Sunday, October 27, 2013

Plotting Sneaking, Walking and Running

So, here is the promised video, with additional "Dead Zone" fix that we are currently working on. Plotted and recorded live, with me using the treadmill.

The green line shows the normal linear interpolation, while the blue shows a "Dead Zone" interpolation we are currently working on. The idea here is to increase the value quicker in the beginning, as normally controllers have a "Dead Zone" area, which basically is an area where the value is still too low to be considered useable (thought of being noise).

Max is 127, and here I sneak, then walk and finally run. I have also just before doing this calibrated the max to be me running, otherwise it would behave differently. Works perfectly. Sometimes there might be some odd jumps, but isn't noticeable in a game. Plot here reflects the quality of the firmware quite well though. :)

With the "Dead Zone" support, it's possible to indicate how many % of the beginning and by how much it's supposed to increase. This naturally needs to be tweaked, which what I'm currently doing. Might not be perfect for everybody, but I like having it available at least.


Wednesday, October 16, 2013

TreadGaming rev6RF Enclosure

Just a quick post about how the new enclosure looks like. I'm quite pleased with how the buttons ended up, as they are "inside" the enclosure, so if you want to tape the whole thing (which I do at the moment to my treadmill), you can, without pressing down the buttons.

I also included the "TreadMan" I had designed into the enclosure. Looks pretty nice.

Here are 2 renders made in Blender:





Sorry about the delay with videos. Will come for sure, as this project is really start to take shape now! Firmware pretty much complete! And, I bought a Elgato Game Capture HD which allows me to record HDMI data. Basically will this allow me to record the gameplay in HD, and I will use my iPad to record the treadmill, so one can see both at the same time.

Friday, October 4, 2013

Finalizing the Firmware - Behind The Scenes!

We are at full speed, or well, the guy over at odesk is at full speed converting and improving the firmware for the new PCB.

I am really impressed by his dedication and work so far! He even got a bonus for it! I asked him to send some photos before I post about the new firmware. Everybody loves photos. :)

So, here are some he send me of his setup:





He bought a motor to spin a CD to test with. Quite the dedication right there. :)

As for the firmware, everything hardware wise seems to work as it should! So, I'm happy that the board itself is working as it should! Haven't tested the nRF24 module though. Might do that later myself, as this firmware upgrade he does, doesn't include that. Should be rather easy to test later I think, as all I need to do is reuse code and try to talk to it.

Anyway, we have gone back and forth for a while now. He makes some improvements/changes, send me the firmware, I flash it on mine here and test.

I test the Nunchucks and using it on the treadmill. And provide feedback.

He made a nice dual solution with serial text output and joystick working at the same time. This makes debugging and testing much easier. He even made a script that plots the data in real-time. Quite handy tool to use when I sneak, walk and run on the treadmill. The data is also send back to him so he can analyze it and improve.

I will put up some graphs and videos soon from my tests.

Thursday, October 3, 2013

High Speed Laser Optical Sensor [ADNS-9800]

I was doing some research and came by this:



http://www.kickstarter.com/projects/1034145369/high-speed-laser-optical-sensor

https://www.tindie.com/products/jkicklighter/adns-9800-optical-laser-sensor/

Neat little thing. He is using the ADNS-9800, while I use the older ADNS-9500, but they are pretty much the same. They are compatible with each other so there is no problem for me using the ADNS-9800. As I understand it, the ADNS-9800 has some filter build in, so it will improve the data, but as I'm getting filtering made in the firmware, the ADNS-9500 will work just fine too.

Anyway, interesting idea using this with a trackball. Nice to see that such a thing can have a following: http:///www.geekhack.org/index.php?topic=44411.180

In there one can see a neat 3D printed DIY trackball:

4

3

2

http://www.flickr.com/photos/8465641@N07/sets/72157633258917941/

Inspires me to design such for mine as well... :)