Mantasisg's wheel rotation

Discussion in 'Technical & Support' started by Lazza, Dec 9, 2021.

  1. Corti

    Corti Registered

    Joined:
    Sep 29, 2014
    Messages:
    1,354
    Likes Received:
    2,495
    New driver & firmware
    269731624_4455952137867768_770166297592802632_n.png
     
  2. pilAUTO

    pilAUTO Registered

    Joined:
    Nov 7, 2020
    Messages:
    665
    Likes Received:
    593
    Thx corti for the info.
    But I'm a lot sceptical.
    Perhaps road effects added are a little arcadish.
    I have realistic settings on the control panel + rF2........ I fear to decrease the FFB realism with new drivers.
     
  3. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    It's a single option, and by default it's off:

    upload_2021-12-22_10-8-18.png

    I also think rF2 doesn't output anything that would contribute to this doing anything, unless it's a DirectX thing or the driver is actually trying to do something different (but for me, "Constant + Spring" says not).

    Hard to tell without blind testing or more details, mind you.


    While I'm here: @mantasisg in another thread you posted a screenshot of your TM software which had blank buttons on it. I got the same when I updated my software through the control panel. Uninstalling ("FFB Racing Wheel drivers") and reinstalling manually via download (with the wheel unplugged the whole time, including reboots) fixed it.

    *(oops, it was this thread - forgot where I was :D)
     
    mantasisg and Corti like this.
  4. pilAUTO

    pilAUTO Registered

    Joined:
    Nov 7, 2020
    Messages:
    665
    Likes Received:
    593
    Thank you for your advice.

    I will not test this setting personally.

    My setting exploits the T300 perfectly in a sense and with a "credible and realistic FFB" objective, and if I want better and more realistic, well I would buy myself an overpriced DD, but not by changing these precious settings harshly found.
     
    Corti likes this.
  5. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    Hi, got back to dabbling with rF2. Starting withthis issue. First of all, maybe you had faster rotation in game because you had selected software as "Rotation Limit Mode". I have found it to be true for me. If rotation limit mode by driver makes steering mismatch to be too slow, software makes it too fast.

    Secondly. Interestingly, I have decided to try Skipbarber, a car that for sure has all these "up to date" hdv steering entries. And when limit mode is by the driver steering is close enough, but there still is something like 10 or 15 of degs mismatch at larger steering angle. But whats more interesting, with "software" option as rotation limit mode, steering is insanely too fast, it becomes incredibly off.

    IDK, but maybe there is some sort of rF2 error, bug...

    I checked my device driver it is up to date.
     
  6. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    I get perfect correlation with Software and 1105, as I said (my wheel itself is set to 1080). The Dissenter doesn't, so has an issue.

    I set 1105 instead of 1080 as that corrects the slight overrotation the wheel itself does. rF2 is doing nothing wrong.

    I think you have something set wrong.
     
  7. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    @mantasisg back here, again, because judging by your RC release thread post I guess you still have problems with your rotation.

    I'm yet to find someone who can't get their rotation right with the correct settings, or identifies that a particular car has an issue and the majority are fine.

    From your post above you indicated the rotation is 10-15 degrees off in device mode, while software mode makes it completely wrong.

    • As I said above, I run software mode on my T500 to correct for the overrotation the wheel does (25 degrees extra, so 12.5 each way - likely correlates to your observed error)
    • If your rotation is massively off on Software, you either have a setting wrong (let's say >50% chance) or something on your end isn't working right. (Driver, or something within rF2).
    I'd put the likelihood of an actual rF2 bug (not a wrong setting) at less than 0.1%. I simply haven't seen anything unresolvable except for rare wheels - which the T300 certainly isn't.

    Try software again, and screenshot your settings (TM settings, and rF2 calibration page). If those look fine then we can aim for broken files or something.
     
    mantasisg likes this.
  8. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    @Lazza Thank you. I restored solving this yesterday. I am thinking about reinstalling rF2, but don't want to do it.

    Yesterday after assigning to release candidate I tried to mix various settings with Dissenter, and it WORKED ! Then switched to one of my cars and it didn't work, then came back to Howston Dissenter and it didn't work anymore. Then I also tried variety of different combinations of switching physics to find where the problem originates, but couldn't detect.

    I'll try to update info, will probably make screenshots. I think the issue could be some weird mix of certain cars physics, steerign settign and device driver. Maybe it is indeed only I who has it, but I actually heard fro ma few other guys thatthey might have this issue while I was participating in HRRC league.
     
  9. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    The Dissenter has issues. Don't bother testing on it. Same for your own cars (for now) because until you're sure your settings are right you won't know if the cars are the problem.

    As a car I'm familiar with I usually advise to test on the Formula ISI 2012. If I recall correctly I've also found the Skip Barber to work fine. Can check for a standard tintop that works if you want.

    Reduce the variables to 1: your game/settings. Use a car that's known good.

    Reinstalling rF2 is a complete waste of time, in basically all situations, and especially this one.


    It's really not difficult, but you need to be methodical and specific.


    **Edit: Oops, so the FISI2012 isn't great either, at least on the RC I'm running. Skip is good, but not a large range
     
    Last edited: Jan 30, 2023
  10. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    Yes, I am a bit struggling to get hold on some rational and useful method to troubleshoot this silly issue. For a year lol Few times I almost uninstalled rF2.

    Today spent some time on it. Speaking of my own cars, at least 1967 endurance pack, it appears like them all seem to default at ~490 degs of steering rotation in game. Any different steering wheel rotation range causes mismatch. I can't find reason for this. And this is what I care the most about - to get my cars to be good.

    I checked Datsun today it seemed to be slightly off too, but with maximum steering wheel range it appeared to be good. Various other cars were good, for example the good old NSX. Yes Skipbarber seems fine. Aston Martin GT3 - good. I agree F ISI 2012 would be good subject.

    I have no ideas currently what could default steering to 490degs, and it seems to be global per whole pack of cars in the mod. In game and in devmode too. And it is physical, for example 911s has 990degrees and game understand that, it is written in settings, and setup, but physically game wheel rotates ~490degs.
     
  11. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    I just checked the FISI2012 and I found it wasn't quite right. So ignore that advice.

    The skip barber is good but only has 409°. Need to find a good tintop with lots of rotation to check against. I don't have time right now.

    You need to check and fix your settings in a car that works, before worrying about your own cars. There's simply no other way to approach it.
     
  12. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    @mantasisg well, annoyingly but probably not surprisingly, there aren't many recent cars with higher amounts of rotation.

    So at this stage I'd recommend 2 cars that operate correctly, in order to verify your settings:
    • Skip barber
    • AC 427 (has 900° by default)
    Both those match my wheel perfectly with my settings (Software, Custom, 1105°, set by vehicle)

    I should mention that I also tested the Ferrari GTE and Vanwall Vandervell, just to make sure there wasn't an old car vs new car thing going on, and my wheel is perfect in those too.

    If I run in device mode, the steering will be slightly off. That's because my wheel physically rotates past 540° in each direction, and keeps registering movement. In theory at full lock it should be exactly upside down, but at full lock (and 100% reported movement) it's actually a bit past that point. I imagine there's a good chance the T300 is similar.


    So: start there. Just check those cars. When you know your settings are right, then maybe we can speculate on what's wrong with specific cars.
     
  13. dadaboomda

    dadaboomda Registered

    Joined:
    May 15, 2016
    Messages:
    712
    Likes Received:
    202
    I confirm that for some cars, there is a total decorrelation between the real and virtual steering wheel.

    With a well-tuned T300 and with a well-tuned CSL DD.

    With rFactor 2 reinstalled cleanly.

    With all of the official S397 cars whose steering wheel angle does not exceed 720 degrees without any problems, so almost all of them.

    The latest Austin mini and Caterham malfunction because of a higher steering wheel angle, these are official and extremely recent cars.

    Cars 67 and 54 (with extremely convincing physics bravo Mantas) are concerned but not all of them I think.

    It is only a visual problem but very annoying in VR, fortunately I believe that the angle of the FFB steering wheel always corresponds to a correct rotation of the wheels (I BELIEVE).

    To be honest I try not to pay attention to it, I've already talked about it several times over the last few years on the forum without it seeming to interest the developers (example : https://forum.studio-397.com/index....-1-now-available-21-10-22.72870/#post-1110757), and in the meantime I've reinstalled rFactor 2 several times and had 2 wheels of different brands, I didn't create a bug report, it's true, but I don't have to do that to be honest.

    I don't intend to invest myself in a bug fix attempt, but I wanted to say that Mantas is 100% right.

    The problem does not come from the user but from the game (I doubt that a guy capable of making cars with such convincing physics made a mistake).
     
    mantasisg likes this.
  14. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    @dadaboomda Thank you, your response is helping me a lot. Although Lazza is right that at first we need to makes sure that there are no problems with settings. However, I am failing to find how settings influence this issue, but it seems like they indeed does. Unfortunately I can't find any consistency.

    TL;DR I don't blame you if you don't read, it is confusing as hell. Long story short - game loads the settings in weird ways, and rotations in most cases can be fixed, or errors reduced, but it is just not how it all should work anyway.

    I just done some testing in game and I was making tiny table on several cars how their steering isn't matching and how it correlates to settings in game. I was pickign Howston G4, AC Cobra, NSX-R. All had mismatch with these settings:
    • Rotation limit mode: software
    • Maximum wheel angle mode: custom
    • Default max wheel angle: 1080
    • Range set by vehicle: on
    NSX-R and Cobra had slower steering in my real hands than in game hands: 360 degrees to ~390 degrees in game. Both cars max steering wheel rotation angle being 900 degrees. And these cars were reaching full steerign lock. Howston steering in game turned ~135 degress, about 45degrees slower than in my hands at 180 degrees.

    Then I decided to test with different settings, and basically I was only alternating Rotation Limit Mode (RLM) and Maximum wheel angle mode (MWAM). First I chose NSX-R. I haven't changed settings yet, and noticed that this time steering is lot more off than previous time. Then I changed RLM and MWAM to "driver" and "automatic". And once I changed MWAM to automatic it showed me 630 degs, which were degs of previous car (Howston G4). Then I realized once I'll load NSX-R in track it will probably be good, and it was good steering degrees were matching pretty much precisely.

    I left these settings and decided to try Howston again, and actually I forgot the whole sequence exactly from there, but it IIRIC Howston again was off, and after alternating RLM and MWAM I got Howston steering "almost" precise, it still was off but instead of 45degs at 180degs rotation, it was now off by maybe 20-15degrees. So it still was off anyway, but somehow apparently it is possible to minimise the error by some shamanistic manipulation of rF2 spirits.

    Then I also tried Caterham Academy, and it was again way off, and after alternating RLM MWAM back and forth it was good. Then same happened with Infinity BTCC car.

    There didn't seem to be much consistency or clear sequence to predict exactly how the car is going to be loaded in game. Pelase notice also how often cars are loaded in game with no FFB (fortunately reset FFB button works every time to call it back). It is just my speculation as I am not a programmer and not a developer (only they can see). But to me it seems like ever since rF2 was moved to new UI, there might be some complex issue regardign to how FFB and steering stuff is initiated while loading up the game, most likely when loading up game into the racetrack/garage. Maybe games default to something if it fails to communicate with driver somehow or what - I could never possibly know, and more importantly it is not my business, my business is to either enjoy simulation or not use it. I am not sure how this issue really relates to cars with larger steering wheel rotation angle and if it really has a meaning for this problem, but maybe it somehow could take part. At the very early stage a year ago I thought that these issues are down to cars not ussing all those later steering physics entries in hdv, like steering shaft coords, steering inner table, "better" steering lock special entries, but after playing around in dev mode it seems to not fix anything, well although devmode runs on old UI, therefore making whole issue even more confusing to mere mortal idiot like me.
     
  15. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    So in there somewhere you did test the Cobra, thank you. And it was wrong. So, simply, something isn't working right for you.

    Can you share your TM settings, controller.json, player.json, and a trace file, from running the AC? Settings as you mentioned: Software, Custom, 1080, vehicle set.
     
  16. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    No it is not simply something on my end. Numerous people already said that they have their steerign off with these certain cars, and you yourself said that Howston is broken somehow. So how is it broken ? Yes AC steering was wrong, but it was also correct, depending on settings although inconsistently. So it does mean that theres is something up besides settings, but it seems to be related to settings. Thankfully AC somehow can be corrected, Howston Dissenter also seems to be somehow managable, but Howston G4 is off, and more than a year ago there never were such issues.

    I just did 10 separate launches during which I was manipulating settings in various ways, but mainly just changing RLMsoftware+MWAMcustom@1080/RLMdevicedriver+MWAMautomatic. I used Cubra, Howston G4, Infinity, Caterham, NSX-R. And there seemed to be no predictability or consistency. Cars either launched with precise steering right on, or it was possible to correct by alternating RLM+MWAM. The only consistent thing was that Howston G4 simply always had its steering wrong, and due to having massive amount of modded physics based on this car it is pretty much only one I really care at the moment, well this and Cobra.

    You, yourself, said before that Howston is broken. Why is it broken ? I can't find why. Could it be that somehow it is not broken, but the way game interacts with it is broken ?

    I can write whole sequence of my todays post (I was making notes), also could post my json files, device controller settings, I just forgot to generate trace file, but it is probably not worth it anymore.

    Also the issue when FFB needs to be reset very frequently is annoying, and it does seem like it happens only with RLM software/MWAM custom.
     
  17. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    How do you solve any problems with such a slapdash approach???

    Yes, some, possibly many, cars don't work right. Many do, consistently. If you can't get any car to work consistently there is something wrong on your end, but testing 12000 different things won't help identify it.

    In this thread that I created for you 14 months ago you've posted 2 screenshots, 1 of your broken TM software, 1 of the old UI and talking about your own car. No files.

    If you don't want help solving it, fine. But please don't post in RC/release threads asking if a bug that you can't identify has been fixed.
     
  18. pilAUTO

    pilAUTO Registered

    Joined:
    Nov 7, 2020
    Messages:
    665
    Likes Received:
    593
    I think the problem affects all users, because as I said, I reinstalled rFactor 2 cleanly several times (for other reasons), installed cleanly with data wipe from windows 10 to windows 11, and I had the T300 and currently have the CSL DD.

    I also think I have realistic FFB settings for my old T300 and CSL DD.

    As I said I have neither the desire nor the time to do tests, to post screenshots etcetera.

    I consider that it is up to the publisher to make these efforts, especially since it has the competence to do so.

    One thing you can do @Lazza if you have the time and inclination is to test the Austin Mini and the Caterham (S397) WITHOUT CHANGING THE DEFAULT STEERING ANGLE (high angles) and without trying to tweak anything it would be.

    If it doesn't work for you either, there is definitely a serious problem with AND/OR the game/cars.

    Lots of cars have this problem, so we can think it's the game that's at fault, not just the cars (not an affirmation).

    By the way don't be surprised, I typed my first message by mistake with a very old nickname (dadaboomda).
     
  19. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,926
    Likes Received:
    3,876
    Well to begin with, problem begun with updates of rF2 without me changing anything. Second, these problems (mainly with cars in question being 1967 endurance mod which originates from Howston G4, which pretty much are ones which I only care about right now) appears not to be only happening to me.

    To be honest I am not too much motivated anymore, and I am almost ready to just step off in peace. Maybe once I share my jsons and trace files you'd be able to offer a solution and it would work. We'll see, but I doubt. Nothing is wrong with my wheel controller settings and game steering settings. I don't think that Json files are relevant here too, but trace file perhaps could offer something. It is the only and last thing that I am willing to do.

    I am adding trace file, I just launched rF2 and entered track with Howston G4, rotated steerign a bit and left. I can't see anything wrong in that file, but I am not a developer.
     

    Attached Files:

  20. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    I know many cars have a problem. I said so myself. I know the Caterham is wrong, I said so in the Caterham thread, and guessed what might be wrong. This isn't about "all cars work fine". It's about "no cars work fine for mantasisg."


    *Just to remind: I said earlier to test the AC 427 and the Skip Barber. That's it.
     

Share This Page