Finding it hard to control oversteer

Your steering angle is completely different from the sim.
Why?
This video just brings water to Spinelli argument.

 

Game Stock Car is isiMotor and to my mind that feels closer to real life than anything IMO. Catching slides in the FormulaV feels completely natural and awesome.

I briefly mentioned this on the other thread, but in rF2 in cars like the Corvettes or Panoz, I seriously feel like it's the front tires letting go, not the rear. Turn the car and give it some gas, and the front tires break away and the car rotates toward the inside of the track. The rear of the car feels like the fulcrum of the slide NOT the front tires.

It's especially noticeable when I'm trying to straiten up from a tank slapper. Whenever I stuff it up (which is most of the time), at the moment the car is straightening up and should be either gripping or penduluming again, it feels as if the front loses it and not the rear.

Next time you go for a spin and are getting it sideways, think in terms of the front letting go and the rear becoming the fulcrum, as oppose to the rear letting go and the front becoming the fulcrum. I know it sounds weird but to me this is how it FEELS. In my opinion some cars have way too little front grip, and too much rear lateral/longitudinal grip. Press the gas and the car is subjected to rotational force, but it's the fronts that let go first.

I've seen this kind of thing happen occasionally IRL, but in rF2 it seems to happen the vast majority of times.

 

It's a replay, not real time. You can clearly see from the telemetry overlay that the steering inputs match my physical steering movement.

I lower the steering rotation when i record the replays like this because it looks weird having 2 sets of hands moving around. If i could remove the in-game wheel and hands altogether from the replay i would prefer to do that.

Here's a similar test with the current version of the car. You can get a decent idea of how they both behave during a slide.

You can claim it's not realistic, or that the behavior is odd, but unless you have some facts or something more to go along with it, then it's just a broad, generalized opinion. I can easily say that it looks perfectly realistic and drives realistically as a pointless counter-argument, but neither claim really matters beyond the words typed out to say it. If people really think this behavior is unrealistic for a car like this, then it should be easy to at least provide some analytical observations of exactly what makes the behavior wrong. Or better yet, you could provide telemetry data compared with the actual car. Except most people around here should know that, that's already been done by someone working in F3 with rF1, and even with that old ISI sim, it matches pretty closely to real-life behavior.

https://drracing.wordpress.com/2014/01/03/how-close-is-close-enough/

So, are people saying that rF2 has gone backwards somehow? Certainly doesn't feel like it to me, but i don't have any telemetry to check, all i have is a bit of experience to compare it to.

@green serpent, What you're describing sounds like what happens to me when i use a higher STS value than what i normally use. Right now i'm using 0.176, i lowered it from 0.181 because i switched from using the AF dampening/friction settings, to using the JSON resistance settings. I found that setting the saturation to 1.0, and then adjusting the coefficient gives the dampening effect a more realistic feel. I have coefficient set to 0.1 right now, and it gives the wheel the same weighted feel as what i was using before with the AF settings, but when the steering torque kicks in it's not slowed down by the dampening effects, so it seems to be what i've been looking for the whole time in this area.

 

Thanks Paul, I think I might start messing around with the STS setting to see what feels best.

In regards to my above post and as I've mentioned before, If I get my throttle input EXACTLY right, the car feels amazing. Blending the throttle between traction, less traction, and back to traction again seems to fix the issue all together, but it's more difficult than I would expect, and most of the time I'm not getting it right. Being too jerky with the throttle or applying the throttle at the wrong time can have huge consequences and make the car do 'odd' things. I'm not saying that anything is realistic or unrealistic, I'm just trying to get the most out of the game by figuring out what's happening at different stages of car behavior.

 

Most of my last post wasn't really directed at what you were saying. I actually agree with a lot of what you're saying, especially about the front letting go, at least in the outdated cars. One difference i've noticed between the new Megane vs the old is that under extreme slip angles, the new Megane responds to steering inputs much better than the old. Even though, with the old car the steering actually feels more twitchy under normal circumstances, it's only when you put the car into a big slide that suddenly the front end just starts to let go and become much less responsive, this is not only in the feel of the feedback, but in the behavior of the car itself. This is the same with all of the older cars in rF2, and it was the same with the FISI before it was updated.

For me the only thing i want is a faster steering wheel rotational speed, but that's an issue with the AF and the fact that it's rotational speed is limited somehow, either due to hardware limitations or firmware. I've confirmed this by testing the NSX with the correct steering ratio and then again with a much shorter, and unrealistic steering ratio. With the realistic steering ratio, the wheel doesn't spin fast enough on it's own to catch large back and forth slides, like tank slappers. With the shorter ratio, it responds to really large slides as i would expect and tank slappers become much easier to control.

 

By rotational speed I assume you mean the self aligning torque? Like I said somewhere else if I let the SAT completely take over then it doesn't do me any favors at all. I need to turn the wheel faster then the SAT does and past that point there's literally no feel whatsoever so it's hard to know how much to counter steer, so I'm guessing more than anything else.

I've never really understood the difference between steering lock (e.g 900 degrees) and steering ratio. Are they independent of each other or if you change one do you need to change the other?

 

The self aligning torque is extremely fast and can keep up with the car within the simulation itself. Where this starts to fall apart is when you have a FFB wheel that simply can't rotate fast enough, that's my problem with the AF, it's simply too slow in extreme situations with cars that have really long steering ratios (like the NSX). With most of the race cars it keeps up perfectly fine in nearly all driving situations, even in big tank slapper situations because the ratios are generally tighter, but with the road cars it's just a bit too slow.

The steering ratio is the amount of steering wheel movement that's required to turn the front wheels. So you have 2 parameters here, the steering wheel lock and the front wheel lock. A long steering ratio would require more steering wheel movement to turn the front wheels a particular amount compared to a shorter steering ratio.

https://en.wikipedia.org/wiki/Steering_ratio

That wiki page explains it pretty well.

The videos below shows why there's a problem if your FFB wheel isn't fast enough. If you keep an eye on the steering wheel, you'll see how quickly it can rotate and this rotation is practically only limited by how quickly the car itself can rotate and still maintain some traction on the front wheels. So, when doing big back and forth slides, the steering can spin very quickly from lock to lock in a fraction of a second. It's understandable that you wouldn't want to fully replicate this for a lot of race cars, since if you have the wheel moving this quickly at 13Nm, then it could be very dangerous, but i would personally rather have the wheel move as quickly as it's supposed to since i know how to avoid getting injured and it would provide for better simulation.

 

@Paul Loatman
Have you tested with wheelcheck what's the max AF rotation speed?

 

Ahhh, OK now I understand.

Now are you replying to Spinelli or to me?
Because, you know, as I stated in other posts, I pretty much agree with your opinion.

 

I think it was around 150rpm, possibly 200, i can't remember exactly and i don't really have the time to do another test right now. I just know from the tests i've done within the sim, that it's just not fast enough with very long steering ratios. It might need to be up to 350-400rpm to cover most road cars.

It was to Spinelli, and to anyone else who makes a broad statement of the realism without providing anything more than to say that it's correct/incorrect. If there's no explicit reason for believing one way or the other, then it's just a credulous belief. The only real reasons i've heard of this unrealistic behavior is that the cars should continue to spin instead of going into a tank slapper, or something like that. I apparently am incapable of understanding it so it was never explained beyond that for me.

 

I told myself I'd avoid any more of this <roll eyes>…

It's long been recognised that sims generally, and rF1 certainly, do a pretty good job of simulating reality up to the point that cars start sliding around. Then however, to quote Dave Kaemmer, there's a real "dearth" of information. But it's still been bloody obvious to everyone including developers that sliding behaviour has not looked natural. It's the reason sims/game devs have been dreaming up all sorts of quasi-physical tyre models. I don't know why you would think that a supposed comparison between sim and RL data for normal race track driving tells us anything about drifting.

The other points about sim wheels being slow etc. have been addressed; if a real driver gets a "death grip" on the wheel and doesn't allow it to turn at all during the recovery stage - or even turns it the wrong way as I've seen fashion models do in drifting lessons - yet the car still exhibits different behaviour to the sim, then it's not steering speed, reactions, ffb, skill etc. etc…

You might like to consider this: some of us, and probably most of the LFS crowd, see a different rhythm in the way both real cars (and LFS) recover from slides when compared with ISI games. So what's more likely: that we're hallucinating a difference that's not there, or that you're not seeing (wanting to see?) one that is? For the record, I've fired up LFS twice (I think) in the last 2 years. I'm not a LFS fanboy or anything. But it does a better job at this than anything else IMO; it's the only one that looks natural during recovery from slides (regardless of inputs! )

Thank you, that's it, done…..

 

I think most people do this have overheated tyres and/or do not give them enough to cool back down.

I have driven LFS for years and there is nothing special about it compared to rF2.

Physics grip steering tyres vary greatly across ISI cars which imho is better then 95 cars feeling the same ( like AC , LFS or pCars ie: you drove one car in each you have drove them all )

I can't believe we still have these debates years on.

rF2 like every engine has flaws but at its best it simply leaves the rest in the wake.

 

Wow so out of reality... But we know by now you like embellishing things; though still amazes me every time.

Maybe you'd prefer if AC, LFS, Pcars had one car of each type of category or class. Then you'd feel big pronounced differences.

 

My point wasn't that i don't see any problems, in fact i do see them, that's why i think the updated cars are better than the old. If i didn't see a problem in the first place, i wouldn't see any improvement either, would i? My point was that i don't see much, if any fundamental differences between LFS and rF2, rF2 is more complicated by nature, and to me, it feels more accurate for the reasons that i've already given, but it's obviously not perfect, and neither is LFS, i think they're both suffering from various issues during a slide. You're only issue with what i'm saying must be the fact that i don't have the exact same opinion as you, where you think LFS is somehow more accurate than rF2 in a slide. Except you've not really explained exactly why though, beyond to say that the behavior is simply more accurate in LFS, which was also my point. Without explicit and targeted reasons to believe either way, there's nothing i can say to that and is therefore not really useful to reply to since there's simply not enough in what you're saying to have a real response for it.

The only way a conversation like that would go is like if two people argue over what type of BBQ sauce tastes better, where one person repeatedly claims: "This one tastes better." While the other repeatedly claims: "No, that other one tastes better."

 

Well I am being honest any car in those titles I can ramp up ai and drive with them.

Try that in every rF2 car you need to be a alien.


LFS I rate with others as in drivers could abuse garage settings and do unrealistic sectors.
I seen it many times in LFS, and all gmotor.


That is where rF2 stands out for me, check any rF2 rank lap there is no bs in them.

The day I see someone beat me in the lowly powered F3 using warped garage settings , I will eat my hat.

In LFS you could do that every day of the week any car and you dare compare it to rF2 , shame. lol

 

I didn't really start this thread to compare rF2 to real life or knock rF2 in anyway. I just wanted to figure out how to control a slide, because I was struggling so much, and it felt very foreign to me and didn't make much sense. My first thought was to find out whether the game was at fault or if it was my skill or hardware.

I came into the situation with a bias because I can drift in games like Forza, LFS, and AC without too much difficulty, plus I have done the odd track day in RL, so I thought that if rF2 was the most realistic sim, it should be the easiest to drift in because I could just apply my real life knowledge. However I have never really held huge slides in real life enough times to closely analyze what I'm doing. I just drive by feel without thinking too much.

I've been spending lots of time getting used to rF2 with about half that time dedicated to going sideways. I analysed it a lot, and I had a pretty good theory about what to do at different stages of the slide, but I still couldn't do it. About 200 hours in, something in my brain just clicked and I got it. I still can't slide racing cars, but I can slide other cars that you would expect to respond well to that kind of driving. I might start a thread to share some things that might be very obvious to some but helpful to others.

It's my personal opinion that games like LFS although they 'feel' better, it's possible that they are overly simplistic in their over the limit handling (especially something like Forza), whereas rF2 attempts to stay true to life, which results in over the limit car control being much more complex than other games. It seems too difficult to control oversteer, and thus unrealistic to those of us who cut our drifting teeth on more basic sims. That's just my hypothesis and I could be completely wrong. RF2 is definately not perfect as we've all agreed upon, I personally think the sim needs more lateral front grip and that the ffb should indicate front grip much more clearly.

But as far as over the limit handling, I'm starting to think it's pretty close to RL. I know people don't think ffb has much to do with it, but it has plenty to do with it. The game feels like whatever your ffb wheel is telling you it feels like. I drive by ffb, not visuals. If anything I adapt my driving style to compensate for inaccurate ffb (i.e deadzone, slow SAT), rather than having to adapt it because of a physics flaw.

 

My observation about spinning was that sometimes very low speed spins and I mean low speed just spin around 240 degrees or more.
IRL a very low speed spin won't go that far and it stops kinda abruptly.
Just feels like it's on ice not tarmac because it swings round so much.

 

Ah right mate.

Even so I think hot tyres is a bigger issue for the uninitiated.

rF2 has real variation in handling with tyre temps.

NO I never said they are realistic.

rF2 makes you take more care and to nurse them a few corners.


AC and pCars I can abuse tyres, and they feel relatively the same, still win races.

That is impossible in rF2 , tyres can fall off a cliff.


Like I said unrealistic maybe but it offers a driver more of a challenge. !@!

AC and pCars no aids are not even a remote challenge to keep tyres good !@!. ...................................... imho.

 

Exactly, same story with iracing. The biggest care is that is challenging and difficult. Realism? '-.-

"iRacing made the sliding/oversteering harder to control, too make the simulation more serious. iRacing is hard to master, because iRacing wants the sim racer to have a competitive simulation. so the sim racers can compete their racing skills to each other. And because its hard to master, people drive in it for years or even for ever and dont want any other racing simulation. Realistic physics ? Yes and No. Lets say the Physics are insanely well made, compared to real life physics. In realife you have much more time to correct a slide and your car has much more grip. sometimes its funny how easy it is in iRacing to spin out. But in the last update they added a new tire physics model. now you have a little bit more grip but its still not the same as in real life. All in all is iRacing the best racing simulator because its interesting and never bores you out."

 

You need to stop taking AC criticism so personal... If you think DD is wrong then post something to support it.