Forum Discussion
FPS Revolution Controller
Hi all. Just wanted to advise of a new product I hope to be ramping into beta-production soon, something I call the FPS Revolution. If anyone has seen the Delta Six controller, it is similar, although I found it after coming up with my idea.
While just getting my teeth wet with the Rift and Omni and wrapping my head around what will be a very definitive and awesome experience with these two matched up, I couldn't help but notice the controller being used was a Wii controller in the very odd shells produced for them that kind of resemble a gun, but the button placement is awkward and touchy. A thought popped into my head that often does when I see something that is less than ideal "I could do better than that" and had a 2-week vacation that came up right after coming up with the concept.
So what was it I came up with? I went to Wal-Mart and found an airsoft gun, cheap spring version nothing fancy. I opened it up to find out what parts inside made it tick. I went to Radio Shack and got an Arduino Uno and a few switches and wire and went to town. Two weeks later after using a Dremel to get rid of a lot of excess plastic that wasn't needed anymore, I had a functional prototype of an AR15/M4 rifle that used to shoot BBs and now shoots data to your PC.
It uses an accelerometer to "feel" the gun turning, a few switches and LEDs to feed input and output and a Bluetooth module to send the data in a simple serial data stream to the PC where a piece of software I wrote takes the data and a configuration that allows you to tailor the software to different titles (COD, BF3, BF4, etc) with different buttons/commands for actions. Pulling the trigger, for example, will left-click the mouse button, swaying the controller will turn in the horizontal direction, while pitching the rifle will turn in the vertical direction, emulating mouse movements. It also has a built-in sensitivity for fine tuning the physical/virtual movement ratio. I also incorporated a "dead-man" switch so that you can engage/disengage the mouse input so that if the virtual movement doesn't match up with the physical movement, you can disengage the movement tracking, move the physical rifle to match the virtual one and re-engage the acceleration tracking.
The biggest issue I've had so far is that the controller, while it works great with a fixed monitor, is kind of like having a mousemat that is too small while playing an FPS where you have to move the mouse to the side of the mat, then pick up the mouse, re-center it, and continue the movement. You basically have to move the controller until you are uncomfortably wrenching your shoulders/arms with the rifle pointed away from the fixed screen, then disengage mouse tracking, point at the screen again and re-engage mouse tracking. While quick and easy to do, it is unnatural and requires a bit of thinking about what you want.
The Delta Six controller overcame that by having the controls work with a fixed screen by leaning the controller left/right to turn. With the Rift and any other HMD, the FPS Revolution will work just like an actual firearm. Point, trigger to shoot, push in the magazine to reload, hit the button that is in the fire mode selector spot to change the fire mode. Also, to go from hip-fire to shouldering the weapon and looking down the sights, there is a button on the pistol grip about where your thumb rests that hitting it will bring up the sights. Depending on how you want it, the software can interact with the game so that you can use the "shoulder weapon" button as a toggle (click to bring up, click to drop to hip-fire) or hold it to bring up the sights, release to drop it. The controls can be set up to counteract the design of the game as well. Battlefield 3 has a toggle sight capability, but Battlefield 4 has gone to a non-toggle hold-to-shoulder method I dislike.
I am in the process of scanning the pieces of the controller into 3D CAD files for replicating, just completed setup of my 3D printer and am hoping by Jan 1 to have a release version prototype. I created an interface board for easy assembly of the controller and have sourced all the parts I will need for the project. I am excited about getting this to market, not for a revenue sake, but because this an exciting time in gaming and I hope to bring a piece of gaming history to a thriving market that needs something to fill the gaps. I hope to have a website up and running sometime, but due to hardware development, have halted site production (I want something I can use with my on-order Rift to fine-tune) and while fine-tuning the project, I can work on the website.
I am working hard on it and am consumed by it (I work on the software, concepts, next steps, research all on my lunch breaks and then back to being a cable guy, crawling through attics, climbing telephone poles with my ladder and digging lines into the ground that have gone bad), but due to a busy work schedule am limited to working on it 2 days a week (my "weekend") and whatever time I don't fall asleep as soon as I exit the shower.
So far, I've designed it with WASD controls built into a foregrip, but due to space available, had to rework the idea to where I'm now using a PSP analog joystick for WASD controls (slightly forward for walk, full forward for run) as well as a series of buttons with custom plastic pieces to press the switches for crouch/prone/jump controls. The fine-tuning I speak of will be assembling the version 2 prototype with the PSP stick and custom buttons. Once I have a rounded prototype, I work with a lot of gamers that are pushing me to finish so they can "beta" the product for me and let me know of any issues in design. One thing I like is the responsiveness of the accelerometer and my program seems to have little to no latency. When I get my Rift, I'll be able to really test the latency and hope to use Arma 3 to test the concept of a disconnected head from body where the Rift controls view and the FPS Revolution controls weapon/body aim.
I have a very loose time-table as development time is limited, but I'm hoping that as far as I've gotten since late September, I will have something functional by January-March. Depending on feedback from my trusted tester group, I might have something to offer mid to late next year. I can't wait to see videos of the FPS Revolution, Virtuix Omni, and the most important piece, the Oculus Rift showing people what getting into the game REALLY means nowadays.
If you've read this far, you're either very bored or really intrigued. I hope those intrigued will be able to take the next step with me soon. Lots of things will change such as the Arduino Uno will be a Micro, the distribution blocks will be replaced by custom-made jumpers (by me of course) all connected to a central board designed by me and produced by Fritzing. There will also be the options for different colored RIS/Foregrip, pistol grip and extendable stock. The main body will remain black, but the furniture will be customizable. Original colors will be Black, Olive Green, Tan, Earth (light brown), White and the optional non-military Navy Blue, Red, Magenta, Purple, Pink, Green and Lime. I'm also looking to try to remove the M4A1 integrated carry handle and put a flat-top rail in its place and put a removable carry handle instead (not that you can see the difference with a Rift on, but I thought it would be a cool idea if I could do it). Also, I plan on making a lefty and righty version where the buttons and switches will be flipped as a righty's thumb rests on the left side of the grip, a button under the palm is far less useful and other differences to allow a lefty to have the same experience as a righty.
Ok. Enough text, now for the pictures (ignore the hairy legs and blue shorts in the photos, forgot to crop them out - and yes, that is my coffee table that I was using on my vacation to make my prototype)..
Early prototype with insides exposed
Flickr Page with Note boxes explaining major pieces:
http://www.flickr.com/photos/77021352@N08/10970982204/in/set-72157637892141526
Early prototype other side showing switches and buttons
Flickr Page with Note boxes explaining major pieces:
http://www.flickr.com/photos/77021352@N08/10971122543/in/set-72157637892141526
While just getting my teeth wet with the Rift and Omni and wrapping my head around what will be a very definitive and awesome experience with these two matched up, I couldn't help but notice the controller being used was a Wii controller in the very odd shells produced for them that kind of resemble a gun, but the button placement is awkward and touchy. A thought popped into my head that often does when I see something that is less than ideal "I could do better than that" and had a 2-week vacation that came up right after coming up with the concept.
So what was it I came up with? I went to Wal-Mart and found an airsoft gun, cheap spring version nothing fancy. I opened it up to find out what parts inside made it tick. I went to Radio Shack and got an Arduino Uno and a few switches and wire and went to town. Two weeks later after using a Dremel to get rid of a lot of excess plastic that wasn't needed anymore, I had a functional prototype of an AR15/M4 rifle that used to shoot BBs and now shoots data to your PC.
It uses an accelerometer to "feel" the gun turning, a few switches and LEDs to feed input and output and a Bluetooth module to send the data in a simple serial data stream to the PC where a piece of software I wrote takes the data and a configuration that allows you to tailor the software to different titles (COD, BF3, BF4, etc) with different buttons/commands for actions. Pulling the trigger, for example, will left-click the mouse button, swaying the controller will turn in the horizontal direction, while pitching the rifle will turn in the vertical direction, emulating mouse movements. It also has a built-in sensitivity for fine tuning the physical/virtual movement ratio. I also incorporated a "dead-man" switch so that you can engage/disengage the mouse input so that if the virtual movement doesn't match up with the physical movement, you can disengage the movement tracking, move the physical rifle to match the virtual one and re-engage the acceleration tracking.
The biggest issue I've had so far is that the controller, while it works great with a fixed monitor, is kind of like having a mousemat that is too small while playing an FPS where you have to move the mouse to the side of the mat, then pick up the mouse, re-center it, and continue the movement. You basically have to move the controller until you are uncomfortably wrenching your shoulders/arms with the rifle pointed away from the fixed screen, then disengage mouse tracking, point at the screen again and re-engage mouse tracking. While quick and easy to do, it is unnatural and requires a bit of thinking about what you want.
The Delta Six controller overcame that by having the controls work with a fixed screen by leaning the controller left/right to turn. With the Rift and any other HMD, the FPS Revolution will work just like an actual firearm. Point, trigger to shoot, push in the magazine to reload, hit the button that is in the fire mode selector spot to change the fire mode. Also, to go from hip-fire to shouldering the weapon and looking down the sights, there is a button on the pistol grip about where your thumb rests that hitting it will bring up the sights. Depending on how you want it, the software can interact with the game so that you can use the "shoulder weapon" button as a toggle (click to bring up, click to drop to hip-fire) or hold it to bring up the sights, release to drop it. The controls can be set up to counteract the design of the game as well. Battlefield 3 has a toggle sight capability, but Battlefield 4 has gone to a non-toggle hold-to-shoulder method I dislike.
I am in the process of scanning the pieces of the controller into 3D CAD files for replicating, just completed setup of my 3D printer and am hoping by Jan 1 to have a release version prototype. I created an interface board for easy assembly of the controller and have sourced all the parts I will need for the project. I am excited about getting this to market, not for a revenue sake, but because this an exciting time in gaming and I hope to bring a piece of gaming history to a thriving market that needs something to fill the gaps. I hope to have a website up and running sometime, but due to hardware development, have halted site production (I want something I can use with my on-order Rift to fine-tune) and while fine-tuning the project, I can work on the website.
I am working hard on it and am consumed by it (I work on the software, concepts, next steps, research all on my lunch breaks and then back to being a cable guy, crawling through attics, climbing telephone poles with my ladder and digging lines into the ground that have gone bad), but due to a busy work schedule am limited to working on it 2 days a week (my "weekend") and whatever time I don't fall asleep as soon as I exit the shower.
So far, I've designed it with WASD controls built into a foregrip, but due to space available, had to rework the idea to where I'm now using a PSP analog joystick for WASD controls (slightly forward for walk, full forward for run) as well as a series of buttons with custom plastic pieces to press the switches for crouch/prone/jump controls. The fine-tuning I speak of will be assembling the version 2 prototype with the PSP stick and custom buttons. Once I have a rounded prototype, I work with a lot of gamers that are pushing me to finish so they can "beta" the product for me and let me know of any issues in design. One thing I like is the responsiveness of the accelerometer and my program seems to have little to no latency. When I get my Rift, I'll be able to really test the latency and hope to use Arma 3 to test the concept of a disconnected head from body where the Rift controls view and the FPS Revolution controls weapon/body aim.
I have a very loose time-table as development time is limited, but I'm hoping that as far as I've gotten since late September, I will have something functional by January-March. Depending on feedback from my trusted tester group, I might have something to offer mid to late next year. I can't wait to see videos of the FPS Revolution, Virtuix Omni, and the most important piece, the Oculus Rift showing people what getting into the game REALLY means nowadays.
If you've read this far, you're either very bored or really intrigued. I hope those intrigued will be able to take the next step with me soon. Lots of things will change such as the Arduino Uno will be a Micro, the distribution blocks will be replaced by custom-made jumpers (by me of course) all connected to a central board designed by me and produced by Fritzing. There will also be the options for different colored RIS/Foregrip, pistol grip and extendable stock. The main body will remain black, but the furniture will be customizable. Original colors will be Black, Olive Green, Tan, Earth (light brown), White and the optional non-military Navy Blue, Red, Magenta, Purple, Pink, Green and Lime. I'm also looking to try to remove the M4A1 integrated carry handle and put a flat-top rail in its place and put a removable carry handle instead (not that you can see the difference with a Rift on, but I thought it would be a cool idea if I could do it). Also, I plan on making a lefty and righty version where the buttons and switches will be flipped as a righty's thumb rests on the left side of the grip, a button under the palm is far less useful and other differences to allow a lefty to have the same experience as a righty.
Ok. Enough text, now for the pictures (ignore the hairy legs and blue shorts in the photos, forgot to crop them out - and yes, that is my coffee table that I was using on my vacation to make my prototype)..
Early prototype with insides exposed
Flickr Page with Note boxes explaining major pieces:
http://www.flickr.com/photos/77021352@N08/10970982204/in/set-72157637892141526
Early prototype other side showing switches and buttons
Flickr Page with Note boxes explaining major pieces:
http://www.flickr.com/photos/77021352@N08/10971122543/in/set-72157637892141526
21 Replies
- Intrigued and impressed. I'm looking forward to hearing how your testing goes next year.
With the batteries stacked in front, how does that feel for balance?
For the foregrip, what sort of configurations and button type layouts have you been thinking of? If you don't mind I'd be interested in seeing a rough sketch of placement and types.
The first layout I imagined was a thumb controlling the WSAD movement, two rows of 3 buttons on the opposite side to be run by middle-pinky fingers. Maybe a single axis joystick set above the normal rest position for the index finger.
- That's awesome aparsons671!
The only reason I created an account here was to start a thread about some inputs on VR firearms but now I saw that I can't create a thread here so I am posting my thoughts in this threads instead.
I've looked at a couple of different VR guns and this is what I think should be changed;
The guns looks to plastic. Need a more realistic view.
It weighs too little, need some weight.
And now the most important thing, it need some kind of recoil!
Think about this; You put on your Araig suit. On this you put on some military clothes that has some weights in it + a backpack with some weights.
You step inside the Virtuix Omni and Oculus Rift. The games start (let's say CoD/Battlefield) and you grab your awesome realistic not-plastic-weight-20gram-gun and go bananas. You would be tired of playing that game in like 20 minutes and you burned a lot of calories and gain some strength + had a very realistic gameplay.
But beside going full military, the gun need some weight and a recoil(!) :) - Does your controller works in Hydra-mode, with just absolute positional and rotational tracking? I'm not interested in any WSAD and sensevity adjustent and "dead man button" and whatnot, I only care for 1:1 in-game interaction.
I could as well just 3d-print a custom pistol-shaped shell for my Hydras. But really though, I'm good with just my stock Hydras, up until there's a VR gun that physically simulates recoil - because THIS kind of things is what ultimately makes the difference between a fancy gimmick and a cool thing worth it's money. It's like Sony HMZ vs Oculus Rift. - If you want a kick in the shoulder you're out of luck, but if you're looking for the feeling of recoil in your hands, you should check these guys out. They've got an active Kickstarter up right now to push out the first batch of dev kits.
viewtopic.php?f=25&t=4671 - Basically, recoil is not something I'm trying to fit in at current due to the space limitations in the gun shell itself. If researched and I find something that can add to it and will fit in the space given, I may get there, but a motor that can deliver the "punch" for recoil is a bit out of sorts with the space in the gun. The new V2 I'm working on will remove a lot of clutter from the gun itself, most of the wires will be jumpers instead of any-which-way-you-can which should clean up a lot of the mess. That would leave the grip and fore-grip open for modifications. I may be able to work something up there, but another issue with the recoil is that the aim is based on the accelerometer's position. Most games have a recoil built in that would affect the actual recoil to where the aiming may not be usable due to both methods of recoil.
Regarding the batteries, I ran through the 8xAAA in about 6 hours testing time and thought the life was a bit too short, so further research yielded a LiPo battery pack and charger that, using a 9v wall wart can charge the 2500mAh battery pack. This pack will be hidden in the magazine as well as a DC port for the charger to plug in. Hopefully, not only will this yield a longer time for gaming, but a bit less costly solution to batteries. The pack itself is somewhere around $20, so a replacement a few years in the future could be achieved.
The position sensing is done by an accelerometer that feels the movement in roll, pitch and yaw. For the moment, I ignore roll and focus on pitch and yaw. The dead-man switch disables this tracking. The idea behind this is that sometimes the physical aim isn't in line with the virtual aim and the need to re-position the physical firearm to match that of the virtual one. If the tracking is 100% on, adjusting the firearm in any direction would yield the gun to always be moving. Lets say an error happens (never on a computer would THAT be imaginable, but lets just say) and you're virtually pointed at the ground while physically pointed horizontally. Without the dead-man switch, you'd have to raise the firearm above level to raise the virtual firearm. That would then be your new "level". With the dead-man switch, you let go of it to disengage the tracking, point the real firearm at the ground to match the virtual one, reengage the tracking via the dead-man switch and you're good to go, synchronized with the view again. This method avoids having to tunnel into the game and know what the pitch and yaw of the firearm is and adjust the game's internal to match the physical positioning which most multiplayer games would see as a hack, getting you banned. - Also, regarding the WASD control, just above the magazine, where if you hold the AR15 without a fore-grip, you wrap your fingers around the mag well. Your thumb naturally sits above the mag well. There would be the analog stick found on the Playstation Portable (see picture below). On the opposite side would be a plastic V-shaped rocker that would be hinged so that pressing either side would interact with two tactile switched underneath, give a different button press for both up and down. This could be mapped to something like adjust stance up/down or up=crouch/down=prone and a separate button next to it or above to allow you to use a spacebar jump or go to standing from current position. This would be on the opposite side of the rifle upper receiver from the joystick.
PSP1000 Analog Joystick
Another project to go along with this would be a colt 1911 handgun (same kind of parts, so similar cost to the rifle). By dropping it (and releasing the dead-man switch) and unholstering the 1911 and engaging it's dead-man switch, this could cause the game to switch weapons so that you don't have to aim a handgun in the game using a rifle if you so choose.
I'm still trying to figure out a melee capability. The Delta Six uses a "attempt stock strike" move where you act like a person is in front of you and you want to use the butt of the rifle to knock them out. It sees that motion and causes the melee action. If someone is using the Omni, they would have a kinnect device looking at them so making a motion the kinnect would see might be more reliable. I have spots to make another button, but I'm trying to avoid too many buttons as it is. - About the recoil http://www.youtube.com/watch?v=k1nuvmoMdYE
- The basic design of recoil is newton's laws about equal and opposite actions and reactions. In order to generate recoil a force has to be applied in the gun. The problem comes that there isn't a lot of space in the gun that isn't already taken up by electronics or switches to move a weight around that causes the firearm to shake. Once I get the Prototype V2 completed, I'll revisit the idea, but the battery life will be short given the torque required to move a mass that will be sufficient enough to generate recoil worth having and that may be a bad move. Looking at the haptic controller linked earlier, the main reason they're having issues going wireless is that the power required for the motor is large and the only method of delivering that power is large battery packs (no room to cram them) or via a cable that feeds power. This project will be wireless and has to be able to run on its own for at least 8-10 hours between charges (my relaxation days are full of 8-10 hours of BF4 with my best friend), so if the power requirement for the motors is too much, that would kill the wireless design.
"aparsons671" wrote:
The basic design of recoil is newton's laws about equal and opposite actions and reactions. In order to generate recoil a force has to be applied in the gun. The problem comes that there isn't a lot of space in the gun that isn't already taken up by electronics or switches to move a weight around that causes the firearm to shake. Once I get the Prototype V2 completed, I'll revisit the idea, but the battery life will be short given the torque required to move a mass that will be sufficient enough to generate recoil worth having and that may be a bad move. Looking at the haptic controller linked earlier, the main reason they're having issues going wireless is that the power required for the motor is large and the only method of delivering that power is large battery packs (no room to cram them) or via a cable that feeds power. This project will be wireless and has to be able to run on its own for at least 8-10 hours between charges (my relaxation days are full of 8-10 hours of BF4 with my best friend), so if the power requirement for the motors is too much, that would kill the wireless design.
Your absolutely right and there for I start searching a bit. The thing we recoil guys want is something called, in the soft air gun business, a CO2 blowback, which uses CO2 to create the recoil.
This gun uses another gas called green gas.
http://www.youtube.com/watch?v=qGvs--HfwK4
