FlyPT Mover

An actuator has a motor inside. It’s jus a matter of encoding. Rotary and linear actuation is supported.

Discovered a fairly big issue in AC which may affect more than the brake.

The brake start on 0 then go to 1 when you start the game. Then instantly go to 0 when you enter the car. You can see the clutch goes from 0 to 1 as i enter the car as well. This causes a massive instant shift on the rig as well as moving on startup

Thanks,

Just added more data in ac/acc. I will look into it, never noticed it.

Super!!!
I tried your settings, and the landing effect was great. Thanks.

Good day Hexpod,
Thank you very much for sharing your setup values. I've never used Basic Pose for Motion or Pose Suspension, you might have opened up a whole new world for me on that
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
My background...
I'm a Private Pilot (SEL & IFR rated, working towards commercial license until COVID hit, that's now on hold). I'm a partial owner of a 2003 SR22 Centennial Edition - N992LK). I'm a Mechanical Engineer (graduated in 1992) and also have knowhow in electronics, welding and light programing. I just built a 6DOF HEXPOD with a single seat and simple cockpit platform (with three attached 43" monitors). I'm in the middle of adding a canopy over the cockpit to remove other visual references (the room).
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Slight issue...
I'm running 2.9.9 and getting this error window when first opening your setup...
Found following errors while opening the file:
Get module "SOURCE :: XPLANE 11 (UDP)" fields (1x)

This source seems to load up, I'm not seeing any impact from the error ? ? ?
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Question on the Pitch & Roll setup...
I see that with both Pitch and Roll, your pose setups are using EMAHP (to slowly re-center them). I believe I understand this logic on the other 4 degrees of motion (inertial catchup). Is that the case also for Pitch and Roll? -- My current perception for Pitch and Roll are these are more a G-Vector/orientation vs inertia. Please let me know your thoughts behind them, thanks !

FYI; here are my setups for them (to manage non-intention rollover)...
• Roll... Range → 180° Filter → LOGISTIC(ROLLOVER(EMALP(EMALP(VALUE;250);250);90);15;1)
• Pitch... Range → 180° Filter → LOGISTIC(ROLLOVER(EMALP(EMALP(VALUE;250);250);90);10;1)

Thoughts on Yaw/flying uncoordinated...
Have you given any thoughts to simulate uncoordinated flight caused from bad Yaw control? -- Perhaps a slight lean and/or tip to put pressure on a butt check?
________________
Best regards,
Mark

I Think the two questions are relevant and interesting but they were clarified many times already.

I could respond on those but as I don’t have much scientific background I would maybe let someone more qualified do the job.

@Dirty, would you take your time plz. and recenter again the motion cueing ideas concerning :

1.) forces in coordinated flight
2.) relevance of HPF filter on pitch and roll velocities ?

If you get an error in mover, you should report the bug with detailed description and I am sure @pmvcda (mover designer) could easily solve it

Thanks

@Mark Pattison ,

The error you get is because @hexpod has a test version a bit different to the current release.
But should not affect the results.
New version is almost ready and will include FS2020 (and F1 2019) really soon.

I play mostly race car games, although I like air crafts also, I lack experience and vocabulary. @Dirty is the expert

Got a question for everyone regarding flight sims and airframe shake on high G/aspect

DCS:
On Warbirds and some jets i get the stall shake and use the noise in mover for that. For high G and while you are very close to stall/lift vector there's a lof of airframe shaking going on in the F14/F16/F18as an example. I have not been able to get that in my rig. Im wondering if anyone has and how to do it? It's the one thing im missing in my flight profile.

Hey @Mark Pattison ,

I have to admit, I do feel flattered that I was @mentioned here I will try to illustrate my view on this.

A little disclaimer: I work with and in Zero-Flight-Time certified LevelD Full Flight Sims, but I do not work on them. So take all of this with a grain of salt. I am not the definitive authority on this, just someone who likes to tinker a lot.

If you read papers on the topic from the 1970s on, you will often hear the tern "classical motion cueing algorithm". It's called classical, because it's pretty much the grandfather of motion cueing algorithms and even the newer variants (e.g. optimal control and model predictive cueing) are as far as I understand them really just variants that improve on some weaknesses here and there. At their core they still use a combination of filters.

There are two main paradigms to motion cueing:

1. Orientation cueing
2. Acceleration cueing

Orientation cueing is what most people intuitively understand and try to achieve. They think that if their home cockpit moves around, it should be in the same orientation as the simulated vehicle. If the plane pitches up, so should their cockpit. As intuitive as it may seem, there are a thousand reasons NOT to do it this way. I have not seen this used in ANY professional flight sim. AFAIK only Cruden use it PARTLY in a semi-professional driving sim and even there they deliver an explanation with it saying basically: We know it's technically wrong, but we found that test subjects complain less and adopt more correct habit patterns when using this algorithm. They call this hybrid "Cruden Cueing" and I find their reasoning behind it totally legitimate! I might even adopt some of it in my own software.

The other approach diverts attention away from the vehicle orientation and instead asks the question: What accelerations does the pilot/driver experience? Hence the name acceleration cueing. Then those accelerations are replicated, as far as this is possible.
The starting point is the motion vector. A set of six numbers that describe the state of the vehicle as far as its movement is concerned:

1-3: Accelerations along the vehicle x, y, z axis
4-6: Angular rates around the vehicle x, y, z axis

These numbers completely describe the motion of the vehicle at any given moment!!!

Sorry, I had to highlight this It is just too important! These numbers are what happens on board the vehicle, nothing else. I hear a lot of people argue like: I want my sim to do this 'n this when the vehicle does this 'n this. But at the end of the day no-one cares what someone wants, only those numbers matter. And for good reasons! Those numbers represent the sum of all forces and the sum of all moments on the vehicle. There is literally nothing else.

--- Oh my god,... this post is getting long ---

Trying to answer your questions:

Here on the left side are those six numbers (pale green).

...and then they are pushed through this combination of HP and LP filters that have the job of eliminating all the information in the signal that fulfill at least one of these criteria:

1. It would drive the rig against its limits
2. It cannot be perceived by a human

What's left is the motion of the rig

It may seem strange, but when done right (!) you can use those filtered numbers to drive a sim that is Zero-Flight-Time certified by the FAA and you get a valid license to fly a Boeing 747 or an A380 around the planet. With a good conscience!

Pitch and roll do both. They are deflected constantly for constant accelerations and are simultaneously used to display the HP filtered signal from the roll-rate and pitch-rate channels. Both exist in superposition, because both are contained in those six numbers.

You don't need to give any special consideration to those. Those accelerations are already present in the six numbers, and they will still be present after the filters because they don't violate any of those two criteria. If you fly un-coordinated, you will have a continuous lateral acceleration present (that's literally the definition of un-coordinated flight) and that is displayed as a tilt in the roll DOF. See here:


I know, it's a lot to digest, but I can only recommend to understand (1.) the accelerations on your simulated vehicle and also (2.) the way this information is filtered. Pretty much ALL questions answer themselves when these two steps are taken.

Just keep asking,.. I totally enjoy this.

Cheers,... Dirty

Thanks @Dirty, no one better to explain it.

I would take the chance to say that in cars...

Traction loss is present in those 6 values, so if you have a rig with yaw and sway, you should get a good representation off TL just using the 6 values and without need to use a fake TL value.

Sorry, out of the aircraft talk, but a question that comes up many times.

Hi Flypt,
Home everything is ok for you and your family .
I downloaded automobilista 2 , is it possible to have a plugin for mover ?
Thanks !

ok no plugin necessary works with shared memory from pcars

@Mark Pattison I am a great fan of Dirty's explanations of the various misconceptions about motion simulation.

That said, I have put considerable time into digesting his explanations, making incorrect arguments, getting set straight by him, and digesting again!

Something I "knew" but never truly understood until recent years is that "Gravity is an acceleration". This is possibly the biggest key to understanding what Dirty teaches us!

Where we get hung up is in assuming that we have to manually create certain situations (like rolling the sim for an uncoordinated turn or knife-edge), where as Dirty explained if the motion software is written correctly they take care of themselves in the simulator for the same reasons they occur in the actual vehicle!

There is a catch here though, I said "if the motion software is written correctly". Until the past couple of years when we got some of this fixed largely thanks to @Dirty none of the available software solutions I'm aware of were working this way! The most obvious symptom was that they were a mess when you rolled through inverted or pitched past vertical. There is still a lot of motion software and plugins floating around the internet that continue to do it the old way and have these problems.

That's because they were using "Euler Angles" for pitch and roll rather than looking directly at roll rate changes, and separately modeling the acceleration of Gravity itself.

If you use Euler Angles for your roll cues and you roll through inverted, when you pitch past vertical you just instantaneously switched from one end of the scale to the other and the sim has to go to the opposite end of it's pitch travel as fast as possible. Feels a bit off when the direction of Gravity barely changed during that same instant! It also made roll cues work backward when flying inverted! Roll left, get a left roll cue. Fly inverted and roll left, and you got a right roll cue!

We were stuck simulating pitch and roll rate accelerations via the change of angular orientation of the aircraft. Now we are able to simulate pitch and roll rate accelerations based on.... pitch and roll rate accelerations!

Watching people like @Dirty @pmvcda , and @SeatTime completely transform the hobby motion flight simulation world from a hacky mess to what it is now has been hugely gratifying. If you use all of the information and techniques you can learn from those three here on these forums in both your software and hardware setups, I think you've ultimately got a setup that is superior to the big money professional stuff! The professional sims generally shoot for a certain bar for "good enough" which we reached in the past two years. @SeatTime and a few others have come up with innovations that go beyond that to really create dramatic improvements in immersion. I think there is a pretty good possibility that we will see some ideas pioneered right here appearing on big ticket sim equipment in the years to come.

The more advanced models of the ACME DMS are the best I've found, and we've got a few really significant improvements they could use. Not to mention more elaborate features like head loading devices they just don't bother with.

Solving motion compensation for VR was the last really big hurdle, and now @Dschadu has come to rescue us from that too!

We live in an amazing time for simulation that's for sure! Things i could never dream of (having funds for) is happening and when i found this software it was truly a game changer. Im not as techy as you guys on this but with combining Roll/pitch speeds (havent tried the accelerations yet) and very slow pitch position i've been able to trick my brain considerably. Sometimes i honestly feel like im inverted coming out of a very low G loop.

There is one thing that is missing though. Thats the knife edge. As a virtual formation pilot with 500 hours or so not including the live shows that's something that is important to me. Lateral Lat acc with gravity does not take vertical acc /gravity into account so they need to be mixed. The more g's the less angle the rig has. Finding a good balance there will probably be hard but if that could be implemented it would truly be awesome. Unless someone already has a solution that is?

In its essence (my opinion) what i am saying is that vertical/acc gravity must have an effect on roll/sway gravity and so on to get a proper motion cuing for flight. It's good now, its just that tiny little bit that's missing

@hexpod
Thank you so very much for the setup... You've given us all a fantastic reference and great example of features. Not sure I'm a fan of the Pose for Noise (RPM), just seems like it's a jack hammer to the gears, but that's only my initial perception, no data at all to back that up, but in theory I love it.

Can you explain what's happening with the Pose for Motion (on Heave)?
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
@pmvcda
Fantastic software... Every update just gets better and better !
Please don't get rid of Singularity... This is brilliant, once I realized what it was doing for me.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -+
@Dirty
What can I say... Okay, you're lightyears ahead of me. I truly appreciate your time in helping me, which you definitely have helped me understand much more. Also, those graphics were fantastic, like a cornerstone/foundation !

Lets see if I understand so far...

Questions: Have you had a chance to look at the setup from @hexpod ? -- Again, thanks to @hexpod for sharing.

From what I believe I see in the charts you provided (image below), I've circled two areas, Blue and Red...


My thinking on this...
The setup seems to return everything (except mixture for sway and surge) back to zero if there is no change in rate or acceleration. I would call this okay (inertial catchup), which I believe is the Blue area I've circled.

I don't believe I see any poses that support the Red circled areas? -- These areas wouldn't return back to zero, I would call this the G-Vector simulated. Example: When you're driving on road with a slant to the right or left, you feel that slant if there is no change in acceleration (or when there is no acceleration).

What are your thoughts? -- I like to think in force vectors, so when one force vector is pulling in a different direction than that of gravity, your body feels the resultant. Sometimes, other force vectors are so great it can seem to reduce gravitational forces to zero, where everything seems to pull perpendicular to the seat no matter what orientation the plane is in (like in a Barrel roll vs an Aileron roll). Below are a few images I've created, to ideally help you see my thinking (which could be wrong). Figure #2 → Figure #4 is what I think is missing from the setup (the Red area I circled on the above graph)...


____________
Best regards,
Mark

@Mark Pattison I wrote this before I saw your most recent post, but this post might help you as well since you have a similar question. Let me know if you want me to clarify or elaborate anything.

If you have your simulator set up correctly you will get the effect you are looking for. If the aircraft is being flown knife edge the 1G acceleration of Gravity IS there and being simulated and will give you that cue because just like the real deal there aren't any opposite forces cancelling it out.

Off the top of my head I'd say the reason for your missing cue is that in your setup you are counting on your "pitch" and "roll" cues angles for your "gravity" cues. Likely via evil, evil Euler Angle telemetry! Anyway, you are theoretically using washout on your roll and pitch cues and so "gravity" gets washed out along with the rotational acceleration cues. If you are using only actual acceleration telemetry your gravity cues are coming from your sway and surge cues instead of pitch and roll cues, and so they are not affected by your pitch and roll cue washout. This way the gravity cue will persist no matter how long you maintain knife-edge flight. On the other hand in a coordinated turn there will be a sway cue from the acceleration of the aircraft changing direction and that will be equal to and opposite the gravity cue so they cancel one another out. Again, just like the real thing.

Get rid of Euler angles completely if you haven't already and rely entirely on roll rate and linear acceleration telemetry. Roll rate gives you PROPER cues for rolls with washout, and the accelerations source for sway will naturally give you the pull of gravity when rolled on your side like that and you aren't generating any opposite force such as in a coordinated turn. The sway acceleration cue should not have any washout. Neither should the surge cue. Unfortunately washout is basically mandatory on heave when using a traditional motion sim. Sustained heave is the domain of G-systems such as G-seat panels and harness tensioners.

If that doesn't totally clear this up for you, please do reply with more specific questions so we can get this working right both in your head and on your simulator!

Side Note- In high performance aircraft maneuvers and in fast race cars that experience quick reversals of turn direction (rally, road course, etc.) most motion simulators are very poorly suited to providing sway acceleration cues. This is one of the places where you really need to count on your G-seat for most of the cue. The reason for this is the process of the simulator changing position to switch from a left sway cue to a right sway cue. It can't switch instantaneously. Whatever time it takes to switch is perceived as a lag or delay in the cue, and what is worse the movement required to switch over in this sudden reversal situation is fast and generates a HUGE false roll cue that also IMO is perceived partially as a sway cue OPPOSITE the correct direction! This is why I abandoned the traditional motion sim concept for G-systems supplemented by real heave. If I had more funds, I would add more "real" motion in addition to the heave, but the G-systems should do well enough for everything else to satisfy me, so I don't feel the need for more motion unless my income improves. =)

For quick reference this should be your overall setup for sources on a MOTION SIM (not G-seat or other G-systems)

Format is:
Desired Cue -> Motion Sim functions used

Roll rate accelerations -> cued via roll movement only and washes out to prevent false sway cues and to prepare for the next roll rate acceleration cue

Pitch rate accelerations -> cued via pitch movement only and washes out to prevent false surge cues and to prepare for the next pitch rate acceleration cue

Yaw rate accelerations -> cued via yaw simulator motions only and should wash out to prepare for the next yaw rate acceleration cue

Heave accelerations -> Cued by heave (and possibly a little bit of pitch if you want to be creative) but requires washout to prepare for the next heave acceleration cue. Supplemental G-seat/G-systems to provide a sustained heave acceleration cue highly recommended!

Surge accelerations -> cued with pitch, NO washout! Optionally actual surge motion can be used if available and if used that should generally wash out to prepare for the next surge cue. Quick changes in required surge cue direction are problematic on motion sims and best achieved with G-seats and other G-systems.

Sway accelerations -> cued with roll, NO washout! Optionally actual sway motion can be used if available and if used that should generally wash out to prepare for the next sway cue.
Quick changes in required sway cue direction are problematic on motion sims and best achieved with G-seats and other G-systems.

Extra Note: In aircraft simulation only, I would personally consider going without washout on the sway rate acceleration and yaw rate acceleration cues.

@Trip I see you have edited your post since so ill leave out the quote.

I thought i was quite clear but let me be very specific with the situation. I agree with you and all the coordinated turn stuff but the software still needs to account for and mix in the changes in (heave) acceleration so that the rig changes its angle depending when its cancelling out the sideways pull For order this is based solely on DCS as thats what ive been testing.

My rig is setup with only accelerations and speeds with LP and HP so it does not hit it limits even when i smack the stick over at 400kn i a clean F16. You can see it in the video on the bottom. Note im also using HP/LP on sway in this but its not in actual use (no movement). Normally it just has a LP but i prefer not to use it and rely on only turn rate due to false cues.

Here's the scenario
Aircraft is at 90 degree.
lateral gravity is at lets say 10 and my rig is tilting 30 degrees
Heave axis is at 0 acceleration
There are no other accelerations.

Essentially that simulates me falling to the earth at 1g sideways. There are no other forces. Now if i pull my stick it will cancel out the 1g acceleration downwards and have 3gs of acceleration pushing me into the seat. Since they are not mixed in the software my rig will still be at 30 degrees cause that's the data its getting from the sim and i will feel as though there's no acceleration in the heave axis cancelling it out. Therefore they need to be mixed with a LP so the transition is relatively smooth or how you wish it to be. Then its up to you if you wish to make use of it or not.

This is also an issue in the G seat. Your left flap will now be pressing a lot more than your right. If you got a slider it matters less as your body gets pushed more or less straight back.

For a quick knife edge bang on stay half a sec and bang back there's no issue but sustained for any length of time and its not correct. So either you use no sway at all or you have a Sway/heave/pitch mix.

Do i agree with you on that it can cause unwanted fast ques and that a moving rig isnt entirely suited to it yes but id rather have the option than not. Especially for G seats where this is much less of an issue.

I see we essentially agree on these things but what we want are a bit different. I still stand firm on of having the option to mix in sway/heave/pitch though

Trust me when i say i can't stand unnatural forces so this is exactly how i am thinking and set things up accordingly. All my profiles have forces the correct way. If i slam the brake the rig will pitch hard but relatively short and throw the wheel in my face. At the same time my G seat back will move backwards. Same on the sway it will run the right way but only very slightly until my G seat takes over. Small sharp correct direction (so not perceived as roll) forces and longer smoother on the G seat.

I edit a lot. =( In fact, I probably fixed some things again after your last read too, so maybe give it a quick look again.

Note: It's very easy to get things mixed up because sometimes we are talking about directions and forces in the virtual aircraft, and other times we are talking about forces in the simulator and real world. This is what a lot of my edits to that post were meant to correct.

I don't believe the behavior you describe here should happen if the sim is set up right. When you are in your knife edge flight as you say the sim will be at your maximum allowed simulator roll angle to simulate your feeling of falling sideways toward earth. This is being driven by the sway telemetry picking up the 1G acceleration of gravity.

Now when you pull back on your stick the sway telemetry remains basically the same, but when you pulled the stick back the heave telemetry coming from the flight simulator will have gone from zero to whatever value resulted from how hard you pulled your stick.

If pulling back the stick from knife edge does not result in the heave telemetry showing an acceleration value from that pull on the stick something is wrong.

So when you go from the only acceleration being 1G of sway from knife edge, to that PLUS let's say 4 G's of heave from pulling the stick back your motion simulator should react with a very significant heave cue.

I think I see where you are coming from now. For now we will talk about only a motion sim, with no G-seat to simplify the discussion. If you are still rolled to a perfect 90 degrees the sway cue value does not change because a heave force is not opposite a sway force. But because of the extreme limitations of a motion simulator in the heave DOF, the heave cue which should be a sustained force four times as great as the sway cue is actually much weaker and only a momentary acceleration that then washes out. So where in real life your 1G of sway would hardly be noticed as you endure a sustained 4 G's of heave, now in your simulator that 1G of sway is the only sustained force and therefore seems totally wrong. Not because it shouldn't be there, but because there should be so much heave force that you don't notice it any more!

Very interesting! You have found a case where the lack of heave capability of a motion sim creates a situation that could potentially have a band-aid put on it to help conceal it.

One thing that I came up with myself that some may not agree with is using pitch for heave cues. I feel that when the sim pitches up, the feeling of being pressed back into the seat can pass as a heave cue, and that pitching the sim down so that you are starting to fall forward out of the seat reduces the pressure on your bottom and back which can pass as a negative heave cue (weightlessness). Perhaps experiment with adding LOW PASS pitch motion to your heave cues. =) I think that will help a lot!

Now add to that as much heave cue as you can get out of your G-seat and you might have something approaching acceptable.

What you are suggesting though, is that when heave cue and sway cue occur together the sway cue be lessened so that you don't feel this big lateral force when it should be buried under a massive heave force that is largely missing. I'm not sure whether or not that could be done in a way that wouldn't cause trouble in many other scenarios. Let me think on that a while. =)

Again @Mark Pattison this may be relevant to your questions as well.