Google Stadia: Phil Harrison answers our biggest questions

As a veteran of multiple console launches at Sony and Microsoft, Phil Harrison has been doing this for the better part of 30 years. By “this,” I mean sitting in a hotel room, being interviewed by a reporter about an Exciting New Thing, answering questions without really giving much away.

In the course of my allotted 30-minute interview with Harrison, I (along with compadre Chris Plante) learned a few things about Google Stadia, a new streaming platform for gaming, that we hadn’t learned at Tuesday’s press briefing.

But mostly, interviewing him is a useful, enjoyable exercise, in that it’s an opportunity to air the discourse, to shake out the conversation and hang it on the washing line.

It’s not game journalist hyperbole to suggest that Stadia is a really big deal. Potentially, it’s the death knell of everything we know about console games, about hardware generations and the hierarchy of platform holders, publishers, developers, media, and players.

Overused words like “disruptive,” and “revolutionary” don’t do justice to Stadia’s potential to upend gaming, to smash the status quo and replace it with something new. Of course, all of this depends on Google’s ability to deliver on its promises; on Harrison’s ability to marshal and steer the future; on the desires and needs of developers, publishers, and, most importantly, consumers.

Here’s our conversation, which has been edited slightly for grammar and brevity.

Technical details

Polygon: So, Phil, is it going to work?

Phil Harrison: It does work. Everything we showed you yesterday was real. We have been testing in private inside of Google for a number of years. We tested publicly with Assassin’s Creed Odyssey in the back end of 2018.

There was a number of technical reasons why we did that test, but one of the non-technical reasons was to shift the narrative a little bit in people’s minds. “Oh, actually, game streaming does work, and it can do justice to the most challenging and demanding titles.”

A little aside: For our presentation yesterday, the live demo — when we went through all of the screen types — when you’re doing a big presentation, you want to de-risk that as much as possible. Our original intention was to bring a dev kit and have it sat in the back of the stage, and run it streamed from that dev kit to all of those devices. We had a technical problem on Saturday. So that presentation live on stage was running from a Google data center in South San Jose, 50 miles away. So it was actually easier for us to run our presentation live on the internet than it was to do it locally with a machine in the room.

One of your pitches yesterday was that Stadia is for everyone. We’ve heard that a few times over the years in the games industry, but what does “everyone” actually mean? Here we are in the highly wired Bay Area, where the internet works pretty well. But if I live in South Dakota or Romania, maybe it’s not so fast? Maybe it won’t work? Is that fair to say?

Yes, of course there will be parts of the world that we cannot reach yet, because connectivity does not reach that particular part of the population. And I’m not going to pick on any individual location or country. Having said that, I did get an email from somebody overnight from Romania saying, “Our internet is amazing. You should build a data center here.” But the point is that there is a rising tide that lifts all boats.

Connectivity is becoming ubiquitous. It’s not yet ubiquitous and I completely accept that, but our goal is to reach everyone. Over time, that will be based on the continuing build-out of fixed-line broadband fiber and other infrastructure to people’s homes.

But there’s a couple of very important technologies that are just over the horizon — principally, 5G — which will further accelerate [the spread of broadband], and give even greater access to even more people. It won’t happen to everyone overnight, but this is the direction of travel.

How do you communicate that to people? How do you say to people, “Here’s this new amazing thing, but it might not be for you because you might not be in the right place?”

The same connectivity challenges that certain physical locations may experience today are the same challenges that prevent them from streaming video, watching YouTube, getting music, playing an online game. While I’m not trying to marginalize those people, it is the reality of the world that we live in. But everything is moving to some kind of digital, some kind of connected future.

So generally speaking, if I can watch Netflix in HD, without any buffering, I’m probably OK to play this?

The example that we use is, if you get a good YouTube experience, you’ll get a great Stadia experience.

How much data will I be using? If I’m sitting home and I’m streaming and I’m playing, what am I looking at for the whole month? And how does that work if I’ve got data caps on my internet service?

So with Project Stream, we recommended and set a threshold of about 25 megabits per second in order to enjoy 1080p / 60 frames per second. In fact, we only used about 20 megabits per second. But we gave ourselves a little bit of a buffer in the calculations.

When we launch, because we’ve made some very significant improvements to our encoder, our streamer, and our compression algorithms, we will get 4K / 60 frames per second in about 30 megabits per second.

And then if you are at a lower resolution, you will obviously use significantly less bandwidth.

That could end up being dozens, hundreds of gigs a month. Some telecoms could be using 5G as a way to bring more data caps in. That would be the concern.

I’d point you to the Verizon 5G tests they’re doing, though. Those have no caps on them. The ISPs have a very strong track record of adapting, based on consumer behavior. When music streaming started, data caps lifted. The caps lifted when video streaming, particularly driven by Netflix and YouTube, became popular. We are confident that they will continue to lift.

What’s the lowest resolution Stadia will go?

720. Technically it can go lower, but we don’t go lower than 720.

What happens if my internet drops, and I’m in the middle of a game?

We have some very clever technology that will maintain frame rate before we drop resolution. We always try and maintain frame rate as best we can. But a lot of that’s proprietary, so we don’t go into the ins and outs of how we do that. We have some very clever technology.

To go back to the question of universal accessibility, the controller is basically a game controller, with lots of buttons and pads. That’s something designed for gamers with lots of experience with complex games. It doesn’t seem to speak to universal accessibility.

I don’t know how to answer the question in ways that will be satisfying to you. But we understand that. I think what our platform allows is for developers to connect the games with players in a way that allows them to try the game in a way that was previously very complex or costly.

Yes, we have our own controller. But if you have an existing USB controller that uses the HID standard, it will work. And I will need to come back to you to confirm this, but I believe that the incredible work that Microsoft has done with the accessibility controller for Xbox will work with our platform. But I don’t know that for sure. So let me come back to confirm that with you.

Josh Edelson/AFP/Getty Images

Stadia and the future

If Stadia works, does it really change everything? Consoles are obsolete. The whole order is up for grabs. Or are you just another thing that’s going to squeeze its way into the status quo?

I wouldn’t make any grand proclamations about the — that one thing has to die in order for another thing to be successful. I think that’s up to people like yourself to make those headlines, not me.

We see this as the direction of travel for the future of games. For sure. It won’t happen overnight. It’s not going to be a switch that gets flipped and everyone moves from device-centric to networks in one second.

But when you see the opportunity creatively, distribution-wise, technically, game design-wise, you know that this is going to be a very dramatic shift in the way that games are made and played.

You’ve worked for Sony and Microsoft. You’ve been in those rooms watching competitors make their presentations. What are those guys doing and saying, right now, do you think? Are they scared?

I don’t know. I would point you to the [Microsoft] memo that appears to have leaked that Phil Spencer sent around as a result of yesterday’s presentation. Draw your own conclusion.

If Stadia becomes the de facto way to play games, that puts you in an enormously powerful situation, individually. Your power would be unprecedented in the games industry. Is that something that you’ve thought about, or talked about?

You know me. I don’t really think about power in that sense. I think about helping developers find opportunity.

I’m not going to answer the question directly. But to give you an example. I very purposefully chose GDC as our moment in time to reveal Stadia to the world, because I wanted all game developers to be able to experience and understand and sense the vision of the future. So that they can build games and they can build their businesses on top of our platform.

[Google CEO] Sundar [Pichai] talked about this yesterday in his opening remarks. In the last four years, we’ve generated $110 billion worth of value for our partners across all of our different platforms. It doesn’t matter what the genre, what the segment is. If we’re not making our partners successful, our platform is not successful.

I was curious that there was no mention of big games publishers like EA and Activision during yesterday’s presentation.

Don’t read it too much into which games we chose, and which games we didn’t choose. We had a series of platform features that we wanted to articulate. We chose a number of games that helped illustrate particular features. But don’t read too much into why so-and-so was there and why so-and-so wasn’t there.

We’ve had deep conversations over a number of years now. We’ve shipped over a hundred development kits already. We’ve got thousands of creatives already underway. So you’ll see a pretty amazing lineup come June.

June. So that’s the next step?

The summer is when we will be next back out in public.

But you’re not confirming E3?

We’re not confirming E3.

Justin Sullivan/Getty Images

Costs, pricing, games

How much does it cost you guys, to develop Stadia? Where is the cost? Is it in cables and data centers and GPU towers? Or is it in research and development?

It’s a bit of both. We don’t break out the specifics as to what the spend is on different categories. But it is a public record that in 2019, Google will be spending $13 billion in infrastructure and capital expenditure. So this is a very significant investment for the company.

It’s unprecedented, in the games industry, that a platform holder owns the means of distribution, the retail component, the hardware, and, if you have a powerful first-party games presence, the product too.

I’m not sure it’s unprecedented. I think that the various flavors of mobile phone ecosystems have had that end-to-end model on which independent developers and publishers are wildly successful. And we would apply the same philosophy here.

We’re doing first party studios with Stadia Games and Entertainment not because we want to have a dominant share of revenue, but because we want to have those beacon and lighthouse experiences that really demonstrate what it means to have the network as your platform, the data center as your platform.

We’re going to have those deep connections with the rest of Google, where we can bring machine learning and artificial intelligence and Assistant and other leading-edge technologies into games for the first time.

We’re going to make some mistakes. We’re going to have some stumbles along the way. But once we’ve got to the great, then we can share that with the rest of the industry. And that will help everybody get to a great point even quicker.

When YouTube TV launched, it was confined to certain geographic areas, within the United States. Do you have similar plans for the launch of Stadia?

Well, we’ve announced that we’ll be launching this year in the U.S., Canada, the U.K., and most of Europe. With Project Stream at the back end of last year, we were live in 12 data centers in the U.S.

We’re already building out the stuff that you can’t see in the infrastructure. It’s known in the industry as “space and power.” So that’s the buildings that you put space into for racks and power and cooling.

Those take a long time to build out. In some cases we’re building dedicated buildings and filling them with racks of computers. We showed a couple of beautiful photographs yesterday. They don’t all look that beautiful. [laughs]

How many jobs are you going to be creating, directly for game developers, with your first-party studio efforts?

I haven’t got a specific answer in terms of numbers of direct employees. But there will be two beneficiaries. There will be studios that we build ourselves, and there will be independent studios that we partner with and invest in. Not at an equity level, but you know, from a support point of view.

We will be helping to grow studios to build games for Stadia. And I think that will be good for the industry.

So Google will be making big, first-party launch games that will need to have the same resonance as a Halo or a God of War. What stage are you up to with them? Do those games even have names at this point?

There are some really deep areas of R&D that are going on that will bring the best of Google to game design. What that looks like and plays like, I think it’s too early to tell. It’s more building fundamental underlying technologies.

We hinted at it a little bit yesterday with some of the things we showed you. Stream Connect and the scalable multiplayer, distributed rigid body physics. You know, some of the stuff that has historically been impossible to do on traditional platforms, we are now able to do. So you’ll see us lean in deeply into that.

Last year, Jade Raymond was serving a group general manager and senior vice president at EA Motive. We spoke with her at E3 about her vision for the future of games, and she emphasized the huge opportunities for companies to make use of in-game data in new and creative ways. She gave an example of a theoretical game that used player data to select a beloved character to be killed off [as in Game of Thrones]. At the time, that seemed extremely ambitious for EA. But if there’s one company that knows data, it’s Google. How much of her design philosophy influenced your decision to bring her on, as head of Stadia Games and Entertainment?

Let me answer in a couple of different ways. I spoke with a lot of different people about who was going to head up our studios. I saw some incredible talent, people with a rich history of hits. But with Jade, we found somebody who could not only imagine the future but could also articulate it in a way that we really liked.

Getting hired at Google is a very complex thing. She is obviously successful in that process. I’m really excited to see what that means for the kind of games and experiences that we build.

To your second point about data, I want to make a really important point here, which is: When gamers think about data, they probably think about their data and their personal data. That’s not what we’re talking about.

What we’re talking about is … in our platform, we have very high-performing server-class infrastructure for memory, and CPU and GPU, but crucially, also for storage. We give developers access potentially to petabytes of storage at a very, very high speed.

That allows developers to write out complex databases that would allow the kind of scenario that you mention become real. In today’s distributed world, it would be very slow and complex to bring all of the states of everybody’s individual game into one central place, run machine learning or AI algorithms on it, and then propagate it back out to every single member of the game. It’s not that it’s impossible; it would just be really, really slow and very, very costly to do that.

But with our architecture, that actually becomes comparatively easy and very fast — like, “microseconds” fast. So that allows game designers to have access to tools or ideas that are going to be pretty radical.

I haven’t asked about pricing because you’ve made no announcements and I suspect you’ve got a canned reply …

So imagine I’ve just given you the canned reply. [laughs]

But if I were a developer or a publisher, I’d want a little bit more than that before I committed to making a huge investment in developing for Stadia. What are you saying to those guys? Are you telling them what your plan is?

We are having conversations that are necessarily commercially sensitive and that are under nondisclosure agreement. They’re not the kind of things that you proclaim from a stage. It’s something that you discuss in a business setting.

Those conversations have been had. We are all really happy with the way that our publishers will be able to be successful on our platform.

Google will sometimes create things without any intention of profitability. You created Google Photos knowing it would lose money, because the machine learning that you gain from Google Photos is staggering. I look at Stadia, and I wonder if this is somehow bigger than just making a ton of money, right now.

We’re a business. There’s no denying that we’re a business. But we’re a business that thinks very long-term, that is probably as long and as ambitious as anybody, certainly in the West.

You can see the evidence of the very public leadership support we have from Sundar, to the investments that we’re making both in capital and infrastructure. We’re in this for the long haul.

Source: Read Full Article