FlyPT Mover

In the new release (3.7) you have the WITMOTIONWT901C as a source.
You can use the received values anywhere, also in the motion compensation.
The sensor is good for rotations and linear accelerations, not for linear position. So motion compensation with this sensor can work for 3DOF rigs, but for rigs with big linear motion it's not good.
But in the new Mover, you can mix any value in the motion compensation, so you can use the values from a pose mixed with values from the sensor.

I think in other softwares they use it for rotations and then for heave and/or surge they use the position sent to the hardware.
This in 3DOF rigs. Traction loss is also a rotation, so no problem for that.
Honestlly, I see no reason to use it, even for other motion softwares. But I might be wrong.

This is good news even for rigs with linear motion. They could make a poor man's Kalman filter where they mix a high pass on the sensor with a low pass on the commanded pose, so that the motion compensation accurately accounts for rig inertia and reaction time, but doesn't drift over a lomger period.

Wow! I had no idea that slip were a no-no in jets. Makes sense after reading this.
In the prop planes I've flown since getting my PPL, performing a Forward Slip is a favorite of mine (only solo or with a CFI). When I got my tailwheel endorsement and aerobatics course several years ago, the instructor had me practice forward slips in the Super Decathlon. Coming from Cessnas and Pipers, I was amazed by the amount of rudder authority this plane has in comparison! Now that I have a 6DOF (PT), I'm on a mission to recreate that sensation. Trying difference profiles and learning, but I've only tried DCS so far and I don't think the planes in DCS slip very well.

Oldies but goldies by Dirty

Hi I'm writing a game is it possible to use custom JSON over UDP as source?

Yes

Great! Any idea how?

Hi Pedro, just wondering if you've heard of an issue with the switch from AIR to GROUND, like in MSFS.. sometimes when landing, and it switches to ground poses, for some reason it just doesn't seem to 'register'.. the air poses shut off but the ground never starts, so i'm rolling down the runway with no motion anymore. If i look at the source window, all the telemetry is coming in, and i can disconnect and reconnect and its fine. Anyway, hope you've seen this before and it can be part of 3.7. thanks!

I've encoded it as bytes now so no worries about that, unless you care to share how for educational purposes. Thank you for this amazing software.

Hi, I'm sorry, I was travelling and made a fast answer on the phone.

You can send it in JSON to UDP or Shared Memory, but in the new release.
In the custom sources, you can receive the JSON with the vars incorporated. I still kept the old custom where we get the value from bytes at a specific position (those are intended for existing games while the new one is for who wants to send specifically to Mover).
I will share a list of vars and units and the JSON is really simple. Something like:

Code:

"LAT_A": 0.04,
"LON_A": 0.02,
"VER_A": 0.23,
"YAW_P": 122.328,
"PIT_P": 17.802,
"ROL_P": 8.02,
"GEAR": 3,
"VEHICLE": "Formula BMW"

We want it fast, so let's keep it simple.
You can send all the vars or just the ones that changed.
I had other ideas, but no time to add a var to set the sub source (like air and ground)

Did not notice. But I remade all the sources.
I will test.

Thanks

I have a problem with setting the direction of movement of the arms depending on how the engines are positioned in relation to each other.
Picture motor position
https://ibb.co/p4Hh6Ym

In this setting, everything works correctly and the movement of the arms is exactly the same as the motion preview in the Flyptmover 3D viewer window

Picture motor position

https://ibb.co/LxqRyMY

When I turn motor #2 and reposition the arm, the arm movement is opposite to the preview in the 3D VIEWER window.
Can I find an option in the program to reverse the arm movement?
I thought about changing the polarity of the power supply to the motor but in such a case the Hall sensor in the SMC3 Utils program works incorrectly ( the green line of movement from the sensor goes opposite
to the movement of the motor) the question is if I can reverse the power supply in the Hall sensor ? like it can be done for potentiometers?

You can reverse the Direction on a check box when setting the rig

Thank you for your reply, but do you mean the checkbox marked with a red arrow in the attached image ?
If I check it, it changes only that the arms move in the opposite direction but not as I need.
I send the signal from Source Loop Pitch speed to the controller and both arms should work up and down at the same time, and in my case engine 1 arm works up while engine 2 arm works down, when on the preview in 3D viewer I see that both arms should move in the same direction.
Maybe my problem is that engine 1 connected to the power supply directly spins to the right and engine 2 connected exactly the same way with poles spins to the left ? Should I use the two engines rotating the same way 2 x left , or 2 x right ?

I mean the check boxes below the crank angles

Thank you very much for your help, you guided me to the solution of the problem

Hi Everyone,

It has been a while since i posted. My current rig is aiming towards a 3 dof with 360 degrees freedom. I am trying to simulate through the rotation of the 3 axis the G force to represent the forces inside a flight simulator or a F1 simulator. I have been trying to understand how to set this up in flypt but with no luck. How can input to is those forces? I know for example the math to convert longitudinal acceleration into a G force based on the distance of the pilot inside the motion simulator, but i cannot drive a single axis with that force. The plan i to combine al the 3 axis to recreate the correct force.

Thanks in advanced.

I don't know if I really understand well what you mean, but here's what I do:
First of all, cars and planes can be treated differently, but the base for all is to make you feel the forces.
So, for example, in a real car, when you accelerate, you get longitudinal accelerations.
There's some noise there when you change gear or go over some irregular terrain.
What it means is that the acceleration has noise, it's not "constant". What I do in Mover is apply a low pass filter to remove that noise and get the "constant" acceleration.
I also apply a high-pass filter to remove the noise.
Now in the rig, if we have surge, it's always limited in travel, and you can't represent the "constant" acceleration.
What I do is tilt the rig up (pitch) to feel some constant pressure in the back, simulating that constant acceleration. In a surge, I use the noise to feel the details.

So, resuming: Low pass on pitch and roll, high pass in sway and surge.
The pitch and roll filters are used in the mix surge/sway in pitch/roll.

The pure roll and pitch are like the orientation of the vehicle in the terrain. If you are stopped in a slope, you can represent the slope in the rig with pure roll and pitch.

Now comes the confusion:
When you mix everything, some forces compensate the others.
For example, in a plane turning and banking, if you put the roll and the constant acceleration together, they might neutralize each other, and then you might be horizontal in the rig.
Or, for example, a car in an oval track.

For your rig, from what I understand you have just rotations so you have to play with them only.
Changing center of rotation (in the rig module) has no effect in your case. You can change it in the source for mover to correct the position in the received data, so the accelerations you get already have the position correction.

Hope it's what you wanted.

Thanks for the reply. So if i understood correctly, i cannot translate in a constant rotation on one axis the longitudinal acceleration? Because the math is kind of simple to convert it, i just need a constant value to multiple it by thr longitudinal acceleration.

To make my self more clear, if i had the possibility to have an option to multiply on a single axis the a constant rotation, if the longitudinal acceleration is 0 i would zero out that constant acceleration, if it is higher i wohld just multiply it by that factor. And then i could apply a washout filter to bring the orientation of the rig to the rest position when i have no force.

Whay I am looking at is basically trying to recreate lateral, negative and posive G forces with pure rotation.

I am quoting myself from an old post:

“Applying a properly constructed classical washout or derivative is valid ONLY under specific conditions :

1. Accelerations have to contain gravity component
2. Rotational velocities have to be available

It might be important to point as it’s not necessarily the case in all titles.

Once you have an access to those, it’s really super simple. You really don’t have to think about specific scenario, whatever you do, the force vector will be applied correctly across heave/surge/sway.”

I think @pmvcda added meanwhile a separate functionality so you can add the gravity component for the racing titles which doesn’t provide it. In this case rotations and centripetal forces in the car should be mixed properly.