Start by turning the in-game gain down, not up. That is the whole verdict for most people who land here. Assetto Corsa EVO ships with a force feedback curve that clips hard on anything above about 8 Nm, and the instinct after a first laggy, muddy lap is to crank everything. Do the opposite. On a Moza R9 I run 100% in Pit House, 55% in-game gain, minimum force 0, road effects around 15, and every “enhancement” slider at zero. The car comes alive because the wheel is finally reporting what the front tyres are doing instead of shouting over them.

The second thing to know is that EVO’s FFB is a different animal from original AC and from Competizione. It is closer to ACC’s physics-driven signal (almost nothing is canned), but the early-access builds have been through at least four FFB revisions since January 2025, and settings copied from a six-month-old Reddit thread will feel wrong. What follows is where things sit now, on the build I have been driving nightly for a couple of months, across a Moza R9, a Fanatec CSL DD 8 Nm, a Logitech G923 and, for one miserable weekend, a borrowed G29.

What EVO’s FFB is actually doing

Kunos rebuilt the tyre model for EVO and the FFB is derived almost entirely from steering rack torque. There is no separate “tyre slip” effect being layered on top like in older sims. What you feel is the self-aligning torque of the front tyres changing as they load up, and the drop-off as they start to slide. That drop-off is the whole point. When the front pushes wide at Brands Hatch through Paddock Hill, the wheel goes light in a very specific way, a half-second before you see the understeer. If you cannot feel that going light, your settings are wrong. Usually it is because gain is so high the signal was already clipping before the tyre let go, so “heavy” and “sliding” feel identical.

The in-game menu has six sliders that matter and a handful that don’t. Gain is the master multiplier. Minimum force adds a floor to the signal so small forces still register on wheels with high internal friction (belt and gear drives). Road effects is the one canned layer, feeding kerbs and surface bumps into the wheel. Then there are “understeer enhancement”, “slip effect” and a damper slider. The damper is a static resistance, and a lot of people confuse it with weight. It is not weight. It is drag.

Two things are worth knowing that the menu does not tell you. EVO has a per-car FFB multiplier buried in the setup screen, and a few of the launch cars (the Alpine A110 and the Porsche 992 GT3 Cup in particular) ship with values that push a direct drive into clipping at the default 100 gain. And the FFB clipping meter in the HUD apps is accurate. Turn it on. A yellow bar that hits red under braking into Turn 1 at Imola means gain is too high, full stop, no matter how good it feels.

The settings, by wheel

I am going to give exact starting values because that is what you searched for, then explain the two or three you should actually adjust. Everything else, leave alone.

Moza (R5, R9, R12, R16)

Moza’s Pit House software is where the real work happens, and the sim-side settings are just a multiplier on top. Start in Pit House:

• Maximum steering wheel torque: 100%
• Rotation: 900° (set it in-game too; EVO reads the base’s angle and uses soft lock, so a mismatch gives you the classic dead zone at the end of rotation)
• FFB reverse: off
• Road sensitivity: 10 (this is Moza’s spring-and-damper interpolation, higher numbers feel like driving through syrup)
• Speed-dependent damping: 0
• Wheel spring strength: 0
• Wheel damper: 5 on the R9, 10 on an R16 (a direct drive with no damper at all oscillates on straights, and the R16 does it violently)
• Wheel inertia: 100
• Wheel friction: 0
• Natural inertia: 100
• FFB reverse polarity: off

In-game: gain 55 on an R9, 45 on an R12, 35 to 40 on an R16. Minimum force 0. Road effects 15 to 20. Damper 0. Understeer enhancement 0. Slip 0.

Yes, the R16 number looks low. A 16 Nm base at 100 gain in EVO’s default profile will throw your hands off the rim in a Cup car at Monza, and it is not more informative, it is just louder. I ran the R16 at 40 for a week and lap times went down, mostly because I stopped tensing my shoulders.

The Moza-specific quirk people argue about on the Discord is the “Road sensitivity” slider. Moza’s documentation is vague. In practice it is a low-pass filter, and below about 8 the FFB gets a sharp, grainy quality on the R5 that some people love and I find fatiguing. Try 8 and 12 back to back, in the same car, on the same lap. You will pick one within two laps.

Fanatec (CSL DD, GT DD Pro, DD1, DD2, ClubSport DD)

Fanatec’s tuning menu on the wheel itself is where most of the confusion lives, because the defaults on a CSL DD are designed to make every sim feel “safe” out of the box, which in EVO means dull. Set the wheel’s tuning menu to:

• SEN: Auto (EVO reports the correct rotation per car)
• FF: 100
• FFS: Peak (Linear is the intuitive choice and it is wrong here; Peak lets the base deliver its rated torque at 100% signal instead of holding some back)
• NDP: 15 on the CSL DD 8 Nm, 20 on a DD1, 25 on a DD2
• NFR: 0
• NIN: 0
• INT: 3
• FEI: 100
• FOR, SPR, DPR: 100, 100, 100 (EVO barely uses the spring and damper channels, so leaving them at 100 costs you nothing)

In-game: gain 60 on the CSL DD 5 Nm, 50 on the 8 Nm boost kit, 40 on a DD1, 35 on a DD2. Minimum force 0. Road effects 15. Damper 0.

The NDP number is the one that generates the arguments. Natural damper is friction added at the base, and Fanatec’s forums are split between “NDP 0, let the sim do it” and “NDP 30, it saves your wrists”. At zero, the CSL DD will oscillate at speed on the Nordschleife straights in EVO, wheel flicking left-right on its own with hands off. At 15 it stops. Above 25 you start losing the tyre drop-off feel that makes EVO worth playing. Put it at 15, drive one lap, and if it still hunts, go to 20.

Also, INT (interpolation) at 3 is deliberate. Fanatec’s default is higher and it smooths out kerb detail that EVO renders well. INT 1 is too raw on a CSL DD and produces a buzz on smooth tarmac, which is a firmware artefact not a physics one.

Logitech (G29, G920, G923, G Pro)

Here is where I talk someone out of something. The G29 and G920 are gear-driven wheels with roughly 2.2 Nm of peak force, a lot of internal friction, and a notchy feel that no in-game setting fixes. EVO’s tyre drop-off is subtle, and a gear-drive wheel has enough cogging that the “going light” signal disappears into the mechanism. You will drive by sight, not by feel, and that is fine for learning the tracks. Do not spend hours chasing settings on a G29 hoping to make it feel like a DD. It won’t.

Still, these are the settings that get the most out of it:

G HUB: sensitivity 50, operating range 900°, centering spring off, spring 0%, damper 0%, force 100%.

In-game: gain 100 (you need every bit of the 2.2 Nm), minimum force 8 to 12 (this is the one that matters most on a Logitech; it lifts small forces above the gearbox friction so you get something during light steering), road effects 25, damper 0, understeer enhancement 20.

Yes, I said leave the enhancements at zero on direct drive. On a gear-drive, the understeer enhancement slider is the only way you will feel the front go, and 20 is enough to notice without turning it into a gimmick.

The G923 is the same story, slightly smoother, and TrueForce does nothing in EVO at the time of writing. Ignore the marketing.

The Logitech G Pro is different. It is an 11 Nm direct drive and belongs with the Moza and Fanatec bases. Set the G HUB force to 100, damper 0, and run in-game gain at 45, minimum force 0, road effects 15. It is a genuinely good base and, at around $1,000 for the wheel alone, an expensive one. If you are buying a Logitech DD specifically for EVO, a Moza R9 at half the price does the same job with better software.

The desk question nobody wants to hear

A desk clamp works for a G29. It does not work for a direct drive in EVO. This sim has a lot of high-frequency content in the signal (kerbs, tyre loading changes), and an 8 Nm base on a flexing desk turns that into a wobble you feel through your forearms rather than your fingers. People post “EVO FFB feels vague” threads and half of them, when pressed, are clamped to an IKEA table. A Wheel Stand Pro or a cheap Next Level Racing GT Lite fixes more FFB complaints than any slider. I spent a month convinced the R9 had a fault before I bolted it to an aluminium profile rig. It did not.

Things that break, rattle or annoy

The car-specific multipliers I mentioned are the big one. If a car feels like it is clipping when others are fine, open the setup, find the FFB multiplier, and drop it to 80 or 90. The Cup cars and the older GT3 Ferrari want this. The road cars mostly do not.

EVO also resets your FFB gain to 100 after some patches. It happened twice in the last four updates. Check the slider after every update before you decide the devs “broke the FFB”. Usually they didn’t.

Wheel rim weight matters more than in ACC. A heavy formula rim on an R9 dulls the fast tyre transitions. The lighter GS rim feels crisper in EVO than the bigger ES on the same base, which was not true in original AC. If you are shopping, this is a reasonable reason to skip the big-money rim.

And the damper slider in-game. I keep saying zero. Some people run it at 10 on a Moza to kill oscillation rather than using Pit House’s wheel damper. Either works, but do it in one place, not both, or you will end up wondering why the wheel feels like it has a wet sock on it.

Quick comparison

How it should feel when it is right

Take the Alpine A110 to Brands Hatch Indy. Through Paddock Hill Bend the wheel should load up as you turn in, then lighten slightly at the apex as the car rotates. Over the exit kerb at Druids it should give you a couple of distinct taps, not a blur. On the pit straight the wheel should sit still with your hands off. If it hunts, add damper. If the kerbs are a blur, lower INT (Fanatec) or road sensitivity (Moza). If the wheel goes heavy and stays heavy through the whole corner with no texture, gain is too high.

That pattern (loads, then lightens) is the entire skill of driving EVO on FFB. Everything else is just making sure you can feel it.

FAQ

Why does Assetto Corsa EVO FFB feel weaker than ACC? Because EVO’s signal has less canned content and a lower average torque, with more of the headroom reserved for the peaks. If you set gain by “does it feel heavy in a straight line” you will clip every corner. Set it by the clipping meter instead, then live with the lighter straight-line feel. After a week it stops feeling weak and starts feeling clear.

Should I use the FFB enhancement sliders? On a direct drive, no. They are artificial layers added on top of a signal that already contains the information. On a gear-driven Logitech or a Thrustmaster T300, understeer enhancement at around 20 is worth it because the mechanism hides the real cue.

What is the best rotation for Assetto Corsa EVO? Set your base to 900° or higher and let the sim’s soft lock handle each car. EVO reads the base rotation on most modern wheels and matches it to the car’s real steering ratio. Forcing 540° in the driver software for a GT3 car is a holdover from older sims and gives you a mismatch between hands and on-screen wheel.

Before you order anything

If you are on a G29 and EVO is your main sim, the upgrade that changes the experience is any 5 to 9 Nm direct drive, not a better rim. The Moza R5 bundle at around $400 and the Fanatec CSL DD at around $350 base-only are the two I would point at, and I would pick the Moza for the software alone. Fanatec’s support and shipping reputation has improved since the Corsair takeover but the forum threads from 2023 have not aged well. Whichever you pick, budget for something to bolt it to. A direct drive on a desk clamp is money spent to feel a wobble.

If you already own a DD, do not buy anything. Open the HUD clipping meter, set gain to 50, zero every enhancement, drive ten laps of Brands Indy in the Alpine, and adjust one slider at a time from there. That costs nothing and fixes most of it.