“One of the first things I wrote in machine code was an expanding starfield. Just literally flying through it, and I found it mesmerizing. I thought ‘It has to be a game.’” That was the moment David Braben, co-creator of Elite and founder of Frontier Developments, changed space games forever. It was video games’ equivalent of the Big Bang, the birth of an idea that would eventually go on to shape four decades of space games and, eventually, Starfield.
In the early 1980s, space games were pretty rudimentary and there was little more to them than flying a ship and shooting aliens. David Braben, who at the time was an undergraduate at Cambridge University, had the idea that a game set in space could be so much more.
“We had Space Invaders, we had Galaxian, we had Williams’ Defender, but they all had a very similar format, i.e. three lives, a score that goes up, you get a new life at 10,000 [points], you get a smart bomb or something games-specific at 1,500,” explains Braben. “Pac-Man was there. And I know it sounds silly, but even then they were starting to become a little bit samey.
“I had played some games like Adventure, Colossal Quest, text-based adventures – the sort where you say, ‘go north, pick up key,’ that sort of thing – and I liked those. And it struck me that these are being played on the same machine, so surely you can [do something] more interesting? I found, with Space Invaders, all I really cared about was whether I got slightly further than I did last time.”
It was around this time, frustrated with the state of sci-fi arcade games and tinkering with some home games, that Braben met fellow programmer Ian Bell, and together shared notes on some projects they were working on.
“We talked about it and we thought, ‘Wait a second, if you had a real spaceship, you’d probably be doing something, you’d be traveling between destinations making money,’” Braben remembers. “And it’s that sort of lightbulb moment where you just start thinking, ‘Wait a second, isn’t score just money?’ And it’s terribly sort of capitalist but from a gameplay design point of view it was fantastic.” It makes sense – the points you score for destroying a ship in Galaxian can be equated to the bounty you claim for shooting a pirate in Elite. But the latter, you also have the added benefit of going through the wreckage and selling on what’s left for additional cash.
By rethinking the elements that made up much of the day’s arcade games in simulation terms, Braben and Bell came upon the idea of Elite, a game where the goal wasn’t to shoot down tiny aliens for power-ups, but to travel the galaxy, fight pirates, collect rewards, upgrade your ship, and continue into the vastness beyond.
However, Braben didn’t just want to travel one galaxy, he wanted to fly across eight. And he wanted them to be filled with planets – 256 in each to be precise. But back in 1984, he was working with a computer that had less memory than a modern day calculator.
“We were targeting the 32K BBC micro but actually only really had 20K of available memory because the screen uses up some [memory], the operating system uses up some,” Braben explains. “And I thought, ‘Well how many locations can you travel between?’ And I was thinking 20, 30…
“That’s not very much data on each one because you’ve got to have all the 3D render, which I’d already written – we knew how big that was. You need all the gameplay, the ship models… And how many ship models can we afford to have in the memory? It became one of those sort of cruel things, thinking, ‘Well, I want lots of those, but I also want lots of those.’”
The solution was a technique that studios like Bethesda and Hello Games – and countless others – have since used to create their own star systems: procedural generation. It’s a tool that’s used a lot in modern day video games, but back in the 1980s it was a relatively new idea in game development.
“I thought, ‘Wait a second, I’ll write a program to generate it,’” recalls Braben. “And it generated so quickly I thought, well, ‘We can just generate it every time – we don’t need to store it.’”
By procedurally generating galaxies Braben and Bell were able to save a tremendous amount of space without cutting back on their ambition. However, it wasn’t always quite as simple as they first thought. In 2013, Braben hosted a TED Talk called ‘Rules Can Be Beautiful’, which detailed the idea of saving space by procedurally generating new planets each time you want to explore a galaxy: “Essentially what it’s doing is, you are constraining the rules to make things that make sense. So making sure that the names of the places were pronounceable, making sure the economies were the right sort of ratios, and then just applying a sort of ‘human logic’ to generating lots of galaxies and just looking if it’s right. It’s amazing how something that is genuinely random can appear quite lopsided. You go, ‘Oh, we don’t really want a lopsided galaxy.’ But it comes naturally out of the random nature.”
As such, some galaxies were not quite as intended, with one planet randomly given the name ‘Arse’. But ultimately the system was a game changer, not just for Elite but for space games that followed in its footsteps. It allowed them to do things they would never have been able to before: “It’s the joy of being able to add something which we couldn’t otherwise add,” explains Braben. “But that was how tight it was. The game literally fitted exactly in memory, with not even one spare part.”
With Elite in hand, Braben and Bell set out to sell the game to the masses. The only problem – a problem they were trying to fix – was that the popular space games were the ones with three lives and missile upgrades, not ones with hundreds of procedurally generated planets to explore.
“Our first disappointment is we went to EMI, the record company. I thought, ‘Oh, they’d be good to sell it.’ […]But EMI rejected it, saying ‘It hasn’t got three lives, and we want a score.’ They thought people weren’t that dedicated and that they want to play a game in five or 10 minutes. I said, ‘No, they don’t. I don’t. I’m a target market that’s not being satisfied.”
The person from EMI asked the pair how long it would take the player to make ‘meaningful progress,’ to get the requisite power-ups and beat the game. Braben and Bell had a simple response: “I said, ‘It doesn’t matter. The aim isn’t getting to the end of the game here. This is a hobby. This is an environment you can live in. This is a world.’ And they really hated that.”
Nowadays games are celebrated for being open-ended or hundreds of hours long. When Elite was being pitched, it was an outlier. “I just thought, ‘Oh my god, what if they’re right? I mean Ian Bell was also… I don’t want to speak for him, but he was keen on this, something that’s just very different,” remembers Braben.
The pair eventually took Elite to Acornsoft and according to Braben, the reaction to seeing Elite was the opposite of what they got at EMI. “The reaction couldn’t have been more different, because they were techie, gamey people like us who were going, ‘Wow, how did you do that? How did you get that working?’” ”
Acorn signed up Elite and released it on September 20, 1984. It went on to sell a million copies across multiple platforms. Back when video games were still an emerging hobby, if you had any kind of game console, chances are Elite was on it.
“I think it was something like 17 platforms that we put the original Elite on? And they all had different CPUs. So the NES for example still had a 6502 variant, as did the Commodore 64. We did a number of Z80 ones, things like the Amstrad, the Sinclair Spectrum. Obviously we did the Apple II and various variants of the Apple II as well, there was an Apple II-C. But it all added up – I think we got a little bit sick of just doing it again and again, so each version typically had a few extra features.”
The different specs of each platform meant Elite was slightly different, and occasionally expanded, with each new release. Among those enhancements is an appropriate musical nod to the film 2001: A Space Odyssey. “We went on to machines like the Commodore 64, which had an unbelievably large memory at 64K. So we added music, we added various other things. There were some extra missions that just added richness to the game basically. But the music [when you dock]; obviously, it has to be Blue Danube when you dock.”
Elite went on to generate numerous sequels, including Frontier: Elite 2, which included even more planets, with advanced physics and realistic chemistry. It also caused an influx of space games that demonstrated the desire to do more than fly and shoot. 1988’s Captain Blood swapped procedurally generated galaxies for fractal landscapes, but it’s clear to see how Elite helped shape its worlds. Similarly, Starlancer and its follow-up Freelancer, which was released in March 2003, were both space trading and combat games, the genre Elite kickstarted almost a decade earlier, but one that continues to this day.
However, 1995 was almost the end for Elite. The third game in the series, Frontier: First Encounters, was released in an unfinished state due to pressures applied by the game’s publisher Game Tek. Its reception was understandably mixed.
“I even offered all the money back to show how bad it was,” Braben explains. “It was very upsetting for me because there was a good game in there, but they did a lot of changes underneath our feet because we were now a company by then.”
Shortly after, Braben decided to take a break from the series. His experience with First Encounters had proved too much. But before long his passion for outer space had reignited and, following a successful Kickstarter campaign in 2012, he was able to re-acquire the rights to Elite and work began on the biggest game in the series, Elite Dangerous.
By the time Elite Dangerous had emerged, the landscape of space games had changed. Narrative-driven RPGs like Mass Effect and shooters such as Halo were the biggest sci-fi games around, though that didn’t matter much to Braben and Frontier during the development of Elite Dangerous.
“I think when you look at games like Mass Effect, which is a great game and none of this is [meant] as criticism, we are different. We are doing something different. And with Starfield, I’m obviously looking forward to seeing how they do it, but I can’t help but think it’s going to be more Mass Effect than Elite Dangerous, because of its nature. It’s possibly more on-foot-focused much like Mass Effect was. They may put in dog fighting, we’ll see. But I think Starfield will be good. The budget is absolutely humongous and I’m sure it’ll be a good game.”
The truth is, Starfield – and Wing Commander, and No Man’s Sky, and countless other space games – likely wouldn’t have existed if it weren’t for Elite. It was the first game that took us to another galaxy from the comfort of our bedroom. It is unquestionably the godfather of space games.
“It sounds arrogant but I think it’s true,” admits Braben, somewhat reluctantly. “It’s certainly the first game to treat that subject matter [properly].There were games in the following years that people remember as well, but before it all there really was were various 3D shooter-type games.”
In 2023, the space genre has never looked so good, but obviously the hardware modern games run on is lightyears beyond the tech Braben and Bell were working with. Despite that, to this day Elite’s primary directive remains the same: to explore and live in the starfield, not just shoot aliens in the sky and collect a new power-up.
“In my heart of hearts, I still genuinely wanted a game where I could go out exploring, a game where I could go anywhere and I could look behind the curtain, and I like to think that’s what we delivered,” says Braben. “Look at all those people who’ve gone out exploring [in Elite Dangerous], and still only a tiny fraction of 1% of the galaxy has been visited by players. Players are discovering thousands of new stars and new planets every day, and that’s even after 10 years. The Galaxy is big – it’s the same size as the real Galaxy and has the same number of stars.”
We’d best get back out there then.
