I agree, I just don’t wan’t muted controls to be the new norm, and get in the way of actual flight dynamic improvements.

1 minute ago, norman_99 said:

I agree, I just don’t wan’t muted controls to be the new norm, and get in the way of actual flight dynamic improvements.

Also agreed...and I don't think anybody really wants that either....especially the 3rd party devs who take their flight dynamics more seriously than Asobo has.

What''s the difference with Prepar3D for example? As long as I have used that sim, and FSX before that, I always had to increase the sensitivity setting and decrease the nulzone. The sensitivity in P3D defaults at 64, which makes control inputs lag behind and then there's that default nulzone of 36! So to make an aircraft fly better (or as it is supposed to) I always had to adjust the nulzone and set a different sensitivity (as per advice from Scott of A2A, min nulzone and max sensitivity). So was I my own flight dynamics designer all the time before MSFS as well?

Nothing new here, only now I have to decrease sensitivity instead of maxing it out.

Edited August 19, 20205 yr by Rimshot

Just now, Rimshot said:

What''s the difference with Prepar3D for example? As long as I have used that sim, and FSX before that, I always had to increase the sensitivity setting and decrease the nulzone. The sensituvty in P3D defaults at 64, which makes control inputs lag behind and then there's that nulzone of 36! So to make an aircraft fly better (or as it is supposed to) I always had to adjust the nulzone and set a different sensitivity (as per advice from Scott of A2A, min nulzone and max sensitivity). So was I my own flight dynamic designer all the time before MSFS as well?

Nothing new here, only now I have to decrease sensitivity instead of maxing it out.

I agree here...

6 minutes ago, EmaRacing said:

Yes, I could do with half the control surfaces travel, I don't do aerobatics and the small planes don't really simulate the loss of authority at close to stall speeds.

I think the the loss of authority is there in MSFS, but the response curve is so exaggerated past the half point of travel (at least for me at -60 setting), that the added deflection offsets what you would feels\ if you had a linear response travel, and then had to pull more because the loss of authority.

1 hour ago, fogboundturtle said:

I disagree. Every controllers have different control sensitivity characteristic. It would only make sense if everyone was using the same controller.  I am sorry Robert and OP but you are wrong.

It's important to understand that I have never said that control sensitivity doesn't play a part in the sim's response to input. Nor have I said control setups are irrelevant. But I have spent more than two decades designing flight models, and I am very familar with control setups and flight model variations. Imagine you are driving a car and that car has a lock to lock steering potential of say 30 degrees each side of dead straight. Ideally I want a linear response, so that the steering doesn't suddenly increase or decrease at a given lock input.

Aircraft are not exactly the same but similar. Light aircraft do not have sophisticated elevators for example. They are nearly always linear in their linkage to stick input. It would be a very bad thing if you set a light aircraft elevator so it hardly moved the control surface at all in the first two inches of back pressure then suddenly and exponentially increased in the next two inches. Likewise it wouldn't be desirable to do the opposite. You need a predictable and steady increase in authority, not a sudden one via some imposed response curve. All that will do is make the aircraft unpredictable and probably dangerous.

In general, when you see the animation of sticks and yokes in simulators, they broadly show the range of deflection. You can see this clearly in MSFS 2020, but the difference here is that if you have to reduce your stick sensitivity SO MUCH to damp down  on  substantial pitch up or roll with a tiny movement, which indeed is the case after having flown a number of its aircraft, this shows very clearly that you are trying to compensate for something else that is quite outside the stick movement itself.

It doesn't matter how you adjust the curves in the response menu. No aircraft should be able to roll that fast with a tiny input and especially not when it is close to stall or approach speed. Relying on control sensitivity as a compensation for excessive flight model control authority is just masking the problem. The problem is that the control authority of moving surfaces is generally set far too high. You can indeed alleviate this a little by making extreme adjustments in the control menu, but requiring up to -70% is plain ridiculous.

You cannot run a flight model just by adjusting stick sensitivity. All that does is attempt to cover up something else. I've been doing this stuff for a long time and I think I know what I'm talking about.

Edited August 19, 20205 yr by robert young

I’m going to have to agree with the issue being plane config/flight model over control sensitivity. I come from the racing sim world where your steering wheel input should match the in game input. If I turn my wheel 280 degrees, the wheel should turn that far in game. If your yoke/rudder input does not match the movement in game, you are just compensating the airplanes response to inputs by artificially lessening your control input. It would be interesting to see some data on maximum roll rates at different input angles (true angle seen in the game) vs the real world results. 

16 minutes ago, Sterlin44 said:

I’m going to have to agree with the issue being plane config/flight model over control sensitivity. I come from the racing sim world where your steering wheel input should match the in game input. If I turn my wheel 280 degrees, the wheel should turn that far in game. If your yoke/rudder input does not match the movement in game, you are just compensating the airplanes response to inputs by artificially lessening your control input. It would be interesting to see some data on maximum roll rates at different input angles (true angle seen in the game) vs the real world results. 

I don't think 280 steering wheel turn equals 280 degrees actual steering lock 😉 I get your point but degrees revolution of a steering wheel is not equal to the actual steering lock, but I see what you are getting at.

1 hour ago, hangar said:

Yes...it's true that the flight dynamics of some (if not all) of the default aircraft are flawed in certain ways...there is NO doubt of this. It's not JUST the sensitivities that are wrong. There are other flight dynamic issues with these planes as well, like thrust/drag and even some engine issues.

But...the only way forward is for 3rd party intervention, period.

Don't expect Asobo to change the flight modeling of their planes as, according to them, it's not likely to happen.

So...I assume that most (if not all) of these things can be rectified by either modifying the flight files or creating new aircraft. This has been the way of things since the beginning. Is it disappionting that we don't get realistic flight models out of the box? You Betchya! But at least, thanks to 3rd party devs, and other FDE tinker'ers there's a pretty clear path forward. 1 year from now all of this will be a non-issue...right now it's all we have and thats a let down I can't deny....but at least now we have a future within this franchise which is something we didn't have before.

When the flight dynamics are so out of wack for each airplane it's not the "3rd party intervention" responsibility to fix it. Yes the 3 rd party will make the fine tuning for the model.

Let me say this, if is true that Asobo "according to them" will not fix their unrealistic coding there is not so much future for them at this point in time.

The idea is also to be able to hand fly the airplane, today is not a proper representation of flight the aircraft dynamics and a big issue.

I'm with OP mostly...and let me preface that I'm not a pilot, but have modeled haptic control surfaces for "real world" simulators of another kind in the past.

I'd imagine any plane that is not fly by wire is nearly linear with an appropriate sensitivity.  There would be 2 deviations to this: 1) a definite null zone with some slack and 2) a potential null zone any time the yoke is held in a position and then goes back to center-center.  When modeling this set up, the dead zone will be most important for where you have to 'take up the slack' so to speak but then should be close to linear.  You would expect to see it become non linear at the very beginning and very end of the movement, but for the broad range of pitch/roll that it would be linear.

 

FBW is another animal - particularly in some of the fancier jets where pitch/roll don't always correlate to elevator/aileron movements as we would expect.  Don't forget, FBW isn't just fancy - often times the commands go to a black box and that translates to make the aircraft behave how we intended - it doesn't mean elevators moved 50%.  

 

FS2020 may have errors in their files for the planes - but theoretically, you would aim to just have your physical yoke be a 1:1 translation of what is seen on screen - pull yours back 10 degrees; the virtual pulls back 10 degrees, and the plane responds as it naturally would.  that could make you go darn near vertical in one plane and still be nearly dead on another.  I don't think the answer is in tweaking the sensitivity.  Getting from 90% to 95% or 95% to 100% - yes the tweaks in sensitiving and dead zone will be the final touch.  But if an aircraft isn't functioning as intended, it is much more likely to be the model and not your gear.

 

Just my 2 cents.

16 minutes ago, killthespam said:

When the flight dynamics are so out of wack for each airplane it's not the "3rd party intervention" responsibility to fix it.

I absolutely agree with you...and I never said it WAS our responsibility. It's too bad that the publisher doesn't feel that making their default aircraft fly more closely to their real world counterparts is important. But the point to my post wasn't regarding right vs wrong, it was about reminding everyone how we've always moved forward from this in the past and that at least we can look forward to the opportunity to even be able to move forward now at all, because before MS announced their return, there was no future in this franchise.

Saying that at least we can now have the opportunity to positively move ahead and develop better flying aircraft is NOT the same thing as denying MS's responsibility in providing a better flying experience out of the box (that's a whole other can of worms I chose not to open here as there would be no end to the arguing of different viewpoints about that).

Edited August 19, 20205 yr by hangar

4 minutes ago, zihmer001 said:

I'm with OP mostly...and let me preface that I'm not a pilot, but have modeled haptic control surfaces for "real world" simulators of another kind in the past.

I'd imagine any plane that is not fly by wire is nearly linear with an appropriate sensitivity.  There would be 2 deviations to this: 1) a definite null zone with some slack and 2) a potential null zone any time the yoke is held in a position and then goes back to center-center.  When modeling this set up, the dead zone will be most important for where you have to 'take up the slack' so to speak but then should be close to linear.  You would expect to see it become non linear at the very beginning and very end of the movement, but for the broad range of pitch/roll that it would be linear.

 

FBW is another animal - particularly in some of the fancier jets where pitch/roll don't always correlate to elevator/aileron movements as we would expect.  Don't forget, FBW isn't just fancy - often times the commands go to a black box and that translates to make the aircraft behave how we intended - it doesn't mean elevators moved 50%.  

 

FS2020 may have errors in their files for the planes - but theoretically, you would aim to just have your physical yoke be a 1:1 translation of what is seen on screen - pull yours back 10 degrees; the virtual pulls back 10 degrees, and the plane responds as it naturally would.  that could make you go darn near vertical in one plane and still be nearly dead on another.  I don't think the answer is in tweaking the sensitivity.  Getting from 90% to 95% or 95% to 100% - yes the tweaks in sensitiving and dead zone will be the final touch.  But if an aircraft isn't functioning as intended, it is much more likely to be the model and not your gear.

 

Just my 2 cents.

This is exactly what I was getting at with my post. Using the racing sim analogy, 280 degrees of steering wheel movement should show as 280 degrees of steering wheel movement in the cockpit in game (I wasn’t talking about actual steering angle at the tires Robert). Basically your yoke/rudder movement should match what you see in game, as you said in the last paragraph. 

You guys are missing one crucial point. Let me try to explain:

I don't own MFS yet (waiting for VR + preordered HP Reverb G2 release + Nvidia 3080 release), but I have simiar issue in X-Plane. I have non-linear curves and still controls are too sensitive, including opinion of real-world pilot who tried my VR rig (I have older veriety of Logitech yoke+pedals). I suspect the issue is not that simple: most gaming controls may have lesser range of motion than the real things. Especially a typical joystick vs real cockpit joystick that is 5 times as long in some planes and therefore has much wider range/arc of motion. I suspect most yokes also have less movement range than the real ones. That means that each degree of rotation cannot be the same as 1 degree of the real in-cockpit yoke, it will run out of range very fast. That's why it's mapped to real range accordingly and that's why there are curves. The curves are a workaround for not having a full range of motion. So the best you can do is set the curve strong enough to map close to 1:1 of the real thing in the central region, then accelerating towards the end, to cover a full range of motion. You may have to limit the far edge of the curve to prevent it from feeling ridiculous, but then some far degrees of control surfaces movement will be lost. Does MFS have enough control over curves and cut-off points? I don't know. You tell me. But even if it does - it is a compromise and you can't set 1:1 mapping unless you have a controller that really rotates or traves the same distance/degree as the real control, which most of them aren't. 

In other words, using the analogy here, if your steering wheel really can rotate 280 degrees, and the real car has the same range of 280 degrees - cool, you can map motion 1:1 and every degree would translate to 1 degree if rotation in the cockpit. But if your wheel only can do 170 degrees, you can't map 1:1. Without curve it will be crazy sensitive, and with curve it will be OK in the middle but even more crazy sensitive towards the edges.

Edited August 19, 20205 yr by romandesign

2 minutes ago, romandesign said:

I don't own MFS yet (waiting for VR + preordered HP Reverb G2 release + Nvidia 3080 release), but I have simiar issue in X-Plane. I have non-linear curves and still controls are too sensitive, including opinion of real-world pilot why tried my VR rig (I have older veriety og Logitech yoke+pedals). I suspect the issue is not that simple: most gaming controls may have lesser range of motion than the real things. Especially a typical joystick VS real cockpit joystick that is 5 times as long in some planes and therefore has much wider range/arc of motion. I suspect yokes also have less movement range than the real ones. That means that each degree of rotation cannot be the same as 1 degree of the real in-cockpit yoke, it will run out of range very fast. That's why it's mapped to real range accordingly and that's why there are curves. The curves are a workaround for not having a full range of motion. So the best you can do is set the curve strong enough to map close to 1:1 of the real thing in the central region, then accelerating towards the end. You may have to limit the far edge of the curve to prevent it from feeling ridiculous, but then some far degrees of control surfaces movement will be lost. Does MFS have enough control over curves and cut-off points? I don't know. You tell me. But even if it does - it is a compromise and you can't set 1:1 mapping unless you have a controller that really rotates or traves the same distance/degree as the real control, which most of them aren't. 

In other words, using the analogy here, if your steering wheel really can rotate 180 degrees, and the real car has the same range of 180 degrees - cool, you can map motion 1:1 and every degree would translate to 1 degree if rotation in the cockpit. But if your wheel only can do 170 degrees, you can't map 1:1. Without curve it will be crazy sensitive, and with curve it will be OK in the middle but even more crazy sensitive towards the edges.

This is correct - if the physical control mechanism isn't close enough to what is being simulated, then it is downhill from there.  Someone posted about racing and steering wheels - if the wheel is same size (or nearly) and the rotation amount is the same (typically -270 to 270) then it will be very close provided they are a 1:1.  If lock to lock is different, or the wheel is significantly smaller in diameter then it will behave differently, regardless of how well the vehicle is modeled.  Flight sims are no different, often times more exasperated with 2 axis instead of 1 axis.  But it is more complex than dropping sensitivity when they don't match in size/travel - this is where the lack of being completely linear comes into play.

And I suspect that's the reason it's so sensitive out of the box and people are complaining so much. Because Asobo has to model 100% of natural range of motion. There ARE people with professional or modified controllers, like huge joystick extentions etc. But most have small ranges and that feels overly sensitive. Maybe if would be smarter to make default curves not 1:1 but smaller. Looking at the videos, first of all SENSITIVITY = CURVE. It's not cut-off, there is no cut-off, the full range is always modeled. 0% sensitivity is 1:1 (not 100%!). So, unless your yoke or joystick rotates to the exactly same degree as in a real plane, you would have to use strong curves that will feel OK in the center, but will be exadurated towards the edges, there is no other way, unless you buy a hyper expensive realistic controller...

If you add a 60cm extension to your joystick, I bet sudennly the controls won't feel overly sensitive at all... I wish there would be an easy way to mod my yoke for larger rotation, but I don't think so. I already modded it with elastic bands instead of  springs for smooth pitch control without a center detent and it so much better already. Adding rotation is not that simple because of potentiometers used, and that's why they make them this way I guess...

Edited August 19, 20205 yr by romandesign