Improving rF2 force feedback?

Discussion in 'General Discussion' started by cosimo, Nov 7, 2014.

  1. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    Excuse me but apparently I have too little tested it the last time and my memories misled me. I'm no longer sure about the uniqueness. Yes maybe it is something unique but not really convincing at all imho.
     
  2. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
  3. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    I already answered that but i have to correct it a bit, and add something to it, because after leaving both aside for a long time i tested AC and R3E again which i didn't since some updates. I have to say that R3E still doesn't works for me mainly because it has still this spongy and imprecise feeling build in, while AC did a much better job for me since i have tested it last. It at least works well with the street cars i've shortly tested after some basic ffb setup.

    AC was very easy to setup, it just need some steps to get it working, while R3E is a mess, even with all the possibility to fine tune it, it was impossible to feel good. AC setup is very easy, setting off spring/damper/self centering in the control panel of the steering wheel under windows, setting off all canned effects in the sim, setting the ffb strength first to 40% and go. The result a quite heavy steering wheel feeling without any clipping, it even can go higher without clipping, and it feels much stronger than rf2 as example which starts to clipping much earlyer, with lots of erratic behaviour on the steering wheel, which unfortunately is impossible to get rid of it just with the possibilitys the sim offers to the user with the UI, and lowering the car multi eliminates just the overall weight, while it eliminates other important information simultaneously as well. The result is a loosy overall steering resistance/wheight, which AC didn't generates at all, because it decreases and increases all the forces more consistent when adjusting the ffb strength in the UI.

    Conclussion: AC isn't that bad as some may think, at least when it comes to the feeling on the steering wheel, it may is even working better and is more user friendly out of the box than rf2 is in some cases. Unfortunately R3E isn't on the same level at this state in my opinion, while it offers really much good content.

    Just as objective comparison.
     
    Last edited by a moderator: Nov 15, 2014
  4. Connor Caple

    Connor Caple Registered

    Joined:
    Oct 4, 2010
    Messages:
    28
    Likes Received:
    0
    I find AC's FFb to be a little less satisfying, mainly because the wheel slip is so 'sudden'. It just lets go in the middle of a corner with little or no warning. I turned slip effects all the way up to get some feeling for what's going on there.

    R3E's FFB took a little setting up, but lets you vary how the slip effects will appear so you can tune in the feeling you like. It's my favourite at the moment.

    rF2 only works for me using the old 'Friction' settings with some serious tweaking in the config files to get some semblance of slip effects like rF1 would provide.

    That said, it's perfectly workable once you get it there and is my second favourite. I think my config needs 'tidying up' as I changed a lot of variables back and forth while trying to find the sweet spot, so I'll probably print it all off this week and do some desk work on it.

    Just my subjective comparison... ;)
     
  5. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
    Yup, rF1 style. Not as technically correct, or natural, or "raw", or as "nice to steer" as rF2, but in terms of finding those couple seconds or even tenths, then the ability to tune it allows the most freedom and customization due to have FFB characteristics connected to all sorts of things from the car, rather than just pure steering forces. I still suffer on braking and turn-in in rF2 relative to rF1 based sims because I can't feel all sorts of things happening with the car during the braking and turn-in phase.


    Example:
    I know that I can brake after the 100 m board for the final hairpin at historic Spa (historic F3 Eve), but because of the lack of specific feel, I have to brake just before it, again, even though I know that I and the car are capable of doing it after the 100m board. It also makes it more inconsistent when really trying to push; sometimes I lock brakes, sometimes I don't, sometimes I release the brake slightly too much (trying to ride the edge of lockup AKA "threshold braking" with practically no threshold feel at all) which can cause me to not slow down enough and run wide (even though I braked at the same spot as always). The driving just becomes inconsistent, and you're therefore always holding back from that little bit more that you know you and the car are capable of doing because when you try to do it, you lock up, or under-brake, or mess up the add-steering-more-and-more-as-you-release-the-brake-more-and-more phase (this can also massively influence lock-up and/or understeer).

    - historic F3 Eve
    - historic Spa
    - no aids, INCLUDING NO AUTO-CLUTCH. Auto-clutch will also enable auto-blip (unless auto-blip is set to "false" in your player.json) and can really aid the driver with corner-entry [braking and turn-in]. It's no different than using traction control, stability control or ABS, even if you don't have a clutch pedal (I don't ATM) - you don't need one, just blip and lift like normal.
    - around 5 laps of fuel
    - b880

    I can do low to mid 4:17s, with a current best of a 4:17.0 (no draft, of course), but because of these FFB issues, I cannot brake and turn-in how I know I and the car can. It's not even a matter of car-setup, but just strictly FFB during corner-entry not telling me tiny pieces of near-but-not-over-the-limit info. I know a low 4:16 is definitely possible for my skill/potential (if I can unleash it like I can with the rF1 FFB), but because of this lack of very specific and minute threshold-grip information, I seem to be stuck with low 4:17s. Also, when pushing for very low "17s", I can be quite inconsistent with brake lockup and/or running wide, again, because of the FFB.


    With rF1-style FFB, though, no such problems. However, most people who just pick-up the game to play (including my friends) say that the rF2 FFB ABSOLUTELY DESTROYS the rF1 style, again, because it's so much more raw, natural, realistic. It really is bloody brilliant, I agree! I don't even make newcomers, or friends, or whoever, play rF1 anymore because of rF2's insanely good FFB, but in terms of pure competitiveness and raw laptimes - once a player becomes used to the sim and learns to understand how to use the rF1 FFB - then I personally prefer a properly tuned rF1 style, even though it's more boring, less raw, less natural, less overall amazing. Why? Because it (rF1-style) allows me to be more competitive and achieve closer to me and my car's true potential (ultimate laptime, and consistency when pushing hard), and that is ultimately really what makes me happy and enjoy the overall driving/racing experience.

    I would never want ISI to ditch the current system, it is the technically "proper" way, and 95% of people (heck probably more) love it a million times more than rF1. All I'm asking for (or wishing for) is a second "style" for the few who understand what I'm talking about (regarding the rF1 style FFB).
     
    Last edited by a moderator: Nov 15, 2014
  6. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    It was fundamentally not wrong, with respect to the vehicle dynamics it was just a different way to provide the user an understanding about. It was a sort or combination of individual results of the vehicle dynamics transmitted to the steering wheel.

    rf2 tried to transmitt this results bundled to the steering shaft which unfortunately does not really work if the physic is not working completely and provides a physical result which is similar to the real physics, and whose output to the steering column corresponds to the real thing, and apart from missing ponents which are necessary to this implementation, such as the steering box, unfortunately it fails at the balance in respect to the hardware, not least the suboptimal tire surface which contributes negative to it.

    I did not want to talk about it this way but the dreamers which more and more pop up the last time and talk things just beautiful cause that I can not restrain myself anymore. No dreamers fanboys and selfish people with no clue out there it is by far not perfect or real.

    Your s****t doesn't helps anybody really, neither ISI nor intrested user or newcomer.
     
  7. TechAde

    TechAde Registered

    Joined:
    Oct 13, 2010
    Messages:
    606
    Likes Received:
    38
    Yes it was.

    For example, with native rF1 FFB increasing caster has no effect on steering weight.

    With rF1+RealFeel or rF2 increasing caster does increase steering weight, as it does in real life.

    Sent from my SM-G900F using Tapatalk
     
  8. I3bullets

    I3bullets Registered

    Joined:
    Jun 8, 2013
    Messages:
    99
    Likes Received:
    0
    Hey Paul,
    I just tried your settings with a Fanatec CSW. The effects are inverted though. Which settings do I need to change? Tried removing the "minus" in the line "Steering effects strength":-10000, but to no avail.

    Edit: Okay, further testing revealed: I had to actually put a "+" so it now reads "Steering effects strength":+10000,
     
    Last edited by a moderator: Nov 15, 2014
  9. speed1

    speed1 Banned

    Joined:
    Jul 26, 2012
    Messages:
    1,858
    Likes Received:
    0
    True that, of course i mean as was tried to convey the emotion, and I was out of the biz before the real feel was published.
     
  10. I3bullets

    I3bullets Registered

    Joined:
    Jun 8, 2013
    Messages:
    99
    Likes Received:
    0
    Didn'T have enough time for testing. Will do so tomorrow. Thanks for your hints which will prove helpful, I'm sure :)
     
  11. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
    Exactly, rF1 without realfeel doesn't. But some people don't care about feeling caster in the wheel. I change caster for car vehicle dynamics (behviour) changes, I couldn't care less if I feel that change in the wheel. Sure it's cool to feel the difference between 7 and 9 degrees of caster, but this does absolutely "squat" for me when it comes to feeling the differences between being 2% slip away from lock-up and 2% slip past/into lockup (just an example, obviously it's more complex than that). Not everyone is looking for a steering rack feel, I used to race real cars, and I, and many others (as the people who like the R3E prove) need a bunch of info, cues, effects (whatever you want to call them) which the rF1-style FFB is capable of sending to the FFB from certain car-physics parameters which are setup in your controller file.

    I just wish it was an option for those who love that stuff. It really makes a massive difference to some.
     
  12. Jamie Shorting

    Jamie Shorting Registered

    Joined:
    Sep 11, 2013
    Messages:
    2,628
    Likes Received:
    3
    Caster has a camber gain effect as well. It also effects your initial turn in response.


    When the Camaro first came out the first thing I did with caster was reduce it to less than 3.0. I hated that super heavy ffb. Now it has changed though. The default was 5 which to me was unusual for such a heavy car.
     
  13. TechAde

    TechAde Registered

    Joined:
    Oct 13, 2010
    Messages:
    606
    Likes Received:
    38
    You are so good at missing my point I'm starting to wonder if you do it on purpose?

    Caster was just one example of how rF1 FFB was disconnected from the car physics.

    Same with Leo FFB, it's output is defined by ini parameters rather than outputs from the tyre, chassis & steering simulation.

    That's fine for a game but not for a simulation.

    Anyway, we know we disagree on FFB implementations so let's not get into yet another argument, been there, done that, worn out the T shirt.

    Do you remember how popular RealFeel was when released, were you around then?

    Sent from my SM-G900F using Tapatalk
     
  14. Connor Caple

    Connor Caple Registered

    Joined:
    Oct 4, 2010
    Messages:
    28
    Likes Received:
    0
    This. Exactly this. It's not a a 'sim' unless it can replace some of those feelings for us with something useful, which is what 'slip' feedback is all about and what is lacking in the standard rF2 setup. I get no inner ear/motion feedback so I need to feel it somehow at the steering wheel because that's the only FFB we get.

    It's the ONLY reason I prefer R3E at the moment.
     
  15. Spinelli

    Spinelli Banned

    Joined:
    Jan 28, 2012
    Messages:
    5,290
    Likes Received:
    32
    I wasn't disagreeing with you, quite the opposite in fact. I was pointing out how much more realistic the pure-steering feel, and how it's implemented into the steering rack/forces, is in rF2. Then I was just going on to make a general point that for some, they prefer the way the rF1 FFB works (once tuned), and that some people - lately - stating that they can feel more of the car with R3E (rF1-style FFB) shows what I'm saying even further.

    Basically, I'm just saying that others would like some of the rF1 qualities in order to feel and interpret car physics in ways that can't be done by just pure steering rack forces. And yes, as you say, caster was just an example. :) . If it wasn't for the rF1 style FFB, I would never have been able to do top 3-5 GTL Rank laptimes, and same with rF1 Rank. And when my specific FFB settings in those sims changed (experimenting with different FFB settings, too much realfeel overpowering the rF1 forces, etc.) then I also wouldn't be able to achieve those same top 3-5 GTL Rank and rF1 Rank. This says a lot about how extremely particular and specific FFB can affect different people.

    Yes, I was around even before RealFeel was released. And realfeel made feeling the steering rack forces much better and closer to rF2, however, (like I've been trying to say for a while now), RealFeel didn't do squat for me in terms of pure laptime gaining info. It just made the steering feel more natural. Sometimes I used realfeel and sometimes I didn't. When I did use it though, I did not disable all the rF1-style FFB like people advised (most advised to disable all the rF1 stuff in order to have an as clean and pure, steering-rack only feel, and only use "low" FFB). I kept the rF1 FFB at either "high" or "full" (depending if I wanted to keep the engine rev vibrations enabled - which could be tuned in order to feel threshold braking 100x better than anything possible in rF2, or just about any sim), and kept my rF1 lines of code in the controller file activated. Then with that, I would introduce RealFeel at around, 50-85% depending on how much RealFeel I could add before loosing out on the original and vital-to-me rF1 grip/slip-threshold effects.

    Again, not arguing, just saying, different preferences for different people. I can make up for a less natural wheel and raw steering feel as long as I get the FFB cues that I prefer regarding specific grip/slip threshold FFB information cues, however, some people may not like those cues and how they relay car physics/behavior, they might not be able to understand (or just not like) how those cues feel and how they relay car-bahviour, and would prefer something much more pure and natural like rF2 (and mostly pure RealFeel for rF1 based stuff). Just saying, different preferences, that's all :)



    Yup. And if it wasn't for this, I would delete rF1, GT Legends, and even R3E from my PC, because their physics aren't as good as rF2 (not even close). But when you can feel a car so much, then that almost makes up for inferior physics because you can drive the car (regardless of how good or bad the physics are) so much better, faster, and consistent - which can equal more satisfaction and enjoyment.
     
    Last edited by a moderator: Nov 16, 2014
  16. David O'Reilly

    David O'Reilly Registered

    Joined:
    Jun 12, 2012
    Messages:
    1,050
    Likes Received:
    756
    Cosimo,
    I will just add the the FFB in RF2 varies a lot from car to car. You mentioned the FIA F2 which is very poor and no caster adjustment to boost it.
    IMO the URD is not "best practice" either. I know you raced the URD recently.
    Have you tried for example the ISI GT cars?
     
  17. Ari Antero

    Ari Antero Registered

    Joined:
    Jul 27, 2012
    Messages:
    1,882
    Likes Received:
    829
    I agree with you and for me in rFactor2, R3E and RealFeel are driving helps similar to TC, ABS etc. and should be penalized. What I mean by this is that if we are racing Nissan 370z GT4 having ABS in real life no penalty, but if TC or any other driving help is used should be penalized, no gain when driving helps are used, "That's fine for a game but not for a simulation."

    Racing Nissan 370z GT4 in real life feel of steering (FFB) is same to everybody and to have it any other way (RealFeel, R3E) "That's fine for a game but not for a simulation." :)
     
  18. Jyyka Pihtari

    Jyyka Pihtari Registered

    Joined:
    Aug 17, 2014
    Messages:
    71
    Likes Received:
    18
    How do you simulate other real life forces then, without a state-of-art motion rig?
     
  19. Ari Antero

    Ari Antero Registered

    Joined:
    Jul 27, 2012
    Messages:
    1,882
    Likes Received:
    829
    G-forces is hard to simulate if no motion rig, but track IR, ButKicker, Real Head Motion plugging, 3D vision and of cause triple screen to get as close 1:1 as possible can be helpful.:)
     
  20. kaptainkremmen

    kaptainkremmen Registered

    Joined:
    Sep 25, 2012
    Messages:
    935
    Likes Received:
    17

Share This Page