WHERE IS LIFT OFF OVERSTEER????????????

Driving the Porsche 911 GT3 it seems that the opposite lock is on, or some kind of stability control, that make it almost impossible to make the car lose control (a car with a rear engine!) there is something very wrong

 

You have just described exactly what official GT3 racing is all about...so the semi-professionals who drive these cars can have a good time and not kill themselves or each other.

Also, a modern road-going 911 is as easy to drive as a go-kart these days and has just about zero connection to the widow-maker/doctor killer 911's of the olden days. I would imagine the race-prepped version would not be a major step backwards, though having never driven one, I can only rely on the accounts of those who do. I am not aware that the 911 is considered inferior to the other cars in the GT3 class, though it certainly has its own character. That happens to be a bias toward understeer in most circumstances, so draw your own conclusions.

 

I think we are using wrong terms to describe one thing: we are expecting every car to have a property for which a reduction of gas application must result in a noticeable increase in the yaw rate. Call it lift off oversteer, mild or snap oversteer, the only difference is the magnitude of the phenomenon we are discussing about.
The problem is everyone is discussing about this as if it was a given because this or that sim has it more or less, but no one is looking at the reason why certain cars will show this behavior and certain others not.
Like I tried to explain there are really two sources of increase in the rate of yaw when the gas is reduced or released both linked to the relevant weight transfer:
1- Aero center of pressure shift due to reduced height at the front and increase at the rear
2-Mechanical settings on the suspensions that will trigger increased sensitivity (i.e. reduction of available grip) to the weight transfer
One smaller contributor is engine braking effect, but it can be felt really only if the car is seriously sharp in terms of setup
It goes without saying that cars like for example Group DTM with no front and rear diffuser have NO aero shift effect and can only experience that wrongly called lift off oversteer if their suspensions setup will promote it and if its rear differential is not having an excessive stabilizing effect compared to the car suspension setup. It's as simple as that.
In particular it is to be noted that until the car setting is that of significant understeering typical of the baseline setups it will be impossible to generate a significant increase in the yaw rate just by releasing the gas. For that to happen the setup must be much sharper and the coast settings in the diff must allow it.
Until we put our minds into what physically is happening to the car we will keep believing that a Porsche 911 with the rear engine and the smallest forward weight shift under braking and turn in of the GT3 group and the weakest rear diffuser should have a strong pitch sensitivity (aka sensitivity to weight transfer) which is obviously wrong. The car that will show more this behavior is the M6 which in fact already in the baseline setup is the less rear stable of the group.
In general it is not true that every car must be able to be driven by modulating the gas application or its release: it's just not physically true. Even those cars who do, can't be driven that way in every corner as in many cases, especially for cars with significant aero in fast corners the aero itself will become dominant and will have a stabilizing effect against the mechanical grip and even the engine braking effect, so no game there. Again, we need to look at the physics behind, not what this or that sim does or what our expectations for having fun are.

While there are for sure certain cars in AMS2 where the diff issues are not 100% resolved, the statement that "liftoff oversteer" cannot be achieved is plainly wrong. Whether we figure it as mild or snap or whatever, the reason for any of these cases of increased rate of yaw are practically the same.
I can achieve as much "liftoff oversteer" as I like, mild to savage on GT1s with no crazy setups like zeroing up differentials or ARBs or silly stuff like that in every track. Actually while racing at Spa I had to reduce that sensitivity as the rear tires were getting to stressed and the car increasingly difficult to drive consistently on a totally reasonable setup.
I've tested the McLaren 720 on an old setup I prepared before the latest diff improvements: I am getting a ton of oversteer in general and a ton of gas application sensitivity (not only coasting but on power as well) even in fast corners which means that indeed the diff has much less stabilizing effect than it had few months ago. It's up to us to dial it the way we like it.
Same I tested the Camaro GT4 (definitely one of the most disliked cars I heard of) and still with all the things that may need a further look and improvement, it can be dialed in quite a bit playing with diff, TC, tire pressures and angles.
This is my honest point of view and experience so far with this sim and the way you can forge the car behavior is one major reason why I like it. Call me fanboy if that fits the definition I really care only about my opinion being honest.
I really find we as community too often keep going the easy path of complaining about every car not driving correctly, being broken and so on, but really spending little effort to work on the setups to make the car do what we would like her to do.
Also so far I have never found a setup I could use to my liking without changes from one track to another on the same car, so I am aware it is a lot of work. And from this stand point I think Steel is doing a great job in serving our community by providing good reference setups.

 

I have driven many a Porsche ranging from a 70s Carrera RS to a late model road going GT3 , but only in a manner befitting that I didn’t want to be responsible for a huge bill
The new ones are super easy to drive for sure, the ones that even driving sedately wanted to kill you where the old 80s turbos. ..nothing nothing nothing then whoa too much .

 

Lots of interesting discussion going on, most of which I don't have time to respond to so will just try to summarise in a few key thoughts:

1. Rotation from lifting (from hereon in "rotation") comes from load transfer from rear to front and a consequent increase in rear slip angles compares to the front. Its magnitude is affected by many aspects: stiffness of the suspension, downforce and aero balance, engine location, weight and wheelbase of the car, and yes, type of differential.
2. The differential alone does not turn rotation on/off, but it can inhibit it.
3. Rotation in AMS2 is somewhat muted by default, and does not appear automatically in every car as soon as you lift even a little (from hereon in "magical rotation").
4. Some cars have "sticky" LSD's, inhibiting rotation excessively in slow corners specifically. They also tend to open abruptly and too much, leading to snap rotation.
5. Other sims cut corners too, R3E diffs are always open even when driving a car with a spool, and AC/ACC diffs are highly idealised to the point of being magical.
6. Other sims have cars that don't rotate either when a poor setup is used. Try the Porsche 911 GT3 R in ACC with the safe preset at Nurburgring GP. Understeer city until you lift off in the Schumacher Esses, at which point you LOO out of control. Not every car in AC has magical rotation either, as mentioned in that RandomCallsign video.
7. A road-going 911 from the 80's has nothing do with a GT3 car from 2017 and should not be used as a benchmark for realism.

 

Well, I'm glad you understand what I'm getting at, so this hasn't been a total waste of time. You say there is only one context, but the references you kindly provided do indicate otherwise (just read the first paragraph of Lift-off oversteer - Wikipedia, which mentions load transfer, how the decrease in vertical load at the rear causes a decrease in the lateral force it can provide and hence an increase in slip angle (though it doesn't mention how the increase in vertical load at the front causes an increase in the lateral force and hence a reduction in slip angle, but you don't have to be Einstein to work that one out).

I agree that forcing the car into oversteer (with the brakes (inc. handbrake!), gears, steering wheel, etc.) is not lift-off oversteer. The term, and the clue is in its name, is that it is caused entirely from lifting off (the throttle).

I don't agree about the thing you think I'm not addressing, which is traction loss. Your Wikipedia reference about lift-off oversteer clearly mentions changes in grip level due to weight transfer, so this is a change in the level of traction (though not necessarily a total loss), and the resultant change in slip angles that your reference clearly explains is what causes the change in yaw angle of the car. I'm not sure why you stated 'the rear is planted and the front just grips more' as this is just a crude way of describing load transfer leading to lateral force changes and resultant change in slip angles. You say that the other sims that give me the rotation I want are not achieving this through traction loss, and this is because they are achieving this with the load transfer described in your reference on lift-off oversteer. You are right in some cases in saying that reducing the front slip angle and increasing the rear slip angle is not lift-off oversteer and arguably not oversteer at all, as doing this may result in the rear slip angle being less than the front slip angle, so the car could still be in an understeering state (even though the line has been tightened). However, most racing drivers like a close to neutral handling car, which means the front/rear slip angles will be similar in magnitude, so any load transfer to the front will tend to put the car into an oversteering condition.

Your final statement that you know and completely understand slip angles but don't want them used in the discussion is puzzling, as is stating that lift-off oversteer is only related to total rear traction loss - read your references (it can lead to total traction loss, but it doesn't have to).

I dug out my Skip Barber "Going Faster" book, which I haven't read for ages, and as is the case when teaching you start with something simplified and then redefine it more accurately, it states that oversteer was originally defined as when traction was lost at the rear first, but redefines it totally in terms of slip angles (which I know you understand but don't want to talk about). So according to the Skip Barber Racing School, oversteer does not have to be total rear traction loss (a sudden rear step out), just a greater rear slip angle than that at the front.

I agree that getting the level of lift-off oversteer is a function of setup, as you want just the right amount and no more. No driver wants a car that likes swapping ends, but no driver wants a car with no lift-off oversteer at all (or lift off understeer!) - the driver needs to be able to modulate the relative slip angles either side of them being equal. I should spend more time with AMS2 setups to see if I can make the ill-handling cars more to my liking (some of them are quite nice as-is, so it's not all bad!).

 

It is not the only car that has this fake behavior, it feels very arcadish that the car auto recover after trying to spin it intentionally

 

Well that's simple to answer - it's because the weight transfer causes it to understeer less! Why are you of the opinion that I think it is always a requirement for a bigger slip angle at the rear in order to get additional turn in? It is all relative, as I'm sure you know.

I understand why it is undesirable to call less understeer 'oversteer', as less understeer is still understeer (no two ways about it!), but exactly the same physics are involved whether it goes from understeer to less understeer, understeer to oversteer, or oversteer to more oversteer. Since it's all relative, I am understanding of the term being used to describe the phenomenon in a general sense because, without measuring equipment, how would you know exactly when to flip terminology to/from LOOS to LOLUS? I can see why the term 'lift-off rotation' is preferable, but lift-off oversteer is surely just a case of lift-off rotation gone too far, and so many people, whether rightly or wrongly, use the term 'lift-off oversteer' to mean the ability to tighten the line by lifting. If we're not careful (i.e. too fussy) would could get into the realms of POUS and POLOS rather than using one general term to describe it all.

 

I was arguing that lifting causes weight transfer and slip angle changes that tighten the line, and in one of my very early posts I mentioned that it could be just a reduction in the amount of understeer, so I have never implied that it is always just oversteer that is created. The Skip Barber bit was to illustrate that oversteer for some is loss of the rear end, but the book clarified that it doesn't have to be this and can be just a larger slip angle than at the front. I wasn't implying that everything is related to oversteer and oversteer alone.

I can assure you that I have heard the term elsewhere, and that for me at least it doesn't originate from Gamer Muscle videos, so it is perhaps more widely used than you imagine. In fact I don't think Gamer Muscle will have thought up that term for himself, so he heard it elsewhere too.

Regarding differentials, these will impact the handling, but how long would an LSD take to react to a throttle lift? Will the primary response be to the weight transfer before the secondary response coming from the differential's clutches reacting? Could it be that so long as the inside wheel isn't spinning up and causing the clutches to engage, that the car dynamics in yaw are a function of the weight transfer alone and nothing to do with the differential?

There is nothing I'd like more (in sim racing anyway) than for AMS2 to behave more like real life, and remember I initially likened its (to my mind) poor behaviour to real life cars, so my criticism isn't all relative to other sims. I would hope it is not a canned effect, as surely weight transfer is something fundamental for other aspects, like camber changes, bumps, roll, etc. in addition to that due to throttle lifts (or application).

 

Hopefully this Prt 2 video should be more useful in demonstrating useable L.O.O. rather than the v extreme examples on display in the first

 

christ so many pages of this
no need to make 100 videos tbh, GT3 work fine.

try it with Retro-F and dont go below 90kmh

 

Just to jog your memory, this was my first post (note paragraph three).

Glass houses! What are you on?

 

Im gonna tune out of this thread like a Ermin Hamidovic video...