Square Enix on Next-Gen and Why The Uncanny Valley Will "Always" Exist
Worldwide technology director Julien Merceron talks to us all about next-gen and Microsoft and Sony's "biggest mistake"
"Next-gen" has been on the lips of just about everyone, as developers and publishers prepare for the inevitable next round of consoles. At E3, however, Microsoft and Sony were quite reticent to talk about any hardware beyond the current generation. Meanwhile, Epic was proudly displaying its Unreal Engine 4 next-gen technology and the folks at Square Enix unveiled "Agni's Philosophy," a beautiful demo leveraging the publisher's Luminous engine, which shows off just how good Final Fantasy might look on next-gen platforms.
GamesIndustry International caught up with Square Enix's worldwide technology director, Julien Merceron, to talk about the Philosophy demo, the arrival of next-gen consoles and what that means for Square Enix and the industry at large, and why Merceron feels that no matter how good next-gen looks, the uncanny valley problem will always exist.
Dig into the full Q&A below.
Q: To start, I'd like to ask about the "Agni's Philosophy" Final Fantasy tech demo and Luminous engine you guys showed, which represents what Square Enix feels is possible for next-gen. Can you tell me more about the technology and what you are hoping to achieve?
Julien Merceron: The way we approached it is really to look into what would be great for us to be able to do moving forward. We tried to go as high-end as we could in terms of content and assets. At the same time, we tried to see how far we could push this when it comes down to real time. As you know, we're focusing on RPGs and games where characters are extremely important. So definitely we needed to deeply think about facial and body animations and to deeply think about dynamics on the characters and how we light these characters and how we integrate them into environments that we want to provide to players in the future. There is probably going to be more content coming in the next few months in terms of the technology we used from the demo, but if we look at it from two years, 'where are we going to land?' we are very, very interested in real-time lighting for many reasons.
First of all, for us it is really important to provide real-time editing pipelines for our artists. So having a real-time solution for lighting, real-time solutions for navigation, real-time solutions for almost every aspect of the game ends up giving a lot of freedom to our artists, allowing them to iterate and polish way faster. We want to push real-time as much as possible and we are obviously deeply working on PCs that allow us to prototype and see what all the best approaches we believe will work or won't work for us.
Q: One of the more solid rumors we've heard about next-gen is that Sony would be moving away from proprietary tech and dumping the Cell concept, moving back to an off-the-shelf solution. Would that be something you and other developers would welcome?
"We have Sony and Microsoft talking about this generation lasting 7,8,9 or even 10 years and it's the biggest mistake they've ever made"
Julien Merceron: Oh definitely. There are two aspects to complex hardware. When hardware is too complex to target, it leads to a smaller number of titles available for launch, and then with fewer titles you end up not being able to sell your platform really well, so you actually start with a big disadvantage. PS3 and 360 haven't found a way to be as successful as PS2 was. On the other hand, if you have complex hardware, although you have problems at the beginning, you do have potentially better longevity for your platform because every year quality is going to increase as developers find out all the optimizations they can use.
But with a simple architecture you do give more chances to everybody, which I believe is very important based on the critical business situation we're in. Games will be more costly. If you start to make the entry bar really high, more studios will die, more publishers will die, there'll be less titles on platforms, etc. If you make it accessible, you give more chances to people, you'll have a better portfolio at launch , but now you also have a problem with your longevity. Studios could be 80 percent perfect one year after the console has shipped, so games for years afterwards will look the same.
Now, I would say that this is not a problem anymore. It was a problem in the generation of offline platforms. Now you don't need to manage longevity by complexity of programming, because your longevity is ensured by your online model. And I would suggest that maybe we don't want long generations. We have Sony and Microsoft talking about this generation lasting 7,8,9 or even 10 years and it's the biggest mistake they've ever made.
This generation has been way too long, and I say this because you have a lot of developers that work on a new platform, and perhaps will not succeed, so they will wait for the next generation, and will jump on that platform. You could not do that with this generation though. So these developers went elsewhere to see if the grass was greener. They found web browsers, they found iOS, they found other things and a lot of them won't come back to the hardware platforms. So you could look at it that thanks to Microsoft and Sony and the length of this generation, it helped the emergence of other platforms and helped them get strong before the next hardware comes out.
Q: What do you think the leap forward will be for the next-gen consoles? How much better will Square Enix's games look on the next-generation compared to now?
Julien Merceron: First of all, in the industry it is a bit divided. You have people that say graphics are good enough and that pushing graphics is not the most important thing going forward. You have other developers that say that they can really push graphics. You look at the Philosophy demo and you can see what more advanced graphics could bring to a game. I'm really talking not only about the technology and how cool it is to work on graphics, but also from the perspective of new technologies enabling directors and how they can provide more immersive environments. Focusing on graphics only would be a huge mistake. You start to have super great graphics, characters look really good and you end up in the uncanny valley, but you don't have animation at the same quality level. Same thing with behavior and AI; it animates well and looks good, but it is making stupid decisions. It simply won't be immersive.
So focusing on graphics only is a big mistake. People will need to focus on animation, and I believe our Philosophy demo is highlighting that, but also on the AI and physics too. So all of these aspects will be extremely important and we can't forget to talk about online because we definitely see that whether it is synchronous or asynchronous multiplayer, it definitely is adding a lot to the games. Whether it is social features with your friends or playing and watching your friends play, if you look at the web, which is an important platform for us, that is definitely a domain where we want to innovate.
Q: You mentioned the uncanny valley, and I think developers have done a good job in recent years as the technology has gotten better. David Cage's Beyond is really showing things off quite well. I think that maybe the uncanny valley is fading away, and perhaps the technology is reaching a point where it is okay. What do you think about that?
Julien Merceron: I don't agree that is it okay. I think Beyond is trying something interesting. For current gen, it is going way further in terms of quality, but it is too complex to implement on next-gen. You would have to sacrifice too many resources in memory. I think Beyond has the right balance, but at the same time what they have is probably not where we would want to be in the future. There are good examples in the industry. Killzone 3 had great facial animations back in the day. There are great teams that, for this generation of consoles, have done good quality that could be considered good enough. The problem I am talking about is, as soon as we ramp up the quality on graphics, this level of quality on facial animations won't be good enough. Some games succeed to be at the limit of the uncanny valley, but … the problem is that as rendering quality will go up, new problems will surface. The quality of the facial and body animations and the acting won't be good enough. So that is why as you evolve, you have to upgrade your physics, rendering and animations.
"Uncanny valley will always be a problem... You can't have these stunning graphics while characters are acting funny on the screen"
Q: So in your opinion, the uncanny valley will always be a problem with developers?
Julien Merceron: It will always be a problem. As you push something, you have to recalibrate and rebalance these out. You can't have these stunning graphics while characters are acting funny on the screen. My point is more in terms of immersion. When we put you out of the game, that gives you a chance to put the controller down and you might not come back. So you don't want to break the immersion. You want to get the players in the world, suck them in and then make sure that there are as little problems as possible to make them stay. The gameplay needs to be involving, the environment needs to be immersive, and the characters need to be full of emotions so you are totally sucked into the experience. This is why we have this role of balancing all these things so we end up with something that is consistent and works well.
Q: Looking at what Nintendo has been doing, I think a lot of people were expecting a bit more from them. Graphically, it doesn't appear to look better than a 360. Have you looked at the tech behind the system? What is your take on their approach?
Julien Merceron: If you look at the recent Nintendo platforms, the DS, 3DS and the Wii, it seems like they are trying to avoid aligning too much with Sony's and Microsoft's strategies. They are trying to take a different path, a different perspective on things. The type of entertainment they provide focuses not so much on graphics or immersion, but more on high-quality and innovative gameplay. They try to be different, they try to find a different market and they try to solve problems that they perceive as being the most important to solve today. So yeah, Microsoft and Sony are trying to solve different sets of problems. Their focus from a consumer perspective is completely different.
I strongly believe that the Wii U is going to provide new ways of interacting with games and there is definitely a community out there that is interested in that. Now there will definitely be another community that won't be too interested in this, and will be interested more in Hollywood blockbuster type of games with cinematic gameplay experiences.
So really I think that if I take a step back on this, I think it is one of the best times for consumers. Today, you can play on iOS, the web, the 360, PS3, Wii, Vita and so many other platforms. Because of the differences, they get a wide variety of experiences. I think it has probably never been as great as today to be a video game player and Nintendo is helping to bring this variety. We saw on the Wii that they helped broaden the market and it benefits everyone. iOS is helping to broaden that market as well. Maybe in five or six years these players will be hardcore players. I would say that all the work Apple and Nintendo are doing today to broaden the market will be one day beneficial to Sony and Microsoft. I hope the new platforms that Sony and Microsoft are making will choose different directions to offer even more variety and diversity for gamers.
Q: A lot of people would say that consoles from Nintendo, Microsoft and Sony may not matter in a few years with stuff like the iPhone, iPad and with cloud technology offering full streaming. These console makers may have to completely change direction.
Julien Merceron: I agree that they will have to look at the market in a different way. I think they understand what is happening these days and I think they understand that the current generation has actually been too long.
Q: I don't know if they had much choice financially; it costs so much to make these systems. Just look at Sony's position as a corporation right now.
Julien Merceron: It's true, it's true. It is all about trade-offs... but I think this is probably the reason why they will do things differently. This is why they announced SmartGlass... This is a sign that they understand what is going on and it is a sign that they understand that they somehow missed something and they are going to catch up, they are trying to adapt. It is true for publishers as well. You build up your IPs, and you then see that things are moving and you have to reinvent your IP. If you see things are changing, you have to find ways to entertain your players because the habits are changing.
These players on browsers and mobile have a lot of social connections and when they go back to consoles they feel alone. That's why social features in home console games are starting to explode. It is because people need different things; they want to share immediately. It is about sharing right now. Everything gets more and more connected and you see these manufacturers reacting to the realities of today. To get back to your question, I think they will adapt, I think they are really smart and understand the world of today and anticipate the world of tomorrow and will act on that.
Q: Does that mean that they will not make consoles after the next round? Are they going to do something different? Are consoles going away?
Julien Merceron: A very good question. If you go back 10 years ago, there was no software platform. It was hardware or nothing. Today, you have many software platforms. Whether it is Facebook or web browser, everyone is going to software platforms. If you look at the trends, the number of software platforms is going up, and in comparison, the amount of hardware platforms is probably going to reduce. It is way cheaper to launch a software platform compared to a hardware platform. Now we are in a world where people want to play from anywhere. They want to be connected always; they want to be connected to their friends, to the games. Even the designers of today cannot think about 'that game' on PS3 or 360. They have to think about the experience of a consumer from when he wakes up to when he goes to bed. He has to be able to stay in the world through mobile, tablet, console, TV and web.
So when you make a game today, you make a game and 10 apps, or 10 games. The way you make a game today is very different from how you made games 10 years ago. So the question is, 'how is that going to play in favor or against hardware platforms?' You do need to carry a device to access something. After that it could be cloud computing, it could be cloud storage, it could be an app or it could be a game. Hardware will always be there, because we need the glass; we need something to touch and interact with. That said, the way these games are designed and played will change dramatically.
"I think these kinds of technologies and experiences on the web browser platform will be very instrumental in the world of tomorrow whether we're talking about software or hardware platforms"
Q: You mentioned the importance of web browsers. Where do you see that going?
Julien Merceron: People talk a lot about hardware platforms and obviously these hardware platforms are instrumental to the type of entertainment that we tend to design at Square Enix. That said, we shouldn't underestimate the kind of experience that you can have in a web browser. Immediately you can play from anywhere. You have browsers all over the place. These cloud storage and computing interfaces will provide better ways for consumers to interact with the experiences they want to have.
The other point is that we've been focusing on the web browser experience for the last two years. We've developed 3D, high-end technologies for the web as well as a classic type of experience as well. There was a very interesting problem we had to solve early on, especially for our 3D high-end experiences. When you make a game on the console, you have to wait for something to load. If you have to wait that long with a web page, people are gone; they don't like to wait that long. We've had to re-architect the engines we are using. We've had to stream things way more efficiently, so you can click and see things, but a lot of content is going to stream in the background. I think these kinds of technologies and experiences on the web browser platform will be very instrumental in the world of tomorrow whether we're talking about software or hardware platforms.
Q: What is your opinion on Flash? The Unreal Engine has been applied to Flash and Epic has been working closely with Adobe, producing some great results. Is that something you've experimented with?
Julien Merceron: You don't need Flash to do DirectX11-type graphics on the web. We've designed a plug-in that lets us use DirectX in a web browser. You don't need Flash. One of the problems I have with Flash is that it's not really multiplatform. Square Enix is a publisher. We focus on many platforms from iOS to high-end PCs. In the middle, you've got probably 12 platforms that we have to support with our own game engines. For us, it is very important to focus on multiplatform engines and scalability of assets and content in the games to really enable our game teams to do a good job. Flash, to me, is platform specific.
When it comes down to the web and a web browser, I think the plug-in approach is the most adaptive to solving problems. That said, we are also trying different things with Google-native APIs, HTML5 and also other games we are making with Java and PHP. So obviously we are trying other approaches to ensure we've done our homework so that we can move in the right direction. I think the plug-in approach for high-end 3D gaming is better for many different aspects.
Q: One thing that Square Enix has had to deal with recently is a bit of controversy first surrounding Hitman and then Tomb Raider. With the technology you produce allowing for greater realism, does the developer now have a greater responsibility?
Julien Merceron: It's a difficult question to answer as a tech guy. There are many styles of movies and games that I like. I'm not sure that I'm good at judging anymore where the line has to be drawn or where a limit has to be. There are things I love about some of the characters because of the context of it. In Hitman, there are these criminals dressed as nuns, and I see them as criminals trying to get rid of Agent 47. I couldn't actually look at it and see how people would react because I didn't see Agent 47 killing nuns or religious people; I saw terrorists disguised as nuns.
And the world of Hitman is a world of disguise... so I saw it as funny and pretty interesting, but definitely when you take it out of context I can understand why some people reacted the way they did. You suddenly realize that some scenes in games are borderline. I think we should definitely be more careful about how we communicate about things, explaining the context and being careful about what we publish. The last thing you want to do is to make people annoyed about things that are supposed to be entertainment. So we have learned from that.