Due to some redesign work, the Stem system is now going to ship in October instead of July.https://www.kickstarter.com/projects/89577853/stem-system-the-best-way-to-interact-with-virtual/posts/827198Originally, Stem consisted of 3 parts: Stem sensors, Stem packs and Stem hand controllers. The Stem pack is the unit that you attach to body parts and contains the force feedback and battery. The sensor is the actual bit that does the tracking, and needs to be plugged into a pack or hand controller. The new design integrates the sensor into the pack and hand controller.That shouldn't affect most of us (I ordered the version with 2 hand controllers, 3 packs and 5 sensors), although the larger versions might have to change a bit (the $1000 version included 5 hand controllers, 5 packs and 5 sensors, 10 devices total but only 5 could have a sensor attached at once).The new design is looking pretty cool (there's pics on the kickstarter update page) and apparently the new design reduces some latency issues, mechanical stability and environmental distortion.

Yeah, I just got the email.A little sad, but not that unexpected. I don't think there has been a single Kickstarter that shipped on time.Hopefully the changes will be worth it.

I do like the new pack design, although the hand controller has gone back to the Hydra style of 2 buttons on either side of the joystick instead of 4 buttons in a standard diamond beside the joystick. I never really liked that style on the Hydra. In theory, a Hydra could be used like a game pad if games didn't support the motion tracking, but the button placement didn't feel good for games that needed more button presses than thumb stick use (especially with the right hand), or dpad use. 2 buttons on either side means there's no clear up/down/left/right button, just 4 diagonals.So I think losing the button diamond is a step backwards. But then again the point of Stem is the tracking, and it sounds like they've made improvements to that.

Now there is no excuse why the design of the controller is so overlarge. Since the tracking system is on the circuit board, they could totally place it to the hand part of the controller like they did with the wireless hydra prototypes aka sixense truemotion. I really find this current double grip design disturbingly ugly.

I think what they really need is a wiimote style wrist strap for the controller. The more immersive the experience, the more likely it is that people will try to throw an in game object but accidentally throw the hand controller instead. :)

" We will work closely with companies like Tactical Haptics to make sure that we provide them with the needed tracking for their products. " Nice!!

Stem redesign and 3 month delay | Meta Community Forums

12 Replies

cybereality
Grand Champion
12 years ago
Yeah, I just got the email.

A little sad, but not that unexpected. I don't think there has been a single Kickstarter that shipped on time.

Hopefully the changes will be worth it.
kojack
MVP
12 years ago
I do like the new pack design, although the hand controller has gone back to the Hydra style of 2 buttons on either side of the joystick instead of 4 buttons in a standard diamond beside the joystick. I never really liked that style on the Hydra. In theory, a Hydra could be used like a game pad if games didn't support the motion tracking, but the button placement didn't feel good for games that needed more button presses than thumb stick use (especially with the right hand), or dpad use. 2 buttons on either side means there's no clear up/down/left/right button, just 4 diagonals.

So I think losing the button diamond is a step backwards. But then again the point of Stem is the tracking, and it sounds like they've made improvements to that.
octaspace
Honored Guest
12 years ago
Now there is no excuse why the design of the controller is so overlarge. Since the tracking system is on the circuit board, they could totally place it to the hand part of the controller like they did with the wireless hydra prototypes aka sixense truemotion. I really find this current double grip design disturbingly ugly.
kojack
MVP
12 years ago
I think what they really need is a wiimote style wrist strap for the controller. The more immersive the experience, the more likely it is that people will try to throw an in game object but accidentally throw the hand controller instead. :)
Frito
Explorer
12 years ago
" We will work closely with companies like Tactical Haptics to make sure that we provide them with the needed tracking for their products. "

Nice!!
kojack
MVP
12 years ago
"cybereality" wrote:
A little sad, but not that unexpected. I don't think there has been a single Kickstarter that shipped on time.

My keychain for supporting a robot boat came on time, I think. :)

Then again, I also ordered an IControlpad 2, but after delays they used up all the money and couldn't continue, so they shut the project down and I've lost $80. :(
(I'm an Open Pandora survivor and still fan, 3 months delay is nothing compared to 2.5 years from payment to shipping)
cybereality
Grand Champion
12 years ago
I really hope the distortion fixes are robust. That was one of the worst problems with the Hydra, especially when each controller would produce different results. It made 1:1 hand tracking very wonky and I bet that's what most people using the Rift want.
owenwp
Expert Protege
12 years ago
The experiments that I have done with distortion correction look very promising, but we need to work it into our protocol and precisely measure its performance with some new hardware before we can really quantify it. I am pretty confident, but I don't want to make specific promises while I am working with hand-assembled prototypes.

But even without it, 1:1 hand tracking is vastly improved over the Hydra if you have a 3 tracker system, the distortion on the head and hands cancel each other out quite well since we use their relative positions, and will be leaving the actual head position alone on the DK2 unless you leave the camera's view. Plus there is no calibration to get wrong.
Anonymous
12 years ago
"owenwp" wrote:
The experiments that I have done with distortion correction look very promising, but we need to work it into our protocol and precisely measure its performance with some new hardware before we can really quantify it. I am pretty confident, but I don't want to make specific promises while I am working with hand-assembled prototypes.

But even without it, 1:1 hand tracking is vastly improved over the Hydra if you have a 3 tracker system, the distortion on the head and hands cancel each other out quite well since we use their relative positions, and will be leaving the actual head position alone on the DK2 unless you leave the camera's view. Plus there is no calibration to get wrong.

Interesting, I thought there was no longer any point to attaching the third tracker to the head with DK2.

What about attaching the third tracker to the torso instead, so that you can move in one direction while looking in another (e.g. walk forward while looking sideways)?
kojack
MVP
12 years ago
"Zoide" wrote:
Interesting, I thought there was no longer any point to attaching the third tracker to the head with DK2.

The camera knows the relative position of the rift, the stem base knows the relative position of the stem sensors, but the camera and stem base don't know each other's relative positions. Having both track one object (your head) lets them know each other's frame of reference.

Maybe you could calibrate it with a stem on your head, work out the camera/stem frames then remove the stem from your head and attach to your chest. As long as the base and camera don't move, it should be ok.

I'm just guessing here though.