I have solved this problem on my Simucube 2 Pro in 100% All is perfect now. How to solve your wheel, i dont know. Try set to minimum your Wheel bandwidth as low as you can. Slow down it.
Torque (and subsequent FFB output %, scaled) is within the game and doesn't depend on the wheel being used. I agree the kerb forces look extreme here, I'm fairly sure the nominal max torque in the DW12 is around 25Nm so I'm not seeing a lot in the way of cornering forces (I'd have to test the car, is caster adjustable? Raising it will generally boost cornering forces). But also Portland doesn't have a whole heap of high speed corners, and obviously the DW12 is made to run on some very high speed tracks. The normal way to remedy this somewhat is to do the opposite of what you've done - raise the vehicle specific to boost all normal forces, while the extreme kerb/contact forces will be clipped to 100% anyway. Lowering the Mult only highlights the difference between the two. I'll go have a quick run and see how it looks.
I've tested with DW12 with values from 10% - 100%. And steering forces are not clipping even with 100% until kerbs are hit (which is fine and expected). But the thing is, no matter what I use, steering shaft tourque is always at least over 30Nm (usually over 50Nm) on some kerbs which is destroying my wheel and rattles like crazy. Only way I can make rattling tolerable is to raise smoothing. But still, I think is very strange that forces are that strong. And it's not just this track/car combo. It's with every car and every track. I can create more telemetry if needed.
Steering shaft torque will always be what it is. You basically can't change that (setup may have some effect, but you can't really change the geometry enough to make a big difference) and it's not dependent on your wheel. It's calculated purely from the physics, so if it's giving 45-50 Nm on some kerbs that's simply what it's going to do. The thing you have some (more) influence over is the FFB Output. Let's say the vehicle has a 'max steering torque' of 25Nm. If you have a Mult of 100%, you're saying: give me full force at 25Nm. Then a 12Nm cornering force will give you about 50%, a kerb strike at 25Nm will give you 100%, and a kerb strike at 50Nm will give you 100%. If you have Mult of 50%, you're saying: give me full force at 50Nm. Then the 12Nm cornering force is 25%, kerb strike at 25Nm gives you 50%, and a kerb strike at 50Nm will give you 100% (see how the strong kerb strike is now 4x stronger than the cornering force, instead of 2x ?) You may actually get better results here by increasing the FFB Mult above 100%, so that the cornering forces use more of your wheel's range and the kerb strikes are clipped not very much higher. The best level to use will be track and setup (see below) dependent. Apart from the FFB Mult, you can also adjust caster in the vehicle setup. Going from 6° caster to 9.9° caster in the setup raised peak force through T1 (no-chicane layout, I'm lazy) from 14Nm to 22Nm. And actually it showed that clipping started around 16Nm, so my 25Nm guess looks to be a long way off reality (I'm also obviously on the road course configuration, ovals may be different). The black traces here are the 9.9° caster. Finally, the nature of the FFB over kerbs probably means you do need to rely on some FFB smoothing to tone it down a bit, because even with the kerbs clipping you'll get a lot of nasty oscillation (and perhaps worse because of the clipping, ironically): (FFB at 400Hz, but I have smoothing on 4 which is reflected in the logging) I believe your wheel does 8Nm, so you're well below the actual forces for this car even if you tailor the level using Mult at each particular track to reach nearly 100% at peak pure cornering force. If the kerb vibrations are an issue you'll need to tune the smoothing and/or reduce your wheel FFB strength (not the in-game Mult, but your actual wheel output).
Tnx for explanation. I am aware and know how to "fix" this rattling, but to extent of losing some info by filtering (I can and do reduce FEI) but lots of data is filtered this way. I was and still am wondering if this forces from kerbs are correct or not. They do seem to be high.
Upon further analysis, the spikes in Steering shaft torque (those v. high values) are from transition from road to grass. Well, most of them are at least. I've tested this on a couple of tracks (both official and 3rd party), new once as well as the old once and with a couple of different vehicles. As soon as the wheel touches the grass (leaves the track) this harsh spikes are registered and the wheelbase wants to kill it self. And the thing is, it doesn't matter at what speed you do this. I was driving without throttle input, around 40 km/h and spikes were still there. Also, these spikes are not registered on the whole track, there are areas where no sudden spikes appear when grass is touched with the wheel. So, my thinking is that this is exaggerated and that this touching of the grass should not produce those forces.
Them rumble strips forces being authentic a good thing. Good thing to have that is. But to not be able to tone them down is not. This practically makes racing at night impossible for me because of the wheel noise. I don't mind it, but my downstairs neighbors do. So simracing is strictly a day sport for me. Living in Finland, the timezone makes this even bigger problem as most action online is very late for me.
Way back... way way wayyyyyy back... i think I wishlisted reinstating an rF1 setting that limited the per-frame (/update) FFB delta, as at the time I thought it would be a better option than smoothing for reducing noise in Logitech wheels without losing as much detail. I still think this would be worthwhile. The problem with situation specific options (like reducing FFB or increasing smoothing when a tyre is on a rumble surface) is both that the transition itself could cause a jolt, and there could still be on-road jolts you'd rather not have. An option that helped globally, with less detriment than high smoothing, would be ideal.
