Ultimate Driving Simulator, on MR Cars

[LS Chiou]

Suspension & LSD revisited.

Thanks to the story of DeltaWing, I learned something.

Tried [toe out at rear] and [very low LSD setting] on the following troublesome cars and got various success:

BTR, F40, Diablo GT & GT2, Audi R8 LMS, Cizeta V16T

To fully utilize the benefit of toe out and low locking, I think it'd better tune the grip balance as rearward as possible. So the suspension should be stiffened in the front and relatively softened in the back.

On these cars, I usually set the front spring rate to the stiffest and let the rear reduced to a similar number. By their 41:59 f/r distribution, the rear end is now very soft, relatively. (and the damping is proportionally following the spring rate, more or less) Also, the ARB is set higher at front and minimal at rear.

The toe out can be -0.2 to -0.5, tune as you like. You'd be surprise they remain stable.

As to the LSD, it's varied quite a lot among different cars. Following is what I feel right, for now (initial torq/accel/decel, for SS or RH):

BTR: 35/10/10
F40: 40/8/16
Diablo GT/2: 18/6/6
Audi R8 LMS: 25/10/10
V16T: 42/8/10

The lock ratios of acceleration side look so little, wouldn't they spin the inside tire when powering out of a corner? No actually, or it's only happening briefly at the slowest corner and the car with highest torque. The tire temp indicator grows yellow for a tiny moment, that's all. So I think it's pretty much enough for the sticky SS and RH.

Low locking makes the cars feel more nimble in direction changing with more natural reaction. They all present progressive lift off oversteer and very stable on gas in the bends. (V16T is still not fully satisfying, but at least OK now... )

As to the relatively high and very diverse initial torque settings, it's for adapting the inherent stability (or the lacking) of the cars.

Say, there's no unit and scale for the initial torque, so we don't know what it actually means. Would it be similar to add some proportion into the latter two? For example, 35/10/10 = 10/15/15, or something like that? Maybe, but I feel they react differently.

For comparing the tuning and proving it's not my own adaptation with driving, I switched the setting back to default and checked. They're indeed very different. I spun out at the first corner I tried with default R8.

Try it, and tell me what you think.

 

[LS Chiou]

Update: toe out at rear on the short wheel base RR (looking at you two, BTR & CTR) is not good on wet track.

Well, anyway, you'll know it instantly when you drive them out into the rain.

 

[JujiroMatsuda]

I had once thought I'd always put one less grade tyre on front on Yellowbird to overcome that oversteer. It solved all the problems and handled just like any other MR car; yes I know the compromise of less braking power due to weaker fronts but the benefit far outweighed the drawbacks. Nothing else was needed to be done, just SH on front, SM on rear, cheapest solution.

However, I thought "What if it rains and I had to wear same rain tyres on front and back?" Well, I didn't need to test to know what it'd lbe like - spin fest. So I had to come up with a setup that is drivable with same tyres on front and back.

Have the car the rigidity improvement done to it then try the below suspension and LSD.

SH on all wheels

100/100
5.75/8.25
4/5
4/5
7/4
0.0/0.0
0.00/0.20
Standard brakes 5/5

LSD: 20/5/20

The stability comes from ARB and LSD and I can drive this car without having to worry about spinning out.
If you want bit more oversteer lower initial and decel. If you want MORE oversteer raise the rear ARB as well.
If you want the understeer do the opposite.

 

[LS Chiou]

Without softened rear end for more grip, I suppose you are a better driver than me -- more precise and consistent in controlling its tail

Also, your ABR setting is much stiffer than mine. It would be disturbed more by the curb, I suppose. So you must be more precise than me.

Anyway, will try and feedback. Thanks for your recipe.

 

[eminem09494]

.

 

[LS Chiou]

Hi JujiroMatsuda,

Your setting is very good, pretty stable in general.👍 I can sense it's softened quite obviously compared to what I'm used to. Dive and squat are more pronounced but it remains pretty stable overall. It makes me think, this is probably more suitable for such a road car, instead of rock hard racing car settings.

I can do very good lap time, too, if I'm careful enough not to step out of the track. As expected, the stiff ABR disturbs the balance somewhat when one side of the car falling out of tarmac. Oh well, I'm not a precise driver.

After a few more laps, I can't help making some tweaks:

100 / 100
6.32 / 8.25
5 / 3
5 / 4
3 / 1
0.8 / 0.0
0.00 / 0.12

BB: 6 / 5

LSD: untouched

Try it if you will

Notes:

• Stiffer at the front and softened damping at the rear make overall grip balance more rearward, thus more tolerable and saveable in oversteering, also less nose dive when braking and less weight transfer.
• A little camber at the front reduces the front grip further.
• Slightly less toe-in at the rear compensate for the loss of nimbleness by the less front grip.
• Softened ABR is more tolerable when stepping on the curb.
• Not able to lock the SH tire on dry by the bias of 5, so I increase it to 6. The rear is kept at (a slightly less)5 to reserve larger margin before possible lock (bumpy surface or off road)
• LSD is spot on

After all, this car is still an RR. Going through a corner, it'd better be early in all operations.


@ eminem09494,

I'm afraid you missed the whole point of this thread, and also missed one of the most enjoyable part of this game. Pitty.

And, nothing is perfectly complete, or entirely broken. Everything is in between. From GT3 upto now, I honestly don't think the physics of GT6 is gone to the side of entirely broken.

Some cars are indeed difficult to drive in their stock form, but that is not necessary a fault.

Cars are very sensitive to tunings, with sensible responses, too. For example, in the case above, the response to the tiny change of toe from stock 0.2 to 0.12 is clearly detectable. Such delicacy didn't happen in GT5. In this regard, GT6 is an improvement already, if not perfect.

 

[sumbrownkid]

@eminem09494: Try Ridox's tune for the R8 LMS. I'd put the link for you but my phone is acting up. He's in the tuning forums.

 

[eminem09494]

.

 

[JujiroMatsuda]

Without softened rear end for more grip, I suppose you are a better driver than me -- more precise and consistent in controlling its tail

Also, your ABR setting is much stiffer than mine. It would be disturbed more by the curb, I suppose. So you must be more precise than me.

Anyway, will try and feedback. Thanks for your recipe.

I was nearly going to say I'm so flattered to hear such compliments but now you realised it's due to a good setting, not me.lol I'm pretty sure the rigidity improvement does a good job distributing some of that undue frontal grip to the rear.

My basic approach to suspension setting is that I first decide how much the total sum of spring rate I'd give to the car. For this instance I decided to allocate 14kg of combined spring rate and I just simply divided the amount relatively to weight distribution of the car, ie 41% to the front and 59% to the rear. Hence, 5.75/8.25. In this way theoretically both front and rear tyres should get equal load when cornering. How much the combined spring rate I should give that I don't know and I wish there is a mathematical method to come up with that number.

If I decide to stiffen the front suspension I'd simply increase the front value and subtract the exact amount from the rear so the total value is always be 14. The problem with this method is that the suspension travel of the front will also change and whenever I change those values the amount of front's dive will always be different. If you want the front to be constant then stick with some value you're happy with and only change the rear. But then I wouldn't know what the good value for the front - how much is too soft? how much is too hard? - it gives me headaches so I had decided to stick to my plan A, the whole sum of spring rates always be the same.

I let the ARB to sort out the handling balance first and when problem still persists then I tweak the F/R spring rates. In this case I gave front the 3 extra value and the car's handling was good enough so I left them at that. I prefer the 7/7 ARB if possible. The value gives me the most grip from all four tyres. If I want to ride over curbs or cut corners I then must compromise to settle at lower ARB value, if cutting corners worth that much that is.

100 / 100
6.32 / 8.25
5 / 3
5 / 4
3 / 1
0.8 / 0.0
0.00 / 0.12

BB: 6 / 5

LSD: untouched

I'm curious how you came up with that front spring rate? Is it done by some calculation? is 6.32 better than 6.30 or 6.35?

 

[OdeFinn]

@JujiroMatsuda, mathematical approach is one nice way to explore car, on spring load there might be also considered to calculate downforce generated by wings.
GT3 GTR generates over 1000kg of downforce on speeds above 125 km/h, problem is that at are those numbers on settings directly downforce on kilograms or some other? I'm guessing at those are kilograms directly, going to test that assumption on mathematical tuning soon.

http://carwitter.com/2014/01/18/building-the-nissan-gt-r-nismo-gt3/

 

[g0ld]

The Diablo GT? I think it handles great, no issues. If you mean the Diablo GT2, part of the issue is due to the very high stock camber the car has, and as you might or might not know, in GT6 camber is currently bugged, so negative camber is actually positive camber. Imagine driving a car IRL with almost 4 degrees of positive camber front & rear. You can also try tone up the downforce a bit.

 

[g0ld]

The Diablo GT? I think it handles great, no issues. If you mean the Diablo GT2, part of the issue is due to the very high stock camber the car has, and as you might or might not know, in GT6 camber is currently bugged, so negative camber is actually positive camber. Imagine driving a car IRL with almost 4 degrees of positive camber front & rear. You can also try tone up the downforce a bit.

 

[g0ld]

The Diablo GT? I think it handles great, no issues. If you mean the Diablo GT2, part of the issue is due to the very high stock camber the car has, and as you might or might not know, in GT6 camber is currently bugged, so negative camber is actually positive camber. Imagine driving a car IRL with almost 4 degrees of positive camber front & rear. You can also try tone up the downforce a bit.

 

[Bo]

Any overly twitchy car on throttle can be tamed somewhat by setting the initial torque fairly high.

 

[LVracerGT]

Any overly twitchy car on throttle can be tamed somewhat by setting the initial torque fairly high.

It can also be tamed with your right foot.

 

[Bo]

It can also be tamed with your right foot.

DS3 users find that one a bit tricky

 

[LVracerGT]

DS3 users find that one a bit tricky

Ah, yeah I forget people still use a controller. I haven't played GT with a controller since GT2

 

[sumbrownkid]

I did try that but the car still wants to spin on corner entry
Same thing happens with the motorsport elise

Hmm, I'd say to add ballast to the front until it becomes more stable but I don't like the idea of adding more weight to a lightweight car.

I'd take this up with Ridox since he knows tuning way more than I do and he can help you out with this car more thoroughly than I can.

 

[eminem09494]

.

 

[JujiroMatsuda]

...some MR cars are now very unstable under braking
For example the MP4-12C imho is perfect in both games

Is it because that car has an active rear wing that works like a parachute under full braking, pulls the car from back end?

 

[LS Chiou]

....
I'm curious how you came up with that front spring rate? Is it done by some calculation? is 6.32 better than 6.30 or 6.35?

That number is just a result of trial & error So I guess 6.30 or 6.35 would be equally good and hard to distinguish.

It's not based on any calculation, just a gut feeling according to the original f/r distribution and the degree I want the front end to be stiffened. The grip bias would incline to the relatively softened side. Or the opposite view is the relatively stiffened side would loose some grip. For this inherently tail happy car and my not-so-good controlling skill, I need more grip at the rear.

One more reason is I'd like to have less nose dive and an overall quicker response, so I stiffen the front, instead of soften the rear.

Your method of fixed total spring rate is new to me and sounds reasonable and clear to evaluate in the tuning process. However, I found the spring rate of default settings (be it OE or fully adjustable race mod) rarely stick to their f/r distribution precisely. There're all kinds of combinations.

In addition to the possibility of tuning for handling response, it can be largely affected by the motion ratio, too. I'm not sure if PD's suspension modeling reached here. There're many suspension designs with the spring connecting somewhere inward on the suspension arm, thus less travel than the wheel (at the end of suspension arm). In such case the spring rate must be contrarily stiffened to have a proper supporting force on wheel.

For example, the spring in a multilink rear suspension can be connected at the half way of the arm, so it needs double the spring rate to get an effective stiffness of that on a MacPherson strut. In a double wishbone suspension, the spring can be mounted in an oblique angle, so it needs higher spring rate than those which are upright, too. As to those push rod designs on race cars, there're all kinds of motion ratios...

If PD's modeling do cover this portion, then the straight numbers of spring rate would not reflect the effective stiffness accurately. And we can only take the relative comparisons to get what we want.

 

[LS Chiou]

@OdeFinn,

Yes the down force is a real problem in spring rate setting. Most race cars needs very stiff spring at the rear to avoid being pressed down to the bump stop. However this affects their behavior in lower speed range. I think one of the problem in Audi R8 is here. Always dilemma.

This is one of the reason I like sport cars more than race cars in GT6.

 

[JujiroMatsuda]

@OdeFinn

Ah that's right I forgot about downforce spring depression. Cars like R8 LMS with its huge rear wing the faster it goes the more depressed the suspensions. You get 4 times more downforce at speed of 200km than when you're doing 100km, right? Like LS Chiou, I haven't driven racing cars lately I'm grinding mainly the super cars nowadays to fiddle around with settings as they are without DF factor, less thing to worry about. I'm still learning the workings of suspension mechanics but I can get the car to behave how I want them now.

For example, the spring in a multilink rear suspension can be connected at the half way of the arm, so it needs double the spring rate to get an effective stiffness of that on a MacPherson strut. In a double wishbone suspension, the spring can be mounted in an oblique angle, so it needs higher spring rate than those which are upright, too. As to those push rod designs on race cars, there're all kinds of motion ratios...

If PD's modeling do cover this portion, then the straight numbers of spring rate would not reflect the effective stiffness accurately. And we can only take the relative comparisons to get what we want.

Very good point there I doubt any spring rate in this game represent direct effectiveness of stiffness. For example if there is a car with 50:50 weight balance and both front and rear suspension rate at 10kg each. The rears have straight forward upright suspensions close to the wheels while the fronts have diagonal ones placed more inward, connected halfway on arms then we wouldn't get the same stiffness, would we? The front springs must be much much harder to achieve the same stiffness as the back ones.

Of course this is only the case if PD has modelled realistically. But you know what? I don't think they are that good. If they were that good they wouldn't neglect that ride height and camber physics still untreated (they shouldn't have released the game in the first place if you ask me) while they've worked their ass off to patch up that money glitch over night So chances are, suspension figures might reflect direct stiffness as if all suspensions are connected on wheel hubs at perfect vertical angle

edit: I forgot about that MP4 beast I just bought one and tuned it. It has almost-too-good-to-be-true handling! I can abuse the car so much just whip the crap out of the poor thing and it just goes faster and faster!

 

[OdeFinn]

@JujiroMatsuda, yes to that for times higher downforce, if I have understood this right:

..more testing waiting.

 

[Rotorist]

Having driven nearly all types of cars, at speed, competitively, I'm going to Eco a couple things I've said in other threads...

Snap sensation (unintended reaction) - falling off the cliff, loosing control quickly "something broke"

Kick sensation (intentional action) - sudden abrupt motion within the performance envelope that is induced by the driver. You try and kick as a reaction to a mistake, and you can over to it.

Drifters kick their car sideways, not snap them. Its a skill, not a reaction.

There is a reason racers say; "the car got away from me..."

And again; "nothing in auto racing 'snaps' unless there is a mechanical failure or oil on the track.

In real life;

FR cars are the hardest as they 'push' power from a lighter rear to a heavy front. There is a sizable envelope to correct. And a balance that is clearly felt through you're butt and adjusting angle of attack can be manipulated rather predictably with throttle. More control is the trade off for more things to balance (racers mostly prefer FR, as control and skill is their main tools)

Thank you for all this info mate. I just want to add that many modern FRs (i.e. Mazdas, BMWs) have almost 50-50 weight distribution that makes them a mild MR without the negative of sudden movements around the centre of gravity.

 

[Lawndart]

Thank you for all this info mate. I just want to add that many modern FRs (i.e. Mazdas, BMWs) have almost 50-50 weight distribution that makes them a mild MR without the negative of sudden movements around the centre of gravity.

Happy to share and glad it's appreciated.

50/50 weight distribution is only 50/50 when the car is not moving.

But the difference in MR vs FR feeling is still vastly different because of where the center of mass is and how it reacts to circumstances.

In the real world, with two comparable cars, one FR, the other MR, both 50/50... the FR car will 'dive' under heavy breaking. An MR 'feels' more like it "squats" in the same situation. This is because of the heaviest part of the car still wants to go forward... One pushes down more directly on the front axels, the other transfers that force through more of the frame and pushes the axels from a lower center of gravity...

I don't remember if I mentioned this here in this thread or a different one but my racing style in an FR car is 'rear wheel' steering... I threshold brake, at some magic point (pure instinct, and always variable) I force the car to rotate as its decelerating... When done right, I have a much more aggressive angle of attack when approaching the apex that enables me to get on the throttle before the apex... Being on the gas seconds before others adds up to higher speeds prior to reaching the next corner...

This style doesn't work the same in an MR car, and completely the opposite of how to drive an RR car... All because of weight transfer, not static weight distribution.

So 50/50 is always a good neutral starting point for a car, but race cars in particular never stay that way after a shake down...

But to your point, that there is inherent instability at the highest rotation point of a car, this is because of the shift of weight through the point of neutrality from one extreme to the other... The idea is to tune a car to be 50/50 at that point so the transfer is as linear as possible and predictable...

I'm not a bit fan of Vettle, but the truth is he is a master at 'low speed corners' and his ability to control that transition better than most. That is his 'trick' that differentiates him from the others... Maybe there is some TC in there too but we'll never know ;p

Here comes a story! Just skip if you don't like reading... It's all based on supporting why I rear wheel steer... Without really mentioning it.

On a more personal note, One maybe you guys can apply to you're own GT game, but based on my real world exp; I race at Laguna Seca... Ironically I'm not a fan of the track in either games or real life but when I win there it's all done in 2 corners. Turn 2 and turn 11. If you as a driver are better at every other corner than me, I'll still beat you because of those 2 corners... Turn 6 is the biggest wild card at Laguna and its predictable that drivers will try to make up most of their time there... And it's for good reason as getting a run on the uphill is significant... But even if I botch turn 6, I can still make up for it in turn 11, and if not then, in turn 2, supported by turn 1 (not really a turn)... Its a one-two punch. When I have a helmet on, I feel that I own those corners...

All because I seem to be able to rotate the car faster and more aggressively without loosing speed... But even if I do loose speed, I forced them to the outside and killed there momentum and all my attention goes to who is behind me, who will try and capitalize on both of us slowing (this is so common)... All I need is to have my nose ahead going into turn 1 and if the other driver is brave, he'll be pushed off the track, if he is smart, he'll lift in fear of me drifting 'over the blind hump' from taking two direct of a line (its phycology)... I'll then steel turn two, again, if brave he will try and dive to the inside, I'll let him have it because he will be forced to brake harder, then I'll rotate the car for the second apex, ensure I own the outside line, he won't get more that 2 inches of space between us, again forcing him to yield as he can't track out as I'm already there, shift to 3rd and take back the racing line... If he is better than me then this will play out differently every lap, if not he/she will be too shook up to want to go through that again without a plan or some advantage to counter each/one move... Or have a faster car...

Am I an asshole? I'm a race car driver... If you get between me and my flag, be ready for the horns... I'm ready for yours. Fear turned into determination is the most powerful human emotion I've ever known... And will likely lead to my demise someday...

Oh, and a tip! - the hardest turn to get used to at Laguna coming from so may laps in games as a kid was turn 4... In game it's like a turn, when you hit it right you track out and go through a bend up to turn 5... Not at all how it is in RL despite that game modeling being quite accurate... I had to learn how to take turn 4 as one massive turn through the bend and plan it before entering turn 4... In games, you turn, level out, ignore the bend as your charging into 5... In real life, you charge all of 4 with the bend being turn exit and it feels like a very short stretch to line up turn 5 before going heavy on the brakes. Try to approach turn 4 like we do in real life and report your results... The difference will surprise you in any pp car... If you find your flying off the track at true turn 4 exit then your almost doing it right, but your not entering 4 early enough.

Can we get an update to shut me up?!

 

[LS Chiou]

👍 We NEED this.

 

[JujiroMatsuda]

@JujiroMatsuda, yes to that for times higher downforce, if I have understood this right:
View attachment 118248

..more testing waiting.

Yup I'm pretty sure you got that formulae right as I have no bloody idea what that means.lol


@Lawndart
Your know-hows and invaluable information coming from first hand experience driving competitively IRL on track look few levels too high for an amateur like me to digest and understand fully:tup: Please keep pounding us with more information.

I'm pretty sure PD is watching as well as this is a free source of information as I sometimes feel reluctant to share some of my findings I'd rather do that through personal emails as I feel like I'm being used as a dedicated debugger, doing their jobs for free. If I ever find such a profitable bug like that money glitch I'd spread it through PMs and tell all not to put it on internet. No, PD, I don't want to do your job for free when you're should be doing it since it's your job and you get paid to do it.