Why does StockV8 have issues with curbs ?

[Jan Mikuž]

Seems to upset the car massively.
The car rotates wildly, often resulting in unrecoverable oversteer in my experience.

Mid speed corners, low speed going 60kmh, doesn't matter.
The car doesn't seem to like the curbs at all.
How realistic is this ? Seems a bit too much to me, a bit artificial.
Maybe it's because of the 14 lock i'm using, that i can't save it...
But the way it reacts does seem very odd.

Weird things like this. All coasting, no throttle :

https://vid.me/XPjKP

 

[Heitor Facuri Cicoti]

That's how the car drives. It's natural. The StockV8 is somehow always understeering on corner entry and mid corner (except when you force it to oversteer). When you go over a curb, there's a sudden increase in front grip, and the understeer turns into lots of lateral force, which the rear isn't capable of holding. The problem gets bigger when we hit the curb with our steering wheel past the point of optimum grip (Slip Angle vs Lateral Force), as you can see in the attached file.

So you have to (instead of just reacting) actively decrease steering angle when you go over a curb with these cars. Because there will be more grip at the front available for you, and with a smaller steering angle you can get the same turn radius.

 

[Spinelli]

I don't buy that. That's a sort of straw-man argument/excuse.

Any car in any sim could exhibit the same behaviour and anyone could then make the same cop-out statement (understeer turning into oversteer).

There's probably quite a few of us here that know that understeer with a sudden gain of front-grip can result in oversteer. That's basic vehicle dynamics/kinematic.

 

[Heitor Facuri Cicoti]

That's not an excuse. That's what happens. And pretty much every car in the sim shows this behaviour. A bit of curb on the inside front wheel, a bit more turn in. The difference is if the car is ready to withstand the extra lateral force without losing the rear. Some are and some don't, with a lot of factors playing a role in this.

StockV8, Montana, Ultima, Boxer, F3 and even some F1 on specific corners can show you that behaviour.

 

[Heitor Facuri Cicoti]

If you think this shouldn't be happening or is some kind of flaw in the sim, that's another question. I'm just answering the question in the title.

 

[Qazdar Karim]

Man, this car can spin in some tight corners with just 5% throttle ...

 

[Spinelli]

If you think this shouldn't be happening or is some kind of flaw in the sim, that's another question. I'm just answering the question in the title.

I partly agree with this and your previous post and, yes, it's exaggerated in the ISI physics engine. But even within just AMS and not comparing to other games or real life, it seems relatively worse with the 2017 Stock Car.

As with all cars and all games, you have to adapt your driving accordingly to the car and game-engine but it's still frustrating especially to newer players who aren't used to the ins and outs of particular physics engines.

Man, this car can spin in some tight corners with just 5% throttle ...

That's pretty common in the ISI engine (FR3.5 in RF2 is a good example) - an often lack of "planting" of the rear unless the rear is overly soft or your're understeering which sort of hides it. It's as if there's no such thing as weight transfer, as if the car's body, suspension, tyres, and everything are 100% static. It's kind of as if there is no weight pressing into the ground. Weird weight and/or inertia behavior sometimes combined with weird or just low low-speed grip/behavior.

The only way to hide this is to make cars really understeery (either by driving style/technique or setup). This stands out more on some cars and less on others but is common to the ISI physics engine.

 

[Marc Collins]

Jan, I have a feeling I showed you this already, but here's an explanation that could make you understand what is going on.

PS: Those being all coasting makes things worse. A bit of throttle helps a lot in those cases. And yes, in real life StockV8s are just like that. The difference is that real drivers have G force as another source of information which helps them react in time. Oh, and they are real drivers.

Edit: PS2: Input lag plays a BIG role in you not being able to catch those slides.

I agree that the 2017 Stock V8 is controllable--if you remember that either the car or the setup or the combination of the two produces the classic "steer with the throttle" race car. If you don't have good quality pedals, I think you would need to use the configuration to reduce the sensitivity. The g-force seat of the pants would make this car completely "normal" and controllable. It's doable without the g-forces, but you need to pay very close attention any time there is a bump or need to switch direction quickly while accelerating.

Heitor, for anyone who wanted to compensate for the lack of real g-forces, what would be the one or two best things to adjust the car to calm it down a bit? The default set-up seems to me to be closer to knife-edge real race set-up than a more forgiving beginner set-up.

I would suggest that the default set-up is designed or flat tracks. Go to a bumpy track like Londrina. See how the default set-up feels and what lap time you can do. Then make one very simple set-up change and try again. Lower the packers by half, from .4 cm to .2 cm. Share the results. For me, it is transformational.

 

[Marc Collins]

Delete

 

[Heitor Facuri Cicoti]

I see the default setup as too understeery, which actually fits well on some tracks (Interlagos at least), and it can be on the safe side for people on an entry level. But that makes the issue with the curbs more drastic.

Personally, I like to get the rear loose, so I can feel (see) when the car starts to go sideways while I still can do something about it. So rear toe goes to 0, front ARB goes a couple clicks down, less rear camber and less rear wing. That also prevents me from turning the steering wheel way past the optimum point.

But your suggestion seems to be a good one. So does Jan's of reducing front toe. A bit more travel for the suspension to absorb bumps is always welcome. Also, making the inside tyre point a little less towards the turn would reduce sudden change of direction on the front end.

My only suggestion for your test is to switch both setups. Default goes after, so you can really tell the previous setup was better, instead of maybe getting faster by practising.

 

[PieterN]

@Heitor Facuri Cicoti : would you mind sharing your setup(s) ? I would like to try your setup for Ibarra.

 

[Heitor Facuri Cicoti]

I don't have one for Ibarra, but there's this from Santa Cruz do Sul which gave me a very controllable car in a twisty track like that.

 

[PieterN]

Thanks for sharing your setup Heitor ! Works quite well at Ibarra Reverse.

 

[Paulo Gomes]

I honestly dont believe that someone is spinning this car with 5% throttle, but ok... And if anyone is also thinking that is down to some problem with the physics engine, take a look at this video and see how easily a powerfull racing car can loose the rear under certain conditions, its a supercar v8 in the video but is a good example.

Another point, is that most players usually use crazy fast steering ratios for a racing car, what takes away much of the sensibility that is needed to drive them on the limit, in my opinion! I think the stockcar v8 in real life will use something like 540/16 of steering ratio.

 

[Marc Collins]

I honestly dont believe that someone is spinning this car with 5% throttle, but ok... And if anyone is also thinking that is down to some problem with the physics engine, take a look at this video and see how easily a powerfull racing can loose the rear under certain conditions, its a supercar v8 in the video but is a good example.

Another point, is that most players usually use crazy fast steering ratios for a racing car, what takes away much of the sensibility that is needed to drive them on the limit, in my opinion! I think the stockcar v8 in real life will use something like 540/16 of steering ratio.

Great video--illustrating how one of the racers with the best driving control of all time (in an AWD totally different car) can easily spin by accident in a RWD V8 high-powered race car...just like we do when we try to drive them in AMS

 

[Spinelli]

That's a strawman argument. Just because you can easily spin a car in real life in certain situations doesn't automatically mean that there can be nothing wrong with a car that spins easily in a video game.

It's quite easy to see, over the past 15 - 20 years, that there are issues with oversteer in the ISI physics engine such as:

- Cars sometimes needing barely any lock to save the slide. If you apply more than a few degrees of lock, the slide abruptly ends (as if inertia / rotational inertia / mass doesn't exist) and snap-corrects the other way. So you have to keep the wheel almost straight and sort of just wait for the slide to end on it's own.

- Cars not wanting to "hold their line". Unless you're understeering, it's like there's an invisible force making the car want to keep turning-in sharper and sharper (as if you are applying more and more steering lock). This is mostly evident anytime you hold some brake even if it's small amounts. This is also why you can often go around corners using, literally, less than half the steering lock of real-life to get around many corners (Eg. RF2, going around Indianapolis using 25% the steering lock real Nascar drivers use with an even slower steering ratio than real life)

- Ending of a slide (how the car "recovers" / "comes back in line". Upon the initial saving of a slide, instead of the front rotating back to where it was before the slide, the rear rotates back inline to where the front is now pointing. (this also contributes to why there is often very little steering lock needed to correct slides and, perhaps more importantly, why you must return the steering wheel almost instantly back to centre upon saving a slide rather than holding the angle like in real life)

- Often, power oversteer wants to make the engine revs (and therefore rear tyres) insanely quickly shoot up as if your vehicle suddenly gained a ton of horsepower. It's like the rubber of the tyres have no mass trying to press them into the ground as they slip, it's similiar (not quite) to as if a rear tyre started spinning on ice rather than spinning on asphalt.

- plus more...

You can replicate this stuff over and over again regardless of car, setup, etc. You don't even have to be playing you can easily see the behavior in videos, replays, anyway. From any car, ever (from what I;m aware of) in the ISI engine going back at least 15 years to F1 2002.



This stuff is easily able to be noticed. Even new players (friends of mine) notice it. They don't know how to put it into words but when I then explain it to them, they always say "Exactly! Yes!".

 

[Marc Collins]

That's a strawman argument. Just because you can easily spin a car in real life in certain situations doesn't automatically mean that there can be nothing wrong with a car that spins easily in a video game.

It's quite easy to see, over the past 15 - 20 years, that there are issues with oversteer in the ISI physics engine such as:

- Cars sometimes needing barely any lock to save the slide. If you apply more than a few degrees of lock, the slide abruptly ends (as if inertia / rotational inertia / mass doesn't exist) and snap-corrects the other way. So you have to keep the wheel almost straight and sort of just wait for the slide to end on it's own.

- Cars not wanting to "hold their line". Unless you're understeering, it's like there's an invisible force making the car want to keep turning-in sharper and sharper (as if you are applying more and more steering lock). This is mostly evident anytime you hold some brake even if it's small amounts. This is also why you can often go around corners using, literally, less than half the steering lock of real-life to get around many corners (Eg. RF2, going around Indianapolis using 25% the steering lock real Nascar drivers use with an even slower steering ratio than real life)

- Ending of a slide (how the car "recovers" / "comes back in line". Upon the initial saving of a slide, instead of the front rotating back to where it was before the slide, the rear rotates back inline to where the front is now pointing. (this also contributes to why there is often very little steering lock needed to correct slides and, perhaps more importantly, why you must return the steering wheel almost instantly back to centre upon saving a slide rather than holding the angle like in real life)

- Often, power oversteer wants to make the engine revs (and therefore rear tyres) insanely quickly shoot up as if your vehicle suddenly gained a ton of horsepower. It's like the rubber of the tyres have no mass trying to press them into the ground as they slip, it's similiar (not quite) to as if a rear tyre started spinning on ice rather than spinning on asphalt.

- plus more...

You can replicate this stuff over and over again regardless of car, setup, etc. You don't even have to be playing you can easily see the behavior in videos, replays, anyway. From any car, ever (from what I;m aware of) in the ISI engine going back at least 15 years to F1 2002.



This stuff is easily able to be noticed. Even new players (friends of mine) notice it. They don't know how to put it into words but when I then explain it to them, they always say "Exactly! Yes!".

You have already made this point dozens of times in multiple threads. Should I repeat again that it is not an ISI engine issue, bit a car modelling issue. We have definitive proof of that as long as we can find at least one car in one ISI-based sim that does not have the unnatural pendulum light rear end. The USF2000 in rF2 does not. 90% of the cars in AMS do not. The StockV8 and SuperV8 may or may not...it's debatable and that's why people are discussing set-up and other issues here. If you can find a set-up that reduces or eliminates the problem--even if it is an unrealistic one--it again proves the point that it is the modelling of the specific car, not a systemic flaw in the ISI engine.

The ISI engine is far from perfect--no one involved with it would ever claim it was perfect. But, there is no evidence that there is a systemic oversteer flaw in it as you have promoted repeatedly for years. There is plenty of evidence that a large number of cars modelled for the ISI engine have been designed with too much tendency to oversteer or exhibit "too light a rear end" as I like to describe it. Some of the problem is related to FFB and not the actual car modelling. That's why there is fewer complaints about this issue in AMS--because it has the most naturalistic and authentic FFB of any ISI-based sim, including rF2.

Being able to feel what the rear end is doing, especially when you have a car designed to oversteer in the extreme in real life like the V8s, is 90% of what is needed to control it. Lower resolution or lower quality wheels cannot reproduce this FFB and lower resolution and lower quality pedals cannot provide adequate input, so some people will never know what the sim is actually capable of.