Objects In Space Was Refinanced Four Times Before It Was Finished

Game development is a brutal, long process. That was especially true for Objects In Space, which quickly became one of the darlings of the Australian indie circuit after people saw the intricate physical spaceship that the developers made. But what most people didn't see was the raw cost: the game's development blew out from two to five years, the creators had to refinance four separate times to get over the line, and the journey had a grave impact on their mental well-being.

Leigh Harris, the co-founder and one half of Flat Earth Games alongside his sister Elissa, gave an honest breakdown of Objects' development from a game that started out as a submarine simulator in space, to one that required multiple dialogue revisions, a complete rework from 2D to 3D assets, the creation of new systems that ultimately required existing mechanics be redesigned to cope with the changes, and all the problems along the way.

The initial scope for Objects was ambitious, but it was the kind of ambition that could have been delivered in a few years. They grew up on a lot of oldschool PC games, including the Microprose simulators of eras past, and they loved open worlds. But learning from the amount of outsourcing they needed for Towncraft, the creators figured that not rendering a live view of space would reduce the amount of art necessary.

Image: Objects In Space

That ended up being the case, but the studio ended up spending more time in development because of the flow-on effects — information had to be communicated to the player in other ways, and the decisions about how that panned out ended up becoming very complicated.

Harris outlined the complication in the most classic of ways: a flowchart. Things began fairly simply with the spaceship itself. It was designed like a sub, and so the team needed the basic systems that go with that: sensors, engineering, power management, and so on.

Because the game was open-world, Flat Earth also had to make quests to give the player a reason to travel from one star system to another. They designed around 120 of those, and about 60 space stations to go with it. The space stations all had to be unique, because that was the main visual connection the player had with the universe outside of their spaceship, and they needed to be able to visually identify the different "sectors" of space — the pirate system, a more corporate system, working-class mining areas.

But it's here that the problems started to emerge.

The space stations, which were spread across 30 star systems, needed to be populated with multiple characters. They had to feel alive, especially since the player character is never really visible in Objects. The creators also wanted Objects to feel more natural, and so they added branching dialogue trees and no player HUD. "We wanted to feel like you weren't playing too much of a video game," Harris explained. "We want NPCs not to be vending machines where you put in a click and they give you information about the game world."

In retrospect, it was possibly one of the worst decisions they made. The branching dialogue trees made sense for the real-time, always progressing nature of the Objects world — everything is always moving forward, so it made sense that the conversation system did too. But it meant that the in-game world, and all of the NPCs that lived within it, had to accommodate that.

And then the problems got worse.

A decision was made early on that the player in Objects would interact with everything through in-game monitors. If you were flying a ship, you were getting the information from a screen that existed in the in-game universe exactly where you'd find it on that ship. It was rustic, authentic, and did a banger of a job echoing that classic spacefarer experience from the late '80s and '90s.

But naturally, that made life hell on earth for the Harris developers later on.

They needed to switch the game engine from 2D to 3D, because the way information was disseminated through the in-game monitors meant those textures needed to exist in a 3D space. And that meant the character design went from 2D to 3D as well, and the space stations needed a total rework along with it.

If those were the only problems, things might not have been so hard. But the natural tension of a sub simulator meant that Flat Earth Games had to create a whole separate system — a safe tutorial space, if you will — where the player could learn all the systems, die without too many consequences, and get to grips with the multiple progression systems without worrying about the constant passage of time.

Player freedom posed more problems. Early player feedback on Objects was that people wanted the ability to say something during conversations with NPCs, and that required more revisions to dialogue scripts so the player had options when they wanted to speak — and also so the game factored in the choices made.

It caused massive complications with the ships, too. The ships themselves in Objects don't have classes or their own stats: they're simply the sum of the parts you choose to put together. That means you're not restricted from building a hybrid fighter/cargo deal, or becoming a merchant and completely swapping all the parts out for something more aggressive, and so on.

But that also means it's technically possible for a player to put together a ship that just ... doesn't work. Logically, Harris explained, no-one in their right mind would do that.

But what if they did?

The developers quickly found out that they couldn't institute a simple flag that prevented players from leaving a space station with a ship that couldn't fly, because that caused too many problems if a player ship got crippled during the middle of a journey.

So they had to build three more systems to account for this. The first was an SOS system, allowing players who couldn't make it back to base to get hauled back in. You can't call the space NRMA or RACV without paying a fee, but what if the player didn't have enough money to get home?

So the second system was made: debt. And if the principle of debt exists in your space universe, then you also need to create some kind of mechanism where that debt gets tracked and collected.

Which is where the bounty system came in.

A shot of the ever expanding Objects in Space flowchart showing the interdepencies between the various systems, and just how intricate and problematic the feature creep became. Image: Alex Walker (Kotaku)

Having the bounty system meant that the quests, scripts and all other kinds of interactions had to be rewritten, because now the game had to factor in this mechanism for punishing the player as a consequence. Transporting people is a large part of the merchant and mission system in the game, but what if you decided to ignore your passenger instead of just ferrying them to their destination? The bounty hunters need to be called in, and if the player needs to do that, then the NPCs you talk to in the in-game world — and any quests they dole out — have to factor it in too.

Even little things became harder work than they first thought. When the game went into early access, players asked the Objects team if they could have filters for buying and selling items at space stations. OK, but small problem: the amount of space the team had to work with was supremely limited and didn't leave a lot of room for UI. A ton of UI work already had to be done for the in-game PDA tracker, which was necessary to help the player keep track of everything, especially all the quests and procedurally generated events which would expire after a certain time.

It's no surprise that the game's development blew out to five years and, consequently, the amount of money needed. The studio initially got by with their own funds and some grant money from the NSW Government — back when Screen NSW was still supporting local game development — but as the game's scope continued to blow out, so did its budget.

Fortunately, the amount of excitement generated around the game's showing at various conventions — a lot generated by the interest in the spaceship-esque controller — meant investors and publishers wanted a piece of Objects in Space, which made the process a little easier. "We found a totally unique selling point and we leveraged it hard," Harris said.

An Aussie Indie Game Inspired Me To Build My Own Spaceship Cockpit

Objects In Space is an open world stealth space trading game currently being built by Sydney-based indie studio Flat Earth Games. When I say built, I mean that literally — to go along with the game's retrofuture, modempunk aesthetic and heavily tactile and complex interface, the team has mocked up its own real-world spaceship control panel, complete with LEDs and VU meters and missile switches.

Read more

The First 45 Minutes From Objects In Space

After years of doing the rounds at PAX Australia and other trade shows, the Aussie-made space trucker Objects in Space is finally out. But if you're not sure what the game is like, here's what to expect from the opening salvo.

Read more

But to keep going, the studio had to leverage future sales not once, but multiple times. And as development continued and the scope creep grew, the burden became more difficult. "I think as we got into that really dark patch, we got there slowly. And as you've seen from the flowchart, we got there one piece at a time, but going and getting back into full time work, I can hold conversations a lot better, I can retain a lot of information better now ... on every level I can see the difference now."

It also affected the relationship between the co-founders. A key skill in any office is the ability to talk about conflict and find a resolution, and Harris explained that while that was something his fellow co-founder and sister excelled at, there is a limit to the amount of conflict humans can work through. "The tension of getting on each other's nerves, we're still trying to recover from that experience," Harris, who has since taken a job with Wargaming Sydney as a senior game designer, said.

The end result is that Objects in Space is a game the creators are extremely proud of — but it's also necessary that they leave it alone for a while. "We had to take on a lot of money from publishers, investors, government grants to get us there ... most of the money the game has made thus far has been recouped by other people, so that's a real tangible cost," Harris said.

Looking back, Harris said there's no way the original vision could have been realised with the original budget or timeline. Knowing the complications and what he does now, the only way to accomplish that would have been to make a combat sub simulator in space with the size and scope of something like Silent Service, but that wasn't the game the studio set out to make.

Still, after five years of development, their Objects in Space vision was fully realised. For now, Harris is happy enjoying the stability of full-time work, a fortnightly pay cycle with the security of annual leave and superannuation, and the freedom of not having a leveraged sword of Damocles over their head.

"We made something we're really proud of, and I'm proud we built it," Harris said. "We should not have made those mistakes. Still cool, though."


The author's accommodation throughout Melbourne International Games Week, GCAP and PAX Australia 2019 was provided courtesy of Airbnb. Update: Some paragraphs have been amended to better clarify the speaker's original intentions.


Comments

    Video game venture capital, give these guys money once they've recharged their batteries. Game's got massive potential!

    Great article, I wonder how it would have turned out if they had made the decision to embrace a more rogue like approach and let players be more able to fail

    I remember these developers from an old issue of Hyper, a couple of brothers who did really well with Towncraft, and it's great to hear they're still doing what they love, hope it all works out for the team.

Join the discussion!

Trending Stories Right Now