Earlier this year, robotics company Anki hired Joby Otero, a 23 year video game industry veteran, to help spearhead its Anki Drive project -- think Scalextric with AI-driven cars, and you're sort of in the right ball park. More importantly, Otero was one of the key people behind the success of the Skylanders franchise. Anki is hoping that Otero can bring his expertise to hone the company's "physical video game" to make it the next big toy/video game hybrid. What's really interesting about Anki Drive is just how much it feels like a video game. The cars are controlled via a mobile app, and can be upgraded via earning points and buying through skill trees -- there's certainly some video game design conventions present here. Yet, it's a physical toy, and as such, it can potentially appeal to those people who may be put off by a more traditional video game. I spoke with Otero about the design behind Anki Drive, and where he plans to take it now that he's onboard. Is the intention to have it feel so video game inspired? Otero: Absolutely! The founders of the company all have PhDs in robotics and AI and machine learning, and they had always been inspired by video games. They don't come from a video game background, but they always had a desire to use their expertise to bring robotics to the masses in a way that felt like a video game playing in the real world. At this time, we have just the foundation. The great thing for me coming into the company, feeling like there was already this foundation there already, but knowing that despite them already having these video game like things under the hood, there's still a lot of headroom to bring to it. That kind of compelling experience that you'd get from a traditional video game, and bringing more of that to the product. For example, right now there's no fiction to it, no stories, no characters to it. Just imagine when we actually do add some characters and voice to it, and give it a sense of a longer term goal and progression building up to that goal. All of the things that we've learned to do really well in the several decades in video games, we can still bring so much of that to it. With the core design, what are the elements that are most important to make Anki Drive feel like something new, and feel like a video game? Otero: When the three guys were making Anki Drive, they didn't have a game designer onboard. This was still long before I came aboard. Although there had been that inspiration to be like a video game that played in the real world, what they had was the technology to let these cars know where they were on the track, and to be able to manipulate their own speed and steering, and move themselves around and communicate 500 times a second with your mobile device. That was the real genesis of it - knowing that you've got this real robot technology where the cars can know where they are in the world within the context of the track, and they can strategise on their own - they know where the other cars are, and they can change their strategies on the fly.

They knew that to get something that felt more like a video game, they'd need to have upgrades, and a sense of memory for the cars such that players can make choices that felt like a video game. Like they were actually upgrading the car, and those upgrades were persistent over time. That's where the propulsion was initially. Once they'd brought in game designer Sean Levatino, who heads up the game design, he brought to it a lot more emphasis on the upgrading and persistent elements. There's still a lot of head-room there though - it has a viscerally discernible impact on the experience, but at the same time, it's not really in a larger structure yet. I guess with something like this, constantly balancing the design must be a hell of a job, especially every time you release a new car, track or game mode. Otero: What's interesting is that, for a company still as new and young as we are, I'm not used to having so much testing support both internally and externally. We're about a 70-person company internally, not counting external contractors, and a good chunk of that is the testing group. It's pretty big, from my experience, for a team this size. It's because of what you just said - when you have all this stuff playing out in the real world, there's a lot of unpredictable things that can happen. There's a lot more opportunity for random events in people's living spaces. Sometimes even deliberate events - because it's physical, you might reach out and grab the car in the middle of a match and do something completely unexpected. The game has to try to figure out what the hell just happened. So it definitely requires a lot of testing, and there are issues like, you might lay out the track on an uneven space... it has to be very flexible, and the AI has to be very adaptable for all those kinds of situations. It's a huge ongoing challenge, but it's also the most exciting thing for a lot of our players - especially as we're finding that family groups want to play it. There's something really magical about this playing out in your living space, compared to a purely digital experience. And even though we know, as long-time game designers, that when a kid is sitting there with their eyes glued to the screen playing a game, that there's a lot going on in their head, they're absolutely not in a zombie state. But for a parent, it can be a bit off-putting, because their kid seems emotionally unavailable, right? So when you play those same dynamics, and have them play out in the real world, suddenly it feels like people are still accessible and physically active, and there's a social engagement that happens that we just don't see with games that are purely digital. Parents seem to find that exciting...[not just] to play, but also because it feels like they can still stay engaged with their kids, while their kids are playing something that is still like a game. Aren't the skill trees, upgrades etc. a bit too complex for the average person? Aren't some of these elements going to be baffling for the average family who perhaps doesn't play so many video games? Otero: Only to the extent that I wouldn't have expected - one thing that I absolutely expected without seeing their user test data, is exactly what you're saying: You give people all of these tools to upgrade, and it can quickly get to the point where it feels complicated - even when you take a step back and compare [Anki Drive's] upgrade tree to a typical game. But what it doesn't have right now is what I'd call a story, for lack of a better term -- some kind of game progression that builds out those features one at a time. The plan is absolutely to have game progression, so instead of having 50 options at your disposal the moment you go to customize your car for the first time, those will be piecemealed out in a way that allows you time to digest it. We're very confident that that direction will allow us to make the game have lasting gameplay, and more depth, but at the same time allow people who maybe don't even play any video games, to ease into it. I love that there's a "boss" car - a car that you need to beat to be able to race them. That feel very video games. Otero: Yeah, and that was a big risk. I have to hand it to the team, they've come to the table with a number of really fun insights that break the mold of traditional toys, and are kinda inspired by video games, but are definitely risky from the point of view of a regular toy purchase. You'd never do that, right? Here's a physical toy, you bought it, you open it up, and so of course you're going to play with it right away.

But something interesting happens when that toy actually has a mind of its own, and we want to begin treating it like a character. Characters have their own decisions that they get to make, and with Corax, the car you're referring to, we thought it'd be really fun to make it like a video game boss. He would require that you beat him in battle to access other functionality. That idea has one foot in traditional video games, and one foot in toys. Toys would never do that, but we took that risk, and it seems to be paying off. We don't get any complaints about it, but we hear a lot of people who really like that - it's an extra level of engagement that they get with that car. It feels like the underlying idea behind this whole thing is that you're looking to make a toy through which there are actually levels of engagement to achieve, rather than just being a toy where you have everything it has to offer from the get-go. Otero: Absolutely. That's it in a nutshell, and we want the cars to be something where any fan of toy cars can pick them up and move the toy around in the real world, letting their imagination run wild. These cars absolutely support that as well as any other toy car out there. But what a traditional toy car doesn't have is a "plan" in the game design that sets up a series of challenges and rewards. That's the core of what a game is, right? Here's the challenge, and if you can beat that challenge, then here's a reward. What defines flow in a game design is how you continue to meet the player at their present skill level. You defeated a level 1 challenger, so now here's a level 2 challenger. And now there's some new tactic that you hadn't confronted before - it's been carefully crafted so that you would be ready for that next level of challenge. And we can do that through potentially 50 or so levels of challenge. That's the sort of thing that can really take this up several more notches. The price: I'm sure you've had many conversations about this internally, but it seems a little steep [$150 for the starter kit]. What is going on with the price? What are you doing to bring it down? Otero: Like with anything, once we've figured out the basic patterns for the hardware, there are some cost optimizations that can be done. Once you know what the product really needs to have, and you've proven that in the market, certain things begin to become obvious. "Oh, we did that last year that made absolutely no difference to the player, so we can tune that price-wise." But even if we did no design changes whatsoever, but suddenly sales went up by 10 times, there are deals we can do with our manufacturing partners to bring the price down simply on that. One of the things we found with Skylanders is that that first point is super valuable. We were able to each year generally make the toys more economical, and sometimes bring new functionality at the same price. Or just improve the profit margins, while still improving the designs in ways that really mattered. So it's a business challenge, and frankly it could very well be that that's just where Anki's position is always going to be. At our heart we are trying to be pioneers, and we are trying to do something that is merging the worlds of toys, games and robots. We're always going to be on that cutting edge, and higher on the price scale. Taking it further - where exactly do you go from here then? Do you go harder on the video game design elements, or are you now looking to take what you have and make them appeal more to people who like physical toys? Otero: It's both actually. Something we found on Skylanders was that once you merge a physical product with a video game, you really have to design on both of those dimensions at the same time. Every time we would try to design a new piece of the video game that didn't feel like it was feeding off of the toy in some way, it felt like we were suddenly designing two different products. It doesn't behave harmoniously any more. If we added some physical innovation to the toy that the game didn't seem to know about, then again it felt like there was this massive disconnect. So for sure, we've got that message deep down in our DNA that every new product and every upgrade to any existing product is always that balance between the hardware and software. There absolutely will be massive improvements that are going to learn from traditional video games, and bring additional dimensions to Anki Drive in that way, and there are major things we're planning to do on the hardware side that will change the dynamic there. The trick will be that we're building up a team that comes from the traditional video game space, big triple-A titles, to sit alongside these robotics experts. Deep down, all of them having this understanding that the two things - that hardware and the software - need to work together. Anybody who tries to do this from the traditional toy space - there's a certain timeline of production that they're used to working on, and it's not the same as what you'd have on a triple-A video game. Having that understanding is going to be a criticial part of allowing us to target longer term, bigger goals for upgrading the product and creating our next product. How many years do you see it taking for Anki Drive to really pick up? I know with Skylanders, it seemed to take no time at all to go from announcement to millions sold. Otero: Honestly, I think it could be by next holiday, or earlier than that, when we finally see it really take off. There is definitely some relatively low-hanging fruit in terms of the marketing and PR side. One of the things is that we didn't spend anything close to what Skylanders marketing budget was to just generate basic awareness of the product. When will it really take off? It's difficult to say because there are challenges around the space needed to properly demo the product. We're working on that, and that's going to take some time. But fundamentally I think the biggest thing is that it is even more of a new category, even beyond what is happening with things like Skylanders and Disney Infinity. With those, the vast majority of the play is happening in a still relatively traditional video game. But there's something about that happening in the real world that takes longer to connect - not when you're sitting and playing it, but more from a marketing standpoint. There are some clever things we need to do on the packaging side, and in terms of our retail setups. And figuring out the best place to be at retail - when you've got a hybrid product like this, where do you put it? Do you put it next to the video games, or do you put it next to the toy cars? Does it belong in consumer electronics? That's sort of where we tried last year, and that wasn't really the best place to put it, because someone who is over by the cell phones isn't necessarily in the mindset to buy a toy at all. Those are the sorts of things we're making steady improvements on -- the messaging, targeting the right audience, and being in the right place at retail. The confluence of those things even just in the next couple of months could see us start to really take off.