Question Raw data output from simtools

Hello people.
I have a problem with the output data from simtools. I am trying to build a 3dof platform and I need the raw data from games. which means I want the exact roll pitch etc values from the game. I have tried to read the data from LFS and simtools both. here are the graphs :
this is data from LFS directly :


and here is the data from simtools: (the left chart is roll & pitch positions and the right one is surge sway & heave accelerations :


these are from fairly the same motions ( ofc had to run LFS twice but tried to keep it similar to each other)
the simtools data is very noisy and actually doesn't seem right. below is my axis settings :


the thing is I tried to check the output and see if the axis1a value is equal to the raw Roll value. so in the "output testing" tab I switched it on and captured the values. I dragged the bar from 0 to -100 then to 100 and back to 0. here is the result:

as you see it is a bit scrambled. what I was expecting is this :


to summarise it I should say generally there are two problems :
1. getting the raw data correctly from the game (i wonder if it is possible)
2. having it in the right format (like the latest picture not the one before it)


one more thing. I took a look at this topic (How to write an Axis Assignments (Math) Plugin for SimTools 2.0 - API documentation) and if the sample code there is the current code of Generic2d option, my axis settings should return the raw values but it doesn't look like that.
I am not a programmer (I still know a little) and I tend to avoid programming as much as possible. it will be very kind of you if you consider this in your suggested solution

many thanks in advance

If the X-axis time unit becomes smaller, it will not look so messy, and then filter, the effect will be better, although some data accuracy is lost, the seat will be softer. I have the same simtools settings as you. (I don’t know if there is an accurate expression of what I mean, because this is translated by Google)

Hey @farhad

Without having even the slightest clue about SimTools, from the graphs you posted alone I get the impression that there might be two problems involved:

1. Unit conversion (degrees <--> radians) and...
2. A well intended (but for a car sim completely pointless) anti-rollover-function.

I have analyzed plenty of motion data in my own software, and these two features in combination could explain those graphs.
I would go one step further and say: If you told me that you were going to use a unit conversion from rad to degrees and are also using an anti rollover function that expects radians, then I would expect EXACTLY those kinds of graphs.

First you should find out in which unit the game is sending values and in which unit SimTools expects then.
Then disable the anti-rollover function if you can. This might not be possible, if it is baked into the plugin by the creator who had the best of intentions.
Set the Min/Max values in Tuning center to a value that is significantly higher than any value you can ever expect to see on that respective channel. Just to be sure this doesn't interfere.

And I hope you will post your findings here, because I would love to solve this.

Cheers,... Dirty

If you look closely, in this image...

...the values range from negative pi to pi (-3.14 to 3.14).

AND they seem to "roll over" from neg pi to pi.

Imagine if you sent values to this function that are 57.296 times bigger (thats the conversion from rad to degrees), you'd see 57.296 rollovers in that image and "57.296 rollovers" is pretty much what I see here:


Does that make sense?

Hi @liuzhimin @noorbeast @yobuddy and @Dirty
Thank you for your answer. Sorry I didn't check this topic after a few weeks of getting no answer because I wasn't expecting any answer anymore

@Dirty I can't really remember what I have done then, but I am pretty much sure I have checked unit conversion, anti-rollover, and all other options that are in software nothing worked :/
So I started writing my own software ,and I am not using simtools anymore

Oh, nice to hear that. I think writing your own software is a great way to learn about motion cueing. I did too Start a thread about it, I'd love to follow this.

sure I will do. but for now, I am really busy, and if I make it the thread would be inactive for a few weeks maybe. but I will definitely do it in a few weeks and would send a post here to let you know