Guide: Optimal FFB settings for rFactor 2 - The key to being in the "Zone" :D

Two choices:

1) Increase filtering (in the UI of the game). Try it one step at a time until the curbs feel OK. If the rest of the FFB still feels vibrant, then problem solved. If the other forces are too muted when you do this, then put filtering back to default 3 and:

2) Reduce the FFB overall until you are OK with the curbs. You can do this with a Fanatec wheel profile globally, or, with the FFB Multiplier in the game just for the FR3.5.

Bottom-line, the RELATIVE FFB between curbs and road surface is fixed by the physics, so you can't fiddle around with it. It is what it is. You have to make global adjustments to solve problems with wheel hardware. (I would recommend this, even though it's more work than just adjusting the FR3.5. There are/will be other cars with exactly the same issue). For me, it was just making sure the filtering was at 3 and not back to 0. In real life, sawtooth curbs can be scary, though, but that's the whole point of them. They are designed to tell you not to drive over them. If they weren't rough, there'd be no point in having them versus a standard curb.

 

Trust me, it's not as simple as that. The wheel does some extremely harsh and sharp affects. I think it's the way the wheel generates those forces, it's not normal. Having to use smoothing at 8 or 10 shouldn't be required. It's not just the Intensity of forces, it's the way they're generated.

Also, putting down the overall FFB power to hide a problem is just that, hiding the problem.

It's literally like a super defined jackhammer going crazy trying to split your entire wheelbase in half.

 

Thanks for the advices, I don't remember now what my filtering setting was, but I'll have a look later today. I'm not going to mess with the global FFB settings, I'd rather go on a car by car basis, because for some cars, like the URD PX revolution it's quite good as it is now, it's not as crazy as for the 3.5, but still could be better.

Spinelli managed to describe very well what I'm feeling, it's like you see on tv when an OW has an accident and the pilot takes his hands of the wheel, it's that violent, but of course a lot less rotation angle. It's like there are no tires in between the kerbs and suspension. I think there is still a problem with how the CSW "sees" these kind of forces, like I said, with the old DFGT you could actually feel the tire deforming when you got on the kerbs, so it's not necessarily a FFB issue.




LE: You'll probably think I'm an idiot but I just tried the FR 3.5 again yesterday on both Silverstone and Malaysia, and the erratic FFB behavior off-track and on the kerbs is completely gone. And I didn't change anything.

Marc, I assume that by filtering you mean FFB smoothing. I had this at 0 and kept it like this, by increasing it I felt that the FFB wasn't as "alive" as I wanted, especially in the high-speed corners on the limit. I don't know what happened in the meantime, but the kerbs feel absolutely great, when you mount them with half of the tire width you can actually feel the tire leaning against the kerb. I did just dropped the FFB multiplier to 0,95 but also with 1,00 the crazy shaking is gone when going off-track. I don;t know what happened, but I'm very happy with how it is right now.

 

I forgot how STC works again. If I have a 27 Nm wheel and set STC to 27 Nm then the game tells my wheel to output 100% force when the in-game car-physics torque reaches 27 Nm (or more). Correct? If so, then wouldn't an Accuforce or Bodnar wheel owner get constant, massive clipping when setting STC to 13 Nm or 16 Nm because there are lots of >13 / >16 Nm in-game car-physics moments??....Obviously that can't be the case...I can't remember how this all works...

P.S. Assume FFB multi is set to 1.0.

 

^Correct.

So for those cases (e.g. AF/bodnar) where setting correct STC values for those wheels shows ffb clipping in techade's plugin when using default car-specific ffb multi of 1.0, then if you want to avoid it you simply have to decrease the ffb mutli until it no longer shows clipping.

 

Oh ya that's right. But then there's also that sort-of random, "artificial" torque value specific to every car that the car creator/modder decides on which, if set higher than the player's STC setting, will mean the STC setting is pretty-much ignored, yes?

 

Yes.

It's called "nominal max steering torque" or something. Your STC needs to be above it for the STC to take effect.

 

Any suggestions what "Steering torque sensitivity" should be used to make T300 (75% force) more linear?

 

Why hasn't this post has been made a sticky yet?

 

Glad you are happy. I only know that I leave everything at default and it feels great on my CSW.

 

The linearity may change by raising the wheel's force (e.g. over 75%). How are you testing for linearity?

 

Wheelcheck, below are results of 75% force:
https://docs.google.com/spreadsheets/d/1PrUVdUivFSiZUyDo_zIALXAXUhnP_m6wTATP2abczPc/edit?usp=sharing

At around 55 input, the output is too high by ~25 units.
"Steering torque sensitivity" setting doesn't look user friendly. I don't know what value I should apply to get the required amount of reduction. Currently trying 0.5 setting.

I think it would be great if rF2 could provide pCars like linearity setting where you select the point of highest point of the curve(knee) and reduction at this point by (scoop reduction) as seen below:


This method shouldn't reduce low forces that much as the reduction is linear from from the knee point instead highly curved (rF2 method) as per example seen in this thread:


Thrustmaster/Logitech wheels already have deadzone/lower forces at start of the curve, reducing them more isn't the best option.

 

double post

 

Have you experimented with the TX/T300RS to see what is most linear for it? I've seen tests showing 72% for the T500RS (while it's true 100% setting is 60%), so I'm very curious what percentage the TX/T300RS is most linear at (true 100% setting is 75%).

 

This thread is an impressive contribution to the community... Not only that you took the time to write it all out, make maps and charts, test and retest and retest some more, but then to answer all the questions and pretty quickly too...

Props man... Seriously.

So, I got a few things to ask the FFB Guru...



1. Have you read this? http://www.race2play.com/homepage/show_posts/7229

2. What are your thoughts on the throttle and brake effect on steering? I believe they're are turned off by default... If you haven't tried them I suggest you do, but do it separately from this next request. because I want you to see the difference in the two separate changes.

3. Recently an Aussie gave me the suggestion to take rF2 to 100% look ahead, -60 exaggerate yaw and 100% on head movement. Do you use anything similar? If not, would you mind trying it for a full fuel run? I'd like to hear your thoughts.

After you've had a chance to digest and share your thoughts, I'll tell you how it went for me; don't want to influence your experiment. I run a DFGT 2nd gen and triple screen 5760x10 with like 24rffov

And again, great thread. I'm still not done reading it all...

Oh... and no centering spring? When I do that it's just so weak and flimsy. Maybe it's my cheap wheel.

 

If you're asking me those questions...

1) No but i will have a look in good time. Got to drive to london in a min so can't for now.
2) Are these artificial effects available in rf2 as well? If so, this is the first i've heard of them. As to whether i would use them or not....would need to test. However i haven't played rf2 since november and won't be able to for some time still.
3) I did experiment with head movements/vibrations a little some time back. I do remember at least halving them all. I would think that whether one should remove them all or not can be very dependent on the persons chosen FOV. The lower (i.e. closer to 1:1) their FOV setting, the less exaggerated g-effects on the drivers head is needed to notice small changes in car pitch/roll/yaw in the moment. The larger (i.e. more fish-eye'd) the FOV is, the harder it becomes to notice small changes in car pitch/roll/yaw and so i can see how exaggerated head movements can be helpful. As example for this, i would look at Baronesbc who makes a number of videos on youtube and you'll see he uses no head movement whatsoever apart from some minor background engine vibration effects. He also uses a pretty low (if not 1:1) FOV as well.

And yeah, DFGT is a weak wheel relative to what else is available. If it feels better for you to use some spring centering...then i guess a little is not so bad. I'd just be a little weary on how that could affect your potential performance or not. I would test with and without for a few days at least with the same car/track combo and see how average performance is affected.

 

T300 is quite linear at ~60% force, however you loose maximum torque, also deadzone increases, which I prefer to avoid.

 

Am I reading the graph right? Does every increase up until and including 100% result in more torque across the board (including max torque)? If so, what happened to the thing about the TM T300RS's/TX's true max power being at 75% and anything over that just boosts the lower-end forces without raising the max (sort of like compression)? Is that all incorrect? Tge T500RS is apparently like that as-well (but at 60% instead of 75%).

Also, does the wheel have a 4% or 5% deadzone with FFB at 100%????

Did you have periodic and constant both set to 100%, and spring and damper both set to 0% for the tests?

 

I believe so, maximum torque at 100% input is noticeable while playing/testing. The engine is producing more noise/whinning while the game gives 100% input to the driver.

It seems to work both as compression and increase to maximum torque, however torque increase isn't large going from 75% to 100%. While 50% to 75% is noticeable.

From my testing it always had ~2,5% deadzone until the wheel starts to vibrate, however forces until 5%-10% (depending on force % set) are very weak and linear/random noise.

Yes.

The graph with comparison of forces isn't made by me, however my testing resulted in very similar results.


Someone made similar tests with T500:


This graph suggests that torque should increase while increasing force. With 90-100% force wheel is almost clipping early, however maximum torque increase is present compared to 50-60% force.

If the behaviour of increased torque while increasing force is correct, then there is a reason to find best possible way to "fix" linearility of these wheels to benefit from increased torque.

 

I think you are conforming what we should all intuit: that there is no reason to ever NOT use linear settings unless our hardware has a problem that you need to compensate for with a software tweak. Am I understanding correctly?

It's no different than using a mixer for music. Why bother? The original artist and recording engineer already picked the optimal settings. But if you have speakers that are crap and can't reproduce certain ranges, you might want to fiddle...at an overall loss of fidelity no matter what you change.