You mention there being a bit of a culture shock – do you have any other examples?
One of the first meetings I had with Nintendo, they were trying to look after us in a meeting and they’d brought some sandwiches for us, for breakfast. And I took one bite of this sandwich and it had mustard in it and I literally spat it out in the meeting, in front of them all – which is probably the worst possible behaviour for a foreign visitor, to spit out the food that you’d just given them, but I had not expected they would put mustard in it! Literally, they spread it like butter inside the sandwich, and apparently that’s the done thing out there, and it’s a speciality. But I couldn’t stand it.
Was that neatly brushed aside?
It was. And all the things you hear about Japanese meetings, where senior people in the meeting might fall asleep and then they’d wake up and carry on with the conversation, as if nothing had ever happened – apparently that’s normal in Japan. All the things you hear about actually happen. And the weird ones, like in the toilets there are no hand towels, because you’re expected to carry around your own handkerchief and dry your own hands with your own handkerchief. And obviously, us westerners don’t do that. We expect towels or a hand dryer. They don’t have those in traditional Japanese toilets.
But there was plenty of amazing stuff, too, obviously. I remember when Nintendo took us to the Nara Deer Park, which is a public park populated by deer, but the deer have become accustomed to bowing in the Japanese tradition to get a biscuit. Literally, you hold the biscuit in your hand and the deer will bow at you and then you give them the biscuit. They said that the deer are a little bit aggressive, and if you've got biscuits in your pocket, they will start attacking you to get the biscuits out of your pocket.
Do you think there was ever any jealousy or animosity between the staff at Nintendo and this group of young Brits invading the company with their newfangled tech ideas?
With the people I dealt with, generally, there wasn’t. At the time, Nintendo of America president Mr. Arakawa nicknamed me 'Jez the Genius' or just 'Genius.' I was this spotty twenty-one-year-old, or whatever. They really looked after me, and I was their golden boy for two years – and my team, obviously – but I was the ideas guy, I was the guy that suggested what we could do and had to dream big, and then my team would back it up with actually doing it and delivering it.
Argonaut employees such as Dylan Cuthbert and Giles Goddard were based over in Japan for long intervals, and would end up relocating there permanently. How much time were you spending in Japan?
I would commute. I would do one week, every month.
Because the actual Super FX chip was made in the UK, wasn’t it?
We were very forward-thinking with the Super FX chip, because what we designed wasn’t a hardwired polygon engine, it was a fully programmable RISC microprocessor. We had the world’s first GPU years before anyone else had done it
Yeah. We did almost all the tech in the UK and Dylan, Giles, Krister Wombell and occasionally some other people – like Carl Graham, Pete Warnes and Ben Cheese – went back and forth. The main team was in London, building the tech, designing the hardware, software and the structure – all that stuff was done in London. The team in Japan were working on the games, and working closely with Miyamoto. We’d have faxes – this was before email – going back and forth as well as regular visits.
So, they had a standard DSP chip that just helped with the maths – it couldn’t help them with the graphics, and it was ultimately still a character-mapped display with sprites, so there was a limit to what they could do. They'd pushed the SNES as far as they could without the right hardware. We were very forward-thinking with the Super FX chip, because what we designed wasn’t a hardwired polygon engine, it was a fully programmable RISC microprocessor. We had the world’s first GPU years before anyone else had done it, because it wasn’t just the polygons – we could do 3D mapping on it, we could do pixel shading on it... we could do anything, except it only ran on 21MHz.
When they gave us the SNES and we designed the Super FX chip, there was no manual, there was no schematics, there was no instructions; we literally had to reverse-engineer the hardware to figure out what we could do. The reason why the Super FX ran at 21MHz was we were putting scopes onto the cartridge pass, and we found that one of the lines had 21MHz on it, and we thought 'that’s convenient.' If only we could design a chip that ran at 21MHz, we could use that clock signal and run our chip without needing our own clock. Then, we could synchronise it with the main system without doing any complex circuitry. That was just from us reverse-engineering the hardware and figuring we could do that.
When we designed this thing, Nintendo had to partner with Sharp to make the chip, and, literally, Nintendo bullied Sharp and said 'you will make this chip, and you will make it run at 21MHz,' – we couldn’t control whether it ran at 21MHz, that was up to the processor. Sharp had to make sure that it ran on 21MHz, so the onus was on them, and it did!
The general understanding that has grown up over the decades seems to be that you did the tech and Nintendo handled the gameplay – is that fair?
No, we did all the programming. The reality was, at the time – and probably even still now, to a degree – Nintendo doesn’t actually employ that many good programmers. They’re mainly creative, and what used to happen was Nintendo would program their games in-house but they’d make them work and they’d make them fun. And then they’d send them to this company called HAL, which would reprogram the games with their good programmers. HAL was an outside company, but they worked only for Nintendo; it was like a crack development team.
The reality is, the way Nintendo used to hire wasn't based on merit or talent, like western companies do. They'd hire graduates out of university and pay them based on age, not on talent or capability. So you ended up with a lot of creative people who did the design and came up with cool ideas to make the games fun. But programming-wise, they’re not the strongest because they didn’t start off life as programmers. It’s kind of hit and miss, whether they ended up being good programmers or not. Whereas teams that only hire good programmers – like Argonaut and like HAL – could do some good code.
On all of the games that we built for Nintendo, we did all the coding, and some of the design. Nintendo did most of the design, and the characters, and the music. So they did more of the creative stuff, while we did more of the technical stuff. It was a very close collaboration. As you know, we had people inside their offices, working alongside their teams; our people were effectively on loan to their team.
Dylan recently said on a Retronauts podcast that Star Fox was originally free-roaming to begin with, but it was Miyamoto who made it on-rails.
Yeah – because Starglider is free-roaming, and Starglider 2 is free-roaming. When we started building NesGlider, which was the forerunner to Star Fox, it was free-roaming, too. And then Miyamoto-san decided that it would be easier to make the level progression gradual if you were on rails. And so we had to simulate the look of free-roaming, but you’re still on rails.
Some people were somewhat disappointed that it was on-rails, but do you think the decision was the right one for a console audience?
The thing is, it’s the same as a platform game, or even the same as a movie. If you could free-roam in a movie, the director has no job. So the director can’t make the narrative and can’t build the surprise and can’t mess with your mind, and stuff like that. So, in order to have the sequence of events happen in a game that delights the player, you have to give the illusion of freedom, but for things to happen in the right times and the right points – and for it to be a predictably enjoyable game – you have to have some kind of guidance. And putting it on-rails was one way to do that. It’s not the only way, but it’s a lot harder to add narrative to the game without putting it on-rails.
Was shifting from free-roaming to on-rails a little bit of a 'eureka' moment?
Maybe. I personally thought the 'eureka' moment was putting characters in the games. Because at the time, in all of our games, the characters were in the story, but they weren’t really in the game. They were perhaps mentioned by name but you didn’t see them – and you didn’t see the action from a third-person perspective, either. Our 3D games were first-person, because that’s what we were used to. Miyamoto’s philosophy was that it had to be third-person, because you have to see yourself get hurt and you have to see yourself get damaged; you have to see the hazard happening, even before it happens, and see the effect it has on you. And that can only be done in third-person. So that was the thing that Nintendo brought to the table.
Moving onto other projects, Stunt Race FX was obviously the next game. Had the relationship between Nintendo and Argonaut changed on that second project?
Not by Stunt Race. We had basically signed on to build three original games for them. Star Fox was the first, Stunt Race was the second, Star Fox 2 was the third. And so, during Stunt Race, I don’t think we had any difficulty with the relationship. The relationship was a two-year contract period, where we were going to be exclusive – although it didn’t actually mention the word 'exclusive', because I think they had legal problems with exclusivity at that point. It basically said for two years we would give them first option on any game that we want to do and they’ll take it.
We had designed the worlds for a 3D platform game, which we called 'Yoshi' something... It was basically taking Super Mario World, and putting it in 3D. They were blown away by it, but they decided it was a bigger opportunity to cut us out and do it themselves
After that expired, we offered them our next game, which was effectively a 3D Mario World with a Yoshi character. We had designed the worlds for a 3D platform game, which we called 'Yoshi' something, and we had a prototype and showed them what it could look like. It was basically taking Super Mario World, and putting it in 3D. They were blown away by it, but they decided it was a bigger opportunity to cut us out and do it themselves. That’s when they started doing Super Mario 64 and we switched our game into what would become Croc.
Obviously, Nintendo have unlimited resources, whereas we were a poor company from the UK that could barely afford to do that, so it took us longer to build the 3D game. We should have been the first out, because we already quite far with the creation of the game, but we weren’t the first out because we didn’t have the funding that they had. And then we raised a bunch of capital money and got the funding and finished the game, and then signed it to Fox – and got a very good deal out of them.
How did that change your relationship with Nintendo?
It was ultimately the beginning of the end. Nintendo wanted us to stay exclusive, but we wanted to grow. We were twelve people at the time. We wanted to grow bigger and do more games, and they didn’t want that. They said ‘no, stay as you are’.
Would this have been around the same time as the Super Visor?
Yes. We had a joint venture company with Nintendo called A/N Software Inc, which stood for Argonaut/Nintendo; it was run by Nintendo. They had 51%, we had 49%, and the patents for the MARIO (Mathematical, Argonaut, Rotation, & Input/Output) chip – or the Super FX chip, as it was called – were held in this joint venture company. After Super FX, Nintendo wanted to do another hardware project with us; we said "we want to do a VR machine", which we were very into at the time, and they said "great, we’ll do a VR machine".
So we started designing the Super Visor. We designed a very cool 3D graphics chip for it. We started researching motion tracking and had a system that worked; Nintendo had introduced us to Texas Instruments, who had this novel concept of what, at the time, was called the DMD – The Digital Mirror Display – but has since become DLP, which stands for Digital Light Processing. Instead of using liquid crystal pixels, the chip has little mirrors and the angle of mirrors can be altered. The chip has the full image on it. It’s a chip like half an inch big, and it has the full, at the time, 320 by 200 full pixels on it. And then you shine lights on it, like red, green and blue lights, and you waggle the mirrors in software, and you get a display.
We were going use this display for the VR headset, and they had just invented this technology. It wasn’t publicly known, so we were non-disclosed by Texas Instruments, and it would have been very cool. But little did we know that Gunpei Yokoi had a side project with this company in America that had a red display. Only could do red, couldn’t do full colour, and it worked in a different way. It had a spinning mirror. And it basically wiped the picture onto your eye, waggling only one line of pixels and then scanning it across your eye – and it would only work with red. But it was cheap.
And then Mr Yokoi nixed our Super Visor project, when it was almost finished, and went ahead with the Virtual Boy – which I call the Virtual Dog. It was just an excuse so it could use that display that he liked, even though the technology was awful; the 3D was terrible and it had no motion tracking. It was attached to the table. You had to put your face into it, on the table. It was just the most stupid idea.
Do you think that the Virtual Boy played into Yokoi's desire to use cheap, existing tech in new ways?
I think that many of the people in Nintendo would have gone with our approach, and would have followed through with having a proper, real VR headset that you could move your head around, but Mr Yokoi thought otherwise. He was very high up; he was the head of R&D1. He ran his division, the Game Boy division, and he basically cancelled our project and went ahead with Virtual Boy and, for us, that looked like a very bad decision, from our point of view. But what can you do? Arbitrary decisions can be made, and there’s nothing you can do to argue. Still, I'm not bitter; working so closely with Nintendo was an honour, a privilege and an amazing opportunity to be at our most creative working for a company that appreciated our tech and innovation."
So they were part-funding that project?
No, they funded it. We got a million dollars for it, which is cheap for designing hardware, but that was a lot of money for us, because we were a small team. Yet again, we under-priced ourselves. We should have charged ten million dollars. Then we would have had the budget to really smash it. Actually, we had a prototype. We had cool stuff working. We had motion tracking.
Was any software created for it at the time, or was it just demos?
No, but we knew we could do it. And then, after they cancelled it, that was kind of the beginning of the end of our relationship with Nintendo. That was a very arbitrary, bad decision, clearly proven by the market of it. Maybe VR was before its time. But you know, VR is catching on a bit more now, albeit it still hasn’t caught on yet. There is always that stigma. And also, there’s still the risk that people will sue you when they do stupid things while wearing a headset, like fall down the stairs or put their hand in a fire, or something stupid, while their eyes are covered by your device.
Yet again, we under-priced ourselves. We should have charged ten million dollars. Then we would have had the budget to really smash it. Actually, we had a prototype. We had cool stuff working
Those product liability laws have burned Nintendo, in the past, and many others. McDonald's got sued when someone spilt coffee on their lap, and so they had to put ‘caution: coffee is hot’. Or peanut companies have to put ‘may contain nuts’ on their packets of peanuts. Product liability laws, in America, are insane, and so maybe Nintendo shied away from creating a headset which allowed people to be able to move around. There will be class-action suits for people hurting themselves while wearing VR and AR gear. It’s inevitable. So that’s probably been a drag onto the products’ success. Because even though the VR products are very good now. Actually, after Nintendo, we designed a VR console for Hasbro.
You couldn’t take anything from that previous project to that new one?
Hasbro had spent hundreds of millions of dollars to Sarnoff Labs, which is a prestigious high-tech company. Sarnoff Labs used to be called Bell Laboratories; it was the original telephone company in America, and David Sarnoff Research Centre is the place connected to Princeton University – a really top place, and Hasbro hired them to build a VR games console; they had tried a bunch of things and failed. And they couldn’t get the 3D graphics chip cheap enough, and they couldn’t get the microprocessor cheap enough, and they were about to can it.
And then I was friends with someone high up in that project, Sandy Schneider, and I told her ‘we could design it’. And then they rushed to London and they talked to us, and we over-promised, as usual, and said ‘oh yeah, we’re designing this microprocessor called the ARC – Argonaut RISC Core, and we could use three of those. We could use one for the 3D, one for the game and one for the sound’ and we called it MatriArc, because there was three of them. And they jumped at it and paid us some money.
But they gave us an extremely unrealistic deadline. They were signing it up for us in June and we had to deliver it in September – it was impossible to design 3D chips and microprocessors that fast. But we signed onto it because we were stupidly naïve and thought ‘we can do this impossible. We can design a microprocessor and a 3D graphics chip in three months’. But we delivered it. The prototype worked and it was all good. And then Hasbro had the fear that they'd spent hundreds of millions designing a console, but it will cost many more hundreds of millions to launch it and market it, and then have to go head to head with Sony and Nintendo. Did they have the stomach for that? And they didn’t, so they canned it.