Camber

[mistersafeway]

Positive camber hasn't been confirmed, it's just one of the theories.

 

[SCClockDr]

Let me get this straight: camber is currently coded as positive, although visually it is negative, although PD's tuning slider has only positive values? IE the slider is in fact correct, but who in their right mind wants positive camber for racing purposes?

Now if they would change the slider to -10 - 0 - +10 then the racers & drifters would have what they need to succeed.

A brilliant coder still needs to understand the real world concept of what they are trying to replicate with the code. This is a very common software development issue and occurs way more often than it should. I suspect the visual group saw the positive camber effect and just reversed the effect & failed to see the issue as originating in the physics & flagging it back to that group for the fix.

IF indeed there is a camber issue! (I suspect there is.)

 

[Verbal]

Let me get this straight: camber is currently coded as positive, allthough visually it is negative, allthough PD's tuning slider has only positive values? IE the slider is in fact correct, but who in their right mind wants positive camber for racing purposes?

We don't (and never will) know what goes wrong. A sign error with the camber values is just one hypothesis, if a quite plausible one that so far could explain everything we're seeing. What we are very sure about is that the camber settings don't work.

 

[Verbal]

I suspect the visual group saw the positive camber effect and just reversed the effect & failed to see the issue as originating in the physics & flagging it back to that group for the fix.

Depends where the error was made. There's basically four modules at work here.

- The gui reports a camber offset to the vehicle model
- The vehicle model computes the wheel position, adding that camber offset to the wheel angle
- The tire model uses wheel load, side slip, longitudinal slip and camber angle to compute tire forces & torques and reports those back to the vehicle model
- The graphics engine uses the same wheel position and angle for the shiny graphics

If the error is made in the gui, the vehicle model (or the communication between the two), the graphics will be affected and the error will be easy to spot.
But a wrong interpretation of the camber angle by the tire model won't affect the graphics engine in any way, so there won't be anything to flag back...

 

[SCClockDr]

@Verbal I concur! There are many other possibilities I am sure but i chose that one for illustration purposes. The fact there are several modules involved here makes finding & permanently squashing this kind of bug quite problematic. Because fixing the actual defect may well throw several working functions/procedures out of whack and generate several bug reports all needing attention. Unfortunately, often is the case where an engineer/coder will see this type of error and correct it in their sphere of influence, preferring to let sleeping monsters remain in their slumber.

I feared this type issue the most just below proper version control, when I worked in this field.

 

[Verbal]

@SCClockDr
Yeah, such bugs can easily happen in a number of ways and in many cases will just as easily be overlooked without thorough testing. OTOH I think this issue might be a very easy fix. I only fear the majority of their target audience outside GTP won't be bothered too much by such things, so even if they noticed it will probably be very far down on the to do list...

 

[Jack Napier]

Tid bit of a helpful hint. I plan to put this is a guide or something but for now Ill just drop a few tips here and there.

Camber and Caster go hand in hand. Basically Caster applyes a neg camber effect to the outside wheels and a positive camber effect on the inside wheels.

To figure out how much Caster you have, put 0 on the toe and Camber setting, then park on the track, turn the steering all the way to a side, grab the camerah and look at your inside wheel. It should have a positive camber angle, add neg camber in the settings and recheck untill there is 0 positive or negative camber on the inside wheel at full steering lock. whatever the camber angle it took to cancel out the Caster IS your caster angle.

Increasing Neg camber will counter the positive camber effect caused by the caster angle on the inside wheel while adding neg camber to the outside wheel.

Things to take into consideration.

The GT350 has 6.2 degrees caster angle, so at a mid level corner the inside wheel is seeing around 3 degrees positive camber.

 

[Polsixe]

It's not a placebo, the car is handling better for you. You have less grip on the front and less on the rear but a slight imbalance towards more rear grip. The car feels balanced for you, but overall it has less grip and you are slower than you might otherwise be with a 0.0/0.0 car, but tuned for the kind of feel you are looking for. You might also try 0.5/0.0 which might get you what you want, with a little more grip.

I'll try that, thanks,

 

[Polsixe]

@Verbal I concur! There are many other possibilities I am sure but i chose that one for illustration purposes. The fact there are several modules involved here makes finding & permanently squashing this kind of bug quite problematic. Because fixing the actual defect may well throw several working functions/procedures out of whack and generate several bug reports all needing attention. Unfortunately, often is the case where an engineer/coder will see this type of error and correct it in their sphere of influence, preferring to let sleeping monsters remain in their slumber.

I feared this type issue the most just below proper version control, when I worked in this field.

Since we only see the graphical user interface front end the math and algorithms in behind will be a mystery. The programmers can easily change the sliders and numbers without changing much else. Changing a + to - is easy peasy.

I suppose it is too much to expect GT6 to accurately model entire suspension dynamics for every car and types of suspensions, wishbones, struts, etc.

 

[claudiorz]

Have to say I haven't noticed any issues with the camber on mine handles exactly as I would expect it too.

If it worked like it should those FF cars would be understeering like crazy with that much rear camber and so little on the front, but I'm pretty sure they aren't because the rear camber is adding oversteer, which it shouldn't.

 

[Holdenhsvgtsr]

If it worked like it should those FF cars would be understeering like crazy with that much rear camber and so little on the front, but I'm pretty sure they aren't because the rear camber is adding oversteer, which it shouldn't.

That amount of rear camber doesn't cause btcc cars to understeer....

It causes them to oversteer.....

 

[BlueShift]

Lots of tests recently, for the FITT chalenge.
It appears the camber effect is really marked in corner entry, like neg camber would, but more camber in the apex or corner exit mean less grip, like pos camber would.

More camber in GT6 = more grip in corner entry, less in apex, lot less in corner exit. Compared to GT5, problems due to camber (braking, accel) is exagerated (don't know if that's GT5 though).

I'm now quite sure there a sign problem somewhere : it seems last patch cured the corner entry phase, but not the other phases.

 

[claudiorz]

That amount of rear camber doesn't cause btcc cars to understeer....

It causes them to oversteer.....

I looked it up and yes you're right, that extreme camber settings causes these cars to oversteer. But my point is that in GT6 any amount of rear camber will reduce rear grip when cornering just as front camber reduces front grip, and that's not how it should be.

 

[UltimateLurker]

I spent the weekend testing my favorite car (all the Lotus Elise variants), and noticed that you really can't tweak the camber from 0 at all without making the car very loose in the rear (even when real life settings are used). However, the Green Motorsport Elise (the fastest Elise on the game) does well with a stock rear -2.0. Interesting that.

 

[donpost]

Rumours abound that camber has been fixed this update. Anyone had a chance to test it out?

 

[Michael88]

So I tried front wheel cambers ranging from + 1 up to +4 degrees and rear wheel cambers from 0.5 up to 1.5 degrees and in ALL cases grip got much worse. I did extensive camber tests with various race cars like the TVR Tuscan Touring car, the Mazda RX-7 touring car, the Bentley Speed 8, the Diablo GT2 and Caterham 7. In ALL cases front camber dramatically increased understeer in all kind of corners and even 0.5 degree camber on the rear wheels noticeably decreased rear grip.
Various spring settings and ride heights didn't make a difference. I'm now running all cars with 0 camber front and rear.

I remember in GT5 front wheel camber definitely increased cornering speed. So whats up with that? Thats surely not very realistic, in the real world many touring cars have cambers up 3-4 degrees to help cornering on tight and twisty tracks.

 

[xSNAKEx]

Have you tried it on a variety of cars with a variety of tyres? Not all cars benefit from camber thats greater than -1

In real life I noticed my cars rear end got much more traction and stability when I reduced the camber from -1.7 to -0.9.

 

[FussyFez]

I'm pretty sure the general consensus for GT6 at the moment is, no, camber does not work as it should. (in 1.03)

I have experienced the same as the OP.


Just read this, which says camber has been adjusted.

So more testing is in order

 

[Michael88]

Have you tried it on a variety of cars with a variety of tyres? Not all cars benefit from camber thats greater than -1

Read my post, I did extensive tests with 5 very different cars.

 

[diptob79]

Up until 1.03, more safe to select either Racing Hard or Racing Soft. Chamber 0 is way to go.

 

[Flying Scots]

Me too have same impression. Camber don't work...

 

[hognhominy]

I'm pretty sure the general consensus for GT6 at the moment is, no, camber does not work as it should. (in 1.03)

I have experienced the same as the OP.


Just read this, which says camber has been adjusted.

So more testing is in order

I have edited the op to reflect most current information, It was always under question, as the poster of info @Trux was not 100% when he posted. If you find a definitive answer one way or the other let me know and I will link to it. For now @Michael88 's testing coincides with what info I can find.

 

[FussyFez]

Just downloaded 1.04 and it seems fixed from initial tests. More grip at 1.5 on the front of an S13 with sports soft. 1.5 all round gave more grip and less snap when losing grip, more progressive grip loss. But a definite improvement in lateral grip with mild camber. Yet to test race cars on slicks, but the effect should be much more noticeable.

ETA

@Michael88 was your testing in 1.03 or 1.04?

 

[Jack Napier]

It's never been broken, just seems peeps have the inside wheel and outside wheel confused when tuning

Use camber in the front to manipulate when the INSIDE wheel will transition from wheels straight with neg camber to steering cranked pos camber from the caster effect, more camber counters the caster camber effect. The higher the camber the better the car will be on tight corners, lowering it slightly will move the transition point closer to when the steering is straight making her better for faster corners . It's always about striking a balance

 

[Jack Napier]

It 100% works, if you think it don't chances are your mixed up.

 

[Z1-AV69]

After 1.04 I'm still getting better cornerspeeds with 0 camber. Also adding front camber adds understeer while adding rear camber adds oversteer. Not fixed imho.

 

[donpost]

It 100% works, if you think it don't chances are your mixed up.

Not sure I agree... a LOT of people have said 0.0 is giving the most grip which is wrong.

Plus camber adjustments are primarily about keeping the OUTSIDE tyre flat against the road in cornering.

 

[FussyFez]

Whilst I can't agree that it was never broken, after some initial tests on 1.04 it feels much better.

I jumped in the last car I was in last night. Took it out as it was with 0 camber all round. Initially felt very similar balance wise, but a slight loss in grip all round, losing about a second and a half round trial mountain reverse.

I then added 1.0 to the front. Immediately the wheel had more weight to it and I could feel the car loading the wheels up when turning in.

The rear now wanted to slide with any throttle application.

I added 1.0 to the rear. 1.0 all round now, and I can get on the power earlier and feel the arse squat and grip, but still power oversteer if I wanted.

I upped the front to 2.0, and although I had more outright mid corner grip, it was easy to brake too late and plough too far from the turn in point (consistent with PD description in game). It was also easier to understeer mid corner by adding too much lock.

I settled in 1.5 all round for the mean time and it feels great, much more overall balance, and you can feel what's going on more.

Very please with initial results but more testing is needed.

I did all this on my maxed S13 track car on sports soft (this has been a tricky car to tune in the past for me, with turn in understeer and snap power oversteer, I enjoy the challenge it presents, and I have learnt a lot from playing with it) . Can post my full setup later if anyone cares enough to recreate my test.

 

[esoxhntr]

Thanks FussyFez, sounds like there's hope.

million dollar question is, was it faster around the track with 1.5 neg camber as opposed to 0.?

 

[Michael88]

I was testing on 1.03, if I had known that 1.04 was coming out so soon I would have waited to test camber on the new update.