Why some old designs are worth revisiting: A Rocket League story

The server-breaking success of Psyonix’s Rocket League is a long-awaited validation of sorts for studio founder Dave Hagewood, who says his team tried to convince the world that a game best summed up as “soccer, but with rocket-powered cars” would be fun seven years ago.

“It was so difficult to communicate the pitch to publishers and platform holders back then,” he told me over Skype last week. “We’d tell them ‘Well, it’s a game about rockets and cars that can fly and jump and...they play soccer.’ And they looked at us like we were crazy!”

And to be fair, maybe they were. Psyonix eventually self-published that game as Supersonic Acrobatic Rocket-Powered Battle-Cars on PlayStation Network in late 2008 to mild acclaim, and moved on to other projects. But they never forgot about the game; in a sense, it was a Rocket League rough draft.

“It’s like we’ve been taking notes for seven years on this game,” says Hagewood, who agrees that Rocket League is basically the UE3-powered car soccer game Psyonix wanted to release back in 2008. “We didn’t want to change the core of what that original Battle-Cars game was; we just wanted to do that game again, the right way.”

How Rocket League got its start as an Unreal Tournament mod

While it’s certainly not uncommon to see a developer polish and iterate on a compelling game concept, that work typically takes place across clearly-branded sequels and spinoffs released a few years apart, at most. Rarely does a studio wait for the better part of a decade before returning to an old idea.

But the core DNA of Rocket League is even older than you think: it almost predates the founding of Psyonix itself, stretching back roughly 12 years to Hagewood’s days as an Unreal Tournament 2003 modder. Like many developers, part of his efforts to break into the industry involved modding popular games; unlike many modders, he designed something compelling enough that Epic itself came calling.

“I added vehicles to UT 2003,” says Hagewood. “Epic had done the initial support, but I made a game mode out of it. They hired me as a contractor to make that game into something for UT 2004, and that ended up becoming Onslaught.”

Psyonix co-developed the vehicle-rich Onslaught mode in Unreal Tournament 2004 with Epic Games

The objective-based Onslaught mode’s big selling point was vehicle-based combat, and it was a key contributor to the warm reception UT 2004 received from players and critics. Working on it as a contractor at Epic taught Hagewood a lot about how a real game development studio operates, and he used that knowledge (as well as his contacts at Epic) to build up Psyonix.

“We built our business on that engine,” says Hagewood. “If you needed anything in Unreal, we were the company to call. But our ultimate goal was to make our own games.”

And indeed, Psyonix has worked on a smorgasbord of games throughout its history -- putting in contract work on projects like XCOM and Mass Effect 3 as well as developing its own games, like Arc Squadron and Nosgoth. But the simply joy of making those Unreal Tournament buggies bounce and fly stuck with Hagewood long after his contract with Epic was up.

“That was kind of my big break in the industry, and me and Ben Beckwith, who was a level designer at the time, were really having a lot of fun taking one of the cars in the expansion pack and making it like, jump and spin in the air and do tricks and stuff like that,” he recalls. “After we started doing our own thing, we were trying to figure out how to make something really cool out of that car stuff.”

They tried all the obvious options -- adding in racing modes, obstacles courses and straight up car combat arenas. Nothing really stood out, until “Someone said ‘Hey, let’s put a ball in here,’” remembers Hagewood.

“I think he was a fan of Deathball at the time. So we started playing it, and within a week, nobody was doing any work -- they were just playing this game.”

The designers at Psyonix found that the physics-based charm of bouncing and rocketing cars around an arena was best complemented by heaving a massive ball in there with them.

“The rocket booster on the cars was really originally intended to just be like, a turbo option,” says Hagewood. “But we discovered, completely by accident, that you could jump, keep bosting straight up, and literally fly across the arena. It was the kind of emergent gameplay we had to keep around, and a few weeks later people were hitting goals by just rocketing straight across the maps.”

Hagewood remembers how much fun the Psyonix team had zooming around the field and smashing into the ball so fast it would fly like a bullet across the field and into the goal, but he also credits that frenetic design with turning off a lot of people to the original Supersonic Acrobatic Rocket-Powered Battle Cars.

Psyonix's 2008 PSN game Supersonic Acrobatic Rocket-Powered Battle Cars

“The original was too hardcore, so we slowed down the gameplay a little bit [for Rocket League],” Hagewood tells me. “And there were issues with online too, because we couldn't afford a dedicated server network back then. So if players happened to connect to someone with a laggy connection, it was a bad experience for everyone.”

Psyonix learned from the experience, and built the cross-platform (PC and PlayStation 4) multiplayer of Rocket League to rely on its tech wherever possible.

"We did a great deal of Rocket League's online services ourselves...we tried to make a lot of it independent and platform-agnostic on our server side,” says Hagewood. “From there, all we had to do was write interfaces to the various platforms so they all talked well together; by the time the data gets to the game's server network, the network doesn't care what platform anyone is on -- it just notes 'oh this person is on Steam, this person is on Sony," and so on.”

This is intriguing for at least two reasons: first, it means that Psyonix had to figure out how to build a cross-platform multiplayer solution that played nice with the unique demands of two very different platform holders.

“The biggest problem is we can't mix friends lists or anything like that. Also, there are certain rules on Sony's platform about age restrictions and content restrictions, things like that,” says Hagewood. “We have to uphold those, and normally Sony can kind of control who on their platform can say what and they can ban people and stuff, and they don't on a cross-platform game. So we have to make sure there's nothing that can cross that barrier of liability to either side. That's been one of the biggest challenges, figuring out how to make it so players can communicate without breaking any of those rules.”

At launch, Psyonix simply sidestepped by not allowing players to communicate cross-platform -- when I spoke to Hagewood last week he claimed PlayStation 4 players could not see or hear messages from PC players in the same match and vice versa, though Psyonix was working to patch in support for cross-platform communication.

Dealing with success

But the studio has been bogged down solving the second, bigger consequence of rolling its own multiplayer solution -- trying to keep its servers online and functional in the face of unexpected player counts. While popular opinion of Rocket League remains high, Psyonix has been hustling to deal with persistent, pernicious connectivity issues following the game's launch.

“It's been crazy. It was very strange having such success, and then having such a big problem to fix at the same time. We were kind of scrambling to get things working,” says Hagewood, sounding tired. “Demand was just so much higher than we expected it to be! We really needed to build a better system to handle that level of load. Funny thing is, we build bigger systems for bigger companies all the time; we just didn't expect to need something like that for this particular game.”

The Internet is littered with high-flying Rocket League clips like this

Hagewood says the team’s greatest challenge has been trying to keep the Rocket League servers up as much as possible in the wake of the game’s launch, even as they try to re-architect the whole shebang to create a more robust system.

The unexpected popularity of Rocket League is due in no small part to the game being made freely available to millions of PlayStation 4 owners as part of Sony’s PlayStation Plus program.

It’s not clear how many of the people who downloaded the game for free would have paid for it otherwise, which means Hagewood can never know how much better (or worse) the game would have sold at launch on PS4 had Psyonix decided not to make a deal with Sony.

“I will always wonder; given how much it has taken off, you have to wonder. You look at the amount of downloads and you go 'Wow, that would have made a lot of money if we'd had even a tenth of that in paid sales,'” Hagewood tells me. “But we knew what we were doing when we made that deal. We knew that was a possibility."

"I'm a very patient person, and I'd rather build this brand and make it become a thing than make piles of cash. I look at it as a long-term strategy: even if we have to wait until we make Rocket League 2, if it's become a thing, that's the most important thing for me. That it becomes this phenomenon. That people realize we've been making this really cool game, and now everyone's playing it.”

And of course, the game has also consistently sold well on PC -- it’s been at or near the top of Steam’s top-selling chart since its July 7th launch. Psyonix also plans to sell paid Rocket League DLC in the future -- after the server issues are straightened out, insists Hagewood.

If there's one thing the founder of Psyonix seems to take from all this, it's that having a game you love not get the attention you think it deserves doesn't necessarily mean you failed -- it might just mean you pulled the trigger too soon.

“Look at how similar this game is to the original one. And yet, look at the difference in its success; I think there's a lot of lessons to be learned there, in terms of how much polishing a gameplay concept can make a difference, and how much your timing makes a difference,” says Hagewood. “This game is coming out right now when there's kind of a lull, and it's also at a time when eSports and YouTube and Twitch streaming are really big. The game has always shined in that respect, even back in 2009, but...all that stuff is so much more important now. “

“So if you have a game concept that doesn't do well, it's not necessarily true that the concept is a failure. It's not necessarily worth dropping that idea. Maybe it just needs to be done in a better way, or at a better time.”