Crytek Interview, Part III of III

MTBS is pleased to be joined by Jens Schöbel, Technical Designer (and stereoscopic 3D wiz) for Crytek, developers of the upcoming Crysis 2 video game and CryENGINE 3 game engine!

Today, the final part of our mystery tour to find out what they have cooking in 3D!  Check out part one and part two if you haven’t already.

1. While I think our industry is filled with companies that are excited about 3D gaming, they are equally unexcited about working together as a cohesive industry.  Sony, Nvidia, AMD, and more are all very proprietary in how they position themselves.  Even when they are announcing 3D specifications or “open initiatives”, they are firmly tied to their brand, and their brand alone.  While the biggest adversaries got behind 3D movies through Blu-Ray, stereoscopic 3D gaming is playing out very differently.  Do you agree with this viewpoint?  Do you see game developers acting as the cohesion for the stereoscopic 3D gaming industry?  What do game developers need that they aren’t getting?

Well, let’s answer the last question first. In my opinion, most of the game developers need more guts. It is quite easy to say, “We never have done this before! It will not work.” To have guts means, to look at your company’s core idea. One of Crytek’s core ideas was and is being technologically cutting edge. That explains why we implemented S-3D into Crysis 2.

This is a technology no AAA title ever has done before, so that was a good option for us to choose.  Nevertheless stereo 3D for us means being and staying cutting edge. That’s what other developers need: A clear goal where they want to be in five or ten years. Overcoming the burdens of the non-cohesive industry is harder than setting your own company vision.

For cohesion, you should remember that stereo 3D gaming is currently a niche. It will and is getting stronger. We are going into S-3D and we clearly know that this is a good bonus. As stated earlier we don’t change the story or game play too much in the 3D version of the game. Our clear goal is it to push 3D. This doesn’t necessarily hold true for other companies as well.

There is too much risk to break down a whole game if you focus on Stereo 3D AAA production only. For me, that seems to be the reason why every company is doing it differently. If the acceptance of stereo 3D in the home rises, the market will simultaneously get bigger. If developers could calculate the risk of stereo development it would be easier for them. From my experience, it is worth trying to add stereo to your game. You learn from your experience even if you don’t ship in stereo at the end.

2. We are starting to see video games exclusively branded by specific 3D hardware solutions.  What is Crytek’s position on stereoscopic 3D branding?  Do you want to support one or all?  Why?

In an earlier question I was talking about missing standards as our biggest villain…but at the same time these missing standards were our biggest ally.  If you focus on one specific brand but there is no standard, there is a very high risk you will disappear if another standard makes an appearance. Remember HD DVD vs. Blu-ray. We all know who made it in the end.

The same currently happens to stereo 3D. Every hardware vendor wants to support as many modes as possible since nobody wants to lose. This is good for us, since we have the ability to focus on widely supported modes. And we will implement all missing modes step by step. We started with dual head and implemented more and more and more. But luckily at every point during development we had S-3D available and could experiment with it. The situation is better now but still there is no 100% clear answer on what will be the future modes. HDMI 1.4a looks very promising, but right now we don’t stick with specific vendors.

3. One day soon, we are going to be in a scenario where 3D HDTVs are much more popular, but still have a majority of 2D HDTVs in the market.  Do you see game developers charging a premium for the 3D mode in future games?  Why or why not?

Since I’m not familiar with the business model behind S-3D, I have no clue at all. I’m passionate about games and game play. If game designers would give me a cool stereo game play experience, I’d pay.

4. When we met at FMX, we were part of a handful of presentations about stereoscopic 3D gaming.  I would bet that by next year, there will be a lot more cooking in the 3D gaming industry.  What risks and problems would you like to avoid with this onslaught of 3D gaming content?  Do you have some ideas on how we can avoid these risks?

I think it’s more a chance than a risk, to be honest. With Crysis 1, we pushed the graphical boundaries for real-time graphics a lot further. I’m pretty sure Crysis 2 will set the benchmark for AAA-S-3D gaming. Maybe we should not try to avoid risks, but instead welcome our possibilities.

The more companies develop in S-3D, the more everyone can learn. Avatar: The Game for example impressed us a lot and showed us also some flaws.  The guys from Ubisoft did a great job in general but totally forgot the HUD. The HUD created exactly the depth violations that I described earlier. This increased our sensibility and of course we’ll avoid that with Crysis 2.

5. You’ve had the chance to check out MTBS’ 3D Game Analyzer.  This is more a gamer-driven Quality Assurance test than an official stamp of approval.  Do you think game developers should get together and develop official 3D QA standards for the industry?  An official stamp of approval?  Under what conditions?

You can answer this question from two perspectives: pure error free visuals caused by wrong programming, and the S-3D experience that changes if you train your eyes.  I’ll give you an answer on both views.

In the latter case it’s a question of opinion. Everyone experiences S-3D in a slightly different way thanks to our eyes and some people can’t see 3D at all for example.  We gained all that knowledge over the  development time. At the beginning we assumed that everyone sees everything in the same way. While developing we always have our glasses lying next to us and every time we implement something new, we can check it immediately.

At the beginning of this process it took our eyes about 10 seconds to accommodate. About one year later my eyes needed about half a second. My eyes were trained to react twenty times faster. At FMX 2010 someone called me ‘The Arnold Schwarzenegger of eye muscles’. This is one of the most important lessons that I learned over time: every eye is different and the 3D feeling changes if you learn to see stereo 3D.  If an official stamp could take this into account, you as a player could easily decide if that game is perfectly suited for your eyes.

The part about error free visuals points into two directions. First, a stamp of approval is good for players who want to buy a game just for the S-3D experience. A list like you have at mtbs3d is very helpful in that case. Second, for us as developers, a stamp of approval would make sense…if we could get it during (!!) production.

Basically you want to avoid risk if you develop an AAA title. In general you get a stamp of approval after you finished the product. Unfortunately at this time all your risk management already paid off or did not. You need something before you finish a game.

Let’s take a closer look to S-3D in this case. That’s a rather new technique in games and at Crytek we want to do it properly. A stamp after development if the game is out would tell us what S-3D glitches we hadn’t solved. But a stamp or guidance during development could help us to avoid pitfalls. If you want to develop a good S-3D game you should follow the latter one. So rather than a stamp of approval, a stamp of “Developed in S-3D” would make more sense for us as developers.

6. Last question, flip a coin!  What will drive 3D to the home: movies or video games?  Why?

My coin would land on its edge. Why? Since both will have its driving force but they are different in details. These details will give us, the gaming industry, a better long term run.

Before answering you need to understand one basic thing in stereo 3D: size matters. That means the depth perception of a 3D picture changes with its size. If you take a 3D picture and scale it down, the depth perception change, and the smaller picture has more depth. That’s a very important thing. The depth perception changes and in a worst case scenario, you get eye strain.

Now let’s go back to the cinema. A cinema screen is big, err…no, it’s HUGE. A film like Avatar is perfectly suited for 3D on such a huge cinema screen. Imagine now you scale that film down to the size of a TV screen, and you have the same problem: the depth perception changes. And to make it more complicated not all the TVs have the same size. There are possibilities to overcome this problem, but they are not easy.

Games have the same problem, but they don’t suffer from them.  In games, everything is interactive and easily changeable – and so is the 3D effect. With one variable you can give it more depth or less depth.

And now you may understand why my coin lands on the edge.  The cinema movies will bring the stereo 3D experience to your home, since lots of people want to see the films they liked at home and in 3D. But the games will drive it to the future. They easily overcome the burdens of changes in depth perception. Games will create the better 3D feeling at home.

Special thanks for out to Jens Schöbel, Crytek, and Electronic Arts for making this interview possible.  Clearly, Jens had a lot of interesting things to share, and now it’s time for your opinion.  In addition to sharing your remarks, please complete The U-Decide Initiative survey.  The results of this study about gamer opinions around 3D could help shape games to come!  A draw of over 50 prizes will happen when the study is complete (free 3D HDTV anyone?).  Both traditional 2D and experienced stereoscopic 3D gamers are welcome to participate.  You do not have to have experience with stereoscopic 3D gaming to fill out a survey.

Share your remarks below!