"Deadlines kill games," Steve Ellis tells us. "Sometimes things come together on plan but if you're trying to do something that is fundamentally a creative endeavour, trying to push boundaries and do things that are new, then things aren't always going to go according to plan because your plan isn't a plan it's a guess."
It's difficult to fault the logic of the man responsible for building GoldenEye's multiplayer mode on the Nintendo 64. You might also know him as the name behind the TimeSplitters franchise, which spanned five years and raked in critical praise by the bagful.
For Ellis, those were the glory days. A time before the onslaught of military first-person shooters, largely triggered by the Call of Duty franchise, took the genre hostage.
But during the development of GoldenEye back in the mid-90s, the fps was still in its infancy. Ellis and his team worked under the direction of Martin Hollis, who would later leave Rare to join Nintendo of America as a consultant on the creation of the GameCube. At that moment, none of them had any that their movie tie-in would explode in the way it did.
"We didn't really know at the time," insists Ellis. "We thought we were making something good, we weren't really following what anyone else had done because there wasn't much to even look at.
"It was a bunch of people who all but one hadn't worked on games before at all and we were just trying to make the best game that we could. We didn't have a strict deadline that we were working to."
And yet it turns out that GoldenEye's most celebrated feature, the multiplayer mode that Ellis designed, was nothing but an afterthought. The team only managed to add it on due to the game being delayed, and neither Rare nor Nintendo knew about it until after it was finished. The game was finally released in August 1997, almost two years after the film.
It sounds absurd, but as Ellis tells us, things were very different back then. "That was one of the good things about Rare," he says. "There was never a pressure that was put on the team, so we were just trying to make the best game that we could. We were massively late compared to when the game was supposed to come out but ultimately it worked out."
"Worked out" is one way to undersell it. GoldenEye's influence on the first-person shooter genre has been huge, and its multiplayer influence perhaps even more so.
"It wasn't until probably the last few years that I properly appreciated the fact that it's going to be considered a classic game," says Ellis. "I spent about ten years wondering when people would forget about it. And just coming to realise that actually maybe it is going to be one of those enduring classics that people do always remember, and that was just an interesting revelation. I didn't quite realise that."
After the success of GoldnEye came Perfect Dark, and then in 1999 Ellis and his team, which included David Doak, took off to form Free Radical Design. They then set to work on their next big hitter, TimeSplitters - a game driven by a thirst to offer something that the rest of the industry wasn't ready to.
"The whole thing about TimeSplitters was that it was to be the antithesis of all these other games," says Ellis.
"It was not about a character, it was about the variety and being able to mix and match characters and backgrounds, kind of like a sandbox. That's what we were going for with it, something where variety was the selling point."
TimeSplitters was unique for a number of reasons but it was its level editor - known as Mapmaker - that really pushed the game ahead of its time.
"The Mapmaker is an interesting one because it's a thing that the publisher always hated because they saw it as a totally non-core feature," says Ellis. "And it required a huge amount of testing which cost them. So they would have all liked us to drop it but luckily we were able to keep it, and particularly with the first one I'm amazed in retrospect that we were able to do that."
Adding a level editor to a console game was a bold step, but in TimeSplitters 2 Free Radical went all out, delivering a game packed with so much content it was more than worthy of sharing GoldenEye's legacy. Unfortunately, the sales figures didn't agree with the critics.
The cracks begin to show
"I think it's a shame that games like TimeSplitters were never as well received as we had hoped, because what we were trying to achieve...we were trying to move away from the tired models of military FPSs [First Person Shooters]."
Times changed. Ironically, Free Radical soon started to afford fewer liberties than Rare did, and Ellis and his team found themselves in a very different industry to the one that gave birth to GoldenEye.
"I always remember when we were finishing Timesplitters Future Perfect getting a visit from Don Mattrick to basically tell us that it didn't matter whether we made the game any good or not," says Ellis.
"We were saying we wanted a couple of extra weeks just to put in some final bits of polish. We realised that the multiplayer had not quite enough attention compared to the previous game and we wanted to finish it off properly.
"And Don Mattrick flew in and explained that if we did that then it's going to move fifty million dollars out of EA's financial year and it's going to hit their share price and actually that's more important than whether our game is any good."
Ellis and his team were learning that business no longer bowed to creativity in quite the same way: "That was my first experience of finding out that your interests aren't really aligned with publishers' interests."
Time to split
But as business went, the proof was in the pudding, and the critical success of TimeSplitters alone just wasn't enough to bring publishers back for more. After Haze for PS3 failed to galvanise enthusiasm, the writing was on the wall for Free Radical and the company went into administration before being bought by Crytek. It's for this reason that TimeSplitters 4 has been dangling in limbo ever since.
We ask Ellis whether he thinks we'll ever see the long-awaited fourth chapter, which Crytek cancelled after it absorbed Free Radical, but there's no good news to be heard here.
"I don't think there's any chance that's going to happen," he tells us. Given his experiences trying to get the game off the ground after TimeSplitters Future Perfect, we can see why.
"I spent a lot of time doing the rounds talking to publishers [about TimeSplitters 4] and the conversations always went the same way. You always got to the point where the marketing person in the room would say 'I don't know how to sell this' because they want a character that they can put on the front of the box."
"In the first half of that company it was easier to get the funding we needed based on critical success because there was always a publisher that thought 'well maybe we can do better than the previous publisher did so we'll take a punt on it," says Ellis.
"Now with the budgets being so much higher you can't make a game for less than eight figures anymore."
Despite this, the TimeSplitters series continues to enjoy a loyal following of fans, many of which continue to campaign for a fourth game regardless (its biggest campaign is still taking place on Facebook).
As for what Ellis would like to see in TimeSplitters 4 were it to ever see the light of day, he's got a few ideas:
"One direction we wanted to take it in was to increase the differentiation between the different characters I think that was one of my key goals for it...to make it less than they are just different skins with slightly different attributes but to give them genuinely different abilities. I think that would be good."
But its hard to say if enough people would use it to justify it, and these days cost is everything. It's always about the cost."
The gaming industry might have reached Hollywood proportions now, but if another GoldenEye ever happens, will it also be of British design?
"There's been a history of a lot of big things coming out of the UK," says Ellis, though he admits that getting things off the ground can be tough, and with so many of the big publishers waving stacks of cash across the pond, it's even more difficult for British developers.
"As a British developer I've often found that it's hard to find investment from other countries," he says. "Particularly if you're talking to American publishers. They'd much rather spend money in America. They like something that's on their doorstep that they can go and see what they're doing."
But far from being turning his back on the industry Ellis has found a comfortable new home on mobile, where he now develops for Crash Lab, the team behind game Twist Pilot. "From a development point of view iOS is the best platform I've ever worked on," he says. "It's just so good to work with. The whole Mac ecosystem, I'm a total convert on that, I don't have a Windows machine anymore."
It's strange to hear these words come out of the mouth of someone responsible for some of the biggest console games of all time. But Ellis learned that while riding the wave of critical acclaim was well and good, it was always going to be a short ride without the backing of the masses.
"It seems that people want games like COD [Call of Duty]," says Ellis. "I've never been a big COD multiplayer gamer, I have enjoyed the single player campaign."
"I don't begrudge them their success. It's just a shame that there isn't room for something that doesn't take itself too seriously."