FlyPT Mover

@Avee Thanks for the explanations and
@hexpod sorry if I am answering only now but I was working.
So from your question about the 180/-180 I believe you had the same problem about the 180 inversion.

Yes, all positions displays (I guess natively) values in deg. from 180 to -180.

As under « speeds » we have similar behavior, those are certainly not velocities. Unless I am missing something, it is quite complex to make a proper conversion and I doubt we can achieve it with what we have in Mover at the moment.

Just curious, why not going DCS path?? A modern game with great physics. If you want to feed your motion rig with something valid, I would advise to switch

I use dcs and msfs and they are both perfect with mover.

Falcon bms is definitely behind them in terms of graphics but it just got vr support and I wanted to try it, specially because it is regarded as the best simulation of a F16 available to the public.

Also for the next update the dev team will release the new terrain engine and pbr textures so I wanted to be ready with a good motion profile in preparation for the next release.

hopefully @pmvcda will have a look at these weird behaviours sooner or later

anyway
Thanks everybody for trying to help

I didn't think of the rollover case. I can't find any function that would deal with it. It could probably be done with trigonometry (sine and cosine don't have the hard rollover) but there are no functions for that.

So I guess it would have to be programmed somewhere in a programming language, best place for that is the game input module Mover-wise. A correct output from the sim would be best of course.

What you also could try, assuming the faulty data you are getting now is actually the acceleration, is integrating the accelerations to arrive at a speed. How well that works depends on how accurate the data output is. The mover function would be INT(acceleration value).

A rollover is a case as a continuous movement that is through mechanical limitation not possible. In that case a crop function or driving limit far away from +/- 180 degree could solve that restriction as the standard HP LP combination the travel restriction. The result is a deflection clockwise followed to middle and deflection counterclockwise, obviously a unavoidable false cue.

Seems I have a bit to read.



A glimpse at if and variables from anywhere in mover.
Here, picking the result of sway in the pose.

I can add Sin Cos... easy to do right now.
Give me some more time. Changes are huge and many things still not working.

Alleluiah

Maybe add a INTANGLE and DIFANGLE function? This would integrate or differentiate angles in a rollover proof fashion. I am no math genius but from stack overflow I got the impression it could be solveld like this
DIFANGLE = ARCTAN2( DIF(COS(Value), DIF(SIN(Value))
What it is basically doing is building vectors out of the angles, differentiating the values for the sine and cosine separately, and then converting the differentiation result (the change) back to an angle with the arctan function with two parameters. I guess that would then give you the difference angle between the samples.

That's at least how I understood it could be done.

New graphics, resizable and we can select colors. Better for some tests I needed:


@Avee , you mean this:


Did not test or think about it, but it's already possible.

Current situation is that I need to rebuild most of the sources yet.
And join all the parts that I worked separatelly. Big mess on USB devices...

That DIF function seems to already have an angle option? It's not in 3.5.3 though.

No, it will be in the new version

Hello everyone.
So I am diving inside the world of mover and let's just say that I am in sponge mode. I am absorbing tons of information and overall I am confused.

So my plan is to use currently flypt to control 3 motors to create this kind of motion sim: https://www.youtube.com/results?search_query=nova+360

I am trying to solve many issue at the moment, meanwhile I am trying to learn flypt.

I managed so far to connect 2 motors and control them with SMC3 using a modified version of the script found here: https://www.xsimulator.net/communit...ial-pid-motor-driver-sketch-w-softstart.9277/ with Sabertooth Packet Serial version.

My big question is now how to I output data to the motors in a compatible way? I don't understand this part yet.

The current plan is to output roll and pitch position to start testing the system on a small scale prototype.

Any suggestion on how to proceed?

N

Nice to see you back FLYPT

Hi, you need to use the rig called "Just Pose".
With that, the "Actuators" are the pose.
You use those actuators on the output.
The actuators use the keys: Yaw, Roll, Pitch...
So when you use those actuators, you are sending the pose.
Something like this:

Thanks for the suggestion. I will try tomorrow and let you know!

@pmvcda I would like to simulate surge and sway on my 2 dof by rolling and pitching the body to use gravity to give the feel of surge and sway. It is essential that the movements can not be felt by the user (otherwise it would be felt as a roll or pitch cue). So the accelerations have to stay below the sensory threshold.

What I need is a filter that limits accelerations, or alternatively that provides a roll-on as well as a roll-off. A filter that smoothly accelerates and slows down. An EMALP smoothes the peaks, but doesn't roll on smoothly. There are other filters that roll on or roll off smoothly, but not both in one filter I think.

Doing some googling, I figured a raised-cosine filter might do the trick. Do you think this might be something you could add?

Hi @pmvcda , I tried what you suggested and it doesn't seems to work. Here the settings:

Not sure what is going on. SMC3 seems to work fine. I can move the motors and change pwr and direction. Any suggestions?

I just posted an example, you need to setup the output to use it with the smc3.
See here: https://www.flyptmover.com/outputs/smc3-and-sps

Thanks for the prompt reply. I will try that tomorrow morning first thing

Sorry for the dumb question, but I have updated to the latest mover and my udp sources stopped working (LFS, Dirt, Dirt Rally). I may have changed something in simtools also. But if I'm not wrong, the udp sources bypass simtools. So, that should not be the problem. Do the games using the udp sources needs to be patched? Is there anything special to do? It's been so long since I messed with Mover (and I'm too lazy to read the 154 pages).
I did had movement with LFS before I updated and Mover works like a charm with Assetto.