Most people who search “iracing settings” are trying to fix a car that feels wrong, and in nine cases out of ten the problem is one of three things: the wheel wasn’t calibrated to the base’s actual rotation, the FFB max force number is set backwards from what they think it does, or a pedal isn’t reaching 100% because they skimmed the “map range” step. None of this requires new hardware. I say that as someone who has bought a depressing amount of hardware.
So here’s the actual answer up front. Calibrate in iRacing itself, not just in your wheel software. Leave every deadzone at zero unless you have a measurable reason not to. Set your FFB in Nm mode with wheel force matching your base’s real output, then use the auto button after two laps instead of guessing. And back up your controls.cfg, because iRacing will eventually forget your bindings the day before a race and you’ll want the fix to take thirty seconds.
Calibration: do it in the sim, and do it right
iRacing’s calibration wizard lives under Options > Drive (the wheel icon). It asks you to center the wheel, then turn 90 degrees to the left, then click. That 90-degree step is where people ruin everything. The sim uses it to calculate your total rotation, so if you eyeball 80 degrees or your wheel software is running a 540-degree soft lock while the base is physically capable of 1080, the math comes out wrong and every car feels either darty or like steering through syrup.
Two rules. First, set your base to its maximum rotation in the manufacturer software (900 or higher on a Logitech G923 or Thrustmaster T300RS, 1080 or more on a Moza R5 or Fanatec CSL DD) and let iRacing manage per-car rotation itself. The sim knows a Skip Barber wants different lock than a Super Formula car, and it will soft-stop the wheel correctly per car as long as you didn’t lie to it during calibration. Second, when the wizard asks for 90 degrees, actually measure it. Put a bit of tape at the top of the rim if you have to. Vertical spoke to horizontal spoke on most round wheels is your 90.
Direct drive owners get a shortcut: most modern bases (Fanatec, Moza, Simagic, Asetek) report their rotation over USB, and iRacing reads it. You still run the wizard once. If the sim then shows the correct degrees in the calibration summary, you’re done and you should never touch it again unless you change firmware. Firmware updates resetting rotation to some default is a recurring Moza Discord complaint, so if the car suddenly feels wrong after an update, recalibrate before you touch anything else.
Belt and gear wheel owners, one extra note. If your wheel has any slack at center (the G29’s gear lash is famous), resist the urge to “fix” it in calibration by off-centering. Center it where the wheel naturally rests. The lash is mechanical and no setting removes it.
Pedals: map range is not optional
The wizard’s pedal step says press each pedal fully. People tap them. Then they spend a month wondering why they’re slow into Turn 1, and the answer is their brake input peaks at 94%.
Press each pedal as hard as you would in an actual panic stop. If you’re on a load cell (the Fanatec CSL Pedals Load Cell Kit, Heusinkveld Sprint, Moza’s CRP2), set the load cell’s max force in the pedal software first, to a number you can genuinely hit repeatedly, then calibrate in iRacing against that. Calibrating against a heroic one-time 100kg stomp means your realistic braking lives in the bottom 70% of the range and you lose resolution where it matters.
Then there’s brake force factor, the setting nobody explains. It’s on the same Options page, and it applies a curve to your brake input. On a load cell, set it to 0. Zero. Linear. The load cell already behaves like a real pedal, where pressure equals braking, and stacking a software curve on top makes trail braking guesswork. On a potentiometer pedal (stock G923, T300 pedals), a value around 1.8 is the common starting point because those pedals measure travel rather than pressure, and the curve fakes some of the progressive feel. Tune it in whole tenths and stop when you can hold a consistent 60% brake through the downhill into the West Bend at Lime Rock without the bar jumping around.
Clutch: if you drive anything with a manual, calibrate it, then find the bite point in the garage screen by watching the input bar. The MX-5 launch off the line cares. Most people can ignore this and map an auto-clutch instead, which iRacing handles fine for the majority of the road racing content.
Deadzones, and why yours should be zero
The deadzone slider exists for broken or noisy hardware. That’s it. A deadzone on your steering means the first degree or two of input does nothing, which you will feel as a car that’s lazy on turn-in and then suddenly isn’t. On a modern base, even a cheap one, the sensor is clean enough that zero is correct.
The exceptions are specific. An aging G29 with a scratchy potentiometer might show input jitter at rest; you can see this in the Windows game controller panel (run joy.cpl) as a twitching axis. If the crosshair dances, add the smallest deadzone that stops it, usually 1-2%, and start budgeting for new pedals because it gets worse. The other legitimate case is a throttle that never quite reads zero when your foot is off it, which will fight you under braking and can trip the pit lane speed check. Small deadzone at the bottom of that axis, problem solved.
What you should not do is copy deadzone values from a YouTube settings video made by someone with different hardware. This is the recycled-listicle disease. Their 5% throttle deadzone is masking their worn spring. Yours is just deleting the top of your throttle modulation.
Force feedback: the max force number works backwards
This is the single most misunderstood setting in the sim. iRacing’s FFB strength, when set to Nm mode (enable “use custom specific to this setup” thinking aside, the toggle you want is in app.ini or simply the wheel force field on newer builds), maps a certain amount of steering column torque to 100% of your base’s output. A lower number means stronger feedback, because you’re compressing less range into your motor. People slide it down expecting less force and get their wrists rattled.
Do this instead. Set “wheel force” to your base’s honest rating: 8 for a CSL DD with the boost kit, 5.5 without, 5 for a Moza R5, about 2.2 for a G923. This tells iRacing what your hardware can do, and it makes the little F meter on screen meaningful. Then drive two or three clean laps in your actual car, somewhere with real curbs and load, Spa in a GT3 works, and click the “auto” button next to the FFB strength. iRacing looks at the peak forces you generated and sets the max force so you get detail without clipping. Clipping, if you’re new to the term, is when the requested force exceeds what your motor can produce and everything above that ceiling flattens into the same maximum shove. A clipping wheel feels strong and dead at the same time. The F meter turning red is your tell.
From the auto-set value, adjust to taste in small steps. Lower the number a touch if the wheel feels vague in fast corners, raise it if curbs are violently loud but mid-corner is fine. On lower-torque gear like the G923 you’ll live with some clipping in high-download cars because 2.2 Nm can’t represent a Dallara’s steering forces; that’s a hardware ceiling, not a settings failure, and no app.ini incantation fixes it.
Damping: 0% on direct drive, and use your base’s own damper/friction settings sparingly instead (a few percent of damping in Moza Pit House or Fanatec’s tuning menu can calm oscillation on lighter rims). On gear-driven wheels a bit of iRacing damping, maybe 5-15%, masks the gear grind. Leave “reduce force when parked” on unless you enjoy your base trying to break your thumbs in the pit box.
One opinion I’ll defend: skip irFFB until you’ve spent a month on stock FFB. It’s a genuinely clever tool, and it’s also the first thing forum posters prescribe for problems that are actually a wrong max force value. iRacing’s native 360Hz FFB is good now. It wasn’t always, which is why the folk wisdom persists.
Button mapping: plan for the unplug
iRacing stores everything in Documents/iRacing/controls.cfg, and its bindings are tied to device order. Unplug your wheel, plug it into a different USB port, and there’s a decent chance the sim sees a “new” device and your fifteen carefully mapped buttons are gone. Everyone learns this the hard way once. Copy controls.cfg somewhere safe after you finish mapping, and plug your gear into the same physical ports forever. A powered hub helps here, both for port consistency and because direct drive bases and unpowered hubs are a flaky combination.
The mappings worth doing before your first race, in rough priority: pit limiter, the relative black box (default is F3, but you want it on the wheel because you’ll look at it constantly), brake bias up and down, and a black box scroll pair for navigating the pit stop screen. If your car has in-car adjustments (TC, ABS, anti-roll bars in some GT cars), map them per car. iRacing has a “use custom controls for this car” checkbox in Options for exactly this, so your GT3 TC buttons don’t squat on bindings your Formula Vee doesn’t need. Voice chat push-to-talk goes on the wheel too if you race in leagues; reaching for a keyboard at 240 km/h on the Kemmel straight is how incidents happen.
Shifters and handbrakes are separate USB devices and calibrate under the same wizard. An H-pattern needs no deadzone thought, but do check that seventh position if your shifter has one, because a mis-mapped reverse in the pits is a classic.
FAQ
Why does my wheel feel different in every car? Mostly because it should. iRacing models each car’s actual steering geometry and per-car rotation, so a Skip Barber and a Porsche Cup car are meant to feel nothing alike. If the difference feels wrong rather than different, re-run calibration and re-do the auto FFB per car; the max force that suits an MX-5 clips constantly in a Super Formula.
Should FFB strength match some number from Reddit? No. The right max force depends on your base’s torque, the car, and even the track. Use the auto button as your baseline every time you switch cars. Copied values from someone on a 15 Nm Simagic are meaningless on your 5 Nm base.
Do I need to calibrate in both Windows and iRacing? Calibrate in your wheel’s own software first (rotation, pedal load cell max), then in iRacing. The Windows joy.cpl panel is a diagnostic tool, useful for spotting noisy axes, not a place to calibrate for the sim. iRacing reads raw input and applies its own calibration on top.
Before you blame the hardware
Run through this list the next time the car feels off: calibration degrees match your base’s real rotation, brake hits 100% in the garage input bar, brake force factor is 0 on a load cell, deadzones are zero, wheel force matches your base’s Nm, FFB was auto-set in the car you’re actually driving, and the F meter isn’t living in the red. That sequence fixes most “iRacing feels bad” complaints for free.
If you get through all of it and the limiting factor is genuinely the gear, the upgrade that changes your lap times is pedals, not the wheel. A load cell brake (the CSL load cell kit at around $60 as an add-on, or Heusinkveld Sprints if the budget is real) improves braking consistency more than any base swap, because braking is the skill iRacing’s safety rating punishes you for lacking. And before any direct drive order, push hard on your desk. If it flexes, a 5 Nm base will walk it across the room, and the honest first purchase is a wheel stand or rig, not more torque.