Oddworld is one of the great gaming stories. Developer Oddworld Inhabitants’ first title was a huge hit on Sony’s first PlayStation console. But while further titles elicited critical acclaim, the team felt poorly-served by a succession of publishers.
Now Oddworld is back, springboarding off a 2014 remake of the original game. And this month the IPs ambitious new title is dropping onto PlayStation Plus and Epic Games Store.
What started out as a relatively modest remake became a five-year epic development, a journey to re-invent the puzzle platformer, and to bring Abe and company to a whole new audience.
Created by a wholly-distributed development approach, born out of a rising global demand for top talent, with the team working around the clock in order to manage the globally disparate talents at its disposal.
It pushed for next-gen, assisted by Unity’s breakthrough in high-fidelity visuals. And it ended up teetering on a technical moonshot to bring it all home. This is the story of Oddworld: Soulstorm.
NEWER AND TASTIER
Soulstorm’s story dates back to that 2014 remake of the original Abe’s Oddysee, titled Oddworld: New ‘n’ Tasty. Working with UK developer Just Add Water, the title proved a commercial and critical success. And its audience was hungry for more. When asked, the fanbase were keen to replay the classics. And that worked with Oddworld Inhabitants’ budget. So the initial plan was to follow up New ‘n’ Tasty with a similar remake of its sequel: Abe’s Exoddus
But the team wasn’t sure that would be enough, co-founder Lorne Lanning recalls: “The landscape of product was maturing so fast, and systems were maturing, and you had new types of titles exploding onto the scene – free to play was changing where a lot of gamers were spending their time, and on how many IPs, and that was concerning.”
“We didn’t think we were going to cut through in the same way we did on New ‘n’ Tasty,” Lanning says. “So what can we do to turn up the volume… very practically and rationally…” he laughs in hindsight.
“We said ‘we’ve got to push it to that next dimension’,” but the question was how to do that in a way “which stays true to what the fanbase wants, and stays true to the originals.”
“We were trying to analyse why people – who could play Red Dead Redemption or Grand Theft Auto, fully realised open worlds – were still asking us for more classic 2.5D adventure platform games.
Accessibility was a part of that “because you’re you’re aiming in two dimensions not three.” But also that the story is being narrated something closer to a movie, with the game controlling the camera, not the player. How then to add that sense of an epic world while retaining that directed feel?
The answer was Oddworld: Soulstorm’s 2.9D approach. A fully realised 3D world but with the player moving Abe along a now curving path through that environment. A path that allows the camera new freedom, rather than it simply panning across a flat world.
“We could do things with the camera that would give us a cinematic approach, towards what our original objectives were in Oddworld, and take advantage of the 2K plus resolution that was becoming available.
Although Lanning felt there was more to resolution than just getting pretty. “We felt like everyone was going to go high-res on their hero characters. And place those characters close to the screen, get a lot more detail. And we said, why don’t we focus on what that resolution gets us in letting characters get smaller.
“Abe was always a character that was dealing with a big hostile world around him. So it made sense to want him to feel really small. And then we could do more things with the gameplay, because we can pull right back… all of which created a whole host of problems.”
BREWING UP A STORM
The vision was there, it worked with the character, it fulfilled the team’s dream to make something more cinematic, and it should help Abe stand out in a far more competitive market. Now it needed to be built, and so we turn to executive producer Bennie Terry III.
“I think we went for about a year and-a-half on the initial trajectory, building on top of what New and Tasty had as a foundation, but still predominantly a 2.5D platformer, with all of the known rules, with all of the known systems, known behaviour.
“But the moment Lorne had the stroke of genius of going to 2.9D, and we’re moving through the world, all of those conventions change. And so the systems that were bespokely designed to work well in 2.5D, truly didn’t work in 2.9D.
“The challenges in refactoring almost all of our systems was significantly more in scale and scope than we were initially anticipating. Everything that’s core to what Abe is and how he functions, none of it was designed to work on a curved path. In hindsight, you think ‘I wish we would have started over from scratch’.
“It was that monumental of a lift for the engineering and design teams to wrap their heads around what we needed to do and how we needed to build… you just don’t know the magnitude of the cans of worms that you’re opening until you open it.”
AN ODD WAY OF LOOKING AT THINGS
Hand-in-hand with the in-game camera pulling back from Abe to better frame him in an epic world, the cinematic camera was coming in for a sumptuous close-up of our protagonist, his allies and enemies.
Lanning explains that going all-in on top-end cinematic cutscenes was one of the earliest choices that the team made. “And this was actually a lot easier to manifest because it dealt with largely pre-rendered, highest res assets.
“The thing we wanted to do is go back and look at the original drawings of Abe,” Lanning explains. And arguably the new cinematics, some of which you can see here, not only do justice to the original art but blow it away in terms of execution. Abe connects with players better than ever through those big expressive eyes of his.
Lorne recalls his wishlist at the time: “I want the eyeball and the eyeball sockets to be able to get bigger. I want the retina to be able to shrink. I want the pupil to be able to shrink. I want those optical effects that happen with the human eye, but greatly exaggerated.
“Aside from animation, the cinematics team was so small, you just couldn’t believe it. A very tiny handful of people,” Lanning looks to Terry, “We put out, what, 40 plus minutes of cinematics in this game?”
“I believe we’re at 50,” Terry replies deadpan.
“Yeah, sorry… the original budget was like 25!” Lanning smiles sheepishly.
“As an executive producer I should only care about time, money and the quality of the product. And however we balance that, that triangle is critical,” Terry says, telling it from his side.
“But Lorne sold me on going big for the cinematics. We originally budgeted 18 to 20 minutes, we had it all scoped out. But when we started pushing the cinematics further, and when we saw what we could do, some of the logical constructs went away for a minute. And we came back to ask ourselves, what is Oddworld? What do we want the company to be? Is it just focused on making games? Or can it be more?”
After all, Oddworld has long had these ambitions, With Lanning focused on how games and Hollywood might merge when the company made Abe for the PS1.
“And if the goal is to evolve the company beyond just making games… What does it take to move that needle outside of games into other realms of media? Because ultimately, Lorne’s a great storyteller. And Oddworld has unique visuals in style and direction. So it cuts through in many ways on that front.
“But what we didn’t have is something that was near film quality, something that was in the realm of what you would sit down in a movie theatre to watch,” says Terry. Well, it certainly does now.
The shift to 2.9D also created uncertainty about the game’s narrative, Lanning explains, uncertainty that could be mitigated by the cinematics.
“How much of the story am I going to be able to tell, in the way I was originally thinking of telling it?” Lanning mused at the time. “Because engineering time is now going to all these things that we didn’t expect… So how then do you still strategically align to a good outcome?” And additional cinematics looked to be the answer, a way of ensuring the team could hold together Lanning’s narrative vision, come what may.
“It was a finite way to measure what something would take. Building the interactive part it’s very hard to measure what it will take unless you’ve done it exactly before, there are so many zillions of ways everything can get slightly askew.
“If you get something wrong, and you’re out of sync, and you’re needing to have it work in real time, you’re going to be screwed,” Lanning adds that it was a way of “hedging our bets… although we ultimately got more [in the gameplay] than we expected.”
Behind the game’s sumptuous visuals – both interactive and cinematic – is the Unity engine. Soulstorm was amongst the early high-profile titles being made with the then fledgling High Definition Render Pipeline (HDRP). With the game and Unity’s new graphics tech being developed side-by-side.
“The technology didn’t exist when we started the project. The rendering quality of Unity and the tools available, from post processing to lighting, all evolved over time,” Terry tells us.
In fact, the change of direction, and the delays that came with it, became something of a boon. It better matched up the project and Unity’s toolset, both in terms of ambition and availability. Plus it was a huge time-saver for the team, says Terry.
“HDRP and other components of Unity’s renderer became a valuable resource. One of the biggest ones I would say is Unity’s tone mapper. It sounds so simple, it’s been in compositing tools forever. But we didn’t have to use a compositor because the tone mapping tools were directly built into the engine. So we could do all the tone mapping, the colour correction, the colour grading, all in-engine. That accelerated our iteration time, versus going out to After Effects, and then bringing it back and forth. It was all in engine, with all the editing done there.
“We also rendered everything out at 4K but when you look at how much data it was pushing in the movies, we got a certain freedom of licence where it didn’t have to compute at 60 frames per second. It could compute a frame a second and you can save out, so we can turn up the resolution dial, we had characters having 8K texture maps.“
And better still, creating and rendering all the cinematics within Unity meant they were a perfect match for interactive segments, allowing for easy transitions, Lanning explains: “When Bennie was cutting together earlier trailers, he was going ‘I’m kind of amazed at how close the game footage looks to the movie footage’. And that was a huge advantage of bringing the two together.”
While New ‘n’ Tasty was created largely by a single development team, Soulstorm’s creation was hugely distributed, with Terry giving us an overview:
“Six time zones and four continents is currently how things are structured. And so we chase the sun here, which is why I was up at five this morning, because I had to connect with Perth, Australia. But everything is distributed development. There’s only Lorne and I in the studio here and Sherry [McKenna, co-founder], at HQ in Emeryville.”
Major outposts include Frima Studios and Sabotage Studio, both in Quebec, Fat Kraken Studios, set up by New ‘n’ Tasty veterans in Leeds and Titanium Studios in Perth. Alongside other freelancers and independents.
This wasn’t the initial plan however, with Oddworld Inhabitants originally looking for a single developer to take on the majority of the work, preferably one fairly close to home.
Engineering started in North America, Lanning tells us, but with the project changing direction, and with the engineering team having other options for work, the Oddworld team “wound up having to move engineering to Australia.” The very opposite of its initial aim to move the work closer to make communication easier.
Lanning explains that it just became hard to find the right people in North America. “The markets were changing and all of a sudden the demand for live art pipelines grew. The success of titles like Fortnite meant they were able to offer good rates for easier work.
“They’ll just lock you up, they have all kinds of talent on a retainer. Just as long as they produce ‘X’ number of drawings or paintings or models a month, they’re doing great. And that made contracting larger companies more difficult, because how do you compete with guys who make hundreds of millions of dollars a month?”
Between the change of scope and the difficulty finding a co-development partner, Lanning feels that at that point a big publisher would have said: “Let’s just kill the project. We can’t really see the end, we can’t tell exactly how much it’s going to cost.
“But as a small developer, you’re like a family that has decided to build their own house. Shit is not necessarily working out how you thought, but you’re in deep, and you have to complete it, you don’t have a choice. Being small can make you persevere and deal with your problems rather than just go ‘fuck it, it’s too hard’.
“And there’s a certain stamina and perseverance and kind of insanity that goes with all that at the same same time, but it’s not an easy road,” Lanning sighs.
“It’s definitely felt like a battle every day,” agrees Terry. “My job is to make sure we have the talent we need and with the vision and the complexity of the project, it wasn’t just something that I could hand over to new developers or even slightly seasoned developers. It was something that required world-class talent to pull off.
“But when your competitors are Facebook, Amazon, Apple, and they’re hiring the engineering talent you need… all of these entities gamifying their experiences, even though they’re not really part of the industry. It became challenging just to hire, because everyone in the top one per cent was gone.
With the best talent in North America becoming increasingly hard to access, The Oddworld team went distributed and built an international team. And that changed radically how the project was managed.
“This is my first 100 per cent distributed development that I’ve ever worked in,” says Terry. “I’ve managed outsourcing here or there, but to have everyone distributed changed my perspective. The eight hour day isn’t necessarily as important for me, but I need to make sure that teams are moving in parallel.
“As one team is slowly going into their night, the other team is picking up where they left off, just keeping that chain unbroken. And that’s given us some great benefits, being able to kind of pass the baton, but also some management and logistical challenges of just keeping the momentum going.”
“So even if there was a simple day for us, it was still a 13 hour day,” Lanning clarifies. “And we at HQ just had to suck that up, or fail. I mean, I don’t want things to be that way. None of us would say, ‘you know what, let’s just do this project that’s going to age us quickly!’”
While the discussion of crunch in the industry is as heated as ever, its increasingly distributed nature makes the issue more complex still. After all, promising that you’ll send your staff home after a full day’s work is one thing, but when they’re freelance contractors, working in their own homes, that’s tricky.
Lanning notes that sometimes he’ll be working and notice that one of the team is also working, but in a very different time zone. “I’m like ‘Tom [Bramall, freelance artist] why are you awake? It’s like 12 o’clock UK time’.” Bramall explained that he saw a mis-positioned eyelid in one of the frames and he was just correcting it. “I’m like, ‘you really don’t need to correct that, we have 50 minutes of cinematics, you just don’t need to do it’.
“But the same was happening across the team,” Lanning admits. To which he tells them: “What we need is to reach satisfactory. Good would be nice. Great would be phenomenal, but satisfactory is what we need right now.
“If you’re in distributed development, it’s a little bit like carrying water in a wicker basket. Because you don’t have all the minds in the same place, looking over each other’s shoulder, having that conversation at lunch, a director that can hit 20 people in an hour, giving feedback and reviews. That’s gone.
“You’re not policing them. And there’s a few reasons for that, even if you wanted to, when you’re dealing with contractors, and you’re dealing with all these different countries, you have to abide by the labour practice laws, contract laws and this and that. We’re not allowed to measure people’s hours. And we don’t,” he says emphatically.
“To my knowledge, we’ve never told any of the staff that they have to work overtime, that they have to crunch. What I find is that because of the distributed time zones and the way things are set up, team members will be on early to sync with me,” Terry adds.
“There’s no doubt that to produce this project, or any game, there are times when things go further into the nine hour day, instead of the eight hour day. But for engineering, for design, for art, cinematics, we haven’t asked them to do it.
“On my end, just just from HQ, and as a EP and Lorne creatively, we spend the weekends together… we spend evenings together… As a leader of the company, I need to make sure that I’m crossing my T’s and dotting my I’s. And I’m in lockstep with all of the creative directives, so that we can build this product and hit the vision.
“I’ve never seen it as crunch for me, because this is what I enjoy doing, this is my passion. You don’t necessarily consider crunch when it’s personal and that’s how I feel when I work with Oddworld.”
A PHILOSOPHICAL ODDITY
There’s something ironic about discussing crunch with the Soulstorm team. After all, Abe’s adventures have long been set within the satanic mills of Rupture Farms, where he and his fellow Mudokons, a slave race, are at best being worked to death, and at worst are themselves the meat for the grinders.
“When Abe released, I remember sort of being attacked as anti-capitalist,” Lanning recalls. “And they don’t say that anymore. Because in a way, it’s almost like everyone’s a little anti-capitalist now.”
“I don’t want to be political. I want to be philosophical. Was Orwell political? Was HG Wells political? Or were we focusing on conditions that the human experience is trying to navigate? And how do we shine a light on that?”
Lanning talks passionately about researching his topics. Whether its politics, economics or industrial design, he loves to look deeply into his subject matter before creating.
“I saw clearly in the 90s that we were being screwed collectively. And people were still like, ‘I love Nikes’. They didn’t know it was made by child labour. And that research really opened my eyes and I wanted to create a property that was a reflection of this dark side of globalisation.
Lanning didn’t want to make a flash in the pan protest though, but instead something that would take on its own life inside people’s heads: “I felt like in the creation of an IP, you had to get into content that would reflect an insight that people would find intriguing as it related to themselves, and have the depth so they could see the creator had much more in mind than you were seeing. Then you have a chance of having something that survives decades, rather than just being timely.
“I didn’t want to say I’m making the game about how McDonald’s is chopping down the rain forest for the sake of 99 cent Happy Meals, that’s not going to fly… I wanted to have the workers, in their everyman’s plight, first have to fight through their ignorance, and then they have the opportunity to fight for something else. But if they don’t beat their own ignorance first, game over, you’re just going to be a cog in someone else’s wheel.” Lanning may not be anti-capitalist, but there’s something distinctly Marxist about that.
“We have a lot of challenges as people and as a planet. And that’s sort of the territory where I take off on Oddworld, I want to always make fun of that. And I think that attracts some of the people that work on the product, they feel a connection to it, because there’s something resonating with them. That it’s not just a cool IP.”
THE GPU MOONSHOT
Oddworld is undoubtedly a cool IP, though. Cool enough to have cut a deal with Sony for PlayStation Plus inclusion, where it will also appear in a fully-fledged PS5 guise. To that end, the team really pushed the envelope and almost tore it in half.
“We were approaching this new console with the PS5, and really all-in on trying to figure out how to make that sing,” Terry recalls. “But in doing so, we were dealing with millions more polygons than we had originally anticipated, because the worlds were so massive.
“To achieve Lorne’s vision, our art team had to assemble a high level of details and quality. We love the way our levels look, but that fidelity came with a significant challenge. There was just too much content. We had too many objects for the CPU to process, too many triangles for the GPU to process, and we used too much memory.
“Standard optimization techniques would improve one area and make others worse. No matter what combination of optimizations we tried, nothing got us close to our performance targets or fitting into memory on last-generation platforms.
“The answer was to cull away what the player never sees. Real-time culling techniques have been around for a very long time. If something is off-camera or behind another object, then don’t draw it. In our case, when these algorithms tried to deal with our massive number of objects, they just made our CPU times worse.
“We went to first principles – what are we rendering that is useful and what is just overhead? Consider this: if you walk along a mountain trail, can you see every face of the mountain, every leaf on every tree, every surface of bark? No. The same applies to moving through Soulstorm. All we had to do was figure out in advance what the player can see and discard the rest. This would reduce our objects, triangles, and memory in one shot.
“The amount of computation was massive. With no time to spare and unsure of whether we’d be successful, we built a set of tools and a sophisticated baking system in three short weeks.
“When we tried running the system on a top-of-the-line [Intel Core] i9, it was going to take over a month per level iteration to run the computations.
And so the team found itself in an extreme version of the same plight as many consumers did last year, trying to hunt down the right card for the job.
“For our full-level culling bakes, we needed the most powerful commercially available GPU with a minimum of 24GB of onboard memory to compute the bakes.” Of course with this being a GPU in 2020, “We could not find one available!”
“The few places taking orders had lead times of more than three months. Even in the face of all those potential project killing challenges, the team had come too far to give up.
“Jason Lee-Steere and his team at Titanium Studios again came through for Soulstorm. They found the only available Nvidia RTX 3090 on the Australian continent. With the RTX 3090 integrated into our automated culling pipeline, and with the computing power of 35 TFLOPS, our times went down from over a month to two to four hours per level, per iteration.
“Soulstorm would simply not be shipping today without extraordinary efforts being executed by our engineering team in Perth,” Terry says in thanks.
AN EPIC AUDIENCE
With the vision delivered and the biggest technical hurdle overcome, the team can finally start thinking about this month’s launch. And what a launch it will be, with millions of certain downloads from PS+ alongside a big push on PC as an Epic Games Store exclusive.
With both of those deals signed, we put it to Lanning that Oddworld Inhabitants might be hedging its bets somewhat on the release, and that the title is already making a good return before it’s even launched.
“I wish that we were in a position that we made something so great, and we made it so cheap, that those deals were fat on the back end,” Lanning replies openly.
“We’ve had long relationships with both Epic and Sony, I think there’s people internally that like the IP, they really liked what was happening. And they knew we could use help.
“And as much as I’d like to say ‘It’s a killer deal! I just can’t tell you the terms.’ Really, they helped us actually get to the finish line. So there’s a lot to recoup on.” And while he seems relaxed about the financials, there’s another interesting upside to the PS+ deal that he’s more keen to discuss.
“We’ve always felt that Oddworld was a tremendously undervalued IP.” And Lanning himself has long spoken out on how he feels previous publishing partners failed the brand. “In the business world. success is measured by the money the product made, but we didn’t give up on Oddworld because we thought we got shortchanged in the marketplace, this shit happens, it’s happened to tons of products. But we held on and said we believe it’s something more.”
And the huge exposure of PS+ gives the IP an opportunity to build a huge audience for Oddworld. The kind of audience that Soulstorm was so painstakingly created for, rather than simply being another 2D title that might sail under the radar. That hugely increased audience does come with some challenges, though.
“One of our problems in legacy was we have to overcome the difficulty barrier, we have to overcome the comprehension barrier. We were notoriously hard, we don’t want to not be hard, but we do want to make the gateway a lot more pleasant,” Lanning explains. “And so we spent a lot of time in this game, making it more accessible,” both in terms of making a story that works for novices to the world, and gameplay that communicates what’s going on to everyone.
“We’re bringing the property to a fan base that may not have zero clue about who we are, just that it’s free online, they’re going to download it and play it,” adds Terry. “You know there’s things you can’t control and there’s things you can, and where we’re focused internally is just making sure we’re ready to react and respond to anything that comes out of that.
“So whether that’s immediate feedback from the community saying they’d love to see this, and these are things that we didn’t consider, we’re on the ball to make sure those things happen, as well as communicating from PR to social and all of those other departments that feed into that.”
THE NEXT ODDYSEE
And overcoming that challenge of a new audience is what will secure Oddworld’s future, Terry tells us: “This is what we wanted. It’s looking at how the company evolves. What do we need to do to remain viable and grow the brand, so that we can do future projects and keep this going? It comes back to: we can’t just launch to our core fan base and say ‘here’s another iteration’, we’ve got to dip our toe in a bigger pond.”
Terry and Lanning both have the look of people who are delighted to find themselves at the end of a very long tunnel, just about to step out into the sunlight, a little dazzled but delighted to find themselves at the beginning of the end. As Terry puts it: “This is our moonshot, if you will. And all of the challenges to getting there have been monumental… we’re at the precipice of shipping and it has been truly a journey. And I hope we never have to work that hard again! But we pulled it together when we needed to, and the team was behind us every step of the way, stuck with us to bring the project home.”
Arguably the road to putting Abe on a curving path ended up being a more tortuous one than its developers ever foresaw, but isn’t that so often the way in game development. “That’s the dance, the challenge of game development, where you’re pivoting all the time, as quickly and as intelligently and as subjectively as you can, and trying to keep the vision intact. And that is one royally difficult dance,” Lanning observes.
And so Oddworld keeps on dancing, when so many other IPs have tripped and fallen. And we sincerely hope that with Soulstorm it finds an audience that allows it to explore all its ambitions, in games and beyond, and that the team can, for a while, put up their feet.