Gaikai Talks PlayStation Now

We recently spoke with Gaikai CEO David Perry about all things PlayStation Now. The longtime developer joined the cloud-gaming company in 2009 and in 2012 Sony acquired it to form the core of PlayStation Now. Check out our interview as we touch on signing with Sony, remote play, what it takes to get a game on PS Now, expansion to more devices, overcoming rough cloud-gaming experiences, and looking to the future.

When you're finished, be sure to check out our interview with Sony's Eric Lempel on PlayStation Now as well.

What initially interested you about cloud-based gaming?

I just got an email out of the blue from a couple of guys from the Netherlands, “Hey, we liked your speech. [Ed—Perry spoke in February 2009 at the DICE conference in Las Vegas about instant access gaming.] We’re actually working on this technology right now." And I was like “Yeah, probably not," because it’s going to be really hard to make this work well. Then they sent me a link. I was suddenly playing World of Warcraft at my house all the way from Amsterdam. And you can imagine the link was terrible because it was halfway around the world, but the fact that I was sitting there playing it... And then they said, "Here we also have Mario Kart." I was playing Mario Kart across the world. And I just went “Man this is exciting.” I cancelled what I was working on at the time and just jumped right into this because I thought this is going to be important.

Wherever I am, I want to pull up any song and listen to it. I’d love to be able to pull up any movie and watch it. I’d like to be able to have a chat with someone about a game and be able to pull it up and play it, or if some game came out at 9 a.m. this morning and I’m able to jump right into it and within one minute be in it. And then be teleporting my friends to me within another minute. That would be huge.

Why was Sony the right direction to go?

We were very worried because we were doing only PC games. With PC games there’s a general problem. There are different systems running those games. You might have Gamespy in the game and Gamespy doesn’t exist anymore, or you might have Steam running a game and Steam adds a security system. So to try and run a game in the cloud it’s like, "Where am I? What’s going on?" Any built-in security systems are going to panic.

Instead of being able to focus on the streaming part and the experience, we were focused on, "How do we make it so that this PC game stops freaking out and will just run properly through the cloud?" And that was becoming a real problem for us because it was massively reducing the amount of games that we could get going quickly. And they expected keyboard and mouse, but we wanted them to work on television sets. We were trying to come up with systems that modify a legacy game in real time. We were actually changing the screens of the games in real time using image recognition and things like that. It was like Pandora’s box, the amount of work. You kept peeling back these layers of problems and trying to fix them. Our fantasy was that we could instead get off this platform that was so complex. We wanted to get one where there was a large set of very high-quality games but with a standardized control system, something that used a joypad that all the games would be compatible with. And it didn’t have crazy random security systems included.

I’ve been a Sony fan since day one, and so it was extremely exciting to us when they started to pay attention to what we were doing. They were at all the cloud-gaming conferences, and I was keynoting quite a few of them so we had a more open dialog with Sony over time. Then we showed them some of the future technology that was never released, but they got to see what we were working on behind closed doors. I think that finally sealed the deal because then they could see where this was going.

Gaikai's David Perry.

How does Gaikai fit within the Sony corporate structure?

You would assume that we would all just have to move to Japan so that we could work with the teams there, but Sony is an interesting entity. We have advanced R&D being done in San Francisco, and if you think about some of their best games like The Last of Us is done in Santa Monica, so they’re kind of distributed already. It wasn’t an enormous shock to the system to say, “Can we stay where we are? We really like where we are. We have a building that has a built-in data center."

To be sitting on top of a real data center is fantastic for development. We can go downstairs and do anything we want to the servers and put new servers in. There’s a 24-hour network operations center, people there all day long looking at screens monitoring everything. To have a data center in your developer’s building was just incredibly attractive to us so we didn’t really want to move.

Furthermore, it’s pretty discrete what we’re doing. It’s very clear it’s this cloud-based gaming thing so it doesn’t step on other peoples’ toes. It’s not like we’re doing the same thing that somebody else is doing. Imagine if we were doing a piece of the PlayStation Store. That wouldn’t work because then we would have to move in with them. By doing something very clearly different I think it’s worked out okay. Our team does fly to Japan quite often. It does mean lots of international phone calls and things like that, but ultimately the communication and the sharing has worked out well.

Just to be very clear, we only do two things for Sony. We focus on cloud gaming and remote play. Remote play was important because it needed to use this system of transferring experiences locally, and then of course the cloud stuff, which was all network based. The network we built with Sony is much bigger and more complex than the one we had as a startup. It’s dramatically more advanced and it’s designed to scale because we have to allow for a lot of PlayStation gamers to play if some new game comes out that they really all want to play. The optimization and capabilities that we’ve got internet-wise are far better than we had before.

What did you learn from developing Remote Play?

Remote play was the perfect way for a new company like us to learn how to work with Sony. Instead of jumping in at the deep end, it allowed us to work on something that to us was reasonably straightforward because we had that capability already here at Gaikai. But to build that into the operating system of the PlayStation meant working hand in hand with the team in Japan. That was a really good, gentle intro into how do we get this code to work properly within the operating system correctly.

And we get it to work on the other side on the Vita. That was prior to the acquisition. That was one of the demonstrations that Sony had seen its state-of-the-art games running literally straight to any device. We showed tablets and phones and televisions and everything, but we also showed PlayStation Vita. To put a high-end game on that beautiful screen, it was something that we loved to show here. I think when they saw that it made it very clear that this needs to be done. It worked out well. It’s very popular. It’s used a lot.

I tried it out on Assassin's Creed IV: Black Flag.

And you gotta remember, this is all first generation. It’s kind of fun to imagine what you can do with it once you’ve got the capability. There’s all kinds of stuff we can do in the future to accelerate it and to offer new kinds of sharing and things like that. Once you get it working and people go, "Yeah, it works. Now what?" That’s the next ten years. That’s the fun part. That’s why I’m still here. I’m still enjoying the creative side of this as well.

Do you ever wish the Vita had some extra shoulder buttons to match the DualShock?

Yeah, that’s what we need. I agree. It’s funny because you know I hope we get involved in more and more conversations as we become integrated into more things at Sony. These are the kind of conversations I really want to have.

How’s the PS4 beta going?

It’s what we expected. There are a lot of people playing and a lot of people are buying games and playing them for long periods of time. I found myself going for hours playing. I just completely forget that this game isn’t local. You’re just playing and it’s just working and you just forget. When I experience that, that’s what makes me go, "Wow, there’s so much potential in this."

This is what we have to achieve for everybody. Everybody that touches it needs to get to that. This is why we remain in beta. We’re constantly thinking of stuff. New ways to do things. And we have a large team here. The team has grown significantly from the way it was before. It’s over twice as big now. And because of that we’re able to actually take the time to experiment and to try to look at what’s happening out there and try to find ways to tune it.

Ultimately, my personal goal is to try to level the playing field for video games versus all other forms of media. This is really annoying for me – people think that gamers will play anything. So it’s okay to take a whole bag of really lame games and sell it to them. In reality, that doesn’t work. And we’ve seen it time and time again. But if you give gamers a selection of games it takes no time at all for them to work out what are the best ones. It’s very important to us that we get really high-quality PlayStation games across all these devices.

It’s like if I go to my television right now, I have a smart TV at home the games on there are dreadful. If you’re mass market and you buy a television like mine, you’re going to be offered really bad bowling and things like that. Netflix has Gravity and Avatar and all this kind of stuff and then you go to the games section and you’re like “Oh man, it’s bowling and darts.” That’s where we need to put The Last of Us and the best that our industry can make. That levels the playing field for the industry, and I think we’ll see it grow. There’s a really good game out there for everyone, whatever their taste is. You just have to help them find it. So if you think about where that could go in the future.

What does it take to get a game onto PlayStation Now?

If we required the developers to come back and modify their game to get it to work, it would massively reduce the library that we could put onto the service, and that’s not acceptable to us. So whatever work we have to do on our end, operating system-wise, we’re going to do so that every game can go on the service. That will go on forever at the development cycle. But the reality is we have to modify the underlying OS instead of saying, “Please find that team." A lot of those teams have moved on. They don’t even know where the final, final, final source code is, so you just don’t want to go that way. You want to always try to think about how you can unlock the past. I would love to be able to unlock every game ever on any 3D platform. That would be great. I can’t do that by requiring all the teams to touch their code.

Is that something that you handle on your end?

No, that would be Sony of Japan doing this. That would be at an operating-system level and a developer level as well. 

Is it possible to change control schemes to work better with modern controllers?

The question that will have to be asked when software runs in the future, something that we’ve never had to think about before as normally it runs on the device you made it for, is "What am I running on right now?" It could be a lot of different things. It has to say, "What is the best control system for the device that I am now running on?"

I like the idea that it's something that would be done down the road. It’s not built in yet. We have a few other things we have to do before we do it. But the idea of having the users able to make suggestions or to create shared control systems I think is a very healthy thing. It’s a very good idea. 

How do you stream multiplayer games?

It’s complicated. You know when you’re playing a multiplayer game you used to sit down and worry about the ping times between different machines or from different machines to a server? If everyone’s playing from the cloud, then all of the multiplayer machines are on the backbone, which means the speed of communication between the different people is just incredible compared to what it used to be. And then, of course, you have your final connection to the user.

I can’t reveal what technical stuff we’ve been working on. We’ve got really cool tech that Sony saw that I think helped close the acquisition. We demonstrated what the future of extraordinarily fast communications between the server and the user to the point where it’s indistinguishable from playing locally. We showed Sony actual demonstrations of it running and it played, the latency was completely invisible. And because of that it makes us very bullish on the potential of all this. The multiplayer’s running on the backbone, and the game feels like it’s running locally. It would make you then wonder which version you’d want to play. Do you want to play the cloud version or do you want to play the local version? The answer is whichever one I win the most in.

Do you see PS Now incorporating smartphones, tablets, and browsers in the future?

It turns out that anything with a screen you can do cloud gaming on. So what we’re going to do is look at each thing that comes along, and we’ll definitely put our service on it but we’ll then test it and see if it works. If it doesn’t work then we’ll just pass on the device. If it’s really cool and the gamers like it, then we’ll unlock it and let people use it. You saw us working on everything [before Sony]. We worked not just on web browsers but also on the gaming sites and YouTube and Facebook and everywhere. There’s nothing that stops you from putting this everywhere online.

Click the image above to watch a tour of PlayStation Now.

How do you plan on winning over skeptics of cloud gaming who are concerned with lag and degraded visuals?

Screen resolution is nothing more than a function of bandwidth requirements, so the bigger the resolution the more bandwidth you need. And if you look at broadband in the United States it’s a wall that’s continually rising. I have enough bandwidth at home now with 100 mbps. It’s huge. I could stream any resolution I feel like, but it depends on how long it takes for this to become universal. We’ve chosen [a level] for the resolution that we think is appropriate that will let an awful lot of people play.

The second thing is there are new compression technologies that keep coming out that people are having all kinds of breakthroughs on. The [H.264 codec] is current, so all of the devices today will stream H.264, but in the future you’ll see more and more devices supporting H.265, and that creates a surprising reduction in the amount of data to the stream experience. So sometimes it’ll be the technology that’ll help. Sometimes it’ll be the bandwidth that’s going to help. It’s really our jobs to choose a high-quality, acceptable resolution a lot of people are going to experience. 

How do you entice players who maybe had a bad experience with cloud gaming in the past?

It’s actually not always that. Sometimes, for example, if you were to plug in an ethernet cable versus using your wifi that might be enough to do it. So it’s education in that situation. That’s one of the reasons we do the stream test. We'll test your connection and make you aware because somebody who thinks that they’ve got a good connection might get a message that says you’re not ready to stream. They’re going to go “Huh? That doesn’t make any sense. What’s wrong with my network?"

We’ve found some really interesting networks out there where people don’t realize that they’ve got some issue with their router or something like that that’s causing them to lose a lot of packets but they didn’t know. By realizing it and fixing it their entire internet experience improves. Sometimes they have a very low bandwidth on their network and it’s because their wireless router is very far away, or it’s old hardware, or there’s some security system.

Some of the old routers people will turn on all the security whatever that is and the chips in those old routers are so weak that the router starts hiccupping data and their entire internet experience is losing packets of data constantly. They just aren’t really aware because the browser just runs slower and they just think that’s their connection. It’s been very interesting as we’ve gone through this process. I think it’s one of our challenges as we have to help people. If it works that’s great, but if it doesn’t work don’t give up just yet. There still might be something simple that you can do here. It won’t just get this working, but will probably improve the internet in your house.

What future plans are you able to discuss at this time?

I’ve been a gamer now for 30-plus years. For me the thing that draws me to games is to experience something new that I’ve never experienced before. When I see something and go, “Wow, that looks really cool.” I’m drawn to try it. I want to see what that’s all about. If you just play forward longterm and start doing thought experiments — What if you could run on any device and run anything in the cloud?

The game’s going to say what am I running on. It’s also then going to have to ask what I want as far as power goes. The game can actually someday ask for as much power as it wants and someday ask for as much hard drive as it wants. You know developers like [Hideo] Kojima, right? He would love unlimited power and storage. If you think about what that will do as far as unshackling the developers that always had to be concerned about limitations. You could just completely let them go wild and free. It’s hard to even imagine, but somewhere, someone at some point will do that, and I’m excited to see what those experiences are going to be like. It’s only going to be possible from the cloud.