Force Feedback Settings by Game: Starting Points for 9 Sim Titles
Set the base driver torque limit once and change only the in-game multiplier per title. On an 8 to 12 Nm base, sensible starting gain is 50 percent in iRacing and Automobilista 2, 55 in rFactor 2 and Le Mans Ultimate, 60 in Assetto Corsa and ACC, and 65 in the F1 titles. On a 12 Nm and above base take roughly 30 percent off those figures. Only iRacing, ACC and Automobilista 2 give a usable in-game clipping indication.
Two settings are constantly confused and they do completely different jobs. The torque limit in the base driver is a hardware ceiling, set once before the first power up and then left alone. The in-game gain is a multiplier applied to the forces a title generates, and it is the number that changes per sim and per car. Every figure on this page is an in-game gain, quoted as a starting point for the clipping test rather than as a final answer. Nothing here should ever cause you to raise a driver ceiling.
Of the 9 titles charted, 3 give a usable in-game indication of when the output is saturating and 6 do not. That single fact decides how you tune each one. Where a meter exists, you can watch the output while you drive. Where it does not, the base driver telemetry window is the substitute, and on a console there is no substitute at all, which is why Gran Turismo 7 is the hardest title on the page to set correctly.
Where does the master force setting live in each title?
Menu wording changes between updates, so treat the paths below as the location rather than as an exact string. What does not change is the architecture: one master multiplier, a small number of genuine physics controls, and a larger number of synthetic effect sliders that are layered on top of the tyre model rather than derived from it. The last column is the mistake that shows up most often in each title, and in every case it is a synthetic layer being used to fix a master setting.
| Title | Engine | Where the master force lives | Built in clipping meter | Start gain, 8 to 12 Nm | Damping to use | Most commonly set wrong |
|---|---|---|---|---|---|---|
| iRacing | In-house iRacing engine | Options, Drive tab, Strength. Shown in Nm when the base reports a torque figure, otherwise as a percentage | Yes, a live force feedback bar that saturates visibly | 50% | Driver. Leave the in-game damper at zero | Linear mode left off, which compresses the top of the force range and hides the moment the front axle gives up |
| Assetto Corsa | Kunos AC engine | Options, Controls, Gain, plus a per car force feedback multiplier on the in-car app | No native meter. The standard route is a community overlay app or the base driver telemetry window | 60% | Driver. The in-game damping slider mostly adds weight, not information | Road, kerb and slip effect sliders raised to substitute for gain, which buries the tyre signal under synthetic texture |
| Assetto Corsa Competizione | Unreal Engine | Settings, Controls, Gain, with Minimum Force and Dynamic Damping beside it | Yes, the in-game readout shows when output is saturating | 60% | Title. Dynamic damping is speed aware and is the one in-game damper worth using | Minimum Force left at a nonzero default on a direct drive base, which adds a constant floor that never goes away |
| Automobilista 2 | Madness engine | Settings, Controls, Force Feedback, Gain, with Low Force Boost, FX and Damping alongside | Yes, a force feedback telemetry page is available in the HUD | 50% | Title. The in-game damping slider is part of the flavour and is meant to be used | Low Force Boost run high on a direct drive base, which lifts the noise floor and flattens the difference between light forces |
| rFactor 2 | isiMotor derived, Studio 397 | Settings, Controls, Force Feedback Strength, plus a car specific multiplier that is saved per car | No native clipping meter. Use the base driver telemetry window | 55% | Driver. Add the in-game smoothing only if the signal is genuinely rough | Global strength adjusted per car instead of the car specific multiplier, so every other car in the garage is wrong afterwards |
| Le Mans Ultimate | Studio 397 lineage, shared with rFactor 2 | Settings, Controls, Force Feedback Strength, with a per car multiplier as in rFactor 2 | No native clipping meter. Use the base driver telemetry window | 55% | Driver. Keep the in-game smoothing low | Settings copied straight from rFactor 2 and never retested, even though the cars and the tyre model differ |
| F1 titles | Codemasters EGO derived | Settings, Controls, Vibration and Force Feedback, Vibration and FFB Strength | No meter of any kind | 65% | Title. The wheel damper slider is the only damping control exposed | Understeer Enhance left on, which deliberately lightens the wheel at the limit and removes the cue you are listening for |
| EA Sports WRC | Unreal Engine, Codemasters | Settings, Controls, Force Feedback, Self Aligning Torque as the dominant channel | No meter of any kind | 55% | Title. Wheel Friction is the closest control and it should stay low | Self Aligning Torque left at the default on a strong base, which saturates on every crest and landing |
| Gran Turismo 7 | Polyphony Digital engine | In-game force feedback max torque on a 1 to 10 scale, plus a sensitivity setting | No meter, and no PC driver layer to watch instead | 1 to 10 scale | Neither. There is no damping control worth the name | Assuming a driver level torque limit exists. On console the limit is whatever the wheel itself is set to |
The damping column is worth reading as a group rather than row by row. Use the title's own damping in Assetto Corsa Competizione, Automobilista 2, F1 titles and EA Sports WRC, because in those titles it is speed aware and forms part of how the sim is voiced. Use the driver's damping in iRacing, Assetto Corsa, rFactor 2 and Le Mans Ultimate, where the in-game control is a blunter instrument. The rule that matters more than either choice is never stack both. Two dampers produce a wheel that is heavy and slow without being detailed, and the usual response is to add gain, which starts the clipping problem the damping was supposed to fix.
What gain should you start at for your torque band?
The table below scales each title's baseline, quoted for an 8 to 12 Nm base, across the four torque bands the site uses everywhere. The arithmetic is stated rather than assumed: the baseline is multiplied by 1.4 for a 3 to 5 Nm base, 1.2 for 5 to 8 Nm, 1.0 for 8 to 12 Nm and 0.7 for 12 Nm and above, then rounded to the nearest 5 and capped at 100. A weaker base needs a larger multiplier to reach the same force at the rim, and a stronger base needs a smaller one to leave the same headroom above it.
| Title | 3 to 5 Nm | 5 to 8 Nm | 8 to 12 Nm | 12 Nm and above | Scale |
|---|---|---|---|---|---|
| iRacing | 70% | 60% | 50% | 35% | Percent of the title's own scale |
| Assetto Corsa | 85% | 70% | 60% | 40% | Percent of the title's own scale |
| Assetto Corsa Competizione | 85% | 70% | 60% | 40% | Percent of the title's own scale |
| Automobilista 2 | 70% | 60% | 50% | 35% | Percent of the title's own scale |
| rFactor 2 | 75% | 65% | 55% | 40% | Percent of the title's own scale |
| Le Mans Ultimate | 75% | 65% | 55% | 40% | Percent of the title's own scale |
| F1 titles | 90% | 80% | 65% | 45% | Percent of the title's own scale |
| EA Sports WRC | 75% | 65% | 55% | 40% | Percent of the title's own scale |
| Gran Turismo 7 | 8 to 10 | 7 to 8 | 5 to 7 | 4 to 5 | 1 to 10 scale, not a percentage |
Read a column downward rather than a row across. Within one torque band, the spread between titles is real: the F1 titles output relatively modest forces and want a higher multiplier, while Automobilista 2 and iRacing generate strong signals and want a lower one. Read a row across and you see the more important pattern, which is that the stronger the base, the lower the number. A MOZA R21 at 21 Nm at $947.16 at 35 percent and a MOZA R3 at 3.9 Nm at $139.00 at 70 percent put a similar weight in your hands. The difference is that the R21 has fourteen unused newton metres above that weight, so a kerb strike has somewhere to go.
What does iRacing's strength figure mean in newton metres?
iRacing is the one title on this page that expresses the master setting in the same units as the hardware, which is genuinely useful and occasionally confusing. When the base reports a peak torque figure, the strength slider reads in Nm rather than as a percentage, so setting it to 5 means five newton metres at the rim, not five percent of anything. Using the band midpoints and the same percentages from the matrix above, the slider figures work out like this.
| Torque band | Band midpoint | Equivalent gain | iRacing strength | Example base |
|---|---|---|---|---|
| 3 to 5 Nm | 4 Nm | 70% | 2.8 Nm | MOZA R3 at 3.9 Nm |
| 5 to 8 Nm | 6.5 Nm | 60% | 3.9 Nm | MOZA R5 at 5.5 Nm |
| 8 to 12 Nm | 10 Nm | 50% | 5 Nm | MOZA R9 V3 at 9 Nm |
| 12 Nm and above | 16 Nm | 35% | 5.6 Nm | MOZA R21 at 21 Nm |
The striking thing about that table is how narrow the strength column is. Every band lands between roughly 3 and 6 newton metres of working force, whatever the base underneath is rated at, because that is close to what a real car actually puts through a steering wheel at racing speed and it is what a forearm can hold for a stint. Buying more torque does not move your working force, it moves the ceiling above it. That is the whole argument made in the wheelbase torque chart, restated from the settings side.
What should you take to zero before you start testing?
Effect sliders are not part of the tyre model. Road texture, kerb effect, engine vibration and collision channels are synthetic layers a developer adds on top of the physics so that a wheel with little detail still feels busy. Raising them makes the rim livelier while making the tyre signal proportionally harder to hear, which is precisely the wrong trade on any base with real resolution. Take them to zero, set gain by measurement, then add back only what solves a named complaint.
| Title | Take to zero first | Leave alone | Add back only if needed |
|---|---|---|---|
| iRacing | Damping, min force | Linear mode on, reduce force when parked on | Nothing needs adding back on a direct drive base |
| Assetto Corsa | Road effect, kerb effect, slip effect, damping | Per car multiplier at 100 percent to start | Minimum force only on gear and belt wheels |
| Assetto Corsa Competizione | Minimum Force, Road Effects | Frequency at the highest the base supports | Dynamic Damping in small steps if the wheel oscillates |
| Automobilista 2 | Low Force Boost, FX | One flavour chosen and left alone | Damping in small steps, FX last and lightly |
| rFactor 2 | Smoothing, in-game damping | Car specific multiplier at 1.0 to start | Smoothing one step at a time on rough cars only |
| Le Mans Ultimate | Smoothing, in-game damping | Car specific multiplier at 1.0 to start | Retest per car class, the prototypes are not the GTs |
| F1 titles | Understeer Enhance, Rumble Strip Effects, Off Track Effects | On Track Effects low but not zero | Wheel damper only if the rim oscillates on a straight |
| EA Sports WRC | Wheel Friction, Engine, Collision | Self Aligning Torque as the working channel | Suspension last, and only a little on gravel |
| Gran Turismo 7 | Nothing to zero, the controls do not exist | Torque ceiling set on the wheel itself | Sensitivity adjusted only after torque is settled |
What is worth knowing about each title specifically?
iRacing
The best instrumented title of the nine. It has a live output display, a strength value in newton metres and an auto routine that proposes a figure from a recorded lap. Use the auto routine as a starting point rather than as an answer, because it optimises against one lap in one car and a heavier car will clip at the same setting. Linear mode is the setting that most often ruins a good rig: with it off, the top of the force range is compressed, so the loudest cue the tyre model produces arrives softened.
Assetto Corsa
An older engine with an unusually pure force signal and no meter to set it by. The per car multiplier is the intended per car control, so leave the global gain where the clipping test put it and adjust cars individually. The three effect sliders, road, kerb and slip, are the biggest trap in the title: they were tuned for gear driven wheels of the era, and on a MOZA R9 V3 at $299.99 they produce texture that masks the tyre model rather than adding to it.
Assetto Corsa Competizione
Heavier and slower feeling than Assetto Corsa by design, because GT3 cars are. The single most common defect is Minimum Force left nonzero on a direct drive base: it was designed to overcome stiction in a gear train that a direct drive motor does not have, so it becomes a permanent floor under the whole signal. Dynamic Damping is the one in-game damper on this page that is genuinely worth using, because it scales with speed instead of applying a fixed drag.
Automobilista 2
The flavour choice changes the character of the wheel more than any slider does, so pick one and stop switching, otherwise you are tuning a moving target. Low Force Boost is the setting people misuse: it lifts small forces relative to large ones, which helps a weak gear driven wheel and actively harms a base with headroom, because it compresses the difference between the light signals you are trying to read.
rFactor 2 and Le Mans Ultimate
These share a lineage, so the architecture is the same: a global strength and a car specific multiplier saved per car. The correct habit is to fix the global figure with the clipping test on one reference car and then never touch it again, adjusting the per car multiplier for everything else. People who do the opposite end up with a garage where every car is wrong except the last one they drove. Le Mans Ultimate deserves its own pass rather than a copied file, because the prototypes and the GTs load the front axle very differently.
F1 titles and EA Sports WRC
Both are Codemasters lineage and both expose several named channels rather than one master. In the F1 titles, Understeer Enhance is the setting to switch off first: it deliberately lightens the wheel as the front axle gives up, which is a helpful cue on a controller and the removal of the exact cue a wheel exists to deliver. In EA Sports WRC, Self Aligning Torque is the working channel and the rest are additive, so raising three of them at once clips far sooner than any one of them does alone. Gravel and snow stages produce far less self aligning torque than tarmac, so a setting tuned on tarmac will feel dead on loose surfaces and is worth a second profile.
Gran Turismo 7
The hardest title to set well, because there is no meter and no driver layer to watch instead. Force feedback max torque runs on a 1 to 10 scale and the sensitivity setting changes how the scale maps rather than how strong it is. On a console, the only true ceiling is whatever the wheel itself is set to, either in its on-rim tuning menu or by its own firmware, which is why the Fanatec CSL DD and GT DD Pro family and the Logitech G PRO Racing Wheel at 11 Nm at $999.99 are the sensible high torque options here. Which wheels authenticate on which console is charted in the wheel compatibility chart.
Why does the in-game number change per car and the driver number never does?
Because they are limits on different things. The driver torque ceiling is a statement about your wrists and your frame: it caps what the motor can do in a wall impact, and it should be set to something you are comfortable being hit with, then left alone for years. The in-game gain is a statement about one car's force output relative to that ceiling, and cars differ enormously. A heavy GT3 car with high downforce generates far more self aligning torque than a light open wheel car, so the same gain that fits one clips the other.
The practical routine is short. Set the driver ceiling once. Run the clipping test in each title on the car you drive most, and write the figure down. When you move to a much heavier or much lighter class, retest gain only, and use the per car multiplier where the title has one. Nothing in that routine ever involves changing the hardware ceiling, and if you find yourself reaching for it, the thing you actually want is more gain in one title, not more torque in every one.
What if the wheel still feels wrong after all this?
Three faults masquerade as settings problems and none of them are fixed in a menu. The first is a frame that moves. If the base is not bolted through every mounting point, or the wheel deck flexes under load, the force reaching your hands is the force minus whatever the frame absorbed, and that subtraction changes with every input. Filtering hides the symptom and costs kerb definition permanently. A Next Level Racing GTRacer 2.0 at $499.99 or heavier is the honest fix above roughly 8 Nm.
The second is the rim. A larger diameter rim changes the leverage at your hands, so the same torque feels lighter and the same gain reads differently. If you swap between a formula rim and a round rim you have effectively changed the gain, which is a good reason to keep one rim while you are learning a base.
The third is the brake, and it is the most common of the three. A large share of the inconsistency people try to tune out of the wheel is actually brake modulation. A potentiometer measures travel, a load cell measures pressure, and pressure is how a real brake works. A set like the Logitech G RS pedals with a 75 kg load cell at $159.99 changes lap times more than another evening of slider adjustment, which is why the first rig buying order puts pedals ahead of the base.
Do not raise the driver torque ceiling to fix a title that feels light. If the F1 titles feel weaker than iRacing at the same driver setting, that is the title's output scale, and the fix is the in-game multiplier. Raising the hardware ceiling changes every sim you own, including the ones that were already correct, and it changes what happens to your wrists in a wall impact.
These figures are not for a desk clamp above 5 Nm. A Logitech G920 at $199.99 is fine clamped to a desk. A 9 Nm base at 50 percent gain is not, whatever the settings say, because the clamp creeps and then releases. Match the base to the mount with the wheelbase torque calculator before you spend an evening on gain.
Safety. Set the software torque limit before the first power up and set it low while you learn a new base. A direct drive base at 9 Nm or more can sprain a wrist or a thumb in a sudden impact, faster than you can let go. Never put your thumbs through the rim spokes, and stand clear during a calibration sweep when the wheel drives itself to full lock.
Once gain is settled in every title you race, the remaining gains are hardware. The direct drive versus belt versus gear guide explains what the drive type changes before any setting matters, and the direct drive wheelbase roundup splits the category into tiers you can actually mount. If the wheel is right and the car still feels inconsistent, the answer is almost always the brake or the frame, in that order.
Frequently asked questions
What force feedback settings should I use for iRacing?
Set the base driver to full scale with filters off, turn linear mode on, then set strength so the meter stops saturating in your heaviest corner and take another 5 percent off. On a 9 Nm base that usually lands near 50 percent, which iRacing shows as about 5 Nm. On a 21 Nm base the same working force is closer to 35 percent. The number changes per car, the driver ceiling does not.
Do force feedback settings transfer between sims?
The driver level settings do, the in-game numbers do not. Each title scales force differently, so 50 percent in one sim can be far heavier than 50 percent in another. Set the base driver once at full scale with damping, inertia, friction and interpolation off, then set the in-game gain per title using the clipping test. Write the final figure down, because driver updates occasionally reset them.
Which sims have a built in force feedback clipping meter?
Of the nine titles charted here, iRacing, Assetto Corsa Competizione and Automobilista 2 give a usable in-game indication of output level. Assetto Corsa, rFactor 2, Le Mans Ultimate, the F1 titles and EA Sports WRC do not, so the base driver telemetry window is the tool. Gran Turismo 7 has neither, because a console has no driver layer to open alongside the game.
Should I use the game damping or the wheel driver damping?
Use one, never both. Assetto Corsa Competizione and Automobilista 2 both have speed aware damping that is part of how the title is voiced, so use theirs and leave the driver at zero. For iRacing, Assetto Corsa, rFactor 2 and Le Mans Ultimate the driver damper is the cleaner control. Stacking the two produces a wheel that feels heavy and slow to respond without feeling detailed.
What is the torque limit and how is it different from gain?
The torque limit lives in the base driver and caps what the motor can ever produce, in newton metres. Set it once, before the first power up, and leave it. Gain is the in-game multiplier applied to the forces the sim generates, and it is what changes per title and per car. Confusing the two is why people raise a driver ceiling to fix a car that felt light in one sim.
Are these numbers final settings?
No, they are starting points for the clipping test, which is the only thing that produces a correct answer for your base, your rim and the cars you drive. A heavy GT3 car and a light formula car generate very different forces in the same title. Start at the figure in the table, run three laps, drop gain in steps of 5 percent until the output stops saturating, then take another 5 percent.
How we choose: we compare published manufacturer specifications, documented torque and travel figures, and verified owner reviews. We do not test gear in person. Everything here is researched guidance, not professional installation advice. Direct drive wheelbases produce enough torque to injure a wrist, and rig hardware carries real weight, so follow the manufacturer's mounting and torque limits.