FF tunes don't make any sense

[oopssorryy]

I've been lurking this thread all day and I thought I'd give it a go, so here is a '99 Celica tuned "realistically". I used my 450pp tune as a base and just messed with the suspension, so if you want ot replicate it go here and copy the parts list.

Toyota Celica SS-II '99

Suspension

Ride Height
Front: -20
Rear: -16

Spring Rate
Front: 6.0
Rear: 15.0

Dampers Extension
Front: 5
Rear: 10

Dampers Compression
Front: 2
Rear: 6

Anti Roll Bars
Front: 1
Rear: 7

Camber
Front: 1.5
Rear: 0.5

Toe
Front: 0.00
Rear -0.20

Fully Customizable LSD

Initial Torque: xx/6

Acceleration Sensitivity: xx/12

Braking Sensitivity: xx/9


Tires
Front: Sports Soft
Rear: Sports Soft

Brake Bias
Front: 6
Rear: 3




Though it doesn't get any lift off oversteer, or really any oversteer at all, but it doesn't understeer either. Why does stiff front and soft rear work well, I don't have your answer, but it is possible to create a good (or at least I think so) tune that goes by the "rules".

 

[Rotary Junkie]

I'm sorry, but some things really are that simple. If you double the spring rate and match the dampers on the rear of ANY real car, (FF, FR, MR, RR, AWD, doesn't matter) and leave the front the same, the car will very noticably oversteer more / understeer less.

Provided the front was already set up properly, yes.

All the other variables do work together, but simply changing that one item should make a very noticable and predictable change.... even when the maximum spring tension on a "race" suspension is only 15kg/mm....

And it does! It never seemed to work out before Spec II but more recently I've tried things like 9.5/14.0 spring rates (or thereabouts) on the premium Megane... And gotten lift throttle/entry oversteer. If I went lower on the front springs, the car would snap into the corner and then scrub wide. Lower still and it just plowed. Higher, it calmed down the oversteer but didn't actually lose much in terms of cornering speed.

But, fine, GT5 does it different... "play the game" you say. I could go along with that, IF everything else didn't have to work together! Stiff rear spring makes oversteer? Fine, so what about the anti-roll bar? Anti-roll bars allow you to stiffen one end of a car in roll without stiffening the car in pitch, or over bumps. Great for bumpy tracks.... well now those have to be backwards to line up with the GT5 physics too! Stiffen up the axle thats loosing traction, thats the GT5 way right? But some have reported they work normally?

I haven't seen an ounce of evidence that truthfully states that a stiffer anti-roll bar actually helps reduce loss of grip at any given end.

In fact, I've tried more times than I can count, and in cars with a tendency to understeer I always wind up with a higher rear ARB than front.

Ok, well then what about ballast? You increase the weight in the front or the rear, you should be causing that end to lose traction first around a corner? Well, wait a minute, how does that common sense fit with the altered GT5 world?

Absolutely perfectly, in fact.

yes, games can differ from reality. Every sim does. But this is taking a core focus of the game, MODIFYING CARS, and turning it into a confusing mess.

Spend enough time with it, get past the initial confusion, then it clicks.

Using your FPS analogy, this is like headshots causing less damage then a shot in the foot. This is like being able to shoot through a steel wall but not a bush.

More like a 5.56 round doing more damage to a solid target than a 7.62 really.

I'm not leaving only because I hope there is an update coming one day.

Seems like, at least to some extent, that update has been ninja'd in. Spec II changed quite a few things in tuning... Not extremely, but just enough to mess with it.

2.02/2.03, somewhere in there, seems to have made another slight change. Can't say what exactly it is, but I do know that my most recent tune was done with spring rates stolen directly from the real world and it came as close as possible to winning the competition it was entered in without being a different car entirely.

By posting in here, I never intended to get into a philosophical debate as to whether or not the tuning physics of the game are correct. I think there's enough documentation on the subject, and this thread in and of itself suggests there IS a problem. It's just that we can't all agree on exactly what the problem IS.

I'm not saying there isn't. Ride height is about as broken as it gets.

Oh and damper extension/rebound is as backwards as it was in GT4. Higher numbers mean less damping/faster extension, reverse for lower.

Not to mention we still have zero camber change with suspension compression/extension.

I'll continue to make tunes that suit my needs, making the car both manageable and quick in the setting I'm driving it. While knowing that my approach is not going to be anything like it would have been in a prior GT game, or in a real life setting.

As will I. Except I still tune almost exactly as I did in GT4 with only very minor differences.

Just to be clear that we are on the same page with regard to this, I'm talking about setting an initial spring balance based on weight distribution and then changing only the spring rate at a single end and not touching any other value.

There's someone else who has a theory directly based on this... And those that have listened to him have seen some merit. I personally threw it entirely out of the window in trade for my usual "I know this should work" followed by tweaking.

Even with stepping through the values slowly I've personally not come across anything consistent enough for me not to believe that a problem exists with GT5's tuning.

Me neither. But I have seen it consistently enough to say that it does work to a certain extent.

In my opinion its claiming to mirror reality, but is falling down enough times to fail to meet that goal.

Sure. I'll accept that.

Another potential issue is that CoG seems to be consistently too low. The only way I've really seen tires get lifted due to cornering force is through excessively high grip (race soft tires) and a raised suspension coupled with the settings one would expect to lift them. Stiff front, soft rear on a RWD car will lift the inside front under these conditions, but if you go a bit lower, no longer. A bit stiffer at the rear, no longer. Same applies in reverse to corner-entry characteristics.

My own experience differs, you know how much time I spent tuning in GT4 (and previous titles) and it had nothing close to the number of issues I keep coming across in GT5.

The only way I can effectively tune in GT5 is to throw real world theory out the window and start from scratch, and for me that's a deal breaker.

This is what I find very odd though...

Tuning FWDs for maximum pace aside I've found reasonably consistent results from using real-world numbers (stiff/soft springs + soft/stiff ARBs is MASSIVELY better on corner exit, although worse on entry than the "normal" method), so I'm not entirely sure where you've gone wrong.

Oh well... There's the entire damper issue and ride height.

The problem I have with this is that you are taking what are very uncommon exceptions and using them to validate what I see as significant problems with GT5 tuning acting in a very inconsistent manner.

I can't say I've personally done any real-life tuning work but I can say that these "uncommon exceptions" happen very often in the sims I've played with. Must not be all that uncommon if I've noticed the issue with several different vehicles.

The general rules and traits of real-world tuning exist because they are just that, the traits that will happen the majority of the time. When you change X in Y direction then Z happens, and it happens in my experience the vast majority of the time in reality. GT5 unfortunately doesn't mirror that, and for me its moved further away from that in comparison to GT4 and a lot of other titles around.

Again, ride height potentially aside, I don't see much of a difference in the "realism" of GT5's tuning vs that of GT4's. I found the same gains from doing the same things in GT4 as I do in GT5... Virtually all of my setups have their roots in a tune from my GT4 days in some way, shape, or form.

When I change X in Y direction and Z doesn't happen, I go the other way and it does (usually, other times it's extremely difficult to make Z happen).

But then again, what do I know? It's not like I know anything about what I'm doing.

I actually quite surprised at myself, that its taken this long to voice these concerns, but the main area I was looking forward to in GT5 was tuning (just as it was for GT4). Once again I invested a considerable amount of my time in doing so, but to be quite blunt about it, the lack of consistency with what I know works in the real world frustrated me to such a degree I find it hard to explain.

And again, I absolutely cannot understand it seeing as I am able to draw exact parallels with GT4 in terms of tuning.

That GT5 is not 100% accurate is no surprise to me, I never expected it to be, what did surprise me is how inconsistent it is in regard to real world theory.

I got over that in GT4.

 

[JackWilson]

No one has answered this yet.

How does having stiff front and soft rear improve a FF car's handling?
It cures understeer, improves turn-in, mid-corner and exit grip, but how & why does that happen someone explain please.

 

[Winters Noble]

I agree that GT5 has it quirks. It pretty much impossible to simulate perfectly every aspect of real world tuning in a game. Anyway, stiff springs in front for ff cars can reduce understeer because we all know that most ff cars have poor weight distribution. For example, I think the dc2 integra has 63/37 distribution. I don't care how light it is, that much weight at the front makes it more prone to understeer. When you brake into a corner, most of the weight settles on the front first causing understeer. Stiffer front springs reduce the speed at which the weight at the front settles giving the car of a nimble characteristic further into the corner. This goes for any car, regardless of drive layout. Pay attention to weight distribution. I'm no expert, It's just my take.

 

[Scaff]

I agree that GT5 has it quirks. It pretty much impossible to simulate perfectly every aspect of real world tuning in a game. Anyway, stiff springs in front for ff cars can reduce understeer because we all know that most ff cars have poor weight distribution. For example, I think the dc2 integra has 63/37 distribution. I don't care how light it is, that much weight at the front makes it more prone to understeer. When you brake into a corner, most of the weight settles on the front first causing understeer. Stiffer front springs reduce the speed at which the weight at the front settles giving the car of a nimble characteristic further into the corner. I'm no expert, It's just my take.

Dampers change the speed at which load is transferred not springs (edited to add - very stiff springs can have a damping effect themselves - but not anything you would find on a road car and certainly not the values allowed in GT5).

I can also assure you that in the real world increasing the front spring rate of a FWD car (and only making that change) will almost always result in an increase in understeer.

Stiff damper settings delay the onset of load transfer (as they resist or damp it) as such the load transfer happens over a shorter period and as the amount of transfer doesn't change it is therefore happening quicker, the opposite being true for softer dampers.


Scaff

 

[JackWilson]

No one has answered this yet.

How does having stiff front and soft rear improve a FF car's handling?
It cures understeer, improves turn-in, mid-corner and exit grip.

Still no explanation to this?

I agree that GT5 has it quirks. It pretty much impossible to simulate perfectly every aspect of real world tuning in a game. Anyway, stiff springs in front for ff cars can reduce understeer because we all know that most ff cars have poor weight distribution. For example, I think the dc2 integra has 63/37 distribution. I don't care how light it is, that much weight at the front makes it more prone to understeer. When you brake into a corner, most of the weight settles on the front first causing understeer. Stiffer front springs reduce the speed at which the weight at the front settles giving the car of a nimble characteristic further into the corner. This goes for any car, regardless of drive layout. Pay attention to weight distribution. I'm no expert, It's just my take.

In any drivetrain layout you want the weight to shift to the front as quickly as possible not as slow as possible. (Under Braking)

In FF cars you want the weight to be on the front tyres as much as possible on entry-mid and exit of a corner to improve grip.

FACT

Increasing stiffness at the front improves grip at the rear. Fact
Inreasing stiffness at the rear improves grip at the front. Fact
But in GT5 that is not the case is it?

 

[SolidSnake7735]

No one has answered this yet.

How does having stiff front and soft rear improve a FF car's handling?
It cures understeer, improves turn-in, mid-corner and exit grip, but how & why does that happen someone explain please.

Because the weight is in the front so there's no need for such stiff spring rates out back. I'm not saying my method is the definitive way to get the best handling from FF's but Mopar-383 tunes some hella fast FF's that always beat mine on the touge and the track so I learned a bit from his methods in regards to FF tuning. Though I can understand why the settings work.

The stiffer front rates allow less body roll up front and allows the differential to work while keeping the car planted and since there is no differential out back the softer rates would allow the wheels to turn wit minimal stress on the tires. Too soft and the tires would slip because of lack of traction. Too stiff and the tires will grip too hard, further increasing understeer and will loose traction during hard cornering because there isn't enough weight on the tires to take advantage of the higher spring rates. WHAT these spring rates are obviously depend on the tires you're using. AND it also matters what your differential settings are because that's what dictates at what point while accelerating or braking the car makes the most use of the tire's grip.

It's all about balance

 

[Scaff]

Because the weight is in the front so there's no need for such stiff spring rates out back. I'm not saying my method is the definitive way to get the best handling from FF's but Mopar-383 tunes some hella fast FF's that always beat mine on the touge and the track so I learned a bit from his methods in regards to FF tuning. Though I can understand why the settings work.

The stiffer front rates allow less body roll up front and allows the differential to work while keeping the car planted and since there is no differential out back the softer rates would allow the wheels to turn wit minimal stress on the tires. Too soft and the tires would slip because of lack of traction. Too stiff and the tires will grip too hard, further increasing understeer and will loose traction during hard cornering because there isn't enough weight on the tires to take advantage of the higher spring rates. WHAT these spring rates are obviously depend on the tires you're using. AND it also matters what your differential settings are because that's what dictates at what point while accelerating or braking the car makes the most use of the tire's grip.

It's all about balance

Not a single thing you have said here has any relation to how real world physics and vehicle dynamics work.


"The stiffer front rates allow less body roll up front and allows the differential to work while keeping the car planted"
Yes a stiffer front will reduce body roll, however that will result is the two front tyres having a greater load differential which will increase the work the LSD has to do.


"softer rates would allow the wheels to turn wit minimal stress on the tires. Too soft and the tires would slip because of lack of traction."
Reducing spring rates will allow the two tyres on a given 'axle line' to carry a more equal share of the load and maximise the traction each can carry


"Too soft and the tires would slip because of lack of traction."
Nope, reduced spring rates provide a more equal share of load and will increase traction on that axle line. The only way what you decribe is going to happen is if you set the rates so low that the bump stops are hit and then you effectively have a solid suspension (and once again stiff spring rates).


"Too stiff and the tires will grip too hard, further increasing understeer and will loose traction during hard cornering because there isn't enough weight on the tires to take advantage of the higher spring rates"
Nope, stiffening the spring rate on a given axle will bias one side of an axle line with more load and the other with less. The phrase "grip too hard, further increasing understeer" also makes no sense at all.

Spring rates alone do not increase or decrease grip in any way, they affect how load is shared along an axle line, that load in conjunction with the tyres grip level (mU co-eff) determine how much grip a given tyre has. Exceed it and you lose traction. Understeer or oversteer will then occur dependent on if we are looking at the front or the rear of the car.

The above is the basic fundamentals of vehicle dynamics and what you are describing has no relation to it at all, take a read of Greyouts guide from the GT4 tuning section as a staring point or the first of my tuning guides (link in my sig):

https://www.gtplanet.net/forum/showthread.php?t=57625


Just about the only thing I agree about is that its a balance. Stiffer spring rates are required to reduce the ride hight as much as you can for a track, that reduces the cars cog which minimises load transfer. That minimises load differentials across the tyres and maximise grip (it also reduces changes to contact patches). However the stiffer rates (particularly if not well balanced) cause issues with load equalisation across the axle lines and reduce grip.

Its finding the right balance between these, particular in drive-trains that cause static and dynamic load issues (in theory anything but 50/50) and those than increase a cars PMI. That's why the ideal racecar in a lightweight MR running as low as the track will allow (low PMI and minimised load transfer).


Scaff

 

[Winters Noble]

Because the weight is in the front so there's no need for such stiff spring rates out back. I'm not saying my method is the definitive way to get the best handling from FF's but Mopar-383 tunes some hella fast FF's that always beat mine on the touge and the track so I learned a bit from his methods in regards to FF tuning. Though I can understand why the settings work.

The stiffer front rates allow less body roll up front and allows the differential to work while keeping the car planted and since there is no differential out back the softer rates would allow the wheels to turn wit minimal stress on the tires. Too soft and the tires would slip because of lack of traction. Too stiff and the tires will grip too hard, further increasing understeer and will loose traction during hard cornering because there isn't enough weight on the tires to take advantage of the higher spring rates. WHAT these spring rates are obviously depend on the tires you're using. AND it also matters what your differential settings are because that's what dictates at what point while accelerating or braking the car makes the most use of the tire's grip.

It's all about balance

It is about balance, but that is acheived through different means. I usually set the springs at the heavier end. This method is used in the real world too, especially at the touge. Yes, a theory says that more weight at an end may increase grip at that end, but another theory says that weight at the front will reduce a cars turning ability. Think of MR cars and think of why some corner so well. Weight at rear. Yes, the drive wheels are at the rear, but the same theories apply to ff's as well. So, what do we do here? There is no simple answer. Not just springs are in play here. In tuning, trade offs are natural when it comes to desirable characteristics. Trial and error is best universal tuning method. Sorry for the long story.

 

[Zeeco_9]

Stiff damper settings delay the onset of load transfer (as they resist or damp it) as such the load transfer happens over a shorter period and as the amount of transfer doesn't change it is therefore happening quicker, the opposite being true for softer dampers.


Scaff

Can you maybe explain the above. I have never really quite understand the physics behind this reasoning. According to Roulle stiffer dampers load and unload slower than softer dampers. Anti-lift and anti-squat comes into play here. Always confuses me

BTW this is one of the more interesting threads I have come across on GT5. Greyout's original post explaining weight transfer (to which Scaff referred to) is still one of the best I have come across 👍

edit: just my 2pence - GT5 does not discriminate between fast and slow damping. You are not able to tweak sprung and unsprung mass seperately. Needless to say you have to rely on springrates and ARB's too much for finetuning transient states. That defeats the main function of springs which is tyre compliance.

 

[Scaff]

Can you maybe explain the above. I have never really quite understand the physics behind this reasoning. According to Roulle stiffer dampers load and unload slower than softer dampers. Anti-lift and anti-squat comes into play here. Always confuses me

BTW this is one of the more interesting threads I have come across on GT5. Greyout's original post explaining weight transfer (to which Scaff referred to) is still one of the best I have come across 👍

Stiff dampers are slower to react to load as in simple terms, the stiffer they are the more they will resist the initial loading (i.e. a stiff damper may require 200lbs+ to be applied before it starts to move, while a soft damper may only require 120lbs to be applied before it starts to move). Now loading of a tyre will always take the same total amount of time and the stiffer damper is initially resisting that loading before moving, then the same loading now has to occur over a shorter period of time.

An example:
Total time to load = 5m/s
Load = 500lbs

Stiffer damper
Resistance time = 2 m/s

Load time (5 m/s) - Resistance time (2 m/s) = 3m/s
Therefore we have 500lbs loaded in 3 m/s or 166 lbs per m/s



Soft Damper
Resistance time = 0.5 m/s

Load time (5 m/s) - Resistance time (0.5 m/s) = 4.5m/s
Therefore we have 500lbs loaded in 4.5 m/s or 111 lbs per m/s

The soft damper is quicker to react to the load (due to less resistance) and the reaction has longer to occur and as a result is spread out and less of a 'shock' to the tyre.

Skip Barber sums it up quite nicely in his book Going faster:

A stiffer bump setting slows down the motion on its corner and speeds up the load transfer. A softer bump setting does the opposite - it allows the suspension to move faster and spreads the changes in loading out over a longer period of time. The same is true of rebound. Stiff re-bound settings will force the suspension system to move more slowly when loads are removed from its corner of the car, but the unloading of the contact patch will be more abrupt. Softer rebound settings allow the suspension to move more quickly and the unloading of the contact patch happens more gradually

Take a look at the diagram below; the red trace represents a damper with a stiffer shim stack than the blue trace. At a shaft velocity of 3 inches/second the soft damper (blue) requires approx 180lbs of force to compress/rebound, while the stiffer damper (red) requires approx 210lbs of force to compress/rebound.

Now as the stiffer damper will resist compression/rebound until the greater load is placed upon it this will delay the reaction of the suspension. However as the total load transfer will take a fixed amount of time, the longer the delay, the less time a stiffer damper has to load/unload the tyres contact patch and the quicker that load/unload will occur. The following figures are illustrative but may help, say it takes 1 second to full transfer a load of ‘Xlbs’, but 0.2 second to transfer our 180lbs (blue – softer damper trace) and 0.4 seconds to transfer our 210lbs (red – stiffer damper trace). This would result in the actual tyre load/unload occurring over 0.8 seconds for the softer damper but 0.6 seconds for the stiffer damper. So for the stiffer damper the time taken for the suspension to react is 0.4 seconds (so longer than the soft damper) but the load itself was transferred in a shorter time of 0.6 seconds.

Scaff

 

[JackWilson]

First things first.
Understeer=when the front tires have less traction than the rear.
Oversteer=when the rear tires have less traction than the front.
Ok we can all agrree on that.

Because the weight is in the front so there's no need for such stiff spring rates out back. I'm not saying my method is the definitive way to get the best handling from FF's but Mopar-383 tunes some hella fast FF's that always beat mine on the touge and the track so I learned a bit from his methods in regards to FF tuning. Though I can understand why the settings work.

The stiffer front rates allow less body roll up front and allows the differential to work while keeping the car planted and since there is no differential out back the softer rates would allow the wheels to turn wit minimal stress on the tires. Too soft and the tires would slip because of lack of traction. Too stiff and the tires will grip too hard, further increasing understeer and will loose traction during hard cornering because there isn't enough weight on the tires to take advantage of the higher spring rates. WHAT these spring rates are obviously depend on the tires you're using. AND it also matters what your differential settings are because that's what dictates at what point while accelerating or braking the car makes the most use of the tire's grip.

It's all about balance

Dude what you said doesn't make any sence at all.

Because the weight is in the front so there's no need for such stiff spring rates out back.

The stiffer front rates allow less body roll up front and allows the differential to work while keeping the car planted and since there is no differential out back the softer rates would allow the wheels to turn wit minimal stress on the tires.

The stiffer the front is, more weight transfer will occur on that axle.

If the front has more weight to carry it must work harder to corner the extra weight, and the rear will work less hard. This means that with the front stiffness increased vs the rear, the front has relatively less traction while the rear has more, so the car will understeer.

Too soft and the tires would slip because of lack of traction. Too stiff and the tires will grip too hard, further increasing understeer and will loose traction during hard cornering because there isn't enough weight on the tires to take advantage of the higher spring rates.

The tires would slip because of lack of traction because the springs were too soft??
The opposite would happen, softer springs allow the tires have more traction.

Too stiff and the tires could lose traction over bumps

WHAT these spring rates are obviously depend on the tires you're using.

Doesnt matter what tires your using, Stiff front soft rear encourages understeer, and soft front stiff rear encourages oversteer.

The only way what you decribe is going to happen is if you set the rates so low that the bump stops are hit and then you effectively have a solid suspension (and once again stiff spring rates).

And I dont think in GT5 that is possible.

Sorry but that's not correct (see the guide I linked to a few posts ago), spring settings do not change the amount of load transferred, they change how it is distributed along an axle line.


Scaff

Ok I will have a look thanks.

Not tonight though its getting a bit late.

 

[Scaff]

The stiffer the front is, more weight transfer will occur.

Too stiff and the tires would lose traction because of the increase in weight transfer and to much stress on the tires.

Sorry but that's not correct (see the guide I linked to a few posts ago), spring settings do not change the amount of load transferred, they change how it is distributed along an axle line.


Scaff

 

[Zeeco_9]

Thank you for your quick response Scaff. It does make sense now, looking at the graph in particular. It confirms that rebound is the tricky one to nail in a setup 👍

 

[XDesperado67]

@Scaff this is probably my own lack of understanding but what I gathered from the in game material PD provides on dampers is that they act to limit the amount of travel available to the springs. I can understand what you've been saying in a real world sense but I'm not sure that's how PD has modeled dampers in the game. If I'm correct in my interpretation of PDs damper implementation then they are acting as nothing more than suspension travel stops, making the springs function not just in their normal roll but also the way you've described for real world dampers. Again its probably just my own misunderstanding of the material PD has provided.

 

[Scaff]

Thank you for your quick response Scaff. It does make sense now, looking at the graph in particular. It confirms that rebound is the tricky one to nail in a setup 👍

No problem at all, its a tricky one to get your head around so I'm only glad to help.

@Scaff this is probably my own lack of understanding but what I gathered from the in game material PD provides on dampers is that they act to limit the amount of travel available to the springs. I can understand what you've been saying in a real world sense but I'm not sure that's how PD has modeled dampers in the game. If I'm correct in my interpretation of PDs damper implementation then they are acting as nothing more than suspension travel stops, making the springs function not just in their normal roll but also the way you've described for real world dampers. Again its probably just my own misunderstanding of the material PD has provided.

Well back in GT4 my own testing showed that damper setting did have an effect on the speed at which suspension reacted, so I would be rather concerned if the same's not true for GT5.

The Caterham Fireblade allows for an easy test.

Set the front springs as soft as they will go and put maximum ballast up front as well. With a soft damper setting you should see the suspension react quickly and with very little resistance to sudden braking and with stiff damper settings the suspension should show an initial resistance to movement under heavy braking.

If I get half a chance I will give these a quick test later.



Scaff

 

[Rotary Junkie]

Let me bring something up here...

Setting up a FWD for maximum traction very nearly makes sense... Apart from one thing.

Front spring rate. Lower seems to provide more traction, despite it theoretically extending further under acceleration, even on the glass-smooth surface that is the test straight at SSRX.

The only reason I can think of is that the softer springs actually reduce the rate at which the front extends, lessening the speed at which load is removed.

Where it gets really strange for me is that the entirety of the "advantage" I see of my method of FWD tuning is found on corner exit; most of my FWDs push a bit going in, then actually tuck in under throttle.

So then... Can someone do me a favor and find an explanation for this? I do not see any reduction of understeer on entry or at mid-corner with my style of FWD tuning (going the opposite way actually works as intended from what I've seen), only at exit. The "proper" route can be made to nicely hold slight oversteer going in if desired but you can't even look at the throttle without getting understeer. My route generally understeers on the way in followed by neutrality or oversteer under throttle, up to the point at which power overcomes the fronts.

Edit: @Scaff: I swear damper extension works in reverse from everything I've seen and felt, so "soft" dampers would actually be 10 ext, 1 compression, and stiff would be exactly the opposite.

 

[Scaff]

Let me bring something up here...

Setting up a FWD for maximum traction very nearly makes sense... Apart from one thing.

Front spring rate. Lower seems to provide more traction, despite it theoretically extending further under acceleration, even on the glass-smooth surface that is the test straight at SSRX.

The only reason I can think of is that the softer springs actually reduce the rate at which the front extends, lessening the speed at which load is removed.

Where it gets really strange for me is that the entirety of the "advantage" I see of my method of FWD tuning is found on corner exit; most of my FWDs push a bit going in, then actually tuck in under throttle.

So then... Can someone do me a favor and find an explanation for this? I do not see any reduction of understeer on entry or at mid-corner with my style of FWD tuning (going the opposite way actually works as intended from what I've seen), only at exit. The "proper" route can be made to nicely hold slight oversteer going in if desired but you can't even look at the throttle without getting understeer. My route generally understeers on the way in followed by neutrality or oversteer under throttle, up to the point at which power overcomes the fronts.

Edit: @Scaff: I swear damper extension works in reverse from everything I've seen and felt, so "soft" dampers would actually be 10 ext, 1 compression, and stiff would be exactly the opposite.

To be honest with you RJ, nothing you have said in this would surprise me, because I honestly believe that GT5 has some serious issues when it comes to how tuning settings work.

Now why that would be is a much bigger question, and a couple of options are:

• Mistakes have been made with the 'form' windows means values are in the wrong place and as such appear to be doing the wrong thing
• The physics are totally shot
• Certain areas of the physics have improved and other haven't resulting in a conflict when some calculations are carried out

Now this thread and a few others have got me thinking about this over the last few days, and while I have nothing concrete to back it up, I'm starting to favour the last of those options.

Let me explain. In my opinion PD have certainly moved some areas of the physics forward, the main one being load transfer, it's certainly better modelled and reacts in a much more realistic manner than it ever did before.

However the tyre modelling doesn't seem to have moved on at all from GT4, and testing others have done certainly seems to back that up - with a very simply incremental increase in grip back purely on compound, with differing tyre widths and profiles having no bearing of lateral-g at all.

In addition I still have a lot of questions about how well differing types of suspension and how the react are modeled. More modern suspensions seem to be OK, however leaf sprung and live axles in particular do not react as they should do (full throttle launches of light, powerful RWD cars illustrate that - no step out occurs at all).

I don't think its beyond the realm of possibility that they factors are, on occasion, causing things to get 'screwy' with the physics and as a result the affect of tuning changes. As I say, this is just me thinking out loud right now, but I believe its an important area to discuss in an open and frank manner. Simply because things are not behaving as they should do if we are to use the real world as a consistent reference.


Scaff

 

[SolidSnake7735]

Well I'm too lazy to read through all that and just say this.. what works in Gran Turismo doesn't always apply to what would make sense in real life. maybe my grammar was off in my explanation. Or it's some sort of language barrier.

I'm not sure who was it that said this but in regards to FF's someone once said "The power is going to the wrong wheels..."

Imagine if you were that car.. use your imagination, would you rather run down a spiral staircase holding a 50lb/22kg box out in front of you or tied to our back? hmmmmm... I wonder?

Or imagine yourself moving a wheelbarrow. Would you rather have THAT tied to your waist as you run along with it or pushing it from behind as intended?


almost forgot... how about dampening? we all seem to have forgotten the big role that dampening rates play in handling. More so than spring rates, i would say.

 

[Scaff]

Well I'm too lazy to read through all that and just say this.. what works in Gran Turismo doesn't always apply to what would make sense in real life. maybe my grammar was off in my explanation. Or it's some sort of language barrier.

Its not grammar or language (I'm English), simply your explication of why certain behaviour happens in GT5 doesn't match reality.

I'm not sure who was it that said this but in regards to FF's someone once said "The power is going to the wrong wheels..."

Imagine if you were that car.. use your imagination, would you rather run down a spiral staircase holding a 50lb/22kg box out in front of you or tied to our back? hmmmmm... I wonder?

Or imagine yourself moving a wheelbarrow. Would you rather have THAT tied to your waist as you run along with it or pushing it from behind as intended?

And this has what to do with what?

almost forgot... how about dampening? we all seem to have forgotten the big role that dampening rates play in handling. More so than spring rates, i would say.

You seem to have missed the last few posts, all of which discuss damping in quite some detail. That aside, noting in regard to damping would change the fact that your explanation of why things are happening in GT5 doesn't match real world vehicle dynamics in any way.

Scaff

 

[SolidSnake7735]

And this has what to do with what?

Scaff

It's a simile/metaphor for drivetrains and which wheels (legs) the power is going to and why FF's aren't as good handling or easy to tune as most Rear-Wheel driven vehicles. with some exceptions.

 

[Rotary Junkie]

To be honest with you RJ, nothing you have said in this would surprise me, because I honestly believe that GT5 has some serious issues when it comes to how tuning settings work.

Now why that would be is a much bigger question, and a couple of options are:

• Mistakes have been made with the 'form' windows means values are in the wrong place and as such appear to be doing the wrong thing
• The physics are totally shot
• Certain areas of the physics have improved and other haven't resulting in a conflict when some calculations are carried out

Now this thread and a few others have got me thinking about this over the last few days, and while I have nothing concrete to back it up, I'm starting to favour the last of those options.

Let me explain. In my opinion PD have certainly moved some areas of the physics forward, the main one being load transfer, it's certainly better modelled and reacts in a much more realistic manner than it ever did before.

However the tyre modelling doesn't seem to have moved on at all from GT4, and testing others have done certainly seems to back that up - with a very simply incremental increase in grip back purely on compound, with differing tyre widths and profiles having no bearing of lateral-g at all.

In addition I still have a lot of questions about how well differing types of suspension and how the react are modeled. More modern suspensions seem to be OK, however leaf sprung and live axles in particular do not react as they should do (full throttle launches of light, powerful RWD cars illustrate that - no step out occurs at all).

I don't think its beyond the realm of possibility that they factors are, on occasion, causing things to get 'screwy' with the physics and as a result the affect of tuning changes. As I say, this is just me thinking out loud right now, but I believe its an important area to discuss in an open and frank manner. Simply because things are not behaving as they should do if we are to use the real world as a consistent reference.


Scaff

This is probably the most reasonable explanation yet.

I'm not entirely positive, but it seems like suspension geometry is extremely basic in most cases... Though it may vary on some vehicles and I've just not done the right tests to see.

I know on the Samba Bus there is absolutely ZERO camber change from suspension compression or extension (where, if I'm not mistaken, it'd show quite extreme negative camber under compression and positive under extension in the real world on the rear, even modelling it as a beam axle would be closer to right) and it's the most obvious vehicle to check for such things on due to how much suspension travel and body roll can be tuned into it.

Now then, my other point of interest with the Samba was that spring rates are definitely correct front to rear. Stiffen the rear, it compresses and droops less, stiffen the front, same deal.

Which is where the confusion hits. The tire model is likely due for a large portion of the blame (issues with tire load vs grip) but I'm not sure it's all of it. See: Drag testing. RWD likes high rear ride height, front ride height is a 50/50 shot as to whether slamming the front provides more grip than running it full up. FWD likes nose down/tail up, with rear ride height again being the dominating factor, AWD likes the same with the same deal again.

 

[Scaff]

It's a simile/metaphor for drivetrains and which wheels (legs) the power is going to and why FF's aren't as good handling or easy to tune as most Rear-Wheel driven vehicles. with some exceptions.

I know that, but what does it have to do with your explanation of why settings work in GT5 not having any correlation with reality?


Scaff

 

[CSLACR]

Just something to pitch in here, I tested the RX-7 TC the other day, and was surprised to find this out.
I was playing with different settings I know for a fact to increase acceleration grip, being high rear ride height, and maximum down force.
Then I tried spring rate, dampers, arb's, etc.

To my surprise (somewhat) setting the rear spring to maximum (from minimum) increased my "G rating" from .95 to .96.
So I actually have every reason to believe increasing spring rate does increase traction.(in GT5, mind you) This would also explain why I run faster laps with a very stiff suspension as well.

BTW - Dampers and ARB seemed to have no effect on grip, from what I could tell. If they do, it's minimal.

Which is where the confusion hits. The tire model is likely due for a large portion of the blame (issues with tire load vs grip) but I'm not sure it's all of it. See: Drag testing. RWD likes high rear ride height, front ride height is a 50/50 shot as to whether slamming the front provides more grip than running it full up. FWD likes nose down/tail up, with rear ride height again being the dominating factor, AWD likes the same with the same deal again.

Of course if you know that raising ride height on the drive wheels adds grip, neither is a surprise.
The weird bit is that lowering or raising the fron you say does have an effect on grip. My test showed nothing, but I've only tried one car so fer.

 

[Scaff]

This is probably the most reasonable explanation yet.

I'm not entirely positive, but it seems like suspension geometry is extremely basic in most cases... Though it may vary on some vehicles and I've just not done the right tests to see.

I know on the Samba Bus there is absolutely ZERO camber change from suspension compression or extension (where, if I'm not mistaken, it'd show quite extreme negative camber under compression and positive under extension in the real world on the rear, even modelling it as a beam axle would be closer to right) and it's the most obvious vehicle to check for such things on due to how much suspension travel and body roll can be tuned into it.

Now then, my other point of interest with the Samba was that spring rates are definitely correct front to rear. Stiffen the rear, it compresses and droops less, stiffen the front, same deal.

Which is where the confusion hits. The tire model is likely due for a large portion of the blame (issues with tire load vs grip) but I'm not sure it's all of it. See: Drag testing. RWD likes high rear ride height, front ride height is a 50/50 shot as to whether slamming the front provides more grip than running it full up. FWD likes nose down/tail up, with rear ride height again being the dominating factor, AWD likes the same with the same deal again.

For myself one of the clearest examples is the Shelby Cobra.

The combination of high profile, thin tyres, leaf spring rear suspension and light weight/high torque, should mean that a full throttle launch should see the rear of the car step out of line to a massive degree. You should have to correct it or risk a spin.

Instead what happens is you simply get wheel spin and the car launches in a straight line, no correction required.

Scaff

 

[SolidSnake7735]

For myself one of the clearest examples is the Shelby Cobra.

The combination of high profile, thin tyres, leaf spring rear suspension and light weight/high torque, should mean that a full throttle launch should see the rear of the car step out of line to a massive degree. You should have to correct it or risk a spin.

Instead what happens is you simply get wheel spin and the car launches in a straight line, no correction required.

Scaff

Right now I'm messing around with the '69 Camaro Z/28 and i get some mild step out while accelerating hard from a corner in 3rd gear (as opposed to second). Meaning, as I've left the corner and the road straightens, not while cornering. But, I don't get any step out while accelerating from a stand-still.

 

[CSLACR]

Right now I'm messing around with the '69 Camaro Z/28 and i get some mild step out while accelerating hard from a corner in 3rd gear (as opposed to second). Meaning, as I've left the corner and the road straightens, not while cornering. But, I don't get any step out while accelerating from a stand-still.

If the word "corner" is involved it's not what he's talking about.

Bold is the problem here.

 

[Greyout]

Well I'm too lazy to read through all that and just say this.. what works in Gran Turismo doesn't always apply to what would make sense in real life.

The problem is with how basic this problem is. Every other racing game I have ever played for the last 10 years that allows suspension setting changes has gotten this right.

Funny - "VIPER RACING" from Serria, with is weirdo graphics, from a good 5-10 years ago, had the best suspension model I've ever seen. Its worth downloading.

The only reason I can think of is that the softer springs actually reduce the rate at which the front extends, lessening the speed at which load is removed.

During suspension movement, you're dealing with the extension / compression resistance of the dampers, not so much spring tension. Springs determine how the car is balanced in a steady state of corner, dampers determine how the car behaves in a dynamic condition, i.e. turn in, turn exit, bump, etc.

Edit:

I didn't know it was still around, but here is a cut & paste from my weight transfer post from a few years ago, explaining why a stiff REAR suspension SHOULD increase traction in the FRONT:

It appears that many people are confusing body roll and weight transfer. Although we associate body roll & pitch with the same forces as weight transfer, they shouldn't be considered the same thing. Softening a spring does not increase weight transfer.

If you were to be riding on a giant sled, with absolutely no suspension, the sled would still exibit weight transfer in a corner, with no body roll at all.

Turning to the left, our little man leaning into the turn.

In steady state corning, what was 200 lbs on each side might become 150/250. The amount of weight transfer is detirmined by how wide the wheelbase is (wider = less weight transfer) and how high the CG is (lower CG = less weight transfer). One can easily see why high performance cars are wide & low to the ground

The reason why weight transfer is important is because of the relationship between how much lateral force a tire can generate and how much downward force is pushing it into the pavement. Obviously, the more downward force, then the more friction, but it does not increase on a 1:1 ratio. X downard force might result in Y friction, but 2X does not result in 2Y - it results in something less then 2Y.

It is because of this that the greatest traction occurs when there is no weight transfer - i.e. when both tires share the weight equally. Anything other then that, and you are making less then optimal traction. If we could, we would make a car that has a CG at ground level (no weight transfer) but obviously that can't be done.

next: Body roll

as we guide our cars around a corner, the suspension allows the body to roll. The axis that the car rolls on, and the amount of roll, depends on the suspension stiffness, the roll center, and the distance of the center of gravity from the roll center.

The roll center is the instantanous point at which the car rolls on. This is detirmined by suspension geometry. It can be found by extending an imaginary line off of the upper and lower control arms to a point at which they intersect (magenta). A line is then drawn from the center of the contact patch of the tire to that intersection point. The roll center is the point at which the last line, blue, crosses the center line of the car. Its also the point at which the two tire lines cross if you are figuring from the both side, which would be used if the roll center was off to one side or the other (again, remember the roll center is constantly moving as the suspension moves).

The grey box is the tire when viewed from the front. The black lines are the suspension arms. The body has been removed.

If you connect the front and rear roll center, you have a line. This line is the axis that the car rolls on. Generally, the rear is slightly higher then the front.

Viewed from the right. The axis that the car will roll on.


As lateral force acts upon the CG, the distance the CG is from the roll center detirmines how much body roll the car will exibit. If the CG was ON the roll center, then there would be NO body roll. the farther the CG is from the roll center, then the more leverage it has on on the body, and the more it will roll.

It is VERY unlikely that PD had time to compute the roll center of each vehicle, and therefore suspension geometry in general is not simulated. This is unfortunate, but expected.

What this DOES help you do is understand why adjusting the springs and sway bars does what it does.

lets take a car that has a perfectly even weight distribution (each number is the weight on each wheel.)

700 700
700 700

in a right corner, we get weight transfer to the left.

900 500
900 500

total weight is still the same.

When we increase the spring rate, we are not preventing weight transfer. The springs react into the body, reducing body roll, NOT weight transfer Stiff suspension does a few things good for us:

- reduces suspension travel, to prevent dynamic changes in camber and toe.
- improves responsiveness
- allows the car to ride lower without bottoming out

so, we increase the spring rate by 50% - but we are still at

900 500
900 500

with the stiffer spring, we could lower the car more, which would reduce weight transfer a little, but we won't include that just yet. now lets see what happens when we stiffen up the rear end more.

850 650
1000 400

but wait - if the rear got STIFFER, why is there MORE weight transfer in the rear?

The reason is because the spring is, again, reacting to the body roll. As the body rolled to the left, the rear spring held up that corner more, increasing the downforce on that side and unloading the inside rear. This results in LESS of a weight transfer up front. The TOTAL weight transfer is still the same, as it should be - 800 lbs difference between left and right.

the front tires are sharing the weight more evenly, so they will be able to provide more traction.

In a FWD race car, with the same weight, it wouldn't be rare to see this:

800 600
1400 0

SWAY BARS:

a sway bar is simply a semi-flexible bar that connects one side to the other. It adds spring rate to one side in roll, but not in bump. So if you have soft springs and a thick sway bar, then the ride quality will still be soft, and you will still have lots of pitch & dive, but the suspension will resist roll. A sway bar increases the downforce on the outside tire by unloading the inside. Thus, a thick swaybar is not a good idea for the front of a FWD race car. usually, production based race cars will not have any front bar at all, and rely stricly on proper spring rates.

 

[Scaff]

Right now I'm messing around with the '69 Camaro Z/28 and i get some mild step out while accelerating hard from a corner in 3rd gear (as opposed to second). Meaning, as I've left the corner and the road straightens, not while cornering. But, I don't get any step out while accelerating from a stand-still.

As CSLACR has mentioned, the corner is still an issue with this. You may not be in the corner, but lateral forces are most likely still acting on the car.

Totally remove them from the equation and then see what happens, give the Shelby Cobra test a go and let me know.

The problem is with how basic this problem is. Every other racing game I have ever played for the last 10 years that allows suspension setting changes has gotten this right.

Excellent post and something I agree with 100%.

In GT and GT2 it was more than understandable that we had these issues, with GT3 and particularly GT4 it started to get a bit annoying, particularly as Enthusia managed to get good a good set of suspension modelling in place.

Range Rover on gravel - notice the wheel droop over the jumps and how the suspensions arc of travel is well modelled.



S15 handbrake turns and do-nuts. They had this working on a PS2, when GT4 couldn't get close and to be honest the physics here still embarrass GT5 at times


That GT5 still faces these issues and they seem to be causing even bigger issues has been, as I have said, a deal breaker for me. I still have GT5, for some reason I still buy the DLC, but I rarely tune now and when I do it rapidly frustrates me. Given that I believe that PD now employ one of the guys that worked on Enthusia's physics just added to the frustration.

I had high hopes when PD stated the engine would be re-built and what was shown with GT5 pointed in the right direction, and do feel very let down in this regard. To the point that when it does arrive GT6 will almost certainly be the first GT title that is not a day one purchase for me.


Scaff

 

[JackWilson]

The stiffer the front is, more weight transfer will occur on the front.

If the front has more weight to carry it must work harder to corner the extra weight, and the rear will work less hard. This means that with the front stiffness increased vs the rear, the front has relatively less traction while the rear has more, so the car will understeer.

but wait - if the rear got STIFFER, why is there MORE weight transfer in the rear?

The reason is because the spring is, again, reacting to the body roll. As the body rolled to the left, the rear spring held up that corner more, increasing the downforce on that side and unloading the inside rear. This results in LESS of a weight transfer up front.

Greyout mentions that increasing the stiffness of springs at the rear increases weight transfer in the rear.

I said that incresing the stiffness in the front increases weight transfer in the front, but you said that was incorrect. I then read what you told me too and greyout says the same thing about the rear.

So I was correct in the first place right?

Range Rover on gravel - notice the wheel droop over the jumps and how the suspensions arc of travel is well modelled.

I'm pretty sure I havent noticed the wheels drop over jumps in GT5, here's a pic i took of a Subaru Impreza on gravel.

The tires arent very clear (This photos was not taken for this purpose), but there is no droop in the wheels whatsoever, they appear to be perfectly plumb, zero camber.

I dont know if thats because the Impreza has a different suspension to the Racnge Rover.

Would the Impreza's wheels droop in mid-air?