Understanding The Deltawing

  • Thread starter diptob79
  • 214 comments
  • 37,262 views
Yes indeed! Nice finding. It's very different between this 3-point configuration and ordinary 4-corner -- 3-point is always on a flat plain.

I think one of the key factor is the corner weight. On 4 wheeled cars, the 4 corners are bumping up and down constantly and randomly on track. Much of the time, there'd be imbalance of loading between two diagonals. It's like the bumps are poking the 4 corners individually (from the ground), and each poke is trying to flip the car.

On race cars, among other things, there're all sorts of dilemmas in suspension setting. For tarmac, it's usually very stiff, and also with high torsional rigidity (for compliance of suspension, not body structure here, although it's rigid too). This is mainly for minimum weight transfer, fast reactions and not crushed by the high down force.

And the situation on Sarthe is difficult. It's too bumpy as the cars are running so fast here. It can't be too soft for other reasons, so inevitably those cars are bumped all over the place -- in the entire track, and at all 4 corners of chassis.

OTOH, the DeltaWing sidestep this problem so nicely by its 3-point basic shape (2 close front wheels considered as one contact patch). There's no diagonals to be bothered, only side to side rockings which do not affect the grip and stability. Another strong point of this car.

Yep! This explains all. I like your analysis very professional. 👍
 
Hi KOGSoldier, Glad you like it. The original tuner ULTRAVIOLENZZ is also a DS3 users.;)

Don't worry about the small adjustments on those numbers. I think, within a certain range, the effect from a tiny adjustment is also tiny, which would probably be largely (sub)merged in the adaptation of driving. Unless the goal is to slash 0.0x sec. lap time, or I won't care too much about the little differences of tuning numbers. I consider the interaction is more important, which is pretty much set by a larger picture of tuning -- the principle of the whole combination.

In this case, the principle is making the rear relatively more grippy, by stiffening the front and softening the rear (in a view of f/r balance related to the default setting). Larger camber in the front reduces its grip even further. This moves the character from oversteer to under. (under is a relative term here, obviously)

Not yet have the time for trying your numbers (it's daytime and I'm working now). The major difference of your setting, as I can see, is the rear toe. This is one of the mystery I don't understand on this car.

According to the description on the left panel of tuning page --also in sync with common sense IRL-- positive number means toe in and negative is toe out. Toe in is good for stability but not for agility... etc. So far so good and this rule applies to most cars nicely, at least from my experiences (in GT and IRL). But on DeltaWing, it seems not the case.

My eyes popped and jaw dropped (as the avatar) when I first saw the -0.5 toe out on both front and rear in the tuning by ULTRAVIOLENZZ. I myself would never choose such a number for stability. But it just works, keeps very stable even when braking deep into the corner on wet track! Odd but true.

In the 24 min Le Mans, I decreased the toe out somewhat mainly for peace of mind, instead of the actual effect. Afterward, I tested more in free run, fiddled with the LSD and toe in Brands Hatch. It seems the effect of toe on this car is very small. Maybe it's somehow beyond my perception, or (sub)merged into the adaptation of driving.

On other (normal) cars, large toe out at rear would let them swing sideways when lift off or braking. At least it adds such tendency. But DeltaWing seems immune to this. I've tried -0.5 to +0.2, nothing significant was found. Odd. Can't explain. (Or I was so confused, lost in all those numbers? :ill: )

As to LSD, such low setting seems more or less like an open diff. By common sense, it can't provide the proper locking ratio to prevent inside rear spinning when accelerating out of a corner, also no (or little) stabilizing pull when decelerating. Again, this special car seems exempt this requirement to a very large extent.

And it seems the low LSD setting violates what I said in post #139. Yes, or maybe no, partially. I sort of figure out a theory for this.

Already very long. More later.

Any comments are welcome.:gtpflag:

It sounds like you know your stuff and you are much more versed in understanding the adjustable parameters in setting up a car...I need to get my game up...I'm always open to learn something I like car and bike racing games so trying to go fast on the limit is my blood...look fwd to chatting more and thanks again LS Chiou and Ultraviolet for sharing helped me out a lot
 
Here is what I think.

Toe first. Ideally, a tire trace precisely true only when the car is going dead straight with toe=0. In most other cases, some lateral slip happens inevitably. Toe in at the rear makes the tail stable because the loaded outside tire is pointing inward to push the tail back in the line. So far so good, but what if the [additional slip angle] by the [toe in] is too much?

A tire's max lateral grip happens at a certain amount of slip angle. This optimal working point is varied by the tire. (Also, different tires have different characteristic curves -- a curve describes relation between lateral grip & slip angle.) Anyway, when the slip angle is larger than the optimal point, grip drops. That's when the car slides too much to be saved. (If the slip is more or less the same at both front and rear, the car would do four wheel drift. Drifting too much or the line is not good, it's going out of track. Unequal slips make it wash wide or tail out etc.)

When a car is driving hard along a curve, there're some slips at all four tires. And the conditions of loaded outside tires pretty much dominate the stance and trace of the car.

With toe in at rear, the steering angle in front must be larger to keep the car tracing the required line. Otherwise, the tail tucks in, the front would be pointing outward relatively. Meantimes, the partially loaded inside rear is pointing outward. Although its grip is less than the much more loaded outside tire, it still contributes to the total lateral grip. Now it's not helping, it's doing the opposite. And the more inward angle of outside tire could be approaching or even going over the optimal slip angle in some violent changes of direction.

So, a toe in rear setting is actually making the 2 rear tires fighting with each other, and also need more help from the front to maintain the line. It could use up the usable slip angle sooner. That's the price we pay for a little more inherent straight line stability.

What about toe out? The much loaded outside tire is pointing outward, it seems the car would be going sideway. It might, but if the grip is still enough to track true, this just alters the stance of car (related to the trace). Similar to 4-wheel steering with opposite phase on f/r. Meantimes, the lightly loaded inside tire is pointing inward, is giving help to the overall lateral grip.

At this stage, I can have a (preliminary) conclusion: toe out is good for cornering (and also the stability in corner), with a condition of ample rear grip to keep the tail in check.

I examined this thought on some tail happy cars -- BTR, F40, Diablo GT/GT2. I tried -0.2 ~ -0.36 on them. Guess what? Spinning like tops? No, they are very drivable with toe out! I adjusted their suspensions as below:

Spring rate --- front: stiffest / rear: softened to be the same as front
Damping--- follow the spring rate proportionally
ARB---- front: 3~5 / rear: 1
camber--- front: 1.0~1.2 / rear: 0

Yes, lift off oversteer is still on them, but not snappy and easy to control. Actually, I like progressive lift off oversteer very much. When in a series of corners, steering by throttle is a real joy. With such setting, these cars are rock stable in corners when you're on gas, tucking in when you lift. Line is freely adjustable by the throttle. Some are more tolerable than others on the degree of lift. Anyway, they are all drivable, enjoyable and not requiring full concentration all time to keep them on track. I love them yet again, by this newly found aspect.

In my distant memory of GT3, those F1 cars have large toe out at the rear. I was surprised at first, but they all drive very well with the help of massive grip of tire and the DF.

I should have thought of it and tried. Thanks again ULTRAVIOLENZZ, if it wasn't you, I won't rediscover this wonder.

And yet, the setting is not completed -- the crucial LSD. That will be in the next post.

Again, any comments are welcome. :)


Edit: it's BTR, I typed CTR previously. They are similar, though.
 
Last edited:
LSD 101.

When a car travels along a curve, the inside wheel rotates less while the outside is doing more. It needs a device to deal with the different rotating speeds, and sending the torque to the left and right driving wheels at the same time. This is a differential.

Most low power road cars have open differentials, by which the two sides are rolling individually and sharing equal torque from the source (engine and transmission), until one of them slips. This usually happens in a corner when the driver push it pretty hard. At this moment, weight transfer makes the partially unloaded inside tire get lower traction, so it might slip and start spinning freely if the torque is large enough. When this happens, almost all torque from the source is 'leaked' through the fast spinning wheel, and the other side shares almost none, so the forward thrust ceases. Simply put, a car cannot accelerate out of a corner when one of the driving wheel starts spinning.

To avoid such 'leakage', we'd better do something to stop such free spinning. One of the methods is to link the two sides mechanically. The most simple link is a solid shaft, like the one on a go kart. It's not very practical for other bigger cars because a go kart actually lifts its inside rear wheel off the ground in a corner with proper speed. (It's by the design of the steering mechanism in front. If its inside rear isn't lifted off ground, it'd be very hard to turn because two tires must be rolling at the same speed by the common shaft -- no differential to allow different rotating speed.)

So, the mechanical link between driving wheels can be open (0 lock), or a solid shaft (100% lock). These two extremes are not good enough for all conditions. We need something in between -- a Limited Slip Differential.

There're several types of LSD. The fully adjustable one in GT is probably a mechanical clutch-type. In which there're three crucial factors about how it works -- initial torque, sensitivity (lock ratio) of acceleration, and that of deceleration. To shorten the already long post, here is a good read. (more on those 3 numbers later... )


Rewind a little, we need some locking between left and right because one of them may let go when:
  • the torque is higher than a degree
  • at the moment, grip of either side is lower than the torque

So, on the contrary, we don't need them locked, or just a little is enough when:
  • the torque is not that high (related to the tire's grip)
  • the grips of both sides maintain more or less consistent, and is above the driving torque

The (partially) lock between left and right is a force (or tendency) which wants to keep both sides rotating at the same speed. But in a corner, they're just naturally different. So the locking is actually a bad thing when we don't need it.

Imagine a go kart which is forced to turn. The inside tire wants to roll less naturally, but it's locked to the other side which wants to roll more. While the outside is doing the opposite. So the two are fighting with each other along the curve. A lot of friction (slip) must happen between both tires and the ground. That's why the resistance is so high.

Thanks to the limited slip (not a solid tie), when using an LSD, the fighting is less. And the friction should mostly happen within the LSD. But it might also happen between tires/ground if the locking force is strong enough (more lock, approaching that of a go kart). Or the two conditions happen simultaneously or alternatively.

Anyway, it'd be worse when any of the friction is getting more. Waste heat and tire wear aside, the locking force is actually interfering the natural rolling behaviors of tires, thus an interference of the handling. Yes we do need such 'interference' for helps in some occasions, but otherwise it'd better be minimal.


Back to the DeltaWing (whew~ finally), the reasons it can get away (or is doing just fine) with low LSD setting are just mentioned above -- the racing tires are grippy, the engine is not that powerful, and the minimal side to side weight transfer keeps the grips on both ends pretty consistent. So it just needs a little lock to avoid slip.

Still some questions:
  • Why is the higher locking setting doesn't work on wet condition?
  • With such 'less is more' concept in mind, does this low locking principle work on other cars?
  • What about the 3 magic numbers - initial torque, acceleration, deceleration?

Already too long, more later...
 
Last edited:
Let's finish this stuff.

Here's what I think.

In my previous setting with higher LSD locking ratio (20-sh to 30-sh), the car is stable, but somewhat 'sticky' in those low speed corners. Such sticky feel is presented as slow reaction at turn-in, and reduced agility in the fast left-right transitions. It's OK, or even useful on dry track especially facing crowded and aggressive competitions where you'd need a car that is thrash-able.

In the 10 lap Silverstone, I used such 'sticky' setting to win, easily. Late brake all the way, very very late at some corners. Also very early throttle out of them. It's doing fast in fast out! And this DeltaWing stick to the ground almost like those X-cars.

But on the wet, such degree of lock between 2 wheels is too hard. It probably makes them slip, like the fixed shaft analogy in the previous post -- outside tire want to roll more naturally but is held back; opposite on the inside. In the bend, they are already fighting each other with some slipping. At this moment, weight transfer to the front by lift off or braking makes the tail floating and the rear tires finally let go, it spins out.

OTOH, less interference of low LSD setting let the wheels rotate naturally, so they maintain traction nicely. This makes perfect sense!

So, does this apply to other cars? Yes, more or less...
 
I always tune my cars in low lsd. I always start at 5/5/5. I always test braking sensitivity in long straights like in Sarthe, the acceleration sensitivity in curvy tracks like Nurb and finding the sweet spot for initial torque that will work for almost all the tracks. Also maybe you want to transfer your thoughts and make a new thread about this? I think most of the people here already unwatched the thread. It's a bit old thread so you will have new readers.
 
Good advice on the setting. It's a back to basic practice. This builds a solid foundation for the tuning. 👍

About a new thread, I did a similar thing, I had another post here for the extension -- applying such tuning concept to other cars.
 
Lower drive lock on the LSD helps, I usually go for about 15-20. Then a bit more, say 25-35 for retardation lock, to aid stability under braking. But the main thing with the Deltawing is just driving style. You can't throw it into the bend slightly too fast and scrub off speed with a bit of neutral-to-over-steer slippage (which is a pretty common style in GT6), it just drifts into the barriers or spins if you have the space. You need to be conservative on the corner entry, find your line (with no double bite at the apex), wait for it to settle fully, and then you can get back on the power. You can get on early, but it needs to be progressive. I've found the Deltawing to be fastish when driven like this, and awesome on the straights - but generally not quite a match for full LMP or Group C cars at the same PP.

But it was all I bought to get me through all the LMP type races in iA and S levels, didn't have anything else. With Power mods and conservative corner entry, she flies :)
 
Try increasing the initial torque on LSD, you'll find it easier and more tolerable in the entrance of corners. Much larger operating window. ;)
 
This is the WORST-handling car in the entire game, at least if you're unfortunate enough to only have a dual shock controller. With a wheel I'm sure it's a piece of cake.

I bought the 2013 model, strapped on a set of Sport Hard tires, and took it straight to Suzuka for my own testing purposes, and no car has made me want to pick up my TV and throw it out my window more than this one. After 15 straight minutes, I managed to get one clean lap, ONE! JUST ONE!

**** the DeltaWing. What a piece of ****.

It's quicker than a Veyron and FXX, and about even with the Aston VGT, if you're good enough to avoid losing control while at the same time not losing too much speed.

My best time was 2:01.683 and I am never driving this "car" ever again. Waste of 2 mil credits.

 
Last edited:
I drove mine for the 1st time in quite a while, and found it to be a dream to drive. After 1.09, it messed it up, but I put a similar setting to the old physics, and it is so stable.

You still have to watch the curbs though.
 
This is the WORST-handling car in the entire game, at least if you're unfortunate enough to only have a dual shock controller. With a wheel I'm sure it's a piece of cake.

I bought the 2013 model, strapped on a set of Sport Hard tires, and took it straight to Suzuka for my own testing purposes, and no car has made me want to pick up my TV and throw it out my window more than this one. After 15 straight minutes, I managed to get one clean lap, ONE! JUST ONE!

**** the DeltaWing. What a piece of ****.

It's quicker than a Veyron and FXX, and about even with the Aston VGT, if you're good enough to avoid losing control while at the same time not losing too much speed.

My best time was 2:01.683 and I am never driving this "car" ever again. Waste of 2 mil credits.



Wait a moment mate! You say that you tried a racing car with SH tires and without any tuning to adjust the change is tires and because it was terrible in the way you tested it -as expected because of its weird weight and grip destribution- you rejected it? Why don't you try it on RH tires and judge it again?
 
Suzuka is a very technical track to first test this car out on.

Slap some racing tires on, adjust downforce, add weight to the front and then try it on a more neutral track like La Sarthe or Monza. Get a feel for how it's gonna act, then tune it for a tough track like Suzuka.

I really enjoy this car and use it whenever I can. It tore up the 24 minutes of LeMon and I just used it to knock out the most recent seasonal event. It kicks a**!

Also, for me, seat time in this car makes a huge difference. It is unlike anything else I have ever driven and I had to adjust to its style.

Seat time, seat time, seat time. Don't bury it in the stockyard. Use it. Good luck to you sir.
 
I've tested 600 cars completely stock (except for SH tires) at Suzuka and haven't had as much of a problem with any of them. Even the Ferrari FXX, Cerbera Speed 12, Veyron, and Zonda R were more manageable on SH tires than this car.

I'm sure with RH tires it's better, but it doesn't belong anywhere near a track that isn't dominated by straightaways.
 
If you had read through this thread and also tried proper tuning (LSD is the key), and it's still not driving properly, I suppose the root cause is not in the car itself, but elsewhere...

In the previous seasonal event of Spa and LMP cars, among all Le Mans racers I've tried, Deltawing is the most capable car in cornering. (And I'm just a mediocre driver.)

It's not a 'traditional' race car, but a low power, low drag, high efficient concept. Its strong points are in different spots.

Try to find it and enjoy it.
 
Try it stock, on SH tires, and tell me it's not the most annoying car you've ever driven.

Don't tell me that the car needs to be modified, because I can handle a stock Veyron, FXX, Zonda R, Speed 12, etc all on SH tires with nowhere near as much of an issue as this car.

A car shouldn't need to be modified to be usable. The fact is the DeltaWing was not designed well. It would have made much more sense to make it a traditional 4-wheeler.
 
Last edited:
It feels better on Racing tires, because, well, race car.

I'm not a tuner, but even I can figure this car out.

I understand that it's better on Racing tires, but even a stock 787B on SH tires is easier to drive.

I challenge anyone with a controller to beat 2:01 with a stock 345 hp Delta on SH tires at Suzuka. It's not me who sucks. It's the car. Because it has only one wheel up front. What on earth possessed them to go with that design? How is it better than having 2 wheels up front? Notice how no other manufacturer makes 3-wheeled racecars? Because they know better.



No need to get butthurt over my hatred for the Delta. If you like it, hey, enjoy it. I just want nothing to do with it, or any 3-wheeler, ever. 2 or 4 wheels for me, thanks.
 
Last edited:
Read the in-game description, then you'll know why it was designed that way.

I have a Fanatec CSW, and there is no way in hell I'll try to drive a DeltaWing on SH tires. It sounds a bit daft to put tires that work on a hot hatch on a sports prototype.
 
I only did it on SH because I needed a single tire choice to cover every car in the game for fair testing purposes. I suppose for the true racecars I'll have to resort to RH tires...but I still refuse to tune/modify them in any way. I like to keep things stock unless I'm drifting.

Anywho, tomorrow I'll retry on RH tires and see if I like it better.
 
... IMHO Sh are the tires to go (3:45 at LeMans)
and please, guys, don't use the so called ABS -
the real Delta Wing hasn't even the real one (!)
(according to Chris Harris in the linked vid) r.
 
Last edited:
This is the WORST-handling car in the entire game, at least if you're unfortunate enough to only have a dual shock controller. With a wheel I'm sure it's a piece of cake.
You have to change your driving style for it.. how you brake with it, your driving line, etc - you can't expect such an oddity to be something of perfection. And with a wheel, it's absolute fun. The car isn't easy. But it's fun.

It's fun, because it's an oddity. It's different.
 
it's best to have two wheel at the front, one wheel at the rear and make it like a water drop aerodynamic design

the delta wing is more aerodynamic than the reliant
maybe it will fly better
 
Yeah, there're 4 wheels in Deltawing, but its layout and operation is almost a 3-wheeler. I'm not sure why there're 2 skinny front tires, maybe it's the regulation or safety concern. ( I could've read it somewhere but forget now...)

Anyway, I believe its character will remain pretty much the same if someone modify it to one front tire.

Some jokes are not harmful, they bring relaxation. But I can't help feeling bad to see people bashing a nice peice of work by some people with huge passion and hard work.

http://www.deltawingracing.com/history/

http://www.topgear.com/uk/photos/top-gear-meets-the-deltawing-2012-05-23

I'm really glad GT6 introduce this car. It means much more to me than those newest hyper cars and such.
 
Yeah, there're 4 wheels in Deltawing, but its layout and operation is almost a 3-wheeler. I'm not sure why there're 2 skinny front tires, maybe it's the regulation or safety concern. ( I could've read it somewhere but forget now...)

Anyway, I believe its character will remain pretty much the same if someone modify it to one front tire.

Some jokes are not harmful, they bring relaxation. But I can't help feeling bad to see people bashing a nice peice of work by some people with huge passion and hard work.

http://www.deltawingracing.com/history/

http://www.topgear.com/uk/photos/top-gear-meets-the-deltawing-2012-05-23

I'm really glad GT6 introduce this car. It means much more to me than those newest hyper cars and such.
I never read anything about three wheelers not be allowed in both the LMGTE and LMP regulations from the FIA (I've read them three times...) but certain parameters can't be broken so it's a bit complex... Mostly why it's a garage 56 prototype test car allowed by the FIA
 
All I can say is t's all in the tune. You have to tune it to prevent the two front tires from becoming the "pivot" of the spin out when turning and braking. I'm looking forward for a closed cockpit version of the '12 and 13' Deltas and new ZEOD in the the next updates.
 
Last edited:

Latest Posts

Back