Interview: Illium VR- Virtual Reality Controllers
Company: Illium VR
Industry: Video Game
Product: VR Controller
Team: Sebastian Sarbora (CEO), Robert Rouhani (CTO), Jazmine Olinger (COO)Education: Rensselaer Polytechnic Institute (all co-founders)
Major: Computer Science
- Product: A realistic gun controller to be used with virtual reality games.
- Funding: Bootstrapping (self-funding) initially, grant money from competitions, eventually received some investment.
- Gaining Initial Traction: Have developers use the controller and make games compatible with it. Once games are made for it, will launch to consumers. Most success in attracting developers came from introductions.
- Manufacturing: For now, ordering parts and assembling them in-house.
- Distribution: Aiming to have product bundled with other products/video games.
- Y Combinator: A startup accelerator that takes a small portion of equity in return for funding and advice. Must apply and get accepted. Helped Illium VR in many ways and is a great resource for the life of the company.
- Getting Investment: Use entrepreneurial resources around you. Connect with other entrepreneurs to learn how to pitch to investors and make valuable contacts. After networking, Y Combinator might be a good option to help you get investment (and help launch your business in general)..
Table of Contents
A. What is Illium VR?
C. Gaining Initial Traction
D. Developing Skills Needed to Make Controller
E. Attracting Game Developers
F. “Y Combinator”
G. Manufacturing and Distribution
H. Getting Investment and Funding
I. Copyrights and Patents
A. What is Illium VR
- What is your product, the “VR controller?”
Sebastian: Our product is a controller that is designed for virtual reality platforms to provide a more realistic experience. Our first VR controller will be a gun because that is the biggest, attackable segment of the game industry. The next biggest one is action games, but for these games it is hard to make one cohesive controller. We developed our own hardware and software to interact with the games and because virtual reality is so new and advanced, you really have to involve the game developers that would want to use our product. We have to send them the controller and allow them to download the software. It is possible to make games that are made already compatible with our VR controller, but it is way more time consuming than simply contacting game developers and asking them if our controller is something they would like to have.
- Will you make other types of VR controllers?
Sebastian: Well, let’s start off with guns. Guns are very scalable in a sense because we can make one gun product, but then make a couple others in a few more years (one that is specifically sniper rifles, one that is specifically rocket launchers). That is more of the medium term.
We would like to bundle our controllers with other games. The long-term beyond that is that we want to make many different types of controllers, including surgical instruments for surgical training (such as scalpels that feel as if they are touching real flesh when touching virtual flesh) or a tennis racquet that feels as if you hit a real ball. We want to be known as the guys who are making the best peripherals for virtual reality—for any experience that you want.
- How did you three meet?
Jazmine: We all came to RPI at the same time. The three of us became friends and started going to hackathons, where we first started working on little projects together. One of the first things we did was an exercise band. It was like a Fitbit but not as good. Then, we went to a hackathon at our school, called hack RPI. That was our first big effort to do something together. Around that time, we started working on VR controllers at a hackathon, but we always intended to continue it as a business. Basically half the time we were in school we were working on this together and we just graduated.
C. Gaining Initial Traction
- What is your strategy to market, advertise, and get traction?
Sebastian: We want to ultimately launch to consumers, but you cannot launch a hardware product to consumers if there is no content for them to use. With our development program, we are reaching out to developers to make games using our VR controller. Once we have some great buzz about different studios partnering with Illium VR to make a game that works with our VR controller, it will be time to sell the controller to the consumer. That way consumers will have a bunch of different games to play with the controller, rather than just one or two.
D. Attracting Game Developers
- How do you get game developers to make games that involve your controller?
Sebastian: There are a lot of components to that. First make a controller and send it out, which means that we have to turn what were previously prototypes into a beta product that can ship to people. Then, there is the side of actually convincing people to use it. We have done a lot of marketing stuff and primarily Reddit (a forum for discussing different topics) is what we use to reach out to a lot of people. There are a lot of big virtual reality sub-Reddits. We also have to make sure that the product is super easy to use. Most developers are on a crunch schedule, so they are only going to spend a couple of hours fiddling with the product before they decide whether they want to use it.
The next part of contacting game developers is developing good leads. We might find a game developer that is making games for virtual reality platforms, but the games are not gun games, or we may identify a developer that is working on gun games, but the games are not meant for virtual reality. Our sweet-spot is gun games for virtual reality platforms. Yet, we found that sometimes people that you don’t think would be interested at all are really interested because of what they are working on behind the scenes.
- Are you focused more on contacting game developers directly or getting the word out and hoping that game developers will contact you and use your controller?
Sebastian: In the past we tried to get any possible developer to use our controller, but we have moved toward asking developers that are more likely to use our controller. It is important that before you move to a targeted approach that you have some content up (website, etc) or those that you are targeting will not take you seriously. Recently, Jazmine has been handling targeting developers, reaching out to different game studios.
Jazmine: Most of our success in attracting developers has come from introductions. We have sent out some cold emails to studios as well, but the response rate on those is pretty bad.
- Do you plan on launching your own game titles as well?
Sebastian: We are primarily not a content provider because we believe we can make controllers better than anyone else and believe that others can make content better than we can. However, we do have some ideas for applications that can be used with our controller. We are working on a couple of games internally. We have a tech demo that we already released. Most videos that are online are showing that tech demo, but we are working on a game for our product launch that will showcase everything that the controllers can do. Since the game is not intended to generate revenue, but rather promote the VR controller, we would give out the game for free.
E. Developing Skills Needed to Make the Controller
- How did you three develop the skills needed to build a VR controller?
Sebastian: We learned the fundamentals from our computer science classes and expanded through extracurricular activities, such as hackathons. At hackathons, you go for 24 hours to finish a project. If you don’t have the skills to make the project, you learn them from your teammates. There is a low barrier to get in and there is no risk if you fail (i.e. losing money, reputation, etc).
Another thing that we have at our school is an undergraduate research program called RCOS (the Rensselaer Center for Open Source) where you can work on anything you want and get class credit for it. It is a great way to build skills while working on something you are passionate about.
At Hack RPI, we also learned about the business end. You have to raise a lot of money to run a hackathon (our school budget is around 70k each year). By getting companies to sponsor, you learn to sell to companies (almost like business to business). You also learn to get users, which in this case are the people who go to the hackathon. That helped us a lot in gaining the skills for outreach and raising money.
Robert: I would say that the other thing is that I went through the game development program at RPI and that was immensely helpful for me, especially with game engine stuff I am working on now.
Jazmine: I agree that extracurriculars were the most important for building your skills, but foundation classes were very good because if you teach yourself to program, you miss a lot of basics. The Freshman and early Sophomore classes are very useful to take, and then start doing stuff on your own. The focus of higher level classes tends to be less on programming and more on being a computer scientist, which is great for graduate school and higher level learning, but not for making stuff.
- What programming languages were most helpful for you to learn?
Sebastian: It depends on what product you are making. From the perspective of virtual reality, almost all hardware and low-level operating system stuff is in C++ and C. There are pretty much no other languages that will go on your hardware/firmware stuff—the stuff that will go on your actual product. A good part of our backend is written in that. But then there is the user facing stuff, like the game engine, which Rob primarily writes in C #. What we are using internally is not necessarily what people are using externally. C # is great because it is compatible with many game developer’s engines and games, but is not good for the low-level stuff because it literally does not work.
Jazmine: I think an important thing that most people don’t get when they are starting out is that it is not really about learning languages, it is more about learning what each language is good for. It is not that hard to pick up the syntax of a new language—once you learn one you can pick up the syntax of another. Knowing when to use Python and when to use C++ , how they work, and why they are better at different things, is something people may not get and it makes it harder when they use a language that is not well suited for the project they want to work on.
- How do you learn what each language is good at so that you use a language that is well suited for the projects that you are working on?
Jazmine: You should do research and look at what languages are used for similar projects. If you have a more formal education, you might understand the backbone of languages and be able to figure it out on your own. Otherwise, you can just Google it or look at articles about the differences between languages.
F. “Y Combinator”- (FAQ List See https://www.ycombinator.com/faq/ )
- What is Y Combinator?
Sebastian: Y Combinator is a startup accelerator. We got accepted into their fellowship program. They are one of the most popular in Silicon Valley. Airbnb, Reddit, a bunch of big name, multi-billion dollar companies have come out of it. It is an important place to start, especially in the tech space.
- Why is Y Combinator so helpful for a startup company?
Sebastian: They are good at figuring out how to solve problems. Every time we went there, they had some good, actionable advice that we were able to implement and got us further. In our case, we were very focused on crowdfunding and fund raising and they said to us that we had prototypes, so we should start selling. It was simple, but we had not even thought of selling yet.
- How do you get accepted into Y Combinator?
Sebastian: They have an application process with two groups of startups a year. Thousands of companies apply and a small portion get accepted. They emphasize good teams and good ideas, which is all you need to get past the application stage. Traction is good, but they may let companies through the application process without that much traction because a lot of companies are just apps that haven’t been made yet or haven’t gotten customers yet.
Once you get past the application stage, they fly you out to Silicon Valley for a ten-minute interview, so you have to boil down your entire company into ten minutes. We applied about three times.
Jazmine: We applied a year ago for the summer of 2015 for the first time and got the interview. A lot of people apply and don’t get past the application stage to get an interview and it is hard to give advice for those people. If you don’t make it past that stage, they won’t give you any feedback, making it hard to know why they did not accept you. I would say adjust your idea or team until they get past that stage.
Once you get past that stage, they give very honest feedback as to why they turn you down. More often than not, what they tell you is “wrong” about your company is probably “wrong”. For our first interview, we did not get in and they told us what we had to work on. We applied again for the following winter and did not get in again, but they offered us a spot in the fellowship program instead. Essentially, this time they said we were not far enough along and the fellowship program was designed for companies that were in the earlier stages. We applied three times, did not get in each time, and might apply again. So just because you did not get in does not mean you should not apply again. People get in on the second, third, or even fourth interview.
Rob: During the interview, it can get stressful. I would say try to stay calm and bring a demo, I think that helped us a lot. It is only ten minutes and it was hard for us to setup in ten minutes, but it was worth it.
Sebastian: Also, Y Combinator is very good at cutting through any fluff and figuring out what you are doing and how well you are doing it, so don’t try to impress them with fluff. Being honest, direct, and succinct is most helpful. Y Combinator is also very well known, so your company gains much validity and publicity by being a part of it.
Jazmine: They can also able to introduce you to pretty much anyone. So not just Y Combinator alumni, but they will probably know people in your industry.
- Do they coach and mentor you throughout your time with Y Combinator?
Sebastian: The fellowship program is different from their standard program in that you can do it remotely, which was great for us because we have manufacturing and assembly where we are. It is difficult to uproot that and move everything to San Francisco and then come back after three months. We did occasionally fly out there and talk to them in person.
When we first got accepted into the fellowship program, there was a twelve-hour kickoff where they showed everyone’s websites and how they needed to be improved. They also talked about how to pitch to investors and a bunch of other topics. Every week, they would have a YouTube stream where they would have high profile people talk to you and let you ask questions. Sometimes they have events in person as well. You can also email the people at Y Combinator at any time and ask them questions.
Jazmine: Even though the program is over for us, we can still email them and ask questions.
G. Manufacturing and Distribution
- How do you handle manufacturing and distribution?
Sebastian: We are operating on a small scale. We buy the different components to the controller (electronic parts, the controller shell, etc) and assemble them ourselves. When you are doing 1,000+ units, it makes sense to go to China. We also have a variety of measures in place to do QA testing and other things that will transfer over when we move to a big manufacturer. We might have them use our software to test all of the boards or test the end product.
H. Getting Investment and Funding
- How did you fund Illium VR initially?
Sebastian: We bootstrapped and I think you have to spend enough money early on to figure out if your idea is worth pursuing and then look for funding after. There is a huge shift from working on a project to working on a business. Usually, right after you cross that point of having a business is when you look for investors. In the past we have gotten funding from grants and investment.
Jazmine: Everyone starts out self-funding to some degree. In our case, we didn’t have much money so our self-funding was very limited. However, if you are self-funding for too long, you can’t really have a business. If you are self-funding in the beginning because you have the money, that is a good way to get started, you need investment, revenue, or some sort of outside money to validate that you have a good idea.
Sebastian: One thing that definitely helps is that we started our business while we were still in school. A lot of our labor costs were lower because we could use the school’s resources to manufacture our product. For other operations, we could have other students help us inexpensively (designing, etc). We were able to enter hackathons and other business model competitions which were for students exclusively, and we had housing, and that offset many of our costs. I think the first year we operated with about $15,000 and that allowed us to develop our product, do outreach, and go places.
- How do you get investors?
Sebastian: You could go to Y Combinator right off the bat, but I would recommend you use the entrepreneurial resources around you. We live in Troy, New York. You don’t hear about it being a central hub for technology, but it still has a very active entrepreneurial community. As long as you are in a reasonable metropolitan hub, there are people who are starting businesses. Learning from them and getting that mentorship is hugely important because investors are fickle creatures and if you don’t know the exact lingo, it can be hard to get their attention, even if you have something that is amazing.
I. Copyright and Patent
- Is a copyright or patent very important for a technological invention like this?
Sebastian: It is in some ways important. In terms of investment, it is super important. A lot of investors don’t want take to the risk of investing without some sort of intellectual property protection. There are defensive intellectual property strategies to prevent people from doing the things that you are doing or to prevent them from preventing you from doing the things that you are doing.
However, for virtual reality, it is a such a fast-moving space that many of the traditional questions of intellectual property don’t come up. You can see that Oculus and Vive are fighting about who stole the other’s technology, but in no way is a lawsuit coming up. They are just saying “Hey, they copied us, you shouldn’t like them.”
Copyrighting is another good thing. Every line of code that gets written is automatically copyrighted. Big courts have just recently ruled that you can copyright API’s. We have an API so that helps solidify our product.
Patenting is one way to protect your invention, but not necessarily the most effective way. We are looking into it, but it is not our focus right now. It is just as important focus on your place in the market. For us, that means getting developers and users. It is kind of a situation where the one who gets there first gets everything. Oculus came first and sold about 500,000 copies of their first CD while Vive sold about 50,000 although, in some ways, Vive is a better platform.
For Further Reading:
Illium VR Compared to Other Systems
- Who would you consider to be your competition?
Sebastian: People initially might think that Vive or Oculus is our competition (platforms for virtual reality games), but really they are not. Sure, they make motion controllers, but they could care less about making realistic controllers for realistic applications. Same thing goes for other big players such as Microsoft (with the Microsoft HoloLens). They are focused on selling their platform. Even the Leap Motion is a companion product. They just track the hands so that is something that could be used with a gun controller.
Most of our competition, such as Trinity VR, has tried things and failed for various reasons, such as they did not have good enough content. For the most part, the field is clear, but there are some players to be worried about, such as PlayStation VR (by Sony). They could potentially be a partner, but they could also compete. They are known for having third party developers, make gun controllers for games, Therefore, a partnership with Sony is something we want to explore before they become our competition.
- Is your VR controller meant to be used for systems such as Oculus and Vive?
Sebastian: Yes, but it will not be included. It will be compatible with both Oculus and Vive. Our Oculus version works in its own way with the Oculus and our Vive version works in its own way with the Vive.
- Do you have any plans for compatibility with the Microsoft HoloLens?
Robert: From what it seems like, the HoloLens is more focused on projecting things in a room and using hand tracking. I am not exactly sure it is our target market right now.
Jazmine: It would be cool for an AR type shooter, but I don’t know if that is going to takeoff.
Sebastian: Also, I think that right now, there is a pretty big distinction between virtual reality and augmented reality. In augmented reality, you have to take the world as it exists and put things into it, and in virtual reality, you are creating a whole new world. They have different challenges and different technology to solve those challenges (in virtual reality, people can throw up if the frame rate and other mechanics are not right). Virtual reality is relatively compatible from our end, but augmented reality, because those challenges are different, is not something that we are tackling right now.
- Can you tell us a little about how the VR controller works?
Sebastian: The key thing that we want to do is make a very realistic controller. A lot of times, virtual reality systems will have very generic controllers and they are meant to do everything not very well. So making the controller physically realistic makes the experience much more immersive.
A lot of the technology is mechanically and electronically simulating a real gun. For instance, most guns have a slide mechanism that allows another chamber to go in, so that has to be represented and the motion of that has to be relayed into the game. We play around with recoil technology a lot. You can have the low-end where the controller vibrates, or the high-end where the controller has a CO2 canister in it and works like a BB gun.
We are also exploring attachments. For example, you can put a scope on your gun and it will be represented in the virtual reality game. Technology wise, all of those are subsets (recoil, attachments, etc) that have their own electronics and connectivity. When you attach them to the controller, that information gets relayed to the game.