This is the profile that lived on my Simucube 2 Ultimate for the entire time I raced the Porsche 911 GT3 Cup in iRacing — exported straight out of the Tuner software before the base was retired in favour of the Simucube 3 Ultimate. Nothing exotic in here: 900° of range, a wheelbase that is allowed to give everything it has, and just enough mechanical feel to keep the rim honest instead of hyperactive.
Simucube 2 Ultimate
iRacing · Porsche 911 GT3 Cup
Basic Settings
The working envelope of the base.
Ceiling for the torque the base is allowed to produce from simulator FFB. At 100 % nothing is held back in hardware — the actual level is dialled in on the iRacing side. Bumpstop forces ignore this ceiling.
Total rotation from lock to lock. 900° keeps the wheel 1:1 with the Cup car’s steering in iRacing.
Mechanical Feel
Always active, with or without an FFB signal from the sim. This is the baseline weight of the wheel — and the main tool against oscillation.
Resistance that scales with rotation speed: the faster you turn, the harder it pushes back. Gives the rim body without smothering detail.
Constant drag, independent of how fast the wheel moves. A small dose settles nervous behaviour around the straight-ahead.
Simulated flywheel mass. It resists changes in rotation speed, so the wheel needs a moment to get going and keeps running a fraction longer once released.
A plain spring back to centre that works regardless of what the sim sends. Off here — in iRacing the tyre model does the centering.
Smoothness
Signal conditioning between simulator and motor. Sim FFB arrives at a limited update rate and can be harsh; these three decide how much of that reaches your hands.
Smooths and extrapolates the low-rate torque stream from the sim up to the internal rate of the base. Low values follow the original signal closely, especially through fast direction changes; high values buy smoothness at the cost of accuracy. With iRacing’s 360 Hz FFB mode active there is far less low-rate signal left to reconstruct, which is exactly why a value of 1 works here — the kerbs stay sharp.
Caps how fast torque is allowed to change. The Ultimate is rated at 9.5 Nm/ms; holding it at roughly half rounds off the most violent spikes while keeping the wheel reactive rather than rubbery.
Low-pass filter on the FFB signal. The lower the frequency limit, the more high-frequency detail disappears — handy against excessive kerb buzz, but a heavily limited bandwidth also introduces audible lag.
Advanced
Compensation filters for specific cases rather than everyday tuning.
Bleeds off constant torque after about a second, so long corners stop turning into an arm-wrestle while sharp road detail still comes through. Disabled here — in the Cup car that steady loading is information, not noise.
Internal signal-processing adjustment that shortens the torque control loop between PC and motor. It does not change the character of the FFB, but a tighter loop lets you run less damping, friction and inertia before the wheel starts to oscillate.
Gamma curve for very small torque signals. Raises or lowers how much the wheel reacts when there is almost no force on it — the usual fix for a lifeless feeling on straights. Extremely sensitive; 1.00 is linear.
How hard the software-generated end stops hit when the wheel runs out of range.
Moves the device bumpstops inside or outside the axis limits — useful when a sim generates its own harsh end stops. At 0 they sit exactly at the ends of the 900° range.
Merges the two digital inputs on the D15 accessory port with the paddle buttons of a Simucube Wireless Wheel. Off keeps them as separate buttons.
Notch Filter
A band-stop filter that removes one specific frequency — the surgical tool against a resonance or one over-enthusiastic vibration. Parked at the bottom of its range in this profile: a 1 Hz centre frequency sits far below anything the FFB signal actually contains, so the filter is effectively inactive.
The frequency being cut out of the signal.
How deep the cut goes at that frequency.
How narrow the cut is. A high Q takes out almost only the target frequency, a low Q removes the neighbourhood with it.
Ultimate Filters
Exclusive to the Simucube 2 Ultimate. Untouched here, since their parent filters are switched off.
Controls how quickly static force is bled away once Static Force Reduction is in use.
Additional damping that fades in as the wheel approaches the straight-ahead position.
How wide the window around centre is in which that extra damping applies.
Game Effects
Gains for DirectInput effects — extras a title can request on top of physics-based force feedback. Modern sims barely touch them; iRacing does not use them at all. Simucube recommends leaving the lot at 100 %, and that is exactly where they stay.
Speed-dependent resistance, requested by the game rather than the base.
Constant resistance, requested by the game rather than the base.
Centering spring effect. Via DirectInput a title can place its centre point somewhere other than wheel centre.
Sine-shaped vibration added to the torque signal.
Square-shaped vibration signal.
Sawtooth vibration signal — the slider covers both the up and the down variant.
Triangle-shaped vibration signal.
iRacing — Sim Side
Half of a force feedback setup never appears in Tuner. These are the matching values on the simulator side, taken from my app.ini and the in-game force feedback options.
Tells iRacing what the hardware can actually deliver. Set it to the peak torque of your base — 32 Nm for the SC2 Ultimate — and iRacing rescales the strength slider around it so you cannot accidentally oversaturate the wheel.
The slider everything else follows from. iRacing prints the resulting Max Force right underneath it — at 8 that works out to 45.1 Nm for this car. Set by feel across a lot of laps, not by a formula.
The steering-column torque at which the base runs out of headroom. Counter-intuitively, a higher number means a lighter wheel — you are telling iRacing not to max out the hardware until the car generates that much. This is a property of the car, not of your wheelbase, so 45.1 Nm carries straight over to a 16 Nm base; only Wheel Force changes.
Adds headroom to the calculation behind the Auto button. Only does anything while Auto FFB is in charge; at 50 % it leaves a sensible margin instead of calibrating right up to the ceiling.
Raw 1:1 telemetry with no low-end boost. The right call on any direct drive base — that boost exists to rescue wheels that cannot produce force in the first place.
Zero. Smoothing tidies up high-frequency noise but pays for it in latency, and the reconstruction filter inside the base does the same job at a far higher rate.
Zero on the sim side. Whatever damping the wheel needs is handled by the base, again at a much higher rate.
Zero. Min force exists to shove weak wheels past their own deadzone. A direct drive base has no deadzone to escape from.
On, at the stock 33 %. Keeps the wheel from sitting in the pit box oscillating against itself.
Off. A static centering force has nothing to do with what the front axle is doing and only masks it.
loadSimucubeAPI=1 in app.ini. Note that enableFFB360HzInterpolated stays at 0 — that is the generic interpolated route for other brands, not the Simucube one.
forceResetBeforeStartup=0, the default. Worth checking, because the 360 Hz handshake between iRacing and the base will not come up if this has been switched on.
How to read these numbers
The panel follows the grouping and naming of the current Simucube Tuner software, but the values are read out of the exported .tdp file. That distinction matters for three of them: Torque Bandwidth, Ultra Low Latency and Bumpstop Feel are stored as internal scale positions rather than the labelled value the UI puts next to the slider. Everything else — percentages, degrees, Nm/ms, dB — is exactly what you would see on screen.
The other thing worth saying out loud: force feedback settings are not transferable currency. Rig stiffness, wheel diameter and weight, in-sim FFB level and plain personal taste all move the result. Take the structure, then tune.
What this profile actually does
- Max Strength 100 %, iRacing does the limiting. The base never becomes the bottleneck. Clipping is managed in the sim’s FFB slider, where you can actually see it happening.
- Damping 17 / Friction 7 / Inertia 10. Low enough to keep the front axle talkative, high enough that the wheel doesn’t go hunting for a resonance every time you lift.
- Reconstruction Filter at 1. This only works because iRacing runs in 360 Hz FFB mode here — there is barely any low-rate signal left to reconstruct. Kerbs stay sharp and the rear stepping out still arrives on time.
- Slew Rate at 4.99 Nm/ms. Roughly half of the Ultimate’s 9.5 Nm/ms ceiling. The most violent torque spikes get their edges taken off; everything else passes through untouched.
- Ultra Low Latency high, mechanical filters low. A tighter control loop buys stability that would otherwise have to be paid for in damping — and damping is paid for in feel.
- Static Force Reduction off. In a car without power steering, sustained load through a long corner is not fatigue to be filtered away. It is the message.
The iRacing side: 360 Hz FFB
Every value in the panel above was dialled in with iRacing’s 360 Hz force feedback mode enabled, and that changes how the profile should be read. Normally iRacing hands the wheelbase a fresh torque value 60 times per second. In 360 Hz mode it packs six samples into each of those frames and the Simucube firmware plays them back in sequence — an effective 360 Hz update rate, through an API that Simucube and iRacing built together. It shipped with the 2024 Season 2 Patch 1 build and needs current Simucube firmware; the switch itself lives in app.ini under [Force Feedback] as loadSimucubeAPI.
The practical consequence sits in the Smoothness card: with six times the resolution arriving from the sim, the reconstruction filter has almost nothing left to invent, which is what makes a value of 0 or 1 usable instead of gritty. The trade-off is that sequential playback costs a little latency — up to one 60 Hz frame — and a handful of drivers genuinely prefer the rawer, more connected feel of the old 60 Hz path. If you run 60 Hz, expect to need more reconstruction than the 1 you see here.
The Simucube 3 Ultimate is a different animal and will get its own page once it has enough laps on it. Until then, this is what the Cup car felt like on the 2 Ultimate — and the rest of the hardware around it is documented on the simulator page.