Header

the apple launcher

COLLABORATORS
Avatar
zachary saale
Project Author
Avatar
taylor guidon
Hardware Engineer
Avatar default
Joseph flasher
Lead Developer
Avatar default
margot mausner
Producer

WHAT IT DOES

To help launch Redd’s Apple Ale, me and my team built an apple launcher and connected it to the internet. In addition, we constructed our own bar, and then asked users to use the Launcher to help us cause as much destruction as possible.


Bonus events like quadcopters, zombie takeovers, and even a dunk tank kept users coming back for more throughout the near two weeks that the live-streamed interactive game took place, and the result was a 12-day long series of hilarious non-stop entertainment!

#LAUNCHREDDS

WHY WE MADE IT

Miller Coors (owners of Redd's Ale) approached us and had a rough idea of what they wanted. They really liked the idea of a shooting gallery of some sort and that's when our design team put together some great mockups of what an interface would look like.   We thought, “What if you got to throw apples in a bar like in the commercials?”


We got back to them with our final concepts, and they loved it!




HOW WE MADE IT

Once Miller Coors greenlighted our idea, we had the daunting task of building an Internet-connected camera set, and then crafting a game like a theme park shooting gallery with what were basically heavily modified softball pitching machines that could be controlled from the Internet. 


We decided to give each connected player 60 seconds of target practice. We settled on that number because we were using Ustream for live streaming and the delay could be pretty heavy sometimes. We wanted people to actually see the shots they had planned out. You would get three shots or 60 seconds, and you could pick all of your shots before it actually started firing.


The set encompassed almost 4,000 square feet. Since it was going to be on camera, we also needed a tall ceiling. None of us at the time really had much 3D design skill so we contracted a set designer to help us figure out what to build. We then hired a set construction company to help us hammer all those details out. It was essentially a 3-wall set because one side was open for cameras, lighting equipment, the launcher and everything else.


We used three Arduino Megas for pretty much everything on set. One acted as the sensor controller, connected to the actual targets. Each one of those targets had a vibration sensor behind it and all those vibration sensors were hooked into an Arduino with an Ethernet shield communicating with our server whenever a target got hit. 


We were first going to custom-build a launcher so we could fire real apples, but then ended up using red street hockey balls instead. We realized as we were doing some camera tests, that the projectiles were just a blur so it doesn't really matter what they were. Also the project ran for two weeks, so we realized that if we were to shoot actual food, it would have been a nightmare in terms of cleaning everything.


After we settled on the hockey balls, we started to figure out that probably the best thing would be to just purchase something that was already built. We found the best pitching machine that we could buy that was also both the sturdiest and the simplest. On top of that, we also found a ball feeding mechanism and modified it using an Arduino so just it could be triggered by users on the Internet. 

WHAT I LEARNED

Overall, the project went really well. Running it for two weeks was really interesting because we were in this dingy warehouse in DC and at times there were power problems. It rained one day, the second floor had crack in it, and we started getting water on all of our components. 


There were times, like during the rain incident, that we would need to turn off the components. We would instead just watch the screen and see where people were aiming, then one of our employees, who was a great pitcher, would be just off camera throwing a ball where the user was aiming.  Generally, the person we had throwing the balls was as accurate if not more so than the pitching machine. 


It was challenging having so many concurrent users that were on the same page of a website but were at times experiencing different things. For instance, you have the people who are in control, the people who are just watching, and then the people who are actively participating and attempting to take control. That aspect was a really interesting learning experience. 

what's next

After this project, we did the Nickolodeon Skill Crane, which was almost the exact same technology and was incredibly simple because of everything we learned from this project.