While configuring control sensitivity on three different PCs, we observed the following:

• Aileron and Elevator: A sensitivity setting between -20 and -25, combined with an Extremity Dead Zone of +0.8, provided consistently good handling and control.

• Rudder: On all three systems, a sensitivity setting of approximately -50 was required to achieve a comparable feel, using the same Extremity Dead Zone of +0.8. The rudder pedals tested were the Thrustmaster TPR, MOZA, and Honeycomb models.

We're interested in learning which rudder sensitivity settings the wider flight simulation community uses.

Any feedback would be greatly appreciated.

Mostly with the Phenom, Raul chose the correct settings.

dd

I use the TM Warthog HOTAS and Logitech (Saitek) rudders.

I make Sensitivity and Dead Zone adjustments for almost every addon - especially helicopters.

Usually a world of difference in terms of reasonable handling.

I tend to apply far more adjustment to Pitch than Roll or Yaw.

For high performance aircraft I try to keep it more on the Dead Zone than the Sensitivity, again especially with helos.

For GA & airliners, Sensitivity usually suffices as combining both can create conflicting effects - and I’m lazy 😂.

Are your settings for all addons or aimed more at specific types?

I’ll note that I usually add an extremity Dead Zone for my “Friction Control” lever as it has notable “chatter” at the extremes. Some engines don’t like their Fuel/Condition levers to be continually micro adjusted 😉

I’m AFK, so I can’t ref my own numbers, but IIRC yours seem reasonable.

I’ll also give these a try on a few and see how it works.

14 minutes ago, UrgentSiesta said:

I use the TM Warthog HOTAS and Logitech (Saitek) rudders.

I make Sensitivity and Dead Zone adjustments for almost every addon - especially helicopters.

Usually a world of difference in terms of reasonable handling.

I tend to apply far more adjustment to Pitch than Roll or Yaw.

For high performance aircraft I try to keep it more on the Dead Zone than the Sensitivity, again especially with helos.

For GA & airliners, Sensitivity usually suffices as combining both can create conflicting effects - and I’m lazy 😂.

Are your settings for all addons or aimed more at specific types?

I’ll note that I usually add an extremity Dead Zone for my “Friction Control” lever as it has notable “chatter” at the extremes. Some engines don’t like their Fuel/Condition levers to be continually micro adjusted 😉

I’m AFK, so I can’t ref my own numbers, but IIRC yours seem reasonable.

I’ll also give these a try on a few and see how it works.

Thanks. I would like to wait to get more details to identify where the problem might be.

I don't want to jump to any conclusions, which is why I'm interested in the numbers used in these settings.

Much appreciated. 

Based on the research I've found online, many of the recommended sensitivity settings above -30 (I've seen up to -70) appear excessively aggressive, creating steep S-shaped curves that result in unrealistic handling characteristics. Users frequently describe the outcome as "twitchy" or overly reactive compared to the predictable, proportional control response expected in real aircraft.

As a general rule, we should avoid using a Dead Zone unless the controller exhibits measurable noise or jitter. Real aircraft flight controls do not have a dead zone, so introducing one unnecessarily reduces precision around the neutral position.

A word of caution regarding the Extremity Dead Zone: values above +0.9 are generally not recommended. Increasing this setting reduces the available control travel, limiting maximum control deflection and resulting in unrealistic aircraft responses. If anyone is interested, you can monitor the flight control synoptic page as you move the controls. With a high Extremity Dead Zone setting, you'll notice that the control surfaces no longer achieve their full travel.

A Sensitivity setting of -50 softens the response around the center, making small control inputs less aggressive. However, as the controls approach their limits, the response curve becomes progressively steeper, making the controls increasingly sensitive near the extremities and bringing back the notoriously "twitchy" feel.

The larger concern is that these exaggerated S-shaped response curves produce a control feel that is neither linear nor smooth throughout the full range of travel. This can lead to overcontrolling, inconsistent control authority, and abrupt aircraft reactions during larger control inputs or precision hand-flying.

This may be worth bringing to ASOBO's attention. A review of the current sensitivity implementation and response curves could lead to a more realistic and intuitive control model for all users.

My VKB Gladiator has 0 dead zone, ailerons axis to -50, elevators to -75.

With default sensitivity, moving the joystick from centre causes excessive control movement, only a slight pressure is required. I've used two joysticks and a yoke in MSFS, and all of them required the same heavy dampening.

IL-2 works fine out-of-the-box, you can use the full range of the joystick without requiring 'S curves'.

5 hours ago, flyingscampi said:

My VKB Gladiator has 0 dead zone, ailerons axis to -50, elevators to -75.

With default sensitivity, moving the joystick from centre causes excessive control movement, only a slight pressure is required. I've used two joysticks and a yoke in MSFS, and all of them required the same heavy dampening.

IL-2 works fine out-of-the-box, you can use the full range of the joystick without requiring 'S curves'.

That is very unusual, especially at -75, IMHO. I've never heard of such values. I'm wondering whether you calibrated your joystick properly using its own calibration software rather than in Windows. Very strange; you have very high-quality Hall sensors there. 

1 hour ago, LRBS said:

That is very unusual, especially at -75, IMHO. I've never heard of such values. I'm wondering whether you calibrated your joystick properly using its own calibration software rather than in Windows. Very strange; you have very high-quality Hall sensors there. 

I checked my settings last night and they are: Ailerons -0.5, Elevators -0.75 for GA & -0.5 for airliners, VKB Rudder -0.5

I don't use Windows calibration and the VKB calibration is default with full, linear movement of the joystick. As mentioned, I don't have to change sensitivity settings in other sims, and this is the third controller I've used in MSFS where I've had to heavily damp the sensitivity to get the controls how I like them.

I've had 4 different controllers since the release of msfs and always found with all of them, settings at -70%, -80% and even -90% are needed in most cases. Even with these settings it's hard to get the sensitivity right. Too low and there is insufficient deflection for certain manouvres and too high just makes aircraft too twitchy to even land.

I would have expected most controlers to need a small adjusment from the default zero, but that has never been the case for me with all the controllers I have used.

For @flyingscampi and @cianpars , I'm glad you're one of the few people who understand and have identified the real issue.

The problem is not simply that the controls are too sensitive around the center. The real problem is the way ASOBO has implemented the Sensitivity adjustment. Instead of providing a linear reduction in control response, it applies a pronounced S-curve across the entire control travel.

While this does reduce sensitivity near the center, it progressively increases the nonlinearity as the controls approach full deflection. The result is that aircraft response is no longer proportional to pilot input. In a real aircraft—or a Level D simulator—the relationship between control movement and aircraft response remains essentially linear, allowing the pilot to accurately predict the aircraft's behavior throughout the full range of travel.

For example, if an aircraft requires a sensitivity setting of -50% to -60% just to feel stable in normal cruise, the first half of the yoke travel may feel acceptable. However, the last portion of the travel becomes increasingly distorted, producing aircraft responses that differ significantly from the nearly linear behavior expected in real-world aircraft. This forces users to choose between excessive sensitivity around the center or unrealistic control response near the limits of travel.

This is the fundamental flaw in the current implementation.

ASOBO needs to understand that large S-curves are only masking the underlying problem. They may improve the feel around the center, but they do so at the expense of realism everywhere else. What the simulator needs instead is the ability to reduce overall control sensitivity while preserving a linear relationship between pilot input and aircraft response across the entire control range.

Until that is addressed, the aircraft will continue to exhibit the "arcade-like" handling characteristics many experienced pilots have been describing.

As you rightly pointed out, this is also why users of other simulation platforms generally do not have such aggressive sensitivity curves to achieve realistic handling.

Edited July 8Jul 8 by LRBS

54 minutes ago, LRBS said:

Instead of providing a linear reduction in control response, it applies a pronounced S-curve across the entire control travel.

I'm not sure whether the aircraft models are too sensitive to control movement, or the controls are too sensitive to controller movement.

Without changing the sensitivity settings, I can fly aircraft normally in MSFS without moving the joystick, I just need to apply a slight pressure. I've compared MSFS to IL-2 before where control is linear and 'natural', but I got 'shot down' for comparing MSFS to a combat sim 😉

29 minutes ago, flyingscampi said:

I'm not sure whether the aircraft models are too sensitive to control movement, or the controls are too sensitive to controller movement.

Without changing the sensitivity settings, I can fly aircraft normally in MSFS without moving the joystick, I just need to apply a slight pressure. I've compared MSFS to IL-2 before where control is linear and 'natural', but I got 'shot down' for comparing MSFS to a combat sim 😉

In my opinion, it's a combination of both. The primary issue is that the control response is not truly linear across the full range of travel, and the underlying flight dynamics are not yet sufficiently refined. The result is exactly what we're seeing.

Some developers have managed to tune or mask the shortcomings caused by the current S-shaped control response, while others have not. However, the common denominator remains the same: the underlying implementation of the control response is fundamentally flawed.

I also find the growing tendency to dismiss or shut down comparisons between products concerning. Honest technical comparisons are essential for improvement. If users cannot openly discuss strengths and weaknesses or point out obvious shortcomings, meaningful progress becomes much more difficult. Constructive criticism is not an attack on a product—it's one of the most valuable tools for making it better.

On 7/6/2026 at 4:37 PM, LRBS said:

Thanks. I would like to wait to get more details to identify where the problem might be.

I don't want to jump to any conclusions, which is why I'm interested in the numbers used in these settings.

Much appreciated. 

This is really going to depend on the hardware you're using. I never have a problem with default sensitivity settings with my yoke, but it's a long-throw Fulcrum One. When I use a joystick, default settings are way too twitchy.

2 hours ago, LRBS said:

For @flyingscampi and @cianpars , I'm glad you're one of the few people who understand and have identified the real issue.

The problem is not simply that the controls are too sensitive around the center. The real problem is the way ASOBO has implemented the Sensitivity adjustment. Instead of providing a linear reduction in control response, it applies a pronounced S-curve across the entire control travel.

While this does reduce sensitivity near the center, it progressively increases the nonlinearity as the controls approach full deflection. The result is that aircraft response is no longer proportional to pilot input. In a real aircraft—or a Level D simulator—the relationship between control movement and aircraft response remains essentially linear, allowing the pilot to accurately predict the aircraft's behavior throughout the full range of travel.

For example, if an aircraft requires a sensitivity setting of -50% to -60% just to feel stable in normal cruise, the first half of the yoke travel may feel acceptable. However, the last portion of the travel becomes increasingly distorted, producing aircraft responses that differ significantly from the nearly linear behavior expected in real-world aircraft. This forces users to choose between excessive sensitivity around the center or unrealistic control response near the limits of travel.

This is the fundamental flaw in the current implementation.

ASOBO needs to understand that large S-curves are only masking the underlying problem. They may improve the feel around the center, but they do so at the expense of realism everywhere else. What the simulator needs instead is the ability to reduce overall control sensitivity while preserving a linear relationship between pilot input and aircraft response across the entire control range.

Until that is addressed, the aircraft will continue to exhibit the "arcade-like" handling characteristics many experienced pilots have been describing.

As you rightly pointed out, this is also why users of other simulation platforms generally do not have such aggressive sensitivity curves to achieve realistic handling.

You’re describing a “speed of light” issue.

I.e., it’s not an Asobo/MSFS problem, it’s a math problem

I.e., it’s the same exact thing in every sim.

31 minutes ago, UrgentSiesta said:

You’re describing a “speed of light” issue.

I.e., it’s not an Asobo/MSFS problem, it’s a math problem

I.e., it’s the same exact thing in every sim.

Very interesting take, but it doesn't align with either my experience or the broader discussion within the flight simulation community.

There are numerous reports from both experienced simmers and real-world pilots describing overly twitchy control responses and exaggerated aircraft reactions in Microsoft Flight Simulator 2024. One of the most common comparisons is with X-Plane 12 and DCS, where many users report being able to fly comfortably with essentially linear control curves, while MSFS 2024 often requires substantial negative sensitivity (S-curve) adjustments simply to achieve a comparable level of controllability.

Describing this as merely a "speed of light" or generic mathematical limitation misses the point. If it were simply an unavoidable mathematical problem, we would expect every modern simulator to exhibit the same behavior. That is not what many users report.

The statement that "it's the same exact thing in every sim" is contradicted by a significant number of user experiences. In both X-Plane 12 and DCS, I can leave my controls essentially linear and the aircraft exhibits realistic inertia, damping, and proportional control response throughout the entire control travel. Small inputs remain progressive, larger inputs remain predictable, and maneuvers such as stall recovery, trimming, and high-angle-of-attack flight feel natural without requiring excessive sensitivity compensation.

In contrast, MSFS 2024's sensitivity implementation often forces users to introduce large negative S-curves to tame the response around center. While this makes the aircraft easier to control initially, it also produces an increasingly nonlinear response toward the extremes of travel. That is fundamentally different from having a truly linear control law combined with realistic aerodynamic damping and inertia.

The fact that so many users independently arrive at similar sensitivity adjustments—and frequently compare the handling unfavorably with other simulators—strongly suggests that this is not simply a universal mathematical limitation, but rather a consequence of how MSFS 2024 currently maps controller inputs and models the resulting aircraft response.