Archive: 6 posts

---

I finally got a Move and tried out the Move Pack, but I found the controls for the Movinator somewhat... non-obvious in their descriptions. I read the Wiki article (http://wiki.lbpcentral.com/Movinator), and while it helped, I had a hard time fully grasping exactly what each one really senses. So, I connected one up to my logic probes and did some trial and error. First off, remember that each signal wire has both an analog and digital component, both of which can be either positive or negative. There's a lot of information about that I won't go into in this post; see the wiki or my logic tutorial level for details on that. The key thing to remember is that, like the joystick sensors on the Controlinator, the digital part stays off until it goes past 75% analog, then stays on until it drops below 25%. Now, on to what they actually do. There are 8 outputs in 2 pairs that I'll be covering here. They are: Vertical Rotation Up/Down Vertical Rotation Front/Back Horizontal Rotation Left/Right Horizontal Rotation Front/Back Turning Rotation Left/Right Turning Rotation Up/Down Twist Left/Right Twist Up/Down However, there are really only 3 different Rotation outputs due to them being duplicated across the pairs. The unique rotation outputs are simply Up/Down, Left/Right and Front/Back. Also, the Twist Left/Right and Up/Down are almost, but not quite, the same thing rotated 90? from each other; there's some kind of threshold involved I haven't fully tested the details of yet. Also, note that what counts as "directly at the screen" depends on the calibration of the Move. So, here's what they actually sense. Please excuse my lack of artistic talent; the graphics were thrown together rather quickly in OpenOffice Draw. Rotation Up/Down is at 0% when pointing directly at the horizon, going to 100% straight up and -100% straight down. http://balorn.com/lbp2/mov-r-up-down.png Rotation Left/Right is at 0% when pointing directly towards or away from the screen, going to 100% straight right and -100% straight left. http://balorn.com/lbp2/mov-r-left-right.png Rotation Front/Back goes from 100% when pointing at the screen through 0% when sideways (regardless of up/down/left/right direction) to -100% when completely away from the screen. http://balorn.com/lbp2/mov-r-front-back.png Twist Left/Right is 0 when the controller is not tilted, or is completely upside down, 100% when tilted 90? to the right, and -100% when 90? to the left. This output does not function if the controller is pointed straight up or down. (The illustration is from behind the controller, pointing at the screen.) http://balorn.com/lbp2/mov-t-left-right.png Twist Up/Down is 100% when the controller is not tilted, 0% when tilted 90? to the right or left, and -100% when upside down. This output does not function if the controller is pointed straight up or down. (The illustration is from behind the controller, pointing at the screen.) http://balorn.com/lbp2/mov-t-up-down.png TL;DR version: Rotation Up/Down = Pitch Rotation Left/Right = Yaw Twist = Roll Rotation Front/Back = Pointing towards or away from the screen

2011-09-24 04:52:00 Author: Balorn Posts: 92

---

Wow, such a great insight! Your images are great, very easy to follow.

2011-09-27 13:16:00 Author: Mr_Fusion Posts: 1799

---

Thanks for posting... I found the lack of info on the move very frustrating, so thanks again and please continue. cheers

2011-10-01 01:12:00 Author: pete007d Posts: 58

---

Glad to hear people are finding this useful. I do plan to do more testing and documenting with things in the Move pack, as well as the new sensors we got recently, and integrating them into my tutorial level (see my sig), but I've kinda been distracted by Disgaea 4 and City of Heroes lately, so I can't promise a timeframe. (Also, I need to redo my logic probe using the new sticker panel material, among other things.)

2011-10-01 03:44:00 Author: Balorn Posts: 92

---

However, there are really only 3 different Rotation outputs due to them being duplicated across the pairs. I made it (by your scale) two and a half different 'outputs' or five discrete analog values - see this post (https://lbpcentral.lbp-hub.com/index.php?t=63273-Move-Does-the-Activate-input-on-the-Move-Cursor-even-work&p=927582&viewfull=1#post927582) for more details. Also, the Twist Left/Right and Up/Down are almost, but not quite, the same thing rotated 90? from each other; there's some kind of threshold involved I haven't fully tested the details of yet. I think you might be missing the fact that each set of paired outputs is a Polar to Cartesian conversion of the angle, so no single output is much use on its own (unless you really want to measure sin θ or cos &theta. The idea is that if you plug both outputs from each pair into a Joystick Rotator, then it'll match the angle of the Move for that axis. Rotation Up/Down = Pitch Rotation Left/Right = Yaw Twist = Roll Rotation Front/Back = Pointing towards or away from the screen Given this Polar to Cartesian mapping, this is actually... Vertical Rotation = pitch Horizontal Rotation = yaw Twist = roll ...and the remaining output "Turning Rotation" is just a composite of Pitch and Yaw, such that if effectively measures roll (relative to the screen) when the Move is held such that you're pointing it at the ceiling. Also, I need to redo my logic probe using the new sticker panel material, among other things. Heh. Mat-changing that was one of the first things I did. Given the connection points don't show in create mode, it makes it so much easier to read. http://i5.lbp.me/img/ft/b452edd2cdf07fb67d627d062f1cd94e5632e170.png Hologram digits (left) vs. Sticker Panel digits (right) There's an interesting 'bug' whereby if you attach a microchip to any material other than Sticker Panel, then mat-change it to Sticker Panel, the microchip becomes invisible in create mode, so you could exploit that to hide the chip on the front of it too... http://ib.lbp.me/img/ft/cebdf7b35b8c41678a04e79d403ffccd0c44a386.png Entire display converted to Sticker Panel... where's mah chip?

2011-10-02 20:34:00 Author: Aya042 Posts: 2870

---

Ah, well, it's been well over a decade since I took Trig, and it was already past when I should have been to bed when I put this together. Still, the pictures should be accurate at least. Edit: And several of my "great ideas" for improving my logic probe using sticker panels turned out to be unworkable when I tried them due to the "transparent" part of sticker panels not really being transparent when not in preview/gameplay mode.

2011-10-03 00:01:00 Author: Balorn Posts: 92

Posts: 1077139    Threads: 69970    Members: 9661    Archive-Date: 2019-01-19

Aus dem Archiv wurden alle persönlichen Daten wie Name, Anschrift, Email etc. - aber auch sämtliche Inhalte wie z.B. persönliche Nachrichten - entfernt.
Die Nutzung dieser Webseite erfolgt ohne Speicherung personenbezogener Daten. Es werden keinerlei Cookies, Logs, 3rd-Party-Plugins etc. verwendet.