FlyPT 6DOF/Stewart/Hexapod Interface for linear and rotating actuators

Yes yes, all fine.

Asking only for precisions what’s behind the code as it helps better understanding and through, a better use.

I like to experiment but with some critical features sometimes, it’s better to know.

As you are very responsive I use the possibility to ask directly.

Cheers

“Auto gain” reduces the intensity while the clipping occurs for better scaling of the effect on degrees of freedom?

It’s something similar to “Capture Max Min” feature in SimTools’s Tuning Center or something different ?

I promise that’s the famous “last one” (question) ;-)

It's just I would rather he spend all his free time coding for now as opposed to explaining it as things will likely change along the way anyway and there will be plenty of time for that later.

Right now, at the current state...

No, it's not similar and I don't think I will go in that direction.
There's no max/min. Just gain.

Gain is there so you can easily lower the intensity of the effects.
This is specially useful in car sims. Motion can be hard after 20 minutes of driving. It really pushes for you. So actuators gain is an easy way to make it less intense. We reduce the stroke by decreasing gain.
At 0.5, all is half the value: stroke, pitch, sway....
For this, we can also use a filter on the actuators.

Now if you where not able of implementing some efficient filtering to achieve a good motion cueing, or if you made adjustments, let's say, to drive a Clio and now you drive a Formula, the effects might cause clipping multiple times.
Clipping in crashes is acceptable. It's caos.
But while driving, we want to keep cueings all the way, up to big (normal) G's.
So the idea of auto gain, is to apply a scale to the output, in such way that clipping doesn't occur.
If you enable auto gain, gain is going to decrease each time the rig clips (does not increase).
After some time, clip might happen only in a crash. So when you see it's stable, uncheck the auto gain.

I would say, that this is also a good measure to test the cueing algorithm/filters.
What logic says is that you should keep a gain of 1.0 and avoid clipping at maximum. Well, just a bit of clipping might not be bad, we can gain in detail for less intense cues.

But:
We want to notice small details, but also keep big ones.
To achieve this, it has to be done with the filters.
Gain helps on a fast approach, but you might end up with just the high loads.

What I want to implement:
An auto adjustment system that works on the filters, not on the gains (but might get info from the auto gain we got).
For that I want more info and resolution from the game
I want standardised info. I want to receive acceleration as it is, 10 m/s^2 or 100 m/s^2.
Rotations in a range of -180 to 180º.
I want to treat values after, not before.
This is why I'm adding direct input.

I even thought of recommending the use of the same values for all games in the SimTools tuning center.
But this has other problems for the health of the rigs right now...

Now let's just put the direct input working well and for two more games... we will see after.

May I ask you your opinion about “approximation algorithm” of center of rotation like this used by Austin Meyer in his X-Plane?

It doesn’t limit the dofs at all in opposite to classical center of rotation.

Is it something which is achievable internally in your software and not being to complicated for the user to set up?

Wow that lasted a little over three hours. Maybe he will find time to finish coding by next year sometime .

Give me time to answer.
End of month, need to pay employees, outsourcing, taxes, send data to the finances and legal entities... I'm full of work (in the computer, so I see the post's).

Give me the link to what Austin Meyer says (complete text), because just that doesn't tell nothing to me.

I know Austin Powers says "Groovy Baby" but yeah that's a different Austin.........................................okay I'll get my coat

Have another on me .

Hello guys. Still I can't figure out why my DOFs are swapped to AMC1280USB. Roll is pitch, yaw is heave... Etc.
I use hexpod interface from @hexpod without any problem so I don't want to change settings on AMC board. What do I make it wrong?

Double check your string:

<255><255><Axis3a><Axis4a><Axis5a><Axis6a><Axis1a><Axis2a><0><0><0><0><10><13>

I've checked it more than double And I've tried lots of different motors order. It's weird yaw and heave are always swapped, but with your hexpod interface everything work well. Honestly I have no idea whats wrong with it. My rig is upside down, I mean front side back, but yaw and heave cant be swapped because of that.

I changed geometry in the actuator setting, like in attachement.

Ok, in FLYpt there is a checkbox to invert the bit output 135, 246.

Output swap swapped. try this

sorry I can't find that. Where is that checkbox?

No hurry, take all time you need.

Unfortunately it’s the integral Austin’s “minimalist” response.

The reason for “(...)” is that it comes from two different mails.

No idea what kind of math he is using for his “CoG approximation”

I you are interested, to have a first idea, you could compare the “graph” output curves (g-load-angular velocities V.S. Motion Platform Stats) by superposing them in xplane.

Last possible pose got broken?

OMG...I found it finaly.. thank you

How can I use it on dual PC config to direct AC, LFS, rFactor, and PCars control? Where can I set IP address to?