The post was however of a big interest by the fact of showing how choosing the right joystick is important when dealing with realism in MSFS.

IMO, it doesn't really matter. I'm quite content with my Saitek X-52 joystck. It center's with the spring, and there is no reason at all, to have it move slightly fore or aft with elevator position. In other words, I'd never even be thinking about my "real life" airplane's stick being slightly one way or the other while in flight.The sense of trim while simming is a case of my "pilot's" mind filling in some of the gaps. All sense of feel is dictated by what I see on the screen, and the joystick centering spring. Yet, I sense that I'm actually trimming an airplance, although it's really just a case of releasing stick pressure little by little, as I push the trim button, until the airplane reamins at the attitude I want. For a desktop sim, this is just fine for me. I don't want force-feeback.L.Adamson

Thanks that's the exact drawing I was referring to ;)Brett it could be that only the vertical CG offset is ignored...I haven’t tested this well.I suppose most real-life weight and balance calculations ignore the vertical axis...so it may have been overlooked.I might try relocating the datum so the vertical CG offset becomes zero – that could help.For now the aircraft does roll on the model’s pivot which is the underside of the wing spar.

That's quite possible.. as stated, I've never applied a ludicrous model orgin, just see if it happens.. lol :( I've always test flown for pitch/roll/yaw, from a locked, spot-view (mainly to make sure coordinated turns appear coordinated) ... All of my models are conventional enough that regardless of loading, the CoG and model origin are always near each other...

IMO, it doesn't really matter. I'm quite content with my Saitek X-52 joystck. It center's with the spring, and there is no reason at all, to have it move slightly fore or aft with elevator position. In other words, I'd never even be thinking about my "real life" airplane's stick being slightly one way or the other while in flight.The sense of trim while simming is a case of my "pilot's" mind filling in some of the gaps. All sense of feel is dictated by what I see on the screen, and the joystick centering spring. Yet, I sense that I'm actually trimming an airplance, although it's really just a case of releasing stick pressure little by little, as I push the trim button, until the airplane reamins at the attitude I want. For a desktop sim, this is just fine for me. I don't want force-feeback.L.Adamson

Agreed, and I have think that it's a non-issue. A good test would be to trim for level flight, and then use the keyboard (or mouse on the panel) to modulate the trim. If the force-feedback stick does indeed account for elevator trim, it will move during this hands-off test.

Hi you all !It is obvious that the issues rised here about MSFS (XYZ axes and elevator defelection angle at trim) are not so visible and have no significance for anyone just interested by flying, because the simulation works fine. For instance, let me tell you that , concerning the pivot issue, even real simulators (those used by airline companies for training) have a fixed one regardless of the cg simulated.However, those who are interested in going thoroughly into FD, to understand how simulation complies with aerodynamics rules and real world flying, may be bothered when they notice that some details have been ignored or simplified by MSFS.As for the MSFS aircrafts Pitch, Roll, and Roll axes, they really correspondent to the visual centre of model. One can check that just by defining a point on his screen (using a transparent paper on a spot view of aircraft) and making it coincide with the visual model centre. Playing with the joystick so that the model moves along the 3 axes, at different cg locations, will show that the model pivots always around the same point drawn on the paper.Now concerning, the issue of joystick and trim, I think we have to understand throughly how MSFS interacts with joystick and vice versa. In other words, how is simulated the link (mechanical and/or electrical) between the elevator and the joystick in MSFS. I am using just a simple spring centered joystick and what I can see is that a given manual deflection of my joystick gives a proportional deflection of the elevator in MSFS. Now how about the reverse process? Does a deflection of elevator (by the trim tab) lead to a proportional deflection of my joystick (ie. several centering and zero resistance positions depending on the trim tab status)? As for now, and as far as my spring joystick is concerned, I see the link only in one direction, that is the direction joystick ---> MSFS ( ie no centering feedback from MSFS). I don't know, may be MSFS has this ability to feedback and recenter joysticks that do no use springs for centering. My joystick centers always at the same centre (no spring resistance, zero deflection) and this situation (joystick centered) imposes, no matter how, zero elevator deflection.

As for the MSFS aircrafts Pitch, Roll, and Roll axes, they really correspondent to the visual centre of model. One can check that just by defining a point on his screen (using a transparent paper on a spot view of aircraft) and making it coincide with the visual model centre. Playing with the joystick so that the model moves along the 3 axes, at different cg locations, will show that the model pivots always around the same point drawn on the paper.

I'm gonna have to disagree (until my testing is in).. This is a perception thing. The focal point IS the model origin, so it's always the center of what you see. As demonstrated in my pivoting at contact-point during rotation... the animated pitch obviousy occurs AT the main gear (not model origin), but as the visual model moves, the point of view orientation moves with it, even if this pitch change happens a million miles away from the model. In other words .. A given degree of pitch change, is ALWAYS seen from a model-centered focal point, and will appear to have happend at that point, because the "camera" follows that point.. See attached image: The big box shows that the model actually swept upward as it reached the new pitch-angle.. The small boxes are before/after of what you see, as the focal-point followed the model.

the FS flight dynamics are mostly a 'black box' where the majority of knowlege about how the game simulates flight dynamics has been found by trial-and-error. the 'nuts and bolts' specifics have never been shared by MS for many stated reasons; the main one being the information in the 'box' cannot be shared as it includes details gained from the aircraft manufacturers simulated planes (ie, boeing, cessna, etc) who do not want the propriety information shared.oveall the game is more right than wrong. the detail in the dynamics is one of the reasons FS is not popular like other computer games ... planes are not 'easy' to fly. this leads to the questions: will MS dumb down the flight dynamics for FLIGHT to make the aircraft easy to fly.--

Try and tell that to an X-Planer!!

However, those who are interested in going thoroughly into FD, to understand how simulation complies with aerodynamics rules and real world flying, may be bothered when they notice that some details have been ignored or simplified by MSFS.

With respect, you may well be one of the very few people in the known universe with this degree of interest... :( Over the last few decades I honestly do not recall many other discussion of the specific points you've raised. :(

ANYway, as long as I propery offset the CoG and wing apex form this bizarre origin, it will pitch on the main-gear at rotation, and pitch from a point that takes the offset apex and CoG into account... it won't pitch around the model center, as though it were on some sort of invisible stick extending way out in front of it.

The major advantage to actually setting the reference_datum_position is that everything else is in reference to this datum.Normally the modeler will set the model's origin to be coincident with the "FlightSim Reference" position, which is typically 1/4 chord, centerline, waterline, but......it need not be! You can in fact -as you've mentioned- shift the model's origin to any value at all, no matter how absurd, since you'd only have to change the reference_datum_position values to compensate.Of course as you've also mentioned, doing so would seriously affect the Preview model's display. :(

However, those who are interested in going thoroughly into FD, to understand how simulation complies with aerodynamics rules and real world flying, may be bothered when they notice that some details have been ignored or simplified by MSFS.

This is a sub-$50 game. Of course many details are simplified or ignored.Anyway, provided the cg is somewhere near the model centre can anyone actually tell the difference?

The major advantage to actually setting the reference_datum_position is that everything else is in reference to this datum.Normally the modeler will set the model's origin to be coincident with the "FlightSim Reference" position, which is typically 1/4 chord, centerline, waterline, but......it need not be! You can in fact -as you've mentioned- shift the model's origin to any value at all, no matter how absurd, since you'd only have to change the reference_datum_position values to compensate.Of course as you've also mentioned, doing so would seriously affect the Preview model's display. :(

I often wondered why the reference point (datum) wasn't always just set to the model's 0,0,0 ... it makes making the config easy, as all coordinates (lights, scrapes, contacts, engines, etc) can be taken directly from withing Gmax .. ala, place the mouse pointer on a beacon, and the coordinates are given to you :( MSFS seems to like having the reference point set out in front of the model (ala all the default models)... and I later learned that that's actually realistic. The first reason given, was that all logintudinal coordinates will be negative, no having to mess with calculations that contain both negative, and positive numbers, and secondly, many home-built plans like to reference a rigid, known point; like the door for the hangar where it's being built.

I often wondered why the reference point (datum) wasn't always just set to the model's 0,0,0 ... it makes making the config easy, as all coordinates (lights, scrapes, contacts, engines, etc) can be taken directly from withing Gmax .. ala, place the mouse pointer on a beacon, and the coordinates are given to you :( MSFS seems to like having the reference point set out in front of the model (ala all the default models)... and I later learned that that's actually realistic. The first reason given, was that all logintudinal coordinates will be negative, no having to mess with calculations that contain both negative, and positive numbers, and secondly, many home-built plans like to reference a rigid, known point; like the door for the hangar where it's being built.

Another advantage of not relating the reference point to a physical point on the aircraft the coordinates don't have to be recalculated if that point changes during design.

This is a sub-$50 game. Of course many details are simplified or ignored.Anyway, provided the cg is somewhere near the model centre can anyone actually tell the difference?

For all intents and purposes.. no, you can't tell the difference.. to the point where it wouldn't matter, even if the model-center and CoG were far from eachother... and don't forget that a wing apex (center-of-lift) figures into pitch movement too... an we know that the loaded CoG is accounted for, because we can put an unrealistically large load in an aft baggage area, and the model will "pitch" back onto its tail (assuming an accurate scrape-point for the tail), while parked.. AT the main-gear contact point (not the model origin)..

Re. the datum.It is definitely seems practical to place the datum at the model’s origin - specifically for measuring all those offsets in aircraft.cfg. Otherwise the manufacturer's datum should be avaiable in the handbook.

As far as I know , for MSFS the model visual centre is the 1/4 MAC of aircraft wing. This is how MSFS works, and this means that the modellers have to stick to this condition.Unfortunately, not all modellers care about the poisition of the centre of the models they built.I think, a modeller must have first the accurate manufacturer's data (from Boeing company, airbus company etc.). Besides the dimensions, he must know the exact MAC length and the exact position of its leading edge so that he can put the center of his model at the right position (that is MAC/4). Same thing for the exact nose and main gear positions to comply with the ground considerations.A visual model (*.mdl) designed taking these considerations matches perfectly with the aerodynamic model (the aircraft.cfg monitored using for instance the Wrench utility). That is , using the same dimensions and the same MAC length and the same MAC leading edge position (besides the same positions for nose gear and main gear), and taking the MAC/4 point as reference, the image of the visual model must superpose (and makes one image) with the virtual model described by aircraft.cfg.Back now to how MSFS "manipulates" the visual model during flight. I remember that I have read many times that " MSFS hooks the model from the visual centre". For me, this means that X Y Z axes cross through this centre and not through the moving cg. It is true however that on ground, the model deals with the external world through the contacts points. Exact then, on ground, the model rotates on main gear at take off or when at rest and not correctly loaded( tailstrike when most aft cg exceeded).

Back now to how MSFS "manipulates" the visual model during flight. I remember that I have read many times that " MSFS hooks the model from the visual centre". For me, this means that X Y Z axes cross through this centre and not through the moving cg.It is true however that on ground, the model deals with the external world through the contacts points. Exact then, on ground, the model rotates on main gear at take off or when at rest and not correctly loaded( tailstrike when most aft cg exceeded).

Even on the ground, where my over-loaded C172 pitches at the main gear... you can zoom in and the pitch will "appear" to have happened at the model origin, becuase the "camera" moves to follow it.

Try this experiment... Execute a normal takeoff, and then replay it from tower view to see what happens..Edit: .. All cfg entries obviously mean all longitudinal coordinates.. :(