Tim Sweeney Criticizes Microsoft/Oculus and Talks Project Scorpio

I got the chance to interview the esteemed Tim Sweeney. You might know him as one of the founders of Epic Games, the developers behind the Unreal Engine. Fun fact: Sweeney wrote most of the code that went into the original Unreal Engine, and it has gone on to become arguably the most popular game engine today. As someone who has been inducted into the Academy of Interactive Arts & Science’s Hall of Fame for his contributions to the gaming industry, he is someone who’s extremely well-respected for his technical brilliance. He’s also an effective communicator and is a straight shooter who pulls no punches.

In my interview with Sweeney, the CEO doles out harsh criticism for Microsoft and Oculus, and suggests that both companies are using anti-competitive business strategies. At the same time, however, Sweeney also defends Oculus’ right to garner platform exclusives.

I also get to pick Sweeney’s brain on a wide variety of topics that include Project Scorpio, the new cadence of next-generation consoles, VR, AR, advanced AI, and a whole lot more.

GameSpot: With Microsoft announcing Project Scorpio, what do you think of it from what you know so far?

Tim Sweeney: Well, upgrading consoles midlife is a great step forward for the industry. The most difficult part of the entire console industry has always been resetting the user base to zero every seven years or so, whereas if we incrementally upgrade the hardware, then we can bring more performance to gamers without wiping the slate clean. That's incredibly valuable. To be able to invest the kind of money it takes to build a high-end game today requires a base of tens of millions of users. By doing these incremental updates, the industry can move forward technologically at a much faster pace without those business challenges. I think it's really the ideal model: bringing the best upgradeability of the PC with the reliability of the console kept, so you're never going to have to deal with driver problems on a console, but you will get the newer hardware.

Upgrading consoles midlife is a great step forward for the industry.

Tim Sweeney

I was just going to say, is it me, or has the console cadence changed maybe to become more like PCs or perhaps more like smartphones?

Yeah, I think this is a fundamental change in the way the console industry operates. I think it's been a long time coming in recognizing this is possible. You could imagine this approach being extended out in the future where once we have a hundred million gamers on the current generation of consoles. No console company would desire to start over from scratch, but would continue to deliver more and more performance with hardware upgrades every few years. I think that's an incredibly smart approach, both to benefit gamers and to make industry economics more palatable for game developers.

Do you see the current generation and the next generation of consoles, as it were, following a similar model to the PC, in that a low-tier PC and high tier PC can play the same games but at different settings?

Yeah, I think that's really brilliant. We're already building games that can scale by a factor of five from low-end to high-end. We have absolutely no trouble taking a game like Paragon built for a 1.8 teraflop [click here for “What is a teraflop?”]PlayStation 4 and scaling it up to a whole lot more performance. Yeah, on the PC, we're already dealing with the Nvidia GeForce GTX 1080, 9 teraflops of performance. The game is awesome on that. As consoles gain performance, we will be eager to take advantage, and we won't need to leave the earlier console gamers behind. We can continue to sustain, continue to run on the current devices for the foreseeable future. Yeah, maybe a decade or more.

With PCs having such a wide variety of hardware variations out there, do you see different graphical settings coming to consoles as well?

Well, already we're going to now have to deal with two discrete levels of performance. I think that's really good. The problem with the PC is it's such a continuum. Pick any performance number and somebody has a PC with that number. It makes it hard for us to make specific choices about what to do on this machine, and what to do on that machine. I think console with two performance points is going to be a very easy target. We can decide exactly what features we can turn on in each of the two cases, and really polish the game heavily for both cases and not have to worry about all of these intermediate points.

Could you see consoles coming out at an even faster cadence, maybe bi-annual or even yearly at some point?

Well, let's see... there's a cost and complexity to an upgrade bidding. Upgrading every year would be ... would create so many points that developers would have to consider and support. That would be hard. I think revisiting console tiers every several years, as there's potential for significant performance gains, is very interesting.

You've previously said some pretty harsh words in regards to Microsoft about monopolizing game development on PCs. You wrote an op-ed with The Guardian. Do you still feel strongly about that?

I'm not a general critic of Microsoft. I just have one specific concern. ...my view is that Windows 10 is an awesome operating system, it's the best yet. Windows Store is a good store, and it's great to have another source of games out there and it's great that Microsoft will be bringing so many of its Xbox games to PC. I think all of these are great steps forward for the PC. My sole concern, and it's a big one, is with UWP (Universal Windows Platform). Microsoft creating a framework by which they could very easily shut off the PC as an open platform in the future. There's a real risk if developers adopt this technology widely and Microsoft flips one switch and suddenly you can't install games from sources other than Microsoft on some versions of Windows. That's a fear I continue to have. It's the reason that Epic has not adopted that, the UWP technology. Despite feeling that there are some great technical features in it. It's a safer Windows API that prevents a ...It runs in a sandbox like iOS and prevents applications from behaving like a virus, yeah. But my view is that Microsoft has to make clear commitments about where this technology will go in the future, if they want developers to adopt it; because if they do not do that, it would be foolish for us to go down that path, which basically gives them all of the rope they need to hang us.

I'm guessing that the controversial Windows 10 automatic updates do not sit very well with you then.

Microsoft has to make clear commitments about where [the Universal Windows Platform] will go in the future, if they want developers to adopt it; because if they do not do that, it would be foolish for us to go down that path, which basically gives them all of the rope they need to hang us.

Tim Sweeney

Well, as a user, I'm not a fan of that. I think Apple has a good balance in making it super easy to update your operating system, but not mandatory. I think that would be an improvement. I think it's perfectly valid to have automatic updates be the default. It is the right answer for most users, but the problem with the forced patches is that they bundle things that fundamentally change the rules of the platform. Now, Apple doesn't do that. Apple's never released an iOS update that takes away fundamental rights you have to use their platform, whereas at Microsoft, they've actually been doing that. One of the forced updates had lock screen ads, and now, your original Windows, which used to be under your control is now just playing ads that you have no control over. Another forced update prevented Windows Pro users from disabling the Windows door on the platform, which these corporations buying Windows Pro thought they had control, you know, their IT departments thought they had control over that aspect of the system, but then Microsoft changed the rules, and that's a really nasty thing, Microsoft giving itself the mandatory power to change the rules on you at any time, and as a user, there's nothing you can do about it. You know, that's wrong. It's that culture that exists within Microsoft. I don't think everybody at Microsoft feels that way, but the culture that's driven by some of their executive leadership believes that it's not your PC, it's their PC, and they can do what they want with it, and they'll do whatever they can get away with. It's constantly figuring out what they can do without users going into open revolt and discovering more and more everyday where that boundary is.

The [Microsoft] culture that's driven by some of their executive leadership believes that it's not your PC, it's their PC, and they can do what they want with it, and they'll do whatever they can get away with.

Tim Sweeney

Moving on, how do you think the launch of modern VR has been, and what's your outlook on the medium's future?

Well, I think it has been a success of the scale that I hoped for, and expected. It's such a new platform with such new techniques required for developing. I think we have to really compare it to the very early launch of the computer revolution. You remember 1980? *laughs* A lot of people might not. Around 1980, 1977 or so, Apple came out with the first mass-market manufactured personal computer, the Apple II. You can still buy these on Ebay for a couple thousand bucks. In their first year, they sold seventeen thousand Apple IIs. Next year, they sold a hundred thousand, and that was the start of the personal computer revolution. We have to realize that VR is starting the same way. It's going to be very small, but what's going to save it in the long run is that it's growing exponentially. I think when Sony comes into the game later this year, it's not going to take very long to get up to a million units from the one hundred thousand HTC Vives and some number of Oculuses in the marketplace today. That number will double, triple, or quadruple every year for the next few years until it reaches many tens of millions. At that point, it will be a huge market that really can sustain and fund AAA game experiences. In the meantime, while it's really solid, this is an awesome opportunity for new developers, especially small and nimble developers to get into a new medium and be one of the founding developers to explore the possibilities.

We have to realize that VR is starting the same way. It's going to be very small, but what's going to save it in the long run is that it's growing exponentially.

Tim Sweeney

In regards to VR, back in March you tweeted, "It's very disappointing that Oculus is treating games like sources from Steam and Epic Games as second-class citizens." Do you still feel the same way today?

Um, yeah. My complaint with the Oculus Rift drivers is the same as with the universal Windows platform at launch. They disable loading of third-party apps by default. As an Oculus user, if you want to play a VR game from Steam, or download a game from Epic, or run the Unreal Engine on VR, you have to go into your settings in the platform and turn on this option to allow sources of software other than their store. That shouldn't happen on PC. You know, your Logitech mouse doesn't try to force you to use the Logitech mouse store. That's crazy. The PC should be open by default. ...I really hate to see all these companies try to usurp control over what users are allowed to do on their operating system, or on their hardware. As a PC user, you should be free to install any software from any source and do what you want. That enables competition where the best companies, the best products, the best services, and the best hardware all win, as opposed to massive corporations trying to force you or compel you to do things their way.

To just play devil’s advocate here, at the end of the day, Oculus, Facebook, they're a business. If they think that they've done studies that shows that they have the highest rate of success doing it that way, what can you say against that? Maybe from a consumer side it sucks, but is that fair from their perspective, if that's what they see?

As a PC user, you should be free to install any software from any source and do what you want. That enables competition where the best companies, the best products, the best services, and the best hardware all win.

Tim Sweeney

Well, if that's your moral framework that anything that is profitable that a big company can get away with is okay, then I think you've got a lot of bad decisions. The fact is that PC gamers aren't idiots. Gabe Newell is the smartest person in the PC industry because he fundamentally realizes it. These gamers are smart; they know what's happening. When companies do this sort of thing, it pisses them off. Everybody wants to have control over the computer. They want to have complete freedom to install anything from any source. They don't want any company's product forcing them to do things against their will. There are things Facebook can get away with on its social network because you're kind of locked in with you and all of your friends being there and it being very difficult to migrate off of that to a different social network, but that's not the case on PC at all. Anybody can download Steam, anybody can download or buy Oculus hardware, use their store. The best [VR] store is going to win, and anybody who interferes with that is just making themselves look bad and pissing off their users, and ultimately losing the high ground.

The fact is that PC gamers aren't idiots. Gabe Newell is the smartest person in the PC industry because he fundamentally realizes it.

Tim Sweeney

On that note, do you think in order for Oculus to take the lead here that they are going to need to re-position themselves to be more open?

The best [VR] store is going to win, and anybody who interferes with that is just making themselves look bad and pissing off their users, and ultimately losing the high ground.

Tim Sweeney

I think that would be really valuable, especially treating it as a platform that's friendly for all sources of software, all developers, all publishers, including publishers that run their own stores, is a great step forward. Also, with Oculus, they have awesome hardware. We just absolutely love it, and we love what they're doing to help nurture the industry to success. They've put a lot of funding into projects to make possible games, awesome games from developers that would not have been economical otherwise. They're making the kind of forward investments in content that Sony is making. I think they're just a few policy tweaks and mindset tweaks away from excellence there.

You mentioned Oculus as providing funding for some games that wouldn't exist otherwise, so then are you okay with them having their own exclusives to their own platform?

Well, I think that's a perfectly valid business model. It's one of the ways to nurture an industry before it's viable itself. If you look at what's led to the success of the PlayStation and the Xbox platforms, it's been the … what was key to their launches were the awesome first party titles that those companies funded for millions of dollars, or tens of millions of dollars, or even hundreds of millions of dollars to ensure that there were great titles available at launch that really defined the platform. Every one of these services has exclusives. Valve's games are exclusive to Steam. Oculus is funding games that are exclusive to Rift. Sony's funding games that will be exclusive to PlayStation 4. It's a sound business model. As a user, you might prefer that every game is available for every platform, but you should also recognize that a lot of these games, especially these early games that are putting bigger budgets into VR experiences could not have possibly been funded by the developers without funding from the platform provider. They adjust because of that model, and I think we're better off with them than with the alternative of not having those games and having a completely non-exclusive set of ecosystems.

As a user, you might prefer that every game is available for every platform, but you should also recognize that a lot of these games, especially these early games that are putting bigger budgets into VR experiences could not have possibly been funded by the developers without funding from the platform provider.

Tim Sweeney

What do you think about augmented reality?

Well, when we think about the long term future of these platforms, I think ultimately we'll see these helmets with high-end VR hardware reduced to the form factor of your Oakley sunglasses, which work everywhere you go and provide a fully immersive experience coexisting with reality. I think that's the ultimate destination of the industry. The ideal that is perhaps fifteen years away, and so I think it's an idea whose time will come, and is rapidly being developed with Magic Leap and Microsoft with Hololens., Right now, they're powering different sets of applications, but over time they'll convert, and over time they will be reduced to such a level of consumer convenience that they'll be indistinguishable.

I think ultimately we'll see these helmets with high-end VR hardware reduced to the form factor of your Oakley sunglasses, which work everywhere you go and provide a fully immersive experience coexisting with reality.

Tim Sweeney

Do you think that we'll ever approach photo-realistic graphics? If so, when, and how many Teraflops do you think we'll need?

You know, we're getting to the point now where we can render photo-realistic static scenes without humans with static lighting. Today's hardware can do that, so part of that problem is solved. Getting to the point of photo-realistic dynamic environments, especially with very advanced shading models like wet scenes, or reflective scenes, or anisotropic paint, though...maybe forty Teraflops is the level where we can achieve all of that.

With humans?

Not with humans. Humans are the harder part, but that's just ...We know exactly how real world physics of lighting work, and so that's just a matter of brute force computing power. Give us enough computing power, and we can do that. We could do that today with algorithms that we know. Humans are a much harder problem, because rendering faces and skin is hard enough, but you quickly realize that the challenge with rendering humans is having realistic human motion to display. Having dynamic human responses in games are reactive to what you're doing, and aren't just pre-baked. As you're interacting with a real human, their eyes are constantly moving with you, the eye contact is super important. You're picking up the emotions on their faces, and they're dynamically responding to you. If you just used a perfect motion captured human, a flawless motion capture with future technology, it would still be uncanny and not feel like it's a real human interacting with you. To do completely photo-realistic rendering of everything, you have to simulate realistic humans and actually simulate human intelligence, emotion, and thinking. It's not a matter of computing power. If you gave us an infinitely fast computer, we still don't have the algorithm. We have no real clue how the brain works at the higher levels. You might understand how one neuron interacts with other adjoining neurons, but the large scale structure of it is still a complete mystery. That could be unpredictably far away. Once we are able to simulate human intelligence, what's going to separate humans from people? You're talking singularity level stuff at that point, but I do think that we're many decades away from having that ability.

Speaking of AI, Is Epic heavily involved with the creation of AI?

Well, you know, we have a very advanced game AI system in the Unreal Engine. It powers games like Paragon and Unreal Tournament and Fortnite. It could do some really amazing things in terms of helping characters locomote around the world. It's doing nothing that resembles simulating a human mind. That's what the intention is with the deep learning research at Google and Facebook and other companies. That's a technology that we're just starting to learn about, and I could see heavily investing in over the next decade. For right now, the rubber isn't meeting the road in that area yet. It's still just a pipe dream for us, and hasn't been at all a practice in the game industry.

Can you envision a scenario in which you get the AI to reach such a level that it almost becomes indiscernible to real people, like a sort of a Turing test, if you will?

That's really hard. Really, it comes down to finding ways to structure these AI algorithms to simulate the brain in the meaningful way. The human brain is very complex. It took, gosh, how many hundred million years of evolution to get from single-celled organisms to us? There's no other animal in nature which really even comes close to doing the sort of things that a human does. Clearly, there are some very hard design problems in that you would imagine that eventually computers should be able to simulate that, but huge amounts of research may be needed to get to that point. But I think that idea is, if you take enough of these deep-learning pipeline stages, you can get more and more of the human level of cognition up and running in a game. Certainly these self-driving and just collision avoidance systems being developed for Tesla, for example, do a very good job of image recognition now, and they're able to solve some of the problems that a brain is able to solve. Overtime, we're just going to have to find ways to solve more and more of those different stages. How do you deal with emotion? How do you deal with facial recognition? Some of those problems are slowly being solved. How do you deal with complex problem-solving? I think as AI researchers are piecing together all of these different deep learning components, they're going to reverse engineer the way our minds work without actually doing any biomedical research in the process. It'll be very interesting to follow.

I’ll leave closing statements with you. Is there anything you’d like to share with our readers?

Tim Sweeney: Three years ago, there were ten thousand users spread among just high-end AAA studios. We opened the Unreal Engine up to the public through a subscription mechanism. Eighteen months ago, there were a hundred thousand Unreal Engine 4 users, and now there are two million. Since we've gone free, anybody can go to our website and download it, we've seen this huge uptake in the engine among game developers of all sizes, including a huge number of indies and a lot of people outside the game industry. It's been shaking up the state of the game industry now. If you look at the eighty titles shown using Unreal Engine 4 at E3, there's a lot of the best titles in the industry, a lot of Microsoft's first party and Xbox titles, like Gears 4, and Sea of Thieves from Rare. Sony's Days Gone, Batman Arkham Asylum for Warner Brothers. Super high-end all the way to some of the best indie games in the industry. It's really a whole different state of the engine industry than there was a few years ago, so that's the thing we're excited about right now.