Looking for a REAL fix for CSL DD grinding / damper loss bug

tl:dr - Fanatec CSL DD still getting gritty / notchy FFB bug all the time with latest firmware, looking for a real fix from fellow DD users and not a sad band aid fix, or a way to hold Fanatec or S397 accountable.

Hello all, this will be a longer one so I'll put a tldr. I bought a fanatec csl dd almost 2 years ago with pretty much the sole intention of experiencing rf2s amazing FFB with a DD. Not soon after first setting it up, I started getting the bug that disables the damper and replaces it with equally spaced gritty notches when returning the car to the garage or loading the next session. Using Reset FFB does not work and it requires the game to be shut down and the wheel to be rebooted to fix it.

Since I enjoy racing the AI at their full potential, I allow the AI to practice on AI learning 2 and also to qualify so they are properly placed on the grid. When I have to shut down the game because of this FFB bug, I lose all the time setting up a race (about an hour) that I have to redo to essentially play russian roulette with this bug occurring or not, but I feel I've lost more time restarting the game and redoing everything rather than actually racing. This also happens when doing an LFM race, and has forced me to quit without being able to rejoin from the same issue.

I used the seemingly universally recommended fix by the community and S397 to disable the damper in the fanatec controller or in the controller json file, but both turn off a good portion of the FFB features that make rF2 so great. It was fine as a band aid for an issue I thought would be fixed relatively soon in 2021, but I played with these settings for so long, I forgot I made those changes and at some point around when the new tire models came out, I started questioning why every car feels boat-like, sloppy, and almost featureless in the FFB.

I took a longer break from rF2 while the AI were being fixed and gave ACC and PC2 a try with my new DD and was surprised at how sharp and responsive the FFB felt compared to what I was used to on rf2, especially with so many people saying ACC FFB is not so great. When I returned to rF2, I heard recommendations to delete the userdata folder for the AI updates. I did so and rF2 instantly felt like I thought it should and so much better than ACC and PC2. Shortly after that, the gritty FFB bug returned and the whole process restarted in a loop for me 2 years later. Also yes, I have the latest fanatec program and firmware on my wheel (v451 ATM) and it changes nothing.

I've even helped people on reddit and the discord that are getting fanatec's official "recommended" settings for rF2 that have the damper disabled right out of the gate because this bug is so pervasive and they ask "why does rF2 feel like garbage". I help them and tell them to re-enable the damper to actually enjoy rF2, but warn them that the bug might occur and every person says it's night and day better.

So now to the point, the only other solution I see from fellow csl dd users is that this issue just "went away on it's own". Not very helpful from a diagnostic perspective. So I'm looking for any kind of help to either find a way to stop this bug from happening on my end without mutilating the FFB that's essentially why I play rF2, or find out what logs to look at and find a way to track down who should be fixing this issue, fanatec or S397. I'm tired of being the helpless customer trapped between two companies just pointing their fingers at each other. Anyone with info please chime in. Thank you.

 

https://forum.studio-397.com/index.php?threads/csl-dd-gt-dd-pro-ffb-settings.72751/page-2

Hi,

I've never had these issues so have never tried to fix them.

My parameters seem really extra (sorry if I have "the ankles which swell" by saying that).

I leave you my parameters in the following link.

Only the force is managed by rFactor 2, in other less realistic simulations the Damper and the Spring are used to reconstruct the FFB signal, but this is not the case in rFactor 2.

Only the force matters, use a controller profile 100% new + a 100% controller file (2 files) + put my parameters in the Fanatec driver (in particular damper and spring on 0%).

It should work.

 

Hello, sorry but the damper and spring values set to 0 in the fanatec controller is what I was doing for 2 years, but it made every car feel spongy and uninformative, especially compared to lesser games like Project Cars 2. I feel you might be in the same boat as me and believed whoever said the spring and damper do nothing in rf2, but it is a bold faced lie or the person who said it tested their theory with a cheap old wheel that didn't translate all the tiny details.

With the damper disabled or on 0, I can feel a generic push in my hand as I drive over a curb, but with it on 100%, I can feel every blade in the rumble strip in my hand. At S397s Daytona, with the damper off, going around the oval it feels very smooth and lifeless. With the damper on 100% I feel every little bump in the road as my wheel subtly jumps around in my hand. I highly suggest if you haven't in a while (or ever) try deleting just your controller.json so the game generates a new one and put your damper on 100%. Try an official dlc track optimally, as they have the tiny laser scanned bumps, and test with the damper on and off while in the game, you don't even have to send the car back to the garage and I bet you'll feel a difference.

 

Also, wanted to add the settings I personally use, as it might make the difference between someone feeling the damper on and off or not.

Fanatec Controller Settings:
SEN - 360 FFB - 100 FFS - Linear
NDP - 25 NFR - 5 NIN - 0
INT - 4 FEI - 70 FOR - 100
SPR - 100 DPR - 100

In-game I use 100% general FFB, 4 - 6 smoothing, and adjust the car specific FFB as needed, usually just down 10%. With these settings the game's FFB is unequaled and feels spectacular, but any of the available fixes seem to disable some feature of the FFB that is very apparent to me. I also have the boost kit for the 8nm.

 

Use my settings I assure you, several things are wrong with your settings.
I just try to help.

For example :

100% FFB strenght in the game systematically causes a lot of clipping.

As I told you, you must deactivate (0%) the spring and the damper in the fanatec driver, but in NO CASE in the controller profile file and the game controller file.

Do not put linear in the fanatec driver but prefer an FFB force value between 85 and 90%, because the hardware correctly supports all the torque, but nevertheless 75% (linear) is insufficient.

I advise you to deactivate NFR and NIN and instead put a little NFRICTION but very little (between 0 and 10%, me 5%).

Interpolation in fanatec driver and smoothing in game 0. No filter...

FOR 100% and FEI 100.

I'm not trying to convince you but to help you.

Also in controller and profile controller set offroad multiplier to 1 (100%) to feel sand and grass as it should.

 

Ok mate, I've been in this situation before on forums where if I don't test your settings 100% the way you explain I will not get any more help so I will do exactly as you say and do a test. I've done a lot of reading on the DD settings, manual, forums, etc. and your settings changes (besides the spring and damper settings) are all pretty much personal preference on how you want the wheel to feel.

So I just finished 10 laps around S397s Daytona with the GT3 Corvette using your settings exactly. I deleted my controller json and my saved controller profiles as you said, then changed every setting in the fanatec controller to what you said. This is what settings I ended up with, I believe this is what you recommended:

Fanatec Controller Settings:
SEN - 360 FFB - 90 FFS - Peak
NDP - 25 NFR - 0 NIN - 0
INT - 0 FEI - 100 FOR - 100
SPR - 0 DPR - 0

In-game:
Car-specific FFB - 88%
Smoothing - 0
Overall FFB - 80%

So all 10 laps were done with these settings, 5 without the damper, 5 with the damper and I did 1 lap without it, pulled over, turned on the damper with my wheel ITM and did one lap with it. Rinse and repeat and....

It felt terrible, way worse than my settings, while also not really producing any more torque that when I use linear. I can see what you were going for with no smoothing at all, but it makes my wheel feel like its passing over catches and sharp edges when hitting even the slightest bump. It does a decent job of disguising the damper being off, so I can see how these settings would be better than mine if you were to keep the damper off. It definitely rumbles a little more over curbs but there is absolutely no debate in my mind that even on your settings there is an obvious change with the damper off.

This is all what I felt just now with your settings in the 10 laps I did: with the damper off, it feels much closer to what my wheel produces with all my settings tweaked in and the damper on but I am still missing feedback from the car. The biggest difference is the subtle load on the wheel as the car's weight shifts around. With no damper, the wheel has no resistance moving it back and forth while stopped or moving, and when going around a corner and the rear starts to rotate, you can't feel the weight starting to come around like you can with the damper on. When I turned the damper on, I can feel the subtle weight in the wheel lighten telling me the rear is starting to lose grip. Again, these are all observations done with your settings.

So now that I have tried your settings, I encourage you to try mine with the same method of deleting your controller json for a fresh one and trying my settings with and without the damper and see if you can feel the subtle difference of the weight of the car in the wheel or not. If you still don't feel a difference, you're not the only one who has said this, but it still answers no questions of mine and makes no sense. Did I get some prototype version of the DD that seems to transmit FFB features other aren't able to feel? Either that or someone's slipping me some strong hallucinogens every day and I've gone completely off my rocker.

 

My settings are NOT a preference.

I'm looking for one and only thing :

Maximum REALISM in the simulation of the steering torque of the real car, whether I like the result or not.

Furthermore, this wheel is very good, is very high-end compared to belt-driven flywheels, BUT ENTRY-LEVEL DD, which means that some slight compromises are necessary but essential (Adjustment of the max torque of CSL DD + Ingame FFB strength [FFB strenght ingame * FFB multi]).

I already had a lot of knowledge before, but even before receiving the CSL DD I did all the precise research to understand all the parameters of the game and the fanatec driver, in order to understand their interactions, and unfortunately only recently the weaknesses of the CSL DD which involve a correction of my initial settings.

My settings are in no way related to my feelings.

I can only give you the information in this post, and then try to find something suitable for you, maybe inspired by it. And hopefully someone will provide further information.




My settings :

1) INGAME :

- Software / Custom / 1440 degrees.
- FFB Strength (ingame setting) = +70%
To avoid rFactor 2 software clipping but also hardware clipping related to the peak settings in fanatec software).
- FFB multi (per car setting) : majority of cars = 100% and ajustable.

- Min force = 0%.

2) DRIVER WHEEL :

- SENSITIVITY = Manual / 1440 degrees.
- FORCE FFB = 85%
- FF SCALE = PEAK
- NATURAL DAMPING = 0
- NATURAL FRICTION = 5
- NATURAL INERTIA = 0
- INTERPOLATION = 0
- FORCE EFFECT INTENSITY = 100
- FORCE = 100
- SPRING = 0
- DAMPER = 0

 

And I checked "damper FFB rfactor 2" on google, first thread proposed by google (forgot this thread BTW) :

https://forum.studio-397.com/index....s-rfactor-2-return-force-spring-damper.72759/

 

Ok, I retested this with NDP on 0 and NFR on 5 and it changes the feeling in a different way, but still the main problem and what I want the focus of this thread to keep on is no matter the settings I use, I can still feel a noticeable difference with the damper on 100% vs 0%. It is this subtle but informative resistance that tells me the rear end is slightly rotating, not completely at its limit, and it also highlights bumps and weight transfer.

The thread that is linked just seems to be another user who has done his own testing and come to his own conclusions, but my own testing is resulting in something different than what he and many others concluded, which is that the spring and damper do nothing. I can't just be mental and I'm not the only person who does feel something is lost with the current damper-off bug fixes. Here is a couple instances just in the last couple months of others saying the same as me, I've removed their names as I don't want to throw them under the bus personally.







So it almost seems some people's wheels are reacting differently than others to these changes, even though we're all mostly CSL DD users as far as I can tell. Thank you for your time and help in offering solutions, but there just seems to be something amiss when I disable the damper and when I don't, I get the gritty FFB bug that requires a full game and wheel restart that would be wonderful if finally resolved.

 

I know it's not the topic at hand, but why SENS on 360?

 

It's probably baked into me from the G27 days but was always how I reliably set the wheel lock back then. If there's a better recommended way to set the lock in rF2 I'd be willing to try it just to see if it affects anything. As for the actual setting of 360, that's just my preference of being able to go full lock to catch a slide if I need without having to start over-under handing the wheel. Once that starts my instincts kick in and in my frenzy to catch the slide, I beat the hell out of the rim 'XD. The CSL DD looks to have a skinny main shaft, so I don't want to take any chances haha.

 

The way pilAUTO has it above will produce the correct rotation: your wheel set manually to 1440 to cover all possibilities, in rF2 calibration set it to Software mode, custom angle, 1440, and tick "set by vehicle".

Most cars in rF2 are set up with more than 360 deg lock to lock (some much more so) so running that fast will give you very twitchy steering. But I doubt it will have any impact on your issue, it's just something to maybe think about adjusting in future. Most modern cars in rF2 still have quick enough steering that you won't need to take your hands off the wheel at racing speeds, other than very slow corners.

 

I have found that forcing the ffb into a smaller range does distort the effects, especially relating to overall steering response. I know it's a more common strategy with some older steering systems but, it doesn't suit DD-wheels very well, from my experience. I had to "unlearn" some old habits when I switched to a DD-wheel.

 

The notion of peak torque and sustainable torque must be taken into account.

Valid for all Fanatec DDs regardless of price, probably all brands.

Officially for the CSL DD the peak is at 8 and the sustainable at 6 nm.

But I really feel that around 88/90% it's perfect (6.8/7.2 nm). I just got home and these values seem more indicated than 85%.

Only very large bumps/kerbs/shocks will have software clipping (70% FFB Strenght) and also cannot be transcribed by the capabilities of the CSL DD set in this way. Well I think.

 

Ok, I'll give that a couple hours of testing and see if I can tell any differences from my default settings. I'll also see if it effects the ffb bug from occurring more or less.

@RaceNut - I've never felt anything that would hint the FFB is distorted, all the bumps and feedback seem what I would expect for what I'm seeing on screen, but I'll definitely do a thorough test and probably just try and adapt to keeping it this way if it's the proper way to set the wheel up.

 

Setting the sens to 360 is the issue. Everything else you may try is hopeless unless you have the rotations set to match the actual vehicle. Set the fanatec driver manually to 1440, then 'in rF2 calibration set it to Software mode, custom angle, 1440, and tick "set by vehicle"'.

 

By distortion, I mean range compression. It's squeezing the ffb into a smaller (than optimal) frequency range. It brings unwanted results in most cases.

 

Hello everyone, sorry for the large gap from the last post. I wanted to post once I had done a large amount of testing but some odd life problems got in the way. It wasn't my plan but I ended up wiping my PC for a fresh install to W11, so my testing is done on as stock / vanilla of an OS as it can be. I did not copy any of my files back from my W10 install, all settings are manually input on a fresh install (Also running the latest Fanatec firmware (v455)).

So using all the settings recommended to me in this thread, including setting the rotation limit to 1440°, still hasn't had an effect on the griding / rocks in the wheel base accompanied by a loss of some of the FFB feeling bug. Once I noticed it was still happening I wanted to test and see what is most likely to cause it happening in the first place to try and at least avoid these situations:

1) Resetting car back to garage while on track, most likely to happen in practice sessions.
2) Pausing the game for more than 10 - 15 minutes. If I pause a longer race with AI to go do something for too long, It seems the game being put in suspension / paused on the CPU creates the same FFB issue when resuming the game.

Since these situations are only avoidable sometimes, I still have issues trying to stay in one online race practice session without having to quit, reboot the wheel, and restart, hopefully before the race starts. Similarly, I haven't done a longer AI race in a while since I'm worried if I do have to pause for any reason, the race is most likely over since the bug will probably happen.

In my time since my last post I've played ACC, AMS 2, Raceroom, ATS, ETS 2 and Snowrunner without this bug happening. This issue seems to specifically happen only in RF2. I would really appreciate some help in getting some visibility and attention on this issue. I understand that most focus has shifted to LMU at this point but I'm worried this issue will also be present in LMU since they are based on the same engine and I can't buy LMU until I can see that this issue isn't happening to CSL DD users. I hope that this can provide some info and help get this fixed so I can get back into RF2 and hopefully keep it from showing up in LMU.

 

It seems that this will not happennin LMU
https://discord.com/channels/1197577617482727514/1205163240938672169/1205569623639199765

 

Awesome, thank you for the link, I just read it.

To anyone who has the power, any chance this can get fixed in rf2? I've got pretty much all the DLC and don't find myself using it much anymore because this bug really kills the night when it happens. Thank you.