Page 8 of 8 FirstFirst ... 678
Results 71 to 79 of 79

Thread: A quasi-objective way to judge physics of sim cars

  1. #71


    Same slide but in car view.

  2. #72
    Member
    Join Date
    Jan 2012
    Location
    France
    Posts
    89
    Try in the inner circle, and do little lefts-rights to have a better control.
    Try to find what inputs variations change the behavior.

  3. #73
    'Little lefts-rights', create huge left-right spins lol. I know what you mean, but in a slide like this once you get it going it's pretty hard to control the line of the car, not impossible but very hard.

  4. #74
    At 11 seconds (steady steering input), a slight left turn of the wheel would reduce the slip angle and increases turning radius. Turning right (more counter steer) would increase the slip angle and tighten the turning radius. Throttle inputs need to be spot on in both cases. It's possible to reduce/increase slip angle while maintaining the same turning radius, but its very difficult.

    Edit: initially during a slide turning the wheel left (away from the direction of the slide) would make the car go left, but eventually it would end up on a larger turning radius. Speed also plays a huge roll in the direction you go. If you give it more throttle and counter steer and you do it perfectly, it can increase your speed and widen your arc. I'm not sure how realistic any of this is because it seems overly hard and complicated. It's just what I've found.

  5. #75
    Member
    Join Date
    Feb 2012
    Location
    Florida
    Posts
    55
    Quote Originally Posted by green serpent View Post
    At 11 seconds (steady steering input), a slight left turn of the wheel would reduce the slip angle and increases turning radius. Turning right (more counter steer) would increase the slip angle and tighten the turning radius. Throttle inputs need to be spot on in both cases. It's possible to reduce/increase slip angle while maintaining the same turning radius, but its very difficult.

    Edit: initially during a slide turning the wheel left (away from the direction of the slide) would make the car go left, but eventually it would end up on a larger turning radius. Speed also plays a huge roll in the direction you go. If you give it more throttle and counter steer and you do it perfectly, it can increase your speed and widen your arc. I'm not sure how realistic any of this is because it seems overly hard and complicated. It's just what I've found.
    It's very realistic based on my own experiences myself with my wheel.

    I honestly don't see the fundamental difference between this and LFS, the only real difference i can see is that in LFS the front end is much more responsive during drifts, so there's not as much of a delay for your steering inputs to affect the car as it does in rF2, and in LFS all of the cars seem to be able to handle massive amounts of slip angles without spinning. I can see why it would be more enjoyable in LFS, since the cars always obey your steering inputs and you basically don't have to worry about spinning, but i don't think that's very realistic, it's not as if the front tires can't ever lose traction, or that any car can hold a huge slide continuously as if they're all purpose built drift cars. In the real world, only cars purpose built for drifting can do the sort of things that you can do with any old car in LFS. The steering response delay in rF2 could be partially due to the tire models as well, but i think a large part of that behavior is very realistic. I think front tire response will probably get a bit better after the tires are updated, but probably not to the level of LFS, and most of the cars should never be able to handle such huge angles of drifting.

    rF2 just seems like it portrays momentum and inertia in a more realistic way, so i can easily imagine if someone has inaccurate FFB that they would most likely have a very difficult time getting into the correct margins to even begin to do continuous drifting in the first place. It all has to be premeditated, you're never going to get realistic seat-of-the-pants feel and generally won't visually see the car react to the steering as dramatically as LFS, so if you can't really feel what the car is doing through the steering in rF2, it will seem as though it's not responding at all or responding unrealistically.

    Back to the actual topic in rF2.

    If the car has too much momentum into a slide, then both increasing or decreasing the counter steer won't help, it will just continue to rotate. I believe this is the behavior your experiencing, and because you probably can't feel the car through the sim very well, you're probably thinking that it should straighten out and decrease the angle of drift, when in reality you're actually not even within the margins to be able to drift the car continuously, thus your act of increasing the counter steer has no real affect, and since you're doing that at the same time that the car is essentially starting to spin, you put 2 and 2 together and get 7. Like i said before, i think this is a simple issue of mistaking correlation for causation, combined with the fact that you need very realistic FFB to drift these cars around like this. It's one thing to do some fast laps with whatever hardware and FFB setup, but to be able to hold slides accurately and to control the car in extreme situations you need some way to know exactly what the car is doing at any given moment and to even take control of the steering wheel at times for very big, quick corrections. The visuals, sounds and the lack of any seat-of-the-pants feel are simply not enough for that, so a good amount of realistic steering feedback is the best way to get that information from the car, but more importantly it's so you get the accurate behaviors, since if you're wheel isn't fast enough to do what the front wheels should do in real life, then it's obviously not realistic behavior of the front wheels.

    Whether the margins for drifting are realistic, or whether the weight distribution is realistic, or whether the traction on the tires are realistic, etc, are all completely different issues related to the car itself and not the underlying physics, these are things i can't know since i've never driven this car. However since i've driven various cars of a similar type in the real world, i can get a decent idea what this would feel like, and i can easily imagine this car existing in the real world and behaving exactly the same way as it does in rF2. Again, i have no idea if this car should be easier or even more difficult to drift, but it is definitely possible and it has to be done in a realistic way too, you can't just force the car to do it like LFS. It's actually not very difficult when you get a feel for it and have good enough hardware and FFB settings.

    I just took the ZR1 to the skid pad track and found it relatively easy to get a big drift going after i got a feel for the car. You have to be within a good range of both the turning radius and speed where the car can drift, but once you're there it behaves just as it should in real life.

    Here's a video of the longest continuous drift i held in the ZR1, i think i did at least a couple of full rotations:



    When i want to correct for more angle, i apply more counter steer. When i want to correct for less angle, i use less counter steer. If i want to apply more angle in the drift then i use less counter steer, not more, using more would mean that you would be correcting too much of the slide and would straighten the car out. Likewise, if i want to apply less angle in the drift, then i simply apply more counter steer to correct for the extra sliding until the car has less of an angle, then i bring back the steering a bit until it's balanced in the drift. Drifting is mostly about being able to balance the car in this way with the controls, and doing it full throttle means you're almost entirely manipulating it with the steering. Again, all of this has to be done within the limits of where the car can drift, it's very much possible to go too fast and to have too much angle in the drift, and just like an actual car, you will end up losing control one way or another in that scenario.
    Realistic FFB settings with Accuforce (responsive mode): Accuforce settings

  6. #76
    It's interesting to compare the two vids. In my video, my throttle inputs aren't very steady. I'm trying to find the balancing point and I'm going up and down with the throttle, and as a result look how the is car wobbling around mid slide. Paul's throttle inputs are smooth and within a much finer margin, the car is balanced, and if you look at it from the outside it look as solid as a rock.

    Obviously it takes lots of skill, but the better the hardware, the more accurate feed back you get and the finer you can modulate you're inputs.

    Most of the time when I try to drift around a track, I've come to realize that the rear tires are either accelerating or decelerating (i.e gripping a little and not spinning freely). I think this causes the the front of the car to push and the fronts to scrub. I think it causes the 'nose of the car toward the inside of the track', and the 'more counter steer spins the car out' abnormalities that have been discussed. It's quite hard to get the tires spinning freely at the right rate, because there isn't much sensation through the pedals (none) and no sensation through the wheel. You can go by sound, but I think a lot of the time when you think you're doing a burnout, really the rear tires are gripping longitudinally a fair bit.

    If you can find the exact point were the rear tires are spinning nicely, and then hold that point, the counter steering should feel and respond naturally. I'm with Paul in thinking that imprecise ffb wheels makes it much harder, and combined with difficulty in holding a steady burnout, I think this is what is causing the strange car behavior and making holding a car sideways feel so hard.

    Spinelli, you've said that you believe that there might be an underlying issue with the isimotor physics in the way that it models inertia and that the inertia mysteriously changes direction toward the inside of the track. This might well be the case, and I can entirely see your point because of the way the cars behave in a slide, but something like inertia should be pretty well known math (or maybe not, I don't really know), however, tires are extremely complex so logically I lean toward the tire argument. I bet if you went into the physics model and you reduced rear tire longitudinal grip, and increased front tire lateral grip, you could slide around much more predictably (like LFS). The point is I don't think it's to do with inertia, I think its tires.

  7. #77
    Member
    Join Date
    Jan 2012
    Location
    Salento (I)
    Posts
    61
    Quote Originally Posted by green serpent View Post
    ...
    Most of the time when I try to drift around a track, I've come to realize that the rear tires are either accelerating or decelerating (i.e gripping a little and not spinning freely). I think this causes the the front of the car to push and the fronts to scrub.... but I think a lot of the time when you think you're doing a burnout, really the rear tires are gripping longitudinally a fair bit.

    If you can find the exact point were the rear tires are spinning nicely, and then hold that point, the counter steering should feel and respond naturally. ...

    ... I bet if you went into the physics model and you reduced rear tire longitudinal grip, and increased front tire lateral grip, you could slide around much more predictably (like LFS). The point is I don't think it's to do with inertia, I think its tires.
    I was thinking EXACTLY the same: too much longitudinal grip while spinning.
    And it must be true because the cars are always accelerating, even when in a "perfect" slide.
    Chris Harris cars don't accelerate that much...

  8. #78
    Member
    Join Date
    Feb 2012
    Location
    Florida
    Posts
    55
    Quote Originally Posted by green serpent View Post
    It's interesting to compare the two vids. In my video, my throttle inputs aren't very steady. I'm trying to find the balancing point and I'm going up and down with the throttle, and as a result look how the is car wobbling around mid slide. Paul's throttle inputs are smooth and within a much finer margin, the car is balanced, and if you look at it from the outside it look as solid as a rock.

    Obviously it takes lots of skill, but the better the hardware, the more accurate feed back you get and the finer you can modulate you're inputs.

    Most of the time when I try to drift around a track, I've come to realize that the rear tires are either accelerating or decelerating (i.e gripping a little and not spinning freely). I think this causes the the front of the car to push and the fronts to scrub. I think it causes the 'nose of the car toward the inside of the track', and the 'more counter steer spins the car out' abnormalities that have been discussed. It's quite hard to get the tires spinning freely at the right rate, because there isn't much sensation through the pedals (none) and no sensation through the wheel. You can go by sound, but I think a lot of the time when you think you're doing a burnout, really the rear tires are gripping longitudinally a fair bit.

    If you can find the exact point were the rear tires are spinning nicely, and then hold that point, the counter steering should feel and respond naturally. I'm with Paul in thinking that imprecise ffb wheels makes it much harder, and combined with difficulty in holding a steady burnout, I think this is what is causing the strange car behavior and making holding a car sideways feel so hard.
    I agree with this assessment. I've previously said that i think there's too much lateral grip, rather than longitudinal like you're saying. My thinking for this is that the cars accelerate forwards more than you would expect because the excess lateral grip forces the car to go the only other way, which is longitudinally, basically to push forwards more than slide sideways. But, i think you're essentially saying the same thing in the end, which is that the balance between lateral and longitudinal grip is somewhat strange which ultimately causes the car to push forwards much easier than you would expect in real life.

    I'm not sure exactly what's wrong with the tires, but obviously the tire models are a very complex thing and even ISI have said that they don't have that aspect working properly yet, and we also know that the latest cars have more accurate tire modeling than some of the older cars, which to me feel a lot better.
    Realistic FFB settings with Accuforce (responsive mode): Accuforce settings

  9. #79
    Spinelli
    Guest
    Nothing to do with FFB, your wheel, your driving skills. It has nothing nothing to do with the driver or his inputs. And it is not even close to the way a vehicle behaves into, during, and out of slides in LFS, AC, NKP, let alone reality.

Similar Threads

  1. Physics of cooling (TMS)
    By Lunchtime in forum Rogue System Suggestions
    Replies: 15
    Last Post: 09-22-2015, 07:41 AM
  2. Some physics questions
    By ggomez in forum Rogue System General Discussion
    Replies: 3
    Last Post: 06-29-2015, 08:56 PM
  3. Urgent - BMW Sauber F1.07 Physics
    By Syed Hasan in forum Modding - Cars
    Replies: 0
    Last Post: 09-05-2012, 01:55 PM
  4. F1 Historics, Classic Touring Cars & Sports Cars - Pictures
    By MarcG in forum Real Racing Discussion
    Replies: 1
    Last Post: 06-02-2012, 10:29 AM
  5. Tyre physics question
    By LesiU in forum Modding - Cars
    Replies: 7
    Last Post: 01-16-2011, 09:28 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •