Amazon: A year in the Lumberyard

Amazon: A year in the Lumberyard
Sean Cleaver

By Sean Cleaver

April 27th 2017 at 11:30AM

Sean Cleaver catches up with Amazon Games Services vice president, Mike Frazzini, to see how the Amazon Lumberyard engine and the company’s gaming focus has come on in the past year

It’s been a year since we last talked to Mike Frazzini about Amazon Lumberyard and we’re already beginning to see the fruits of its labour. After a solid months of hires and development, there are a few games already out, including new multiplayer brawler game, Breakaway. One of the things he talked heavily about then was the purchase of Twitch and its integration in to the Lumberyard engine and the games Amazon are making.

“We look at twitch and are amazed by how the twitch community has changed the way games are experienced and what we think is next is for the twitch community to change the way games are made,” says Frazzini. “We start with this idea of ‘what would it mean to build a game for Twitch’, Twitch for broadcasters, for viewers and our designers said, ‘well what do you mean?’ I don’t know, you tell us what that could mean. And they came back with whiteboards full of ideas and these really interesting ways for what it would mean to build a game that was fun to watch.

“We believe in creative autonomy, so we allow the designers to say ‘hey, if you love Twitch and you’re interested in creating online multiplayer community based games,’ then from there you go what do you have in mind, what do you think you want to do.

“And so with Breakaway the team was very interested in creating a game that was competitive where you could see the entire game map from one camera angle. And then they just start iterating. So what you hear hopefully from a lot of developers now is that you start off with an idea, you get it in front of customers as quickly as possible and they say ‘we really like this and this, we really don’t like these other things’ and you go ‘ok’. And maybe the things they don’t like you tweak a little bit and sometimes it’s a lot more than that. Breakaway is out view of saying how can we make a game that has to be fun to play, obviously, but maybe we can make it when we think about it that’s also fun to watch.

The question is when is the right time, how many people do we let in... We'd like you to play and tell us what you think
Mike Frazzini, vice president, Amazon Game Services

100 PER CENT OF STATISTICS

Frazzini has previously spoken to us about the use of ChatPlay and JoinIn with the Lumberyard engine. But with Breakaway the development team are trying a new technology in Metastream, which provides broadcasters of the game live and accurate stats of peoples in game performance, as Frazzini explains: “Whenever you’re watching a competitive sport or game, statistics are more fun. They make it more interesting. You watch World Series of Poker, this guy has this per cent chance of winning. So within Lumberyard, we created this thing called Metastream, which kicks out data from the game and serves it up to broadcasters.

?“Then a broadcaster can take that data and tell stories with it. This is an example of where we say that, how can we make a game that’s built for twitch and broadcasting. The game has to be fun, if it’s not then none of that matters. But if you go from there, there’s definitely a lot you can do to amplify it.”

Amazon has moved on apace from this time last year with a string of hires including former Daybreak Studios CEO John Smedley, Westwood Studios co-founder Louis Castle and EA’s Rich Hilleman as well as opening two new studios in San Diego, California and Austin, Texas. Amazon always refers to to everyone as a ‘customer’. Whether you’re a developer using the engine, a broadcaster using the software or a player playing the game, everyone is a customer in some way shape or form. And Frazzini and his team are very much behind the idea of getting things out to customers as quickly as possible to generate feedback and encourage iteration. In fact their entire ethos has been behind three things: the creation of communities, the service to customers and the ease of iteration.

“We launched Twitch Prime last August at Twitchcon, that’s done extremely well for us,” says Frazzini. “We launched Lumberyard last February, with nine releases since launch and over 2000 fixes and feature enhancements in that time. And so the development on the team is accelerating and us getting Lumberyard out early in to a beta was totally the right thing to do, because we have customers who are coming on early and they’re getting more support from us, they’re helping to shape the roadmap and then what ends up happening is you end up in a situation where the development teams are getting rich feedback from customers. They’re not guessing what they think they should do; customers are saying ‘this is what we need.’ Ok, and then in two weeks you build it and give it to them.”

BREAKAWAY SPEED

This pace of iteration is something that Amazon has taken in part from early access for games and has started to apply it to everyone in a game development pipeline and it all sounds rather organic. “We really believe this is how games will be developed over time,” believes Frazzini.

“We’re a game developer and a big part of Lumberyard is that we wanted it to be modular. We want game developers to be able to change things that they want to change, so we give them full source and then we built this component system in it that makes it really flexible. So that if they want to build something unique for their game because they want to create unicorns that spit rainbows out of their mouths and they need a particular system that does that, they can build it and integrate it in to the engine.

“That is a goal for us we want to make it really easy for them. And the reason is, to your point, which is when you’re making a game and you’re listening to customers, you’re going to be wrong around what you thought the game should be. Customers just go ‘I don’t like that’ and then you go ‘crap’ we’ve got to make some changes. And so you want to have that flexibility to be able to do that in development. You can’t do that utterly otherwise you’d never finish a game. But there is some degree where you have to listen to the big swings and those big opportunities to make the corrections.”

EARLY DOORS

This kind of pipeline eventually ends up in the hands of players, and when it comes to releasing games early, or possibly incomplete, Frazzini tells us that the benefit of this is an oft discussed point for Amazon. “We wonder about that,” Frazzini says. “We debate this internally because I play a lot of the early access games, they’re a lot of fun and I’m glad that I got an opportunity to do that. And we’ve debated it so I don’t know if I can give a good academic compare and contrast. What I can say is the idea of getting games in front of customers early and often is totally the right way to think about it.

“The question is when is the right time, how many people do we let in, how public can it be or not. If you make it super public really early, it doesn’t look very good, the gameplay is probably clunky in places you didn’t know about. With Breakaway and with all our games we said at Twitchcon, certainly what we want to do is find the nucleus of what’s cool and awesome and when we do, say ‘here it is, and we’re just letting you know we’re early and we’ve invited you in but we think we have enough here, we’d like you to play and tell us what you think.”

CUSTOMER-CENTRIC

Amazon’s thoughts though are not to compete in the space of game engines and distribution, but to be in the forefront of it for their customers and communities and to get the best team behind their products. “I think the interesting thing is in Amazon is that we’re customer-centric not competitor-centric,” Frazzini continues. “And if you want to be inventive and you’re competitor- centric, you have to wait around for your competitor to do something. And when you’re customer-centric, you don’t have to do that.

“The average tenure of the team is 10.8 years now. So we have people that know how to make games [and] that have made game engines. We know how to do a lot of this stuff but the inventive stuff is the stuff that no one knows how to do because it hasn’t been done yet. And for us, our focus is two fold.

“First of all he engine has got to be great so we take that very seriously. But then in terms of what’s unique about it, the cloud is today an incredibly important part of game development and its only going to increase over time. And when you talk to game developers what they tell you is the same thing. It’s really important hiring and retaining cloud engineers. It’s hard.

“The second area is communities. The most successful games now are all about communities. We had a statistic - 85 per cent of revenue on PC and console is from multiplayer games. It’s crazy. You look at mobile it’s the same thing. You look at the top sellers – how many of them have multiplayer community based features to them? Most of them.

“Twitch is one of the biggest most vibrant places where those communities form. We brought Curse, which is a huge community platform, wildly popular gaming site, and communication channels. That’s another example all about community. So we think ‘ok, what are the cool things happening in these Curse communities and how do we build tools in lumberyard to help game developers to create this type of content.’

“The cool thing is if you think cloud and community, for us, that’s totally the areas for extreme invention. Our render? We’d like to think it’s going to be, if it’s not, the best in the world. It’s got to be otherwise what’s the point? And our animation systems and our physics systems and things like that. So we take those things extremely seriously, but then we think ‘Ok, what’s the real point of frontiers that we’re going to be inventing?” For us it’s around the cloud and it’s around communities and events.”