“Steering torque capability” and “steering torque sensitivity” in controller.json file.

Discussion in 'Technical & Support' started by Wesa, May 1, 2019.

  1. Wesa

    Wesa Registered

    Joined:
    May 1, 2019
    Messages:
    1
    Likes Received:
    0
    So after searching a bit I’m still not certain what to do with these two settings found in the controller.json file. Apparently they affect ffb to some extent.

    Has anyone determined how each setting affects ffb and what ideal setting should be?

    My wheel has a torque of about 3.9 no, so I’ll assume it needs to be 3.9 in the “torque capability “ setting.

    What to do about sensitivity?
     
  2. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    Capability is for powerful wheels. Leave it.

    Sensitivity can be personal preference, generally leave it. If you like strong forces all the time you could raise it, but weaker forces shouldn't be too bad on a 3.9Nm wheel (in gaming terms).

    *I'll explain this more later, when I have some time. I know a prescriptive "do A, don't do B" isn't very helpful with understanding.
     
    Last edited: May 1, 2019
  3. Emery

    Emery Registered

    Joined:
    Oct 24, 2010
    Messages:
    3,035
    Likes Received:
    1,654
    Steering Torque Capability was introduced in build 906, back in Dec 2014.
    https://forum.studio-397.com/index.php?threads/rfactor-2-build-906-now-available.46649/#post-776411
    Added a Controller.JSON value “Steering Torque Capability” in order to specify the strength of strong FFB wheels (like direct drive). If this is set properly, and it exceeds the NominalMaxSteeringTorque for a given vehicle, it allows a 1:1 correlation between the calculated torque and the actual FFB torque. (However, note that we still don’t model power steering, so the calculated torque is stronger than it should be for some vehicles.)
    NominalMaxSteeringTorque is a value in the car's physics. For most cars, I believe its in the 10-14 Nm range, but for some specific cars with large aero (e.g. DW12), it can be as high as 30 Nm.

    So in practice, unless your gaming wheel produces at least 10-14 Nm, "Steering Torque Capability" will not change how your FFB is affected.
     
  4. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    Steering Torque Capability

    Emery has covered the detail above, but let's put some numbers on it so we can see what it does and when to (not) use it.

    Let's say we have 2 cars:
    • A: Track day car, light, quite narrow tyres, no downforce to speak of. Has a nominal max steering torque of 9Nm.
    • B: Modern formula car, big tyres, lots of downforce. Max steering torque around 25Nm.
    Note: Nominal max for each car is chosen based on 'normal' inputs. That usually won't include things like a 45° kerb strike at 320kph, since that would generate huge forces and everything else would pale in comparison. This nominal max is actually an important parameter to get right, the recent Reiza bundle just had adjustments in this area to help the content better compare to the existing rF2 content.

    In each of those cars the game scales the calculated steering torque in accordance with the nominal max value. So assuming your wheel does 4Nm, if car A has a current steering torque of 6Nm, the game can see that's 66% of the car's maximum, so it outputs 66% FFB to your wheel, which then produces* 4 x 66% = 2.66Nm. If that same car produces 9Nm at a particular moment, your wheel will go full power and produce 4Nm; if the simulated car produces 12Nm because of an unusually large deviation or impact, your wheel will still produce 4Nm (it can't go any higher).

    *how much force your wheel actually produces we'll get to in the second section

    This is all fine, and generally how things work with normal consumer wheels. Whether you jump into car A or car B, you will experience a range of forces (depending on the situation) ranging from 0 to 4Nm at your wheel, based on where things currently are in relation to each car's maximum. Let's step it up a notch.


    You go out and buy a 25Nm direct drive wheel. Fantastic.

    You drive car B, and it's transformed. You're feeling the actual forces the (simulated) car is generating; road surface + tyre + steering geometry produces 5Nm? You can feel it. High speed cornering producing 22Nm? You can feel that too. Your wheel matches the car and you're 1:1 with what you'd feel in real life (excepting game inaccuracies).

    What about car A? Well, not so good. Those calculated forces in the range 0-9Nm (in the car) are now stretched over 0-25Nm by your wheel. You catch a lazy slide, overcorrect, the car rocks back and instead of the solid 5Nm you'd get in reality your wheel is hitting you in the wrist/jaw with nearly 14Nm. Driving a passenger car is like doing a workout, and if you think you're going to nudge anything you better take your hands off the wheel or you could injure yourself.

    Solution? You could adjust the vehicle-specific FFB Mult down, so the forces better match reality. Hope you know what sort of forces you'll encounter so you can get the right value. And then do that for every car, occasionally forgetting you haven't yet and getting a shock when you jump in.

    The better solution? Change Steering Torque Capability to 25. In car B (or anything with even more force) the game will let you feel the full power of your wheel. But jump in car A and the game can see its nominal maximum is only 9Nm, 36% of what your wheel can do. It will then scale all forces by a factor of 9/25, so you feel everything correctly in car A without needing to change anything else. (added bonus: higher forces that would have been lost previously [ >9Nm], are now sent to your wheel)

    Any car with a higher nominal max than your wheel can do, will get scaled 0-100% like before. Anything lower will be scaled down so that it produces correct forces, instead of being too powerful and not matching reality.


    What happens if you adjust this parameter with your consumer wheel?

    As above, if you set it to 4 nothing will change. I think there are no cars in the game that produce a maximum as low as 4Nm, so the game will continue to scale everything to your wheel and all cars will potentially produce your full 4Nm at the wheel, but generally a little less (so you can continue to feel detail).

    If you think that cars scaling sounds like a good idea and put it on 25, your wheel will produce 4 x 9/25 = 1.44Nm in car A at full power. Most of the time somewhat less, sometimes a lot less. Now you can barely feel any detail, and overall forces are weak.


    So: if you have a powerful wheel and want cars that should feel lighter than your monster wheel to actually feel lighter, set the Steering Torque Capability to suit your wheel. Otherwise, leave it.


    Steering Torque Sensitivity

    First, this defines the way the range of 0-100% force is scaled to your wheel. The standard is a straight line, so 50% maximum force (in the car) is given to you as 50% strength at your wheel (assuming 100% is full force from the car; these are all intertwined a bit). If you raise it, 50% in the car will produce more force at your wheel, and the opposite is true for lowering it. So higher sensitivity will give you generally more force, lower will generally produce less force, while in both cases not changing where 0 is and what 100% means. (same concept as the input axis sensitivity in the game)

    There are two main reasons to adjust this:
    • To change the general level of force,
    • To correct shortcomings in wheel behaviour.
    The first is as described above: if you have a weak wheel, you might find that during mid-corner with mid-level forces you aren't really able to feel changes as you lose and gain traction at either end of the car. By dialing up the sensitivity you can give yourself more force (and therefore bigger fluctuations, easier to feel) in normal situations, while not losing detail up near 100% because you haven't simply scaled all FFB up (though it will flatten out a bit, so it's personal preference on what sort of compromise you'd prefer). Lowering sensitivity to adjust overall FFB probably isn't as common by itself, and certainly not with a consumer wheel.


    The second reason is a topic of much debate. There's a theory that your wheel doesn't produce forces in a linear manner. The game might say 50% but your wheel doesn't produce 50%. Maybe it's only producing very low force at lower percentages, and doesn't catch up until you reach full force (so all intermediate forces are too low). Or it might try too hard to give lots of power up the top and make 90% hardly any different to 100% (at your hands), making it difficult to discern details at high loads.

    There was a program made which pushed the (unloaded) wheel with varying forces and measured how quickly it moved, and used these results to produce a graph of wheel linearity. By careful analysis of that graph and adjustment of the sensitivity and other parameters (like minimum torque, effectively raising the 0-level of your wheel) you can make it more linear and give better awesomeness in its feedback. In theory.

    On the other hand, electric motors (which wheels usually have) tend to be quite linear when static, as they're based on the principle of induction. Half of full power will produce very close to 50% of full force, much closer than suggested by the aforementioned program and probably close enough that you won't feel any difference. Note this is when you're holding the wheel still.

    In practice, you'll generally be holding the wheel but it'll often be moving a bit. The wheel's response may change in that scenario in comparison to holding it still, though how much is anyone's guess (I've seen a static test, but I don't know if anyone's gone to the trouble of measuring a moving wheel's response - and this would be motor dependent, so there's not much chance you could find someone testing your wheel so you don't need to).

    In my opinion, for hands-on driving, the wheel's generally moving slowly enough at times when you want to feel the forces that just leaving the wheel (and sensitivity) as-is will probably suffice. If you have a fast responsive wheel and do a lot of drifting, and want the wheel to rotate properly, you may want to look at that program (it's called Wheelcheck) and adjust the sensitivity so that its 'free' movement is closer to reality.

    If you feel that FFB forces tend to be too weak on your wheel, and increasing overall FFB doesn't do enough (make sure you don't go too high, or you will lose detail at high forces which isn't any fun), you could try increasing the Steering Torque Sensitivity so that your wheel feedback is generally stronger.

    While you're there, consider adjusting the minimum torque (that one's in the game, too) and see if that improves your feel.

    Wow, that got a bit lengthy. Sorry about that.
     
    Ernie, I_Bellett, Jack7793 and 9 others like this.
  5. Staenker

    Staenker Registered

    Joined:
    Apr 30, 2019
    Messages:
    20
    Likes Received:
    5
    Please don't change the detail/length of your posts, if you wanna change something, stop feeling sorry for crafting very helpful resources ;)
     
  6. Supa

    Supa Registered

    Joined:
    Dec 27, 2017
    Messages:
    151
    Likes Received:
    68
    Correct setting for DD wheels is to always run it at %100 strength in Wheel software to take advantage of all that fidelity. Makes it interesting when FFB multiplier is also at 1.00 ingame with a 26nm wheel. I usually have to run FFB multiplier as low as .40 for GT3 cars.
     
  7. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    @Supa that's the kind of situation the Steering Torque Capability is designed for.
     
  8. Supa

    Supa Registered

    Joined:
    Dec 27, 2017
    Messages:
    151
    Likes Received:
    68
    Yes but what happens to the FFB signal to the wheel when I set FFB multiplier to say .40 instead of leaving it at 1.00 as it should
     
  9. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    That feels like an open question with a low chance of picking the right topic; what do you think happens?

    (this is veering off-topic a bit, since manually selecting a mult of 0.40 in the game and letting the Steering Torque Capability do it for you will both end up doing the same thing at the wheel, but hey...)
     
  10. Supa

    Supa Registered

    Joined:
    Dec 27, 2017
    Messages:
    151
    Likes Received:
    68
    Well im picking it reduces all forces evenly across the board.....
     
  11. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    @Supa I'm not sure what you're getting at. I'm not even sure if you're advocating leaving everything on full strength or reducing it (manually, or automatically) to 0.40 when appropriate. Is reducing all forces good or bad in your opinion?
     
    Last edited: May 6, 2019
  12. Supa

    Supa Registered

    Joined:
    Dec 27, 2017
    Messages:
    151
    Likes Received:
    68
    Well I mentioned Wheel software always at %100 and just reduce ingame to suit ie for RF2 would be FFB multiplier. AMS is just a strength slider on top of Realfeel
     
  13. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    @Supa Yes, but again, that's what Steering Torque Capability does. If the car you're driving produces a nominal max of 15Nm, and you've told the game your wheel does 26Nm, it will use an FFB mult of 15/26 = 57.7% automatically. So you can save yourself a step if you set that parameter.
     
  14. Supa

    Supa Registered

    Joined:
    Dec 27, 2017
    Messages:
    151
    Likes Received:
    68
    Hmmm Ive always set steering torque capability to 26nm but still needed to reduce FFB multiplier ingame. Am I missing something else?
     
  15. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    Well, I guess there are a number of factors involved:
    • editing the right file
    • each car's nominal max set to a reasonable value
    • driving each car in a way that reveals its full range of force
    • being able to feel the difference
    You can check the first one by setting it to a large value like 200, and seeing if forces are much weaker.

    The second one you can probably cover by driving two very different official cars; perhaps the skip barber and the indycar. (skip has a max of 9.5, there might be other content with lower, I'm not sure. I imagine the indycar is nearing mid 20s.)
     
  16. Supa

    Supa Registered

    Joined:
    Dec 27, 2017
    Messages:
    151
    Likes Received:
    68
    Yep Skippy has reasonably weak FFB even with the DD wheel. Dallara and GT3s too much needing me to reduce multiplier.
     
  17. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    @Supa This is probably where we get back to the power-steering or no power-steering issue. The Dallara might be quite close to reality as it has no power steering. In the past I've seen FSAE data of close to 30Nm on turn-in, 11Nm mid-corner.

    The easy way to fix that would be to set the nominal max higher on cars with power steering, but it's something ISI and S397 seem unwilling to do. I don't know if they have some sort of power steering implementation in the plans for (much) later.
     
  18. davehenrie

    davehenrie Registered

    Joined:
    Jul 6, 2016
    Messages:
    7,454
    Likes Received:
    4,369
    rF1 often had cars with a Power Steering strength option in the upgrade area. (if I'm not scrambling my memories!)
     
  19. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    Yes, and you can do the same in rF2, using either the same steering system as rF1 or the newer rF2 one. It is just a simple rescaling of forces rather than a simulation of a power steering system, which is why I wonder if there's a reason they don't generally do it. One of the content releases even mentioned power steering at one point I think, though it might have just been in a post somewhere too (definitely was by staff though).

    For game content it would require an option or parameter, as with a 2.5Nm wheel you don't want even power-steering cars to have reduced FFB.

    Actually I've just answered part of my own question: you can't just ramp up the nominal max, as then the Capability parameter won't work properly anymore. So even if they decided to implement 'simple' power steering by scaling the existing forces (probably the sensible approach, though the hardcore sim nerds will complain it's not realistic) they'd need another parameter detailing that scaling, rather than using the existing one, so that the 'power steering' is only applied when appropriate.
     
  20. Supa

    Supa Registered

    Joined:
    Dec 27, 2017
    Messages:
    151
    Likes Received:
    68
    I checked my controller file and found my STC was only set to 16nm so I changed it to what it should have been 26nm. I then tested with wheel at %100 wheel software and game FFB multiplier at 1.00 and while it takes some effort the FFB in GT3 and the new Reiza stuff was awesome. Trouble is there are alot of Mods where they havent setup the FFB correctly so you cant use this setup for all RF2 stuff unfortunately.
    Going to be interesting doing a 40 min race in GT3s hehe

    Cheers for the helpful tips!! :)
     
    Lazza likes this.

Share This Page