mind that people would care
about facial animations. Or that people would
dislike my Fuse characters, like what's wrong with my Fuse characters? They look great. So I just wasn't expecting people to even notice at all in the first place. And then people did notice. By the time I actually knew that people would've liked it to be higher quality it was too late in development to make changes like that anyways. But as far as like the broader scope of those sort of decisions,
I guess kind of my main decision-making process
is like is somebody going to write a negative review
if I don't add this feature? Or are they going to
write a positive review if I do add this feature? And anything else, it's like I still want to have like a little bit of extra. You know, I want to meet expectations and go a little bit above. But a lot of developers,
they think that a feature is really important that actually isn't. So it's also just like, for me personally, I like simple games. I like a game that I can
sit down and start playing and I don't need to spend five hours getting into it for it
to, you know, get good. Most of it can be boiled
down to fetch quest. Get this item from this place. But I tried to make it so
like there was, you know, oh, this one has a goofy
character, or this one takes place in a slightly
different environment, or, you know, this one
you have to do parkour. You know. I'm a very strong proponent of tools down like a month before launch. And so leading up to
launch, I basically had a month where it was
just emailing, you know, content creators, reaching out to, you know, just sending keys to people. Working on YouTube
videos, dev logs, TikToks, just the whole marketing campaign stuff. Ultimately, I mean it went
pretty well leading up. I was able to get done most of the stuff that I wanted to get done. I did a live stream where I actually launched the game on stream. And you know, for the
past couple weeks I'd been promoting like the specific
hour and the time zone. Everybody knew exactly
when it was dropping. So I, you know, on the live stream pressed the launch button. Release. My app. We press release now. There were probably four
or five points throughout development where the
pressure got really high. It was one of these things where I knew no matter when I released the game people were going to play it. And the big thing that
I had to worry about is were they going to like it? There were various points
throughout development when I would be playing the game, or I'd have a friend playing the game, and I'd realize oh, like
this, this really sucks. This is actually not a
good game at this point. Pressure went up and down
and then I would like fix a bunch of stuff and I'd be feeling good about it and I'd have somebody else play and it's like oh, it still sucks. By the time it got to launch
I was pretty confident in it. I'd spent plenty of time, you
know, probably three months just in polishing, I just
had hundreds of people play testing it and
giving tons of feedback. So I was pretty happy with, you know, the feedback that I was getting by launch. So the pressure on launch day. I mean it was high, but
it was pretty manageable. - [Danny] You had some confidence in it. - Yeah, I had some confidence. We just intently watch the spinny box. Let's start reloading the Steam page. Hey, I think it's out. I think it's out. Guys, go check it and let
me know in chat is it out? Is it out? Closed down the stream. Went and started checking numbers. I went to YouTube immediate, you know, I sent out codes to like content creators like two weeks beforehand,
so a flood of videos like immediately as the game came out. Started watching those. Started checking the
player count on SteamDB. Because, you know, you don't get the sales figures until
a little bit later. And it started, like the player count just started going up and I was like oh, that's already higher than
I was expecting it to be. Just in the first few minutes there. And then I was kind of just
able to sit back and watch it. There weren't really any major bugs. Like that was kind of my
big fear going into it. I think that's like every
developer's worst nightmare is that there's gonna be a
game breaking bug on launch. Thankfully I had gotten all of that stuff under control during kind
of the testing phase. And there were no issues. I was able to take the
whole weekend and just watch everything unfold
which was very very nice. Yeah, I mean like I was
watching playthroughs and stuff. Trying to make sure that people weren't getting stuck anywhere. You know, I was checking, you know, reviews and the community hub on Steam trying to make sure that there
wasn't anything going bad. And there were a couple small things that I ironed out over
the following like week. Yeah, for the most part
it ended up being fine. All throughout the
development process since, you know, it was gaining steam. No pun intended. All the way through, you
know, for the whole year. So I had plenty of content
creators that reached out to me and I had a list of, you
know, email addresses and all the people who were interested and who had talked about
the game or who had, you know, tweeted about the
game and all this stuff. So I had a list of people who I already knew were interested. And actually I think I only emailed people who were already
interested in the game. It ended up, you know, going pretty well. I think pretty much everybody that I was hoping would play it did. All my favorite channels that, you know, that I've been watching forever. And probably the one that I was most excited to see was John Wolf. I've been a fan of his
channel for quite a while. I probably credit him with a lot of my game design knowledge, honestly. His critiques on stuff
'cause it's a very honest levelheaded approach to
like critiquing stuff. And so watching all of his
horror game playthroughs over time, that's been a big help. So seeing him play through
it was really cool. - Yeah, it's really cool. I'm really enjoying it. Oh oh oh, oh oh! I saw some scrap. (laughs) - Some of them, you know,
have played my previous game, so I've kind of experienced it already. You know, No Snake Hotel
was a little game that I did on my YouTube channel
for sort of a challenge. That got played by a bunch of people. Same with My Beautiful Paper Smile. That one got played a bit as well. So it wasn't a foreign feeling, but it was really really cool to see them enjoying another one. Yeah, the reaction from
fans was phenomenal. The people who I wanted to
enjoy it enjoyed it thoroughly. And that has me very very happy. I mean that's where, like
any stress that I had leading up to launch, that's where it was, you know, that's where it stemmed from. Seeing, you know, my subscribers and the people who had been following me throughout the development journey, seeing them really happy with it. That was awesome. Reception from like press. I can't say I was expecting good scores. I wasn't really expecting
press to like it. It's one of those things
where it's like, you know, for a game as like
lighthearted and just like, that doesn't really take itself seriously. Like it's like a half meme
game, half serious game. I just knew that press
wouldn't be into that. So I wasn't too worried about it. - [Danny] Were you personally offended by the four out of 10
from IGN or was it like a- - No, not really. I actually, I avoided
watching it for quite a while. Eventually I did and I
was like oh it's not... It's whatever. I mean he's got some valid points. I mean four out of 10 is a little low. A little lower than I was expecting, but. - [Danny] It seemed like a game that they often wouldn't usually review. - Right, that was the weirdest thing to me is that as many press folks
reviewed it as they did. Like it got tons of press. It's a spider train. Why are you even looking at this? That was the part that
confused me the most. - I was very excited. Very relieved for him because I wanted it to be well received. And I wanted it to, you
know, the IGN rating. That bummed me out. But it didn't seem to bother
him too terribly much. - [Danny] And obviously the sort of wider reception was overwhelmingly positive. - Yeah, it was great to see
the high reviews on Steam and watch playthroughs and
seeing people really like it. So it was good to see that
it appeared that it was gonna be commercially successful and that it would give him some breathing room to be able to, you know,
survive and, you know, be able to work on another
project and keep doing this. This thing that he wants to do. I've been watching him
work for a long time. And you know, he was
working on Choo Choo Charles while he was still living at home with us. And, you know, he was kind
of a workaholic about it. But, you know, I watched
him learn all the different facets of the gaming
industry and so I knew if anybody could do it, he could do it. And I knew he was being very systematic about the game development
and the marketing and fully understanding the big picture of how to be successful. - Basically as soon as
I started making games, I knew that that was what I wanted to do. I was like no matter how long it takes, eventually I'll have a successful game. And whatever time I spent in college, whatever money I spent in
college, it'll become worthless. So even if it's 10, 15
years down the line, eventually this is gonna
happen or I will die. One of the two things. - [Danny] How has the
success sort of affected your day to day now and while
you're looking at the future? - Kind of the main thing is, you know, it it gives me breathing room. You know, it gives me time to, you know, get the porting figured out. You know, I'm able to outsource that. Yeah, ultimately it gives
me time to figure out the next project and do
the next project right. And do it better, you know? It's like the game has done well enough and the YouTube channel
has done well enough that like, you know, like I don't need to make crazy big games or anything. As I've continued
researching game development and just doing game development, I keep on finding like oh wait, I could do something cooler. Like I could improve on this mechanic or that mechanic or all
these different things. I just, I want to see how
far I can push it, you know? I feel like there's a lot
more that I can do and, you know, a lot more
games that I could make. So it's like I still
feel like I need to be, you know, working as hard as I, you know, did on Choo Choo Charles
for the next games. So, you know, I mean there's
probably gonna be some, a little bit more audience carryover since I do have like the
YouTube channel and stuff. But it doesn't, I feel like it doesn't get too much security. Like every game kind of has
to stand on its own two feet. (dramatic music) (train engine running) (Charlie snarling) (keyboard typing)
