[PIMAX] All about the VR headsets

I will contact them.
I have a rebate of 170$ because I sent my Vive. The rebate is for a new headset. I tried it on their system and it did not work.
I guess they consider we are doing an upgrade and it does not apply.
I also paid for a cable, that I will ask to be set as a coupon instead.
I could communicate via e-mail until the announcement. Since that, I have no reply.
I will make a ticket.
But now I'm broke. I can't afford a headset. It would not be wise.

I would expect they are very busy. It is often a mistake to give too many options . I took the risk of ordering the full 8KX system as part of the Kickstarter and have some money in the kitty from other rebates, so if I am prepared to wait for the Pimax 'knuckles' no extra cash needed. It has although been two years and I am still to have all this tech in my hands .

How it works in terms of tracking devices ??

Are they supposed to be attached on the moving platform or you have to fix them to the wall ?

Someone said they will go open source with their drivers. Does it means we will be able to fix all motion compensation issues in order to fit it to our needs ?

Oh, that's what I was afraid to hear.

I remember a talk I had with @Dirty about the VR development, when I was expressing my frustration on the improvements oculus did on his dk2 development kit for the common user, destroying at the same time the experience with motion platforms.

Indeed back than in 2014, one had just to fix a tracker cam on the platform to get all the motion perfectly compensated. I remember he said :

"In the VR industry the whole motion community is so insignificant as a grain of sand on the beach."

I didn't got much into the topic since stone age DK2 time, but just seeing hundreds of posts here about the compensation issues I want to ask a question … is it not simply possible to get back to what was working very well 5 years ago ?

Best.

My issue with the motion cancellation tech, is that most are not capable of accurately tracking a very fast moving object, as the manufactures 'use case' likely specifies that they do not need to. These 'fast' movements can be caused by either harpic devices - buckkicker etc, or high speed servos and will cause a loss of track, with the data then 'lost' until it gets a good 'lock' again. This was yet another reason why I have moved to a mainly G based motion/pressure system. Don't know if this will be fixed in the future, but I'm not banking on it.

So far, cancellation with Pimax is simple... Almost.

You just fix one of the controller to the rig, so both lighthouses can see it, idealy.

In PiTools, there is a motion cancellation drop down. In there, you have all the active controllers. You select one, and the world locks to it. Then, in SteamVR setting, you reset the seat position.

But, there is a few things I need to test more. And last weekend, I was at home, but without electricity (wind, trees, broken wires everywhere) and I could not test anything.

The little I did, I found out that if you set a controller to cancellation, adding a new controller will reset the cancellation. Also, the controller need to be set a specific way as seat position do not reset the rotation. Once I got it right, I had glitches all the time. Could be something else. I need to test it with, and without, to see if I get ride of the glitches (sudden moves, like lost of traking). It could be controller position. I missed a whole weekend of testing, snif, snif. I will do it later and send my findings to Pimax. They are the only ones who did something about this issue. Maybe there is something more about it in the new Pitool. Will test that too.

If I understand our needs correctly, we want to keep all the 6dof movements of the head and cancel all the 6dof movements of the platform... right ?

I don’t understand why we can’t achieve it with a simple on-platform cam.
The simulator doesn’t move in the room. I don’t understand why we should need the off-platform trackers system.

@Pierre Lalancette , if you are in touch with the developers, could you possibly ask them if there is a possibility instead of using external reference (off-platform) for platform movement subtraction, to simply use on-platform tracking cam and take the platform as a reference for local 6dof head movements ?

If the pimax goes open source I think @pmvcda should be aware of it. I am sure he can solve the interface issues or adapt a more convenient and reliable one.

Best

Long story short:
The headset has to take the platform for motion reference and not the room.

It would be so much simpler IMHO.

Room reference is terribly complex as it has to take the sum of head and platform movements first, than mesure with a separate tracker the platform motion only, than subtract one from another. Well, I doubt we can achieve a reliable functioning with this approach.

I would be interested what @pmvcda thinks about it.

I use motion compensation with pimax, attach the vive tracker on the simulator, over my head to the seat, turn on iracing, reset position and go a dozen laps on nordschleife without any problem, I have no tracking failures. It works without any problems, I turn on and play, I am very happy with Pimax and PiTool

Is your rig a 6dof ? Are you using a tilt coordination for surge and sway acceleration?

Sure, just thinking that issues could be accentuated or simply not noticed if the rig has very little angular capability.

First, there is a difference between Vive and Occulus.

Occulus camera can be attached to the rig, and that's what they do. I don't have that. I've heard that there is a slow deviation over time that could interfere in the long run. I'll let those with a Occulus confirm that affirmation.

Pimax use Vive Lighthouse. It is great. They are fixed to the wall and once calibrated, they stay that way for the rest of the universe space time... almost.
Fixing them to the rig would be incredible, but we can't. It's two rotary lasers that calculate the time it takes to reach each headset trackers in the UP and WIDE axis. It is very sensible (and precise) and this is why they are fixed to the wall. Putting them on a rig would simply destroy them.
The inverse matrix calculation of the rig is the mathematical way to go. And it works great. I have little glitch right now, but it could be tracker placement.
This is why I need to test more... One day... In a distant future...

Image comparison :

Yeah, I know. I want the 8KX now. But I'm so broke...

We have also this. Is it a device which could be used with motion platforms ?

I'm pretty impressed with the Pimax motion compensation, their software is generally pretty good now (I have the 8K), but this is still my biggest problem. I can either use motion (with compensation) but no transponders, or both and fly around in the car... I wish I had the brains to fix it myself. I was hoping it would be as simple as setting a deadzone in the motion compensation software so that high frequency small vibrations are effectively ignored, obviously this would introduce a little lag, it's such a shame this hasn't been sorted yet.