Forum Discussion
£10 running-on-the-spot input device trials
Hi all,
Yesterday I took a cheap wireless mouse, removed the case and buttons, and wired one after the other the following three inputs (see photos) to the PCB to test. My aim was to try using the setup with an AutoHotKey script I had created, so that I could jog, run or stand still using the input device - and for my character to do the same in any FPS game. (For this purposes, I actually tested in Quake and google earth - but in front of my home projector as I'm long-awaiting the rift consumer release)
1. A pressure pad from a dance mat.
image[4].jpeg
https://developer.oculusvr.com/forums/download/file.php?id=3059
result: worked pretty well actually! I could successfully pace,pace,pace with variable speed on the spot. As such, I took a jog down the Las Vegas strip in Google Earth 3D street level, which was really cool! And with the autohotkey script I devised, when you stop running (catch your breath) your character stops running too - even though you're still stood on same spot.
2. An audio-sensor switch taped to the floor to pick up vibrations (from running anywhere in the area on a hardwood floor).
image.jpeg
https://developer.oculusvr.com/forums/download/file.php?id=3056
result: disappointing - but I expect this needed further adapting to work really well. For this, I took a cheap sound-activated LED light, which was originally wonderfully sensitive to sound - especially when the sensor was taped down to a surface and I knocked anywhere on the surface. So I took the audio sensor out and wired this into the mouse PCB input. However, suddenly the sensitivity of the input sensor was incredible weak. I had to really KNOCK the sensor hard with my knuckles to get any input through. So alas, this taped to the flood produced almost NOTHING. But I question why it worked extremely well when part of the LED light setup? I know very little about electronics, but could it be that it required also some resistor or other component to make it more sensitive? ...I'd love to get this working effectively, it would mean that you could potentially be less limited to the area in which you moved, so long as sound vibrations were picked up.
3. A PIR movement sensor ripped from a cheap motion-activated LED device.
image[3].jpeg
https://developer.oculusvr.com/forums/download/file.php?id=3060
result: hit and miss. On some occasions I could face this towards by body and run, and my on-screen counter clocking the inputs would correspond. But lack of sensitivity seemed to be an issue again in all positions I tried it. I think similar to the sound-input device, something else is needed to boost sensitivity here. And of course, best just to monitor motion at feet-level, as firing a gun you don't want to trigger any 'forward' motion.
So, check the photos, let me know your thoughts, or any other input switches which might be interesting to try to measure jogging/walking on the spot. (How about a pedometer input device attached to the mouse? Then just put the it in your pocket)
A final note: There are people obviously designing treadmill motion simulators etc, but in my opinion, just running on the spot really doesn't feel much different an experience from actually moving in a particular direction, especially when you're focused on the graphics from the game moving around you. And this requires only a small space and a quick setup - allowing you to do it anywhere easily - and cheaply. So I think there is actual potential in making an effective but simple on the spot input device.
Yesterday I took a cheap wireless mouse, removed the case and buttons, and wired one after the other the following three inputs (see photos) to the PCB to test. My aim was to try using the setup with an AutoHotKey script I had created, so that I could jog, run or stand still using the input device - and for my character to do the same in any FPS game. (For this purposes, I actually tested in Quake and google earth - but in front of my home projector as I'm long-awaiting the rift consumer release)
1. A pressure pad from a dance mat.
image[4].jpeg
https://developer.oculusvr.com/forums/download/file.php?id=3059
result: worked pretty well actually! I could successfully pace,pace,pace with variable speed on the spot. As such, I took a jog down the Las Vegas strip in Google Earth 3D street level, which was really cool! And with the autohotkey script I devised, when you stop running (catch your breath) your character stops running too - even though you're still stood on same spot.
2. An audio-sensor switch taped to the floor to pick up vibrations (from running anywhere in the area on a hardwood floor).
image.jpeg
https://developer.oculusvr.com/forums/download/file.php?id=3056
result: disappointing - but I expect this needed further adapting to work really well. For this, I took a cheap sound-activated LED light, which was originally wonderfully sensitive to sound - especially when the sensor was taped down to a surface and I knocked anywhere on the surface. So I took the audio sensor out and wired this into the mouse PCB input. However, suddenly the sensitivity of the input sensor was incredible weak. I had to really KNOCK the sensor hard with my knuckles to get any input through. So alas, this taped to the flood produced almost NOTHING. But I question why it worked extremely well when part of the LED light setup? I know very little about electronics, but could it be that it required also some resistor or other component to make it more sensitive? ...I'd love to get this working effectively, it would mean that you could potentially be less limited to the area in which you moved, so long as sound vibrations were picked up.
3. A PIR movement sensor ripped from a cheap motion-activated LED device.
image[3].jpeg
https://developer.oculusvr.com/forums/download/file.php?id=3060
result: hit and miss. On some occasions I could face this towards by body and run, and my on-screen counter clocking the inputs would correspond. But lack of sensitivity seemed to be an issue again in all positions I tried it. I think similar to the sound-input device, something else is needed to boost sensitivity here. And of course, best just to monitor motion at feet-level, as firing a gun you don't want to trigger any 'forward' motion.
So, check the photos, let me know your thoughts, or any other input switches which might be interesting to try to measure jogging/walking on the spot. (How about a pedometer input device attached to the mouse? Then just put the it in your pocket)
A final note: There are people obviously designing treadmill motion simulators etc, but in my opinion, just running on the spot really doesn't feel much different an experience from actually moving in a particular direction, especially when you're focused on the graphics from the game moving around you. And this requires only a small space and a quick setup - allowing you to do it anywhere easily - and cheaply. So I think there is actual potential in making an effective but simple on the spot input device.
6 Replies
- also just ordered a vibration/tilt switch sensor from ebay to test.
http://www.ebay.co.uk/itm/10PCS-Vibration-Sensor-SW-18020P-Electronic-Shaking-Switch-Hot-Selling-OT00307-/291156031501?pt=UK_BOI_Industrial_Automation_Control_ET&hash=item43ca40780d - This is a great idea for a cheap input device for walking or running. Are pedometers actually good at counting steps? Do pedometers usually contain something like that vibration sensor you got on ebay or are they different? I wonder if we couldn't find a cheap pedometer hooked up to an arduino or something to give us simple forward movement input for games.
- Well, I'm off to Spain in a few days, but on my return my vibration sensor should have arrived (from china). So I'll let you know the result of using that as an input. I'm excited too, as when hooked up wirelessly and potentially chucked in a pocket, you've got freedom to roam, run on the spot or run in any direction, and the input should be registered the same.
Here's a nice tilt switch video:
https://www.youtube.com/watch?v=_sgU7vi9w6M
The main issue I can see from the outset will likely again be some other circuitry again needed to bolster the sensitivity of the input. And if even then the device correlates will to the movement of a foot-pace. (you could put the sensor in the middle of your body so that vibration was picked up evenly no matter which foot you paced.)
If the device is too sensitive, this should be able to be compensated for in software. For example, if during one foot pace, the sensor registers lots of inputs (lots of micro vibrations) then you could:
- accept a percentage of inputs per second
- or force a 'sleep' or 'wait' after 1 input is received, before being allow to process another. (without queuing received inputs)
Side note: The device here would probably end up being similar to the Wii JOG or uMOVE... but I could NOT find anyone who made something similar as a PC input sadly.
https://www.youtube.com/watch?v=HGZOiyU0LSI
So, fascinated to test it. I believe the pedometer could unleash freedom of movement. (and then if you want to run on the spot on a polished surface in socks, you've also got an in place treadmill). - Awesome report there pretzelzzz, will for sure keep an eye on your trials!
I would imagine that a tilt switch sensor will work really well to capture running/walking on the spot. One thing that concerns me though is that running on the spot usually doesn't stop oneself from actually moving forward (or backwards, especially when "blindfolded"). I would at least try having something keeping me inside a safe zone, like three backward facing chairs too lean on if needed. :) - Cool idea. I wonder if you took the sensors out of the mat and attached them directly to your shoes they'd trigger a bit more accurately and you'd be able to turn as well.
- What about using a simple wiimote, or even just a regular android or iphone smartphone in your pocket and a FreePIE script. It is not actually "cheap" but considering everyone has a smartphone...
I think there are already a couple scripts around for the wiimote walking/running and even jumping and crouching. Adapting those for a smarphone should be pretty straight-forward.
Edit: BTW I've just reallized that most smarphones have a compass so this could be a great idea. You could potentially measure walking, running, jumping and crouching and even the direction you are going to. The only con is most smarphones' compass is not a 3Dvector one so the yaw measure starts jumping when the device is pointing straight up.
