FFB re-centre issue after hard hits in rF2? Pull to one side

I’ve posted a similar note in the rf2 thread - but as that’s mainly a ‘tuning’ thread, I thought I better raise my head here just in case it’s a wider issue…

I have a weeks-old Pro and updated the TD and firmware last night before c. 90 minutes of racing in rf2.

The track/car combination I was using needs a lot of use of the kerbs - which obviously give quite heavy feedback.

On 5 or 6 occasions, after a heavy kerb hit, the steering exhibited a noticeable pull to one side when trying to subsequently drive straight up the track. No car damage had occurred.

The in-game FFB reset mappable control resolved the issue each time.

I’ve had this issue literally once or twice with the Pro since getting it - but nothing like as frequently as last night. I don’t know how/if the in-game FFB reset action interacts with the hardware…

It’s almost as if the wheel is ‘stunned’ by the sudden forces generated by the kerb hit and forgets where centre is by a few degrees - leading to the pull…

Any ideas?

@Mika - any ideas? Does this sound like a sim issue, or could the SC2 be ‘blinded’ to the sudden rotation caused by FFB kerb effects?


Difficult to say. The in-game FFB reset does reset the FFB, that is, it clears all FFB effects and creates them again into device.

The device does not use any FFB effect data when it produces the USB report back to PC, so at least it can’t affect the steering angle.

@MrFox: Just a (slightly weird) idea … Is the damage model in rf2 switched on or off?
If on, just try with this switched off.

This is to eliminate the (rare) possibility of your car suffering of a misaligned steering after hitting the kerbs hard, wrongly reset with the FFB reset. I’ve been a programmer for quite some time, so I know everything is possible. Just to rule this out. :wink:

@Mika - thanks. I’ll do some wider investigation when I get the chance. In retrospect, the easy way to check if this was a sim or device issue will be to quit out of the sim and see if TD reports anything other than a zero degree offset when the wheel is in the straight-ahead position.

@Joern - damage effects were turned on but no damage was reported in the car status display (or via CrewChief). I don’t believe damage was the cause but I can easily include invulnerability in the testing I will try.

I’ve been getting more time in on the rig, but still not had dedicated time to test this phenomenon.

I have had at least one more instance - and at a completely different car/track setting (Porsche 911 GT3R at Portimao). On this occasion, it was a right hand bend and the kerb bounce was not dramatic, but the wheel had a constant mild pull to one side (right, this time, I believe) upon landing.

Absolutely, definitely no car damage and the wheel corrected itself immediately upon an in-game FFB reset that I initiated after the problem continued through the next corner or so.

I’ve had the impression on previous occurrences that the effect can (sometimes?) sort itself out if you can keep driving for several seconds - but usually, I just hit the FFB reset button as I’d prefer to keep racing.

Are there any TD filter or Controller.JSON settings that are likely to ‘hold on to’ a force for several seconds? It’s almost like a Constant Spring force is being sustained when it should’ve decayed within a fraction of a second…

There is nothing in True Drive or in fact in the Simucube that would cause such effect. I wonder whether this is a rFactor 2 physics engine bug.

Any Direct Input filters active in your True Drive profile?
I had similar issues with ForzaEmuwheel and DI friction and SineWave, sometimes the forces keep pushing to one side because the DI signal offset is applied.

@Mika - thanks. I have been wondering if it’s something to do with the JSON rumble strip settings - e.g. the ‘pull factor’ and update frequency. It’s clearly not as simple as that though - otherwise I’d get this effect ‘all of the time’. I might try going back to the track/car combo where I originally found this as it was much more frequent and therefore easier to test.

@Alfye20 - I’ve not touched those filters… …partly as I seem to remember reading Mika saying that they were sim dependant and not always actually used(?)