VR Gun Imagn' Games

It's kind of funny that your controller turned out nearly as similar in design as mine did. When I started, I was using an M4 shell, but due to space limitations and the number of buttons required to make the controller useful, I ended up using a SCAR-L instead. I had to make a few circuit boards using Fritzing and have to solder the components on as part of the production line, but the biggest issue I have with the project is going to be taking my physical object and turning it into a 3D CAD file for export to a machine shop which will make the injections molds I'll need to mass produce the shell.

Glad to see there's a market and other ideas out there. The DeltaSix will never be what we need for a moving un-grounded screen like the Rift or other VR headsets. I'm working on the final button arrangements for my controller now and hope to be getting some feedback from my friends who are gamers to make design changes for a final prototype, but I'm on #3 now after some control and input issues.

The other side of things that I find just as important as the controller itself is the software that makes it configurable and interact with the games. It has been a real trip making this project and I hope that your endeavors end as well as you envisioned.

Oh, and here's a picture of my project...the cord coming from the barrel is a temporary one that has yet to be activated, but I didn't want to lose it during my last modifications...I am adding a D-Pad on the right side on the mag well for actions such as crouch/prone/jump and a forward/backward action that might be something like spot/run commands which need to be at hand, but out of the way from the standard controls.

Well, the code is kind of the heart of the project as well as the components and how they all work. The idea on how it works is also nothing new, so the parts are rather standard for making this item, but the magic is how they all work together and the software, like a conductor of an orchestra, that takes the random bits and turns it into a lovely work of art for all to enjoy. Hell, once I make them and ship them, anyone with a screwdriver can disassemble it and get the parts list, so I don't hold it near and dear to my heart. When I get the controller with all the buttons on it (have a 3-set and a D-Pad to finish) I'll probably take another picture of the insides to compare to the picture I attached.

I attached a picture of the insides of my proof-of-concept prototype I made first. Once I knew it could be done with the right parts and testing, I heavily changed the design of it. I attached a few pictures of that proof of concept design just to show you how ugly it usually starts. Note the 2 4xAA battery holders in the barrel, the Arduino UNO in the magazine, the +5V distribution block in the grip, the L3G4200D above the trigger, 15-pin VGA connectors for quick disconnect of the UNO from the rest of the gun and the N-size DC power components to disconnect the batteries from the gun (for assembly, it was necessary that the barrel be removed from the gun at times). Even at 22ga, those wires take up a lot of room in the bottlenecks.

It's been nearly or over a year (off time only - I have a regular day job that takes up a lot of my time) and about $6000 spent on various tools (high-end soldering iron, work bench, organization bins, 3D printer for prototyping buttons and knobs) and materials (ICs, switches, resistors, pins and headers, screws and nuts). When I first started, I thought I'd take my idea to Logitech or Mad Catz/Saitek and market the idea to them, but then I remembered the way the world works with beancounters who make certain decisions and I would undoubtedly lose control over the end product, I couldn't turn it over and walk away. I had to see it though myself.

The list is pretty simple, though...
Arduino Micro for the brains and heart
L3G4200D accelerometer for sensing movement
JY-MCU for serial bluetooth communications to a standard bluetooth receiver on the computer
2-Axis joystick with select button and breakout board for WASD controls
resistors/headers/CD4021BE for the buttons to serially be input to the micro (16 buttons with only 3 pins)
resistors/headers/74HC595 for the LED outputs for status indications (not yet fully implemented but ready)

The hardest thing about this project is finding the space for all of it. I had to design a circuit board that takes all the black/red wires from each button and puts it on a 8x2 set of headers and then include headers for the jumpers that connect my intermediate board to the Micro on the appropriate pins. Then there are headers where the L3G4200D and JY-MCU connect to my intermediary board. The intermediary board makes it simple to connect everything together in a very repeatable way to minimize time constructing. The trigger is a lever snap action switch with a black wire of Ymm and red wire of Ymm (X and Y are lengths), the Grenade button is a 3/4" momentary pushbutton with a black wire of Amm and red wire of Bmm (A and B are lengths) so that I can buy a ton of the switches, cut the wire to the proper length, put the connectors on the end of the wire, then snap it into the appropriate location on the header of the intermediary board.

This also helps with parts lists and figuring out pricing. If it takes me 3 lever switches that cost $1.35 each, 5 3/4" pushbutton switches that cost $3.69 and 8 tactile switches that cost $0.30 each, uses 8 feet of black wire, 7.5 feet of red wire, then the other lengths of wire added in, the cost of the pieces of the shell to be replicated, the Micro/Accel/JY-MCU/Joystick, my circuit boards and parts for them, the screws and nuts, the battery, battery charger circuit and parts, then a bit added in for the time it takes to put it together for recouping the expense and time for development, it will give me a number that I put on it for the cost of making it.

If I order enough to make 100 of them at once, the bulk pricing will bring down the cost of the materials, the shipping can be spread over each gun so that $80 for shipping is $0.80 added per gun or I could eat that cost out of the profits I'll receive from making them.

As I work on it more and more, it gets closer to being done. Just the other day, I put on my DK1 and paired my controller with my gaming/test bench computer (its a test bench as it has no development software or items on it like most users wouldn't - ensures my software will work as installed with no extras needed that somehow installed on my development machine that I forgot about) and played the Test Range on Battlefield 4 and went to town on the dummies. It was like I was 5 again on Christmas morning, laughing and giggling like an infant. Having the software finally turning the raw data from the controller into inputs for a mainstream game, the buttons, battery and Bluetooth all working as desired. I didn't have all the buttons fixed onto the chassis yet, but it inspired me to keep working on it more and more as it is almost there. The finish line is just around the corner. But the finish line is just this race. The next race will be manufacturing and reproducing this thing which I have very little experience at doing. I'm hoping some simple rubber molds will work. While inexpensive to make the molds, the mold itself is time consuming to set up properly and has to be redone as the mold will eventually wear out. If the rubber molds don't work out as planned, the more expensive (possibly requiring massive capital which I don't have) route would be injection molding. Before I can have someone make the injections molds, I have to get the physical object into 3D space first.

One of the reasons I had to make several different prototypes is that when I change a part like the joystick (originally was going to use a thin PSP thumbstick), the new hole for the replacement part didn't line up properly and ended up with a weird cutout that hindered with the functionality testing of the new part, so I had to throw out the old shell and make a new one exactly as the original minus the old hole. Sometimes I have to re-position an item and the old holes are still there, but until they interfere with something, I'll deal with them later. The biggest change was going away from the Colt M4A1 to the SCAR-L. I wanted a switch on the butt plate that you'd hit when you shoulder the rifle so it would shoulder the weapon in game, but the M4 gun I was using had a detachable stock which would make the button and wire a very difficult affair to deal with. Then there was an issue with space in the gun. The M4 is about 3/4 the volume of the SCAR and the forward handguard of the M4 made it impossible to do much with. The SCAR allows me to mount the battery and battery charging circuit board and still have room to spare for the wires and comfort.