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

Today for the first time I noticed that Logitech profie has the Global setting and the Spechific setting.What isd the diffirence and can you post your setting of both global and specific setting?
Thanks

 

If you create a Profile for rF2, you need to adjust Specific Settings. If you don't use a profile, the game will use the Global settings.

 

Ok, thanks.

 

What do you think?

 

Thank you DrR1pper for this valuable advice to be in the "Zone"!
Now that we are almost there, I think it is natural that new doubts arises, specially if we consider that most users (like me) don't know in depth how each factor affects FFB in our old controller.ini or in the new controller.json

So, we have for the G25 and G27:

"Force Feedback":{
...
"Steering torque minimum":0,
"Steering torque minimum#":"Minimum torque to apply in either direction to overcome steering wheel's 'FFB deadzone' caused by friction",
...
"Steering torque zero-speed mult":0.45,
"Steering torque zero-speed mult#":"Multiplier at zero speed to reduce unwanted oscillation from strong static aligning torque",
...
},

and I observed that lowering S_t_z-s_m from 0.45 one can prevent the stationary oscilation of the wheel.

My question is how these two factor interact? If any. And how their role to be in the "Zone"

Thank you!

PS: wajdi77 interesting post, never saw something like this in years.

 

You're welcome newton.

To answer these questions, let's start with why we have need of a "Steering torque zero-speed multiplier" and i think i can be best explained it by explaining how and why stationary oscillation occurs in our ffb wheels in the sim.

1) With the car stationary, you input a steering wheel disturbance by giving the ffb steering wheel some momentum followed by letting go of it.
2) As the ffb steering wheel rotates under the momentum you've just given it, the ffb steering wheel position changes and this is registered by the sim as steering input changes over time.
3) The steering input changes over time causes the front tyres to turn. As the tyres turn, frictional forces are generated (from tyre scrubbing with the road surface) which acts to resist the tyres changing direction. This also results in a counteractive force at the steering wheel and is sent back to the ffb wheel.
4) This counteractive force sent back to the ffb wheel slows down the momentum of the ffb steering wheel over time.
5) After a few iterations of this feedback loop between the sim and the ffb wheel, the ffb steering wheel has no more momentum to stops rotating (in whichever direction you initially set it on) but at this same moment in time the ffb motor is still producing a counteractive force sent from the sim.
6) Ordinarily, the sim should stop sending a counteractive force to your ffb wheel because if the steering wheel is not moving then the front tyres are not turning. If the front tyres are not turning then there's no friction force produced at the tyre. If there's not frictional force at the tyre then there's no longer a counteractive force at the steering wheel and the steering wheel and tyres stop turning. However, the sim is not aware that the ffb steering wheel has actually stopped rotating because the sim only receives positional data about the ffb steering wheel and not it's velocity. It will take 2 whole physics refreshes of identical ffb wheel position readings by the sim before the sim then knows that the steering wheel has actually stopped rotating and to cease sending a counteractive force to the ffb wheel.
7) But ceasing the counteractive force at this point is too late and should have happened 2 physics engine refreshes earlier, where your ffb steering wheel had physically stopped rotating.
8) Due to this delay in turning off the counteractive force by the sim, your ffb steering wheel has had enough time to build up momentum to now start rotating in the opposite direction and the entire process starts all over again causing a self perpetuating oscillation.

The principle reason why stationary oscillation never happens in the real world is simply because the steering wheel is directly connect to the front tyres, meaning the steering wheel turns instantly when the front tyres turn. There are also other factors at play here (such as the ffb steering wheels internal friction and moment of inertia especially in less powerful and less responsive ffb wheels which are also undesirable steering input factors that don't exist in the real world).

Leo Bodnar actually wrote an article (here: https://dl.dropboxusercontent.com/u/17548791/FFBdontwork.pdf) that i didn't fully understand but because of your question it made me think hard about it and i think i now understand fully what he was saying since it's related in part to why stationary oscillation can occur in sims (e.g. positional data and not velocity data and also how the wheel and driver inputs affect the moment of inertia of the steering system in the car). Leo goes on to explain how such a problem can be fixed but would require a paradigm shift in ffb design in both hardware and software but should be completely doable.

Going back to your original question, how do the "steering torque minimum" and "steering torque zero-speed multi" interact?. They don't, or at least their not supposed to. "Steering torque zero-speed multi" is simply to reduce/remove the oscillation effect of our ffb wheels only when the car is stationary (and is not wanted when in motion or else you'd just feel weaker forces). "Steering torque minimum" is to remove the initial ffb deadzone of our wheels and if you set it too high causing oscillations in the wheel, although you can remove the stationary oscillation completely by lowing the "Steering torque zero-speed multi" it is not going to fix the oscillations when in motion. What you want to do is work on removing the oscillations when in motion by finding the "steering torque minimum" sweet spot for your particular wheel and then if stationary oscillations still occur, go ahead and lower the "steering torque zero-speed multi" until you've found the sweet spot for that too. It never dawned on me to do this until you asked these questions btw.

Regarding how they affect being in the zone, "steering torque zero-speed multi" has none (only removing any annoying stationary oscillations you may have which incidentally i do have with the karts, so thank you for helping me realise the solution ) but "Steering torque minimum" has a baring on being in the zone. In my opinion, not as much as avoiding heavy ffb clipping but non the less i strongly believe it plays a role.

 

not to be ignorant, and theres too much to read through. But having played GCS13 i found the FFb to be better..
in a nutshell what are considered the major FFB settings to change (in rf2), to get a better feeling. over kerbs grip etc etc.

dan

 

You need to read through the OP at least.

http://isiforums.net/f/showthread.php/19034-Optimal-FFB-settings-in-rFactor-2-guide-The-key-to-being-in-the-Zone-D?p=260089&viewfull=1#post260089

 

To be fair I use the default settings only adjust rotation and FFB Multiplier to my liking. Also I use 0 FFB smoothing even with my old Logitech G25 and also with my new Fanatec CSW.

For me it is great and I have won a lot of competetive races with it. Depends on your taste, but I never bothered messing around with it and I am very fast with it.

 

thanks guys..

i really wish the car specific FFb would stick across all cars in that model. Doesnt it stay with the car you choose , by number?

bit silly lol

 

Hello DrR1pper,

thanks for your guide. I think it is one of the main topics in simracing which is worth to take some time to find the right settings. I know some points already from the iracing forum...so like you said, nothing really brand new.

But there is one thing which I like to discuss. Maybe it was already a topic in this thread. I read only your guide not all further posts. My question is about the steering force output on the wheel. I get it that the ingame ffb should be set like there is no ffb clipping. So I use the "rF2 Pedal & FFB Overlay Plugin" to check my maxima. But here is the point. The plugin uses the Internal rF plugin which provides the data for it. But that means also that this plugin shows only the rF ffb force itself but NOT the steering force output on the wheel.

Like you already mentioned is a G25 not the state of the art and especially very weak (input lag) at the neutral point or rattling very loud if the output is to heavy. If I use your setting 100% at the Logitech profiler for Overall effects and decrease the ingame FFB mult to get it in the right zone, my wheel effects are especially in low speed corners and at the neutral point much to less. But if I increase the Logitech profiler for Overall effects after I found the right ingame FFB mult with the "rF2 Pedal & FFB Overlay Plugin", I have more FFB strenght but the values in the plugin are still in the right zone.

Means also you should never reduce the ffb strength on the wheel profiler and let the ingame FFB multi on the same wrong parameter, which produces always the ffb clipping. But if you have the right value ingame according to your guide, you can still increase the ffb Overall effects at the wheel profiler if the effects are to weak.

Is that right or am I missing something?

 

Hi Sebastian and you're welcome.

I think i understand your question but just in case i have not, instead of giving you a yes/no answer i will reply with an explanation that should hopefully answer your question.

rFactor 2 ffb multiplier (i.e. in game "car specific ffb multi" setting) and your ffb wheel profilers "overall effects strength" are completely independent of one another. And because of this, there are actually two potential sources of ffb clipping, from within rFactor 2 and from within you logitech profiler.

What that means is if you kept the rFactor 2 ffb multiplier at any constant value and vary your ffb wheel profilers "overall effects strength", the force outputs from the sim are still going to be the same. But before the final force output from the sim reaches your ffb wheel, the ffb output signal is modulated by the settings you have chosen in the logitech profiler. So in the example you gave, where you lowered the rFactor 2 ffb multiplier to avoid ffb clipping (which you can see via the "rF2 Pedal & FFB Overlay Plugin") you also experienced a reduction in force output that was most noticeable in the low end forces (as lowering the rf2 ffb multiplier will naturally lower the entire strength range of the output forces). You made up for this reduction in the low end forces by increasing the "overall effects strength" in the logitech profiler which increased the ffb strength of (most notably) the low end ffb forces. This is the correct way of going about the problem up to a point.

Increasing the "overall effects strength" past 100% is ok up to about 110-112% because going higher than this on a g25/27 starts to cause profiler-side ffb clipping effects in the high force end. I don't have a g25/27 graph to show this but here is my force output response curves with changing "overall effects strength" on a t500. The default value for a t500 is actually 60% instead of being called 100% like it is on a g25.



As you can see, 60% ffb response curve is the curve with the highest peak torque output whilst still remaining linear across the majority of the ffb range. As we go higher in "overall effects strengths" there is a slight increase in peak torque which is nice however the response curve starts becoming non-linear with a flatting of the high end forces which is effectively causing ffb clipping. Now, in rfactor 2 with the "rF2 Pedal & FFB Overlay Plugin" you will not see the modulation of this response curve with the rf2 ffb output signal because the "rF2 Pedal & FFB Overlay Plugin" only shows the ffb output that rF2 is producing. Which means, even if you have set the rF2 ffb multiplier low enough to avoid ffb clipping in game, depending on the "overall effects strength" in your logitech profiler, the final ffb output sent to your wheel may have ffb clipping occurring (as shown in the graph).

Now, so long as your not going over 112%, this won't be a problem and you can keep using your chosen value. For myself with my t500, i use the 70% value because if you look back at it, there are 2 main benefits. Firstly, the peak torque in actually increased a little and secondly, although it's hard to tell from the graph, the initial ffb deadzone of the t500 is reduced by half (from 8% to 4%) and the low end forces are all stronger (seen by the fact that the 70% curve is steeper than the 60% curve).

Now when you then combine this with some "steering torque minimum" (as explained in part 2 of the guide), you can remove pretty much all the initial deadzone and further improve the strength of the low end forces (because removing any initial deadzone will shift the ffb response curve to the left meaning all the forces sent to the wheel correlate to a stronger force output at the ffb wheel.

I hope that made more sense than confusion. If you still have any questions, don't hesitate to ask.

 

Thank you for this detailed and understandable explanation. It does make totally sense that also the wheel itself has it's limit and it's also worth to check here ffb clipping. But how do you produce this graph? Is it the csv values of the "Step Log 2 (linear force test)"?

About part 2: I tested it but even with Steering Torque Minimum = 0, my wheel is oscillating when stationary. So should I take some tests also with different settings for...?

 

Dr. R1pper
So are there curves like these for the G25 and G27?

 

Correct.

So was your wheel always oscillating when stationary or did it start when you set the "Steering Torque Minimum" = 0? I've not been on rf2 now for about 5 months (as i'm abroad) so i don't know if this has something to do with the new JSON file format or something. Also, did you "Steering Torque Minimum" to = 0.00000 in the controller.ini file? It requires a 5 decimal place value to work.

 

The default value is 0...so was it set. This already produces on my G25 oscillating when stationary.

But the atteched 0 seems not be important anymore. I set it like mention at your guide to 0.03000 and after starting rF2 again, the value is changed to 0.03.

 

Ok, so turned out there is a graph for the G27 by Marc Coyles in post #460

Here's his wheelcheck.exe graph (same way i made the one for my t500): http://isiforums.net/f/attachment.php?attachmentid=13556

As you can see, the g25/27 has a very different natural ffb response curve. It's naturally not a very linear response curve, even when the overall effects strength is reduced down to 60% but this is not something i would really worry about because in all the cases because there isn't as strong of a flattening at the high end forces as there was for the t500. But immediately apparent is the very large initial ffb deadzone of the default 100% setting. Just increasing the overall effects strength from 100% to 110% reduces tho initial deadzone by 75% which is a huge reduction. Like i said earlier with my t500, removing the initial ffb deadzone has the effect of increasing the strength of the low end forces by shifting the response curve to the left which is why the wheel feels so much stronger in the low end when you increase this value in the profiler. In the extreme case, 120%, you do get an undesirable effect of a very steep initial response curve that very sharply tappers off at the very early stages of the ffb output (10%). In the case of the g25/27, a good compromise is around 110-112% overall effects strength.

 

hmmm. Before you ever tried anything with the controller.ini file, was your g25 oscillating when stationary ever before reading this guide?

hmm, someone else will have to weigh in on this, i think it has something to do with the json format (or something) that i'm unfamiliar with.

I missed you earlier question about...

You can leave this on default (should be 0.45) or reduce it some more if you still get rattling when stationary. However, if your getting a lot of oscillation when stationary, it usually means you're also getting some oscillation when driving which is not right and needs to be resolved. Something is causing it that should not be there.

Here is what i mean by oscillating when stationary (which in my case only happens if i set the "steering torque minimum" value too high.

ok, i must go to bed now it is very late here. I will respond to any replies in 9 hours.

 

With DFGT in the Logitech Profiler I run with FFB at 98%. I spent dozens of runs with wheelcheck program to come to that conclusion.
What ever wheel you run, and you are finding oscillating wheel with stm at 0%, you would have to accept that the FFB percentage you are running in yours wheels setup program is too high. (yes?)

 

I have found that with my G27 it was better to lower FFB to 98% was most linear for me and set minimum torque to 0.12 (12%) The min torque moved the dead zone out to a point that was all but zero dead zone. (used wheelcheck.exe to check)

Profiler settings the same as you have shown except spring effect is set at 100%. I have tried this setting all the way from zero to 150 and as a personal feeling have settled on 100.

The last part of the puzzle for me is the FFB multiplier which is set for each mod. Good thing is that game saves the setting for each mod so once set no more need to change. The pedal overlay is the tool I use as my guide for this

setting the FFB multi down until no red clipping evident. I have settings from 0.30 V8 super cars to 1.24 mazda Miata. Just takes time to find what the user likes best.

My wheel will oscillate when I hit race if I don't hold the wheel a little but as long as I just hold the wheel a little the no oscillating happens and I test by driving a very slow speed down pit lane and again no oscillating so all good.
As said a million times this area of the game is all about what the person using it likes.

While in reality trying to get the most out of the gear we have (that's me) is a nice goal in the end it does not matter if the person using it is happy with something less than some prophesied ultimate setting.