Two years of cancer and crafting: The making of Crashlands

Butterscotch Shenanigan's Seth Coster on hard years

Two years of cancer and crafting: The making of Crashlands

Despite numerous delays pushing its release from summer 2015 to January 2016, there was still a huge amount of anticipation surrounding Butterscotch Shenanigan's craft-heavy adventure, Crashlands.

Its blend of action-RPG, crafting and survival elements, unique art style, and bizarre sense of humour helped keep fans interested, even as the release was pushed further and further back.

To better understand the reason for the delays, we had a chat with Butterscotch Shenanigan's programmer, Seth Coster, about the game's tough development.

PG.biz: Where did the concept for Crashlands originate?

Seth Coster: In September of 2013, a few months after launching Quadropus Rampage, we were trying to come up with our next game idea.

Normally we had infinite ideas and the difficulty was picking one, but suddenly we were coming up empty.

Sam was getting really tired during design discussions, and he became very irritable and negative.

He couldn’t seem to get out of bed before 10am, and by the time 2pm rolled around, he was so frazzled that he would just go home for the day.

I was worried that he was depressed, or that he didn’t want to make games anymore but didn’t know how to break it to me. In any case, it was a massive change in behaviour for him.

Butterscotch Shenanigans, from left to right: Sam, Seth, Adam

To try to kick things back into overdrive, we did a thing called the Butterscotch Jam, during which we made a game every day for 5 days, looking for something that had enough of a hook to it that we could flesh it out and release it as our next game.

The project that came out of this was a sort of endless platformer called Extreme Slothcycling. It was pretty goofy and had the potential to be fun, but it was a very shallow project in terms of ambition and story.

We worked on it for a few weeks, but Sam was really struggling, even more so than before. He started going to different doctors, trying to figure out what was wrong.

Eventually, he was diagnosed with Stage 4b Lymphoma (the last stage), which is cancer of the lymph system, at the end of October 2013.

Sam took one look at Extreme Slothcycling and just flatly said, I don’t want this to be the last game I make before I die.

They took a tumor the size of a kidney out from his left chest wall and started him on powerful chemotherapy within a week.

Had he waited another week or so to start treatment, he probably would have been dead on arrival.

After the first chemotherapy treatment Sam was feeling almost fully back to his normal self. Even with the intense discomfort that comes with chemotherapy, it had already eradicated so much of the cancer that it was a huge net gain for his overall well-being and mood.

After a month of tests and treatments, we sat down to work on Extreme Slothcycling. Sam took one look at it and just flatly said, “I don’t want this to be the last game I make before I die.”

Turns out he had started making a prototype in Game Maker of what would eventually become Crashlands.

At the time, it was a single screen that had some leaves on it, and you controlled a weird flat robot thing that hoovered up the leaves like a Roomba.

It was about as rough as a prototype gets, but he explained the vision he had for the game.

It was to be a huge, open world, full of stories and characters, tameable creatures, crafting, combat, building, and anything else you could think of.

The earliest surviving screenshot of Crashlands, December 2013

The idea was insanely huge and well beyond what we were capable of doing at the time.

But I figured, if he had to go through all these cancer treatments for who-knows-how-long, I should do my part to give him something big to keep working on and keep his mind off all the pain.

So I agreed to do it, and that’s where Crashlands started.

How long did the game take to the develop, and was it really just the three of you working on it?

We started development in early December 2013. Sam was in and out of the hospital a lot, going through stem cell transplants and chemotherapy, and in 2014 we also spent a couple months developing BscotchID and integrating it into our older games.

Once you subtract those delays, Crashlands took something like 18-19 months to complete.

Sam did all the art, I programmed the game, and Adam developed web tools and utilities to speed up the dev process.

Adam also built a tool called the Crashlands Creator that allowed him and Sam to create the game’s story and quests, while I coded the game side of that system.

The only part of development we didn’t do was audio, which we contracted to a local studio called Fat Bard, who specializes in video game sound effects and music.

First build of 'build mode', February 2014

Since we had no money to offer, they agreed to do the entire soundtrack and set of hundreds of sound effects for a small share of the revenue, which was pretty dang generous of them.

What was the biggest obstacle you encountered and how did you overcome it?

Besides Sam almost dying multiple times throughout development (he was declared in remission in March 2014, but the cancer resurfaced in February 2015), money was tight the whole time.

Our studio philosophy had always been to try to make mid-size games really fast, but Crashlands flew in the face of that idea.

Adam and I took no paychecks for the first year of the project, since our wives could support us, while Sam was paid a very meager monthly salary.

We dreamed that we could make Crashlands big enough for a dual release on desktop and mobile.

Despite that, we still ran out of cash and had to launch Flop Rocket on iOS in February 2015, which allowed all three of us to get paid for a little while, albeit poorly.

Adam and I had to take pay cuts again after 6 months. And even with those pay cuts, as of Crashlands launch time we had only 2 months of cash left in the bank.

Was the game designed as mobile-first before the decision was made to launch on Steam, or was it always intended for multi-platform release?

We dreamed and hoped that we could make Crashlands “big enough” for a dual release on desktop and mobile, but we weren’t sure if we had the capability to pull it off at the beginning. Still, we always had our sights set on that.

We were acutely aware of how difficult it is to make a fun, enjoyable game when you only have a smooth pane of glass as your control scheme, but we figured that if we could make something that played well on mobile, it would also play well on PC.

Early version of the world map, April 2014

As development progressed, we realized that the biggest point of frustration in crafting games is the bulky UI and overwhelming game systems.

Shuffling stuff around in bags, building boxes and chests to store all your stuff, losing track of your materials, getting overencumbered, memorizing recipes, and so on.

So about halfway through development we just looked at all that stuff, and decided not to do any of it.

That streamlined experience made both the mobile and PC versions of the game better, so we definitely think it was the right call.

We knew that Crashlands' presence on mobile would cause it to lose value for some of the PC crowd.

But on PC we also got to add some nice bonus stuff, like hotkeys for combat, that could take advantage of the extra control options the typical PC player has at their disposal.

Were there any specific problems with developing the game simultaneously for both mobile and PC?

The biggest hurdle from the mobile side were the hardware restrictions of mobile devices. On a PC you can afford to be a bit loose with your optimization, because the machine running the game tends to have extra processing and RAM lying around that you can lazily grab.

Mobile is not so forgiving - most devices have extremely limited RAM and a weak GPU to boot, which means doing graphically intense calculations (like, say, a mini-map) basically implodes the device.

But someone sent us a picture of Crashlands working on a $10 phone they got at Walmart the other day, so we’re pretty happy with where it ended up.

The most difficult aspect of developing for PC was less on the technical or game design side and more on the marketing and positioning side.

We knew that Crashlands’ presence on mobile would cause it to lose value for some of the PC crowd, who would view it as tainted by a lowest-common-denominator.

We worked really hard in the messaging to angle the game, and future games by our studio, as platform agnostic-games that exist outside the territories of either PC or mobile.

This worked pretty well with our trailer, but there’s still a lot of doubt that a game can be truly native to both platforms simultaneously and not lose anything.

We’re hoping to push against that idea and earn the trust of the PC gaming crowd as time goes on.

What do you think are the big differences between the game's mobile and PC players?

Honestly, I don’t think there is one! Here’s how I see it.

When the iPhone first blasted onto the scene and made everyone want a smartphone, a lot of developers realized the power of that platform as a gaming device.

The problem was, the original iPhone had 128mb RAM and a 412mhz processor, which meant all the early games had to be really small and simple. “Time-wasters,” in other words.

The average playtime for people who only own the game on iOS is over 7 hours.

Within the first couple generations of smartphones, we started hooking analytics into all these tiny time-waster games. And--surprise!--people were only playing them in short bursts.

Developers and marketers around the world looked at these analytics and said in unison, “MY GOD! People only play mobile games for 2.5 minutes at a time!”

Developers just took it at face value that people are only going to play mobile games for 2.5 minutes at a time, 2-3 times per day, and they designed their games with that in mind.

But in retrospect, there’s one really obvious fact here: if you design a game to be played for only 2.5 minutes at a time, then people will only play for 2.5 minutes at a time.

The game had a blockier art style initially, but this was changed mid-2014

With Crashlands, we started with the intention of making a game that people could get lost in for hours, and we didn’t assume that people would only want to play it for 2.5 minutes. We wanted it to be fun whether you played it for 2.5 minutes or 2.5 hours.

And it definitely worked. At the time I’m writing this, Crashlands has only been out for 10 days, and the average playtime for people who only own the game on iOS is over 7 hours.

That’s 43 minutes per day on average. This is approaching the average daily playtime that people put into console gaming, which is around 50 minutes.

So I don’t think there’s a difference between PC gamers and mobile gamers. I think there’s a difference in the assumptions developers tend to make about those groups.

Can you provide any data on downloads etc?

I won’t give specifics on our download numbers, but it was interesting to see the relative share of the pie that each platform claimed, since it’s rare for a game to launch on Steam, iTunes, and Google Play simultaneously and get some sort of feature treatment on all three.

Crashlands managed to get the iPad prime feature spot, a decent Google Play spot under “New & Updated” (though no big banner), and a Steam pop-up ad on launch.

This gave us a tremendous amount of eyeballs and attention.

Steam started off at a blast and was in the lead from a revenue standpoint for the first 3 days, at which point the pop-up ad had run its course.

By Tuesday afternoon, iTunes had overtaken it and maintained a steady lead, likely due to the tremendous visibility of the App Store’s feature spots.

Google Play sits in third place at about half our Steam revenue, but still commands a viable enough share to easily make it worthwhile.

A lot of games tend to leave the writing until quite late in development, but Crashlands seems quite heavily built on the humour in the writing. At what point during development did you start to focus on this aspect?

The ridiculous tooltip descriptions were in production the whole way through - I’d usually crank those into existence after a heavy day of coding and possibly some whiskey, so they tended to fly off the rails quickly.

We had a general sense of the story for most of development, but we didn’t get into details until the Crashlands Creator was finished, which was in late August of 2015.

Sam worked on the story while getting his Stem Cell transplant and plowed through the majority of it, establishing the main beats of the narrative. Adam then circled back and edited and elaborated where needed.

One of us said “What if... you never had to manage your inventory?” And everything just exploded from there.

Sam then wrote the side quests, and again Adam circled back to edit and spice up. The whole thing, a solid 50,000+ words of dialogue, was written and encoded over the last 3 months of the project, and then edited and elaborated based on feedback from the Beta.

I was just adding and tweaking game mechanics, working on balance, and continuously coding and debugging while Adam and Sam wrote the story, and I just left them to their own devices.

My first playthrough before beta was a first-time experience of the story Sam and Adam built. It was really cool to see it all for the first time!

When did you know you had a game that you were really happy with?

The biggest moment was shortly after Adam joined the team. We had never made a crafting game, and as a result, the early versions of Crashlands did a lot of mimicking other crafting games.

When Adam joined, we got a fresh pair of eyes on the project, and as he was doing his first playthrough, he just made an offhanded comment about being sick and tired of digging through his inventory trying to find stuff.

And one of us just said, “What if... you never had to manage your inventory?” And everything just exploded from there.

I think once we started down that track, it started to really feel like we were doing something new.

You were quite open with your development process, producing dev blogs and podcasts about the process. How helpful was that to the development of the game?

It probably didn’t affect development on a nuts-and-bolts level, but it did help us keep our sanity.

Working on the same thing in isolation for almost 2 years is a really weird thing to do, and it can mess with your mind.

Doing things like dev blogs, giving talks, and doing a weekly podcast allowed us to break that isolation. It let us remind everyone that we still existed, and it allowed our players and fans to keep cheering us on as we neared the finish line.

How was it running a beta test?

We knew the beta was going to be difficult for our testers, because we were short on time and Crashlands is pretty huge. So we had to make sure we were getting people who were “all in”, so to speak.

We whittled our tester pool down from 1,100 people to 180, and the beta ran for 4 weeks.

To manage feedback, Adam created a web tool that is similar in function to Reddit. Players could submit a bug report, and other players could comment on that bug, attach screenshots, or just “upvote” it to indicate that they also saw the same thing.

The first boss, the 'Baconweed Fairy', added in February 2015

This allowed us to focus on the most-prevalent bugs by just working our way down the list from the top.

And any time someone submitted a bug report, our system would create a duplicate of their save file, which we could then pull down into our own games to more rapidly diagnose any bugs they were reporting.

Over that time, those 180 people put in well over 4,000 hours of gameplay and produced over 2,000 points of feedback, which culminated in us creating about 25 pages worth of patch notes. It was a very intense four weeks.

Have you been surprised by any player feedback?

So far the only feedback that has been surprising to us has been the adamant call for controller support, on both PC and mobile, or the ability to move your character with WASD on the PC.

We designed the game to be played best a certain way, and in our minds, using these other control schemes would just lead to a more cumbersome and awkward gameplay experience.

And then comes the big one, which will be the Crashlands Creator patch.

We could be wrong, but it’d sure be a lot of work to redesign the game to find out.

That’s not to say we’ll never do it; it’s just that we’d rather put our time toward stuff that’s going to benefit all of our players, not just those who play with controllers.

Do you have anything planned for future updates?

Yep! We just released a bugfix patch that’s addressing some of the odd things that have cropped up since launch.

Next we’ll be taking a couple day breather, and then we’ll be working up what we are calling the “Quality of Life” patch. It’s basically going to try to smooth out some of the remaining rough edges of the player experience, and clean up any remaining bugs that might have escaped.

And then comes the big one, which will be the Crashlands Creator patch.

It’s a patch that will open up player access to the Creator, so people can make their own campaigns, complete with mutators and the ability to create stories, characters, and hand-designed locations in the world.

We’ll be trying to add all kinds of fun things that players can put into their campaigns, although I’m not totally sure what we are capable of here, so I can’t divulge too much of what we are planning.

From there, we’ll start working on whatever’s next, though we hope to periodically revisit Crashlands and spice it up with new items and whatnot.

Can you tell us anything about any upcoming projects?

Honestly, your guess is as good as mine!

We’ll definitely keep pushing the multi-platform concept, and we’re hoping to have our next game incorporate some kind of multiplayer element, even if it’s a minor one.

But beyond that, no idea!

You can buy Crashlands for $4.99 from the Apple App Store and Google Play.


Ric is the Editor of PocketGamer.biz, having started out as a Staff Writer on the site back in 2015. He received an honourable mention in both the MCV and Develop 30 Under 30 lists in 2016 and refuses to let anyone forget about it.