Observations on suspension settings

[nomis3613]

Fascinating thread!

I agree that the front and rear ride heights are the wrong way around (sounds like this is the general concensus anyway).

I'm braking in to a turn (trail braking) and the front of the car is nailed to the line I want, as I get close to the apex I ease off the brake to allow the car to coast a while... but as soon as I release the brake the car starts to run wide of the line I was on.

What's happening here is as soon as I release the brake I get weight transfer towards the rear of the car, unloading the front tyres and causing understeer.

Now this just doesn't happen in real life unless you're driving something with a very rear baised weight distribution (such as a 911).

Maybe it's related to us relying on ABS to prevent us spinning out every time we touch the brakes. I tried testing this, but I suck too much to drive without ABS!

Toe is definitely right... I checked this on a couple of other cars last night.

Still not 100% sure about Springs or ARB's... difficult to tell as the cars seem to get better the harder you set these!

I find that stiffer "front" springs give better turn-in response. Stiffer "rear" can slightly reduce exit understeer and make the car less stable in braking. While this doesn't conform to the textbook "stiffer front = understeer, stiffer rear = oversteer" theory, doesn't mean it's necessarily backwards because maybe the situation is more complicated than just diagonal weight transfer. So I dunno...

 

[Basilea]

I'd say bound and rebound is correct! Yesterday I tested extreme settings on dampers and 10/10 front and 1/1 rear is a hard/stiff front. The other way around the front took every bump with ease and felt soft.

 

[dylansan]

Has anyone tweeted this to Kaz yet? I'd say this is kind of an important glitch.

 

[PiRacingChamp]

Glad I ran across this thread. The suspension setups have seemed a bit screwy compared to earlier versions (1-4). Do we have any indication that PD is aware of this? Would be nice if they used the psn downtime to work on this issue. This seems to be more important than just about any other fix, as this affects the basic physics model. It would be nice to have a definitive answer, one way or the other.

 

[C-ZETA]

Well, that's just...amazing. Of course it's only observations but if you've seen it with ride height, then maybe it is correct.

 

[praiano63]

Ferrari enzo against the security rail under the "pit" panel, RIDE HEIGHT F-20 R+40

Ferrari enzo against the security rail under the "pit" panel, RIDE HEIGHT F+40 R-20

Nothing wrong. The ride height is OK ,like it's showing on pics. Try the same and see yourself...

 

[HunRotten]

But that's the problem you see. Even though it looks like as it should, it doesn't handle as it should.
See, on the first picture You set it so the nose is down and the back is pointing up. It should have a great deal of oversteer but it doesn't.It drives like as if You set it up the exact opposite.

 

[Synwraith]

Ferrari enzo against the security rail under the "pit" panel, RIDE HEIGHT F-20 R+40

Nothing wrong. The ride height is OK ,like it's showing on pics. Try the same and see yourself...

What I'm saying is this...

If you change the front ride height to max and the rear to min, the car displays this visually when you look at the replay, but the car behaves as if the front was min height and the rear max.

All the best
Maz

 

[Mike Rotch]

I've moved this to the Tuning Forum as the topic is a very important one for the wider GT5 community and a lot of people are either unaware of it or are not fully understanding of it.

So I ask new posters to:

A) Read the last 4 pages. They contain content that demonstrates tuning and racing acumen that we should acknowledge. If you cannot add constructively to the discussion (this means posting statements that are not substantiated with fact/ analysis/ experimentation results), do not post.

B) If you have read the last 4 pages and still have questions, do not be afraid to ask . We are all here to help each other and the current contributers are the most knowledgable people on this topic.

 

[Talon]

So, the first post in this thread was after PSN went down. Is there any way this "glitch" is an offline-only thing? I never really bothered tuning for Aspec, but have done quite a bit online, and never felt anything was totally wrong. I always just thought I didn't know what I was doing if a setup didn't work. This would finally explain the online/offline physics difference.

 

[smalldave]

So, the first post in this thread was after PSN went down. Is there any way this "glitch" is an offline-only thing? I never really bothered tuning for Aspec, but have done quite a bit online, and never felt anything was totally wrong. I always just thought I didn't know what I was doing if a setup didn't work. This would finally explain the online/offline physics difference.

That would make sense, as if you lower the ride height at the front for online (common practice really) and the car is more over-streery, offline when everything is the opposite way round, you would have understeer! Well done sir You may have cracked it.

 

[Z1-AV69]

So, the first post in this thread was after PSN went down. Is there any way this "glitch" is an offline-only thing? I never really bothered tuning for Aspec, but have done quite a bit online, and never felt anything was totally wrong.

As I'm more into A-Spec: I never felt offline was totally wrong.

Just make some tests: Tune a car with 0/0 ride height and 5/5 compression, 5/5 extension. Use the spring rates and camber to balance the car out. Now fire up de la Sarthe '09 with this nice big bump right before the breaking zone to the first chicane of the long straight. Now alter the damper compression or extension front or rear to 1 or 10, only 1 setting at a time. You can easily feel the effect when going over the bump or while diving in under breaking. Altering the front setting changes the front and rear is rear. Nothing wrong here.

Ride height is more complex in it's effects. But imo budious is right here:

As for the ride height itself, I'm quite convinced that it is an extension of suspension travel in addition to a physical ride height change in the car. I've noted in the past where a car will handle as though a +5 addition of ride height equates to a %5 increase in suspension stiffness. ...

In your case, by my finding, +15/-30 would result in a car with 15% stiffer front and 30% softer rear in effect to the given kgf/mm measurement, assuming an example with the game's stock motion ratio. That would result in a FR car having softer rear end and thus better traction and stronger steering at the front, then also consider any weight transfer in rake angle, but I'm not sure how specific or accurate the game's physics are in that regard, as you have pointed out as a possible issue here.

 

[Stotty]

Heh, my thread has been moved

So you are saying the rake angle transfers weight in the opposite direction as expected, but not necessarily that the ride height relationship to spring rate is wrong? I'll have to take a look at that...

As for the ride height itself, I'm quite convinced that it is an extension of suspension travel in addition to a physical ride height change in the car. I've noted in the past where a car will handle as though a +5 addition of ride height equates to a %5 increase in suspension stiffness. You can find an example of the formulas demonstrating this possible relationship in the OTI post in my signature, just scroll down to the bottom and look at example 2-A regarding the MR2 '86 NA and MR2'86 Super Charged tuning and handling comparison.

In your case, by my finding, +15/-30 would result in a car with 15% stiffer front and 30% softer rear in effect to the given kgf/mm measurement, assuming an example with the game's stock motion ratio. That would result in a FR car having softer rear end and thus better traction and stronger steering at the front, then also consider any weight transfer in rake angle, but I'm not sure how specific or accurate the game's physics are in that regard, as you have pointed out as a possible issue here.

Yes, that's exactly what I'm saying.

Ride height shouldn't make any difference to spring rates or suspension travel (within reason) as ride height adjustment on race suspension is independent of spring length.

Having a front high/rear low set up on a real car would cause understeer... as well as you wanting the front splitter as close to the ground as possible to give maxium aerodynamic benefit, the weight transfer caused by a front high/low rear would be a nightmare.

In addition to high front/low rear reducing understeer in GT5 settings, it also gives an increase in straight line speed. Not a big increase, but noticable never the less... I was getting 1-2mph over the line at Indy Road in the NASCAR. High front/low rear would not have this effect in real life as more air would get under the front of the car increasing drag and reducing straight line speed.

I find that stiffer "front" springs give better turn-in response. Stiffer "rear" can slightly reduce exit understeer and make the car less stable in braking. While this doesn't conform to the textbook "stiffer front = understeer, stiffer rear = oversteer" theory, doesn't mean it's necessarily backwards because maybe the situation is more complicated than just diagonal weight transfer. So I dunno...

The 'soften front to reduce understeer' idea (and the extentions of that), are all relative. Soft is a race car is not soft as we understand it from road cars! And as you say, the softer you go on front springs/dampers to try and reduce understeer the less reactive the front of the car becomes and turn in ends up being slow and unresponsive.

---------------------------------------------------------

I'm no tuning expert, merely sharing my experience of tuning the NASCAR for that particular WRS round.

Since I posted this I've also spent some time in a JGTC NSX at Suzuka. I continue to believe that front & rear ride heights are transposed in the settings menu, though so far with the NSX I haven't made up my mind if I still think spring and dampers are transposed in the same way... to do this I'd need to spend a bit more time in the car.

Anyone wanting to test/try this themselves... take a car and track you know really well (somewhere realtively flat is best as it removes the need to tune for bumps in the damper settings). Try flicking between high front low rear and low front high rear... you see easily enough that understeer is less if the front is set high and the rear low... exactly the opposite of what should be the case.

I'd be interested to hear others experience of spring and dampers rates... though if you're going to post your own experiences I'd appreciate it if you kept it as scientific as possible... ie; share car/track/settings in full AND the process you followed along with what you found.

Would be nice to get to the bottom of this as cars that understeer less are so much more enjoyable to drive

Cheers,
Chris

 

[budious]

Yes, that's exactly what I'm saying.

Ride height shouldn't make any difference to spring rates or suspension travel (within reason) as ride height adjustment on race suspension is independent of spring length.

...

As I replied on same subject in another discussion a moment ago:

Right, physically in the real world it shouldn't. However, programming is a very different realm so cause and effect can be disassociated with expected behavior. I just try to model an equation that explains my findings, I don't try too hard to place any rational on the functioning of the equation itself other than it explains it my findings.

I'm implying it is an intentional trimming mechanism, not a programming glitch. It doesn't have any real world correlation as you noticed most of the tuning features don't; the tuning menu is just a mini-game in itself to be reversed engineered for those with OCD.

 

[Motor City Hami]

I don't buy the conspiracy thoery. I believe that these aren't glitches -only people who are using tuning extremes to compensate for other flaws in their tunes - like LSD tuning. Could they really have f-ed up the driving engine that badly in 6 years of programming. And have gone as far as listing in-game descriptions if they knew they'd be that far off. The Gran Turismo series would lose ALL credibility if the tuning had this many glitches in them. Again, I believe that the glitches are really people reaching to understand why they can't make a car handle.

 

[Stotty]

I'm implying it is an intentional trimming mechanism, not a programming glitch. It doesn't have any real world correlation as you noticed most of the tuning features don't; the tuning menu is just a mini-game in itself to be reversed engineered for those with OCD.

'Tuning OCD' 👍

I don't buy the conspiracy thoery. I believe that these aren't glitches -only people who are using tuning extremes to compensate for other flaws in their tunes - like LSD tuning. Could they really have f-ed up the driving engine that badly in 6 years of programming. And have gone as far as listing in-game descriptions if they knew they'd be that far off. The Gran Turismo series would lose ALL credibility if the tuning had this many glitches in them. Again, I believe that the glitches are really people reaching to understand why they can't make a car handle.

Go try it for yourself.

It's easy enough to test ride height seperate to LSD settings... just minimise all the LSD settings or just fit an open diff. You should easily be able to feel the effect of low rear high front in reducing understeer

I've been playing GT since GT1, and I spent a lot of time with the tuning in GT4 and GT5P. So although I don't claim to be an expert tuner, I do have a fairly decent understanding of how the different variables interact.

 

[No_OBsT33R]

Been a few oddities I see. The ride height thing is annoying. There seems to be an issues with the physics.

It seems they wanted to use the offset heights to effect drag. Increasing the rear ride height angles the body more, It generates a lil more downforce at the cost of extra drag. They also wanted to prevent cars from flipping over from catching too much air underneath them (remember the CLK flipping at Le Mans) so they minimized the effects of air getting underneath the body.

Seems like programer solutions.

IDK it's a tough call at this point.

I still refuse to have my cars set up retarded like that using glitch advantages it kills the tune relying on glitches. Can you Tune or do you just know how to take advantage of the various glitches.....

 

[Stotty]

It seems they wanted to use the offset heights to effect drag. Increasing the rear ride height angles the body more, It generates a lil more downforce at the cost of extra drag. They also wanted to prevent cars from flipping over from catching too much air underneath them (remember the CLK flipping at Le Mans) so they minimized the effects of air getting underneath the body.

Drag isn't programmed in GT5... setting downforce at minium doesn't give you any increase in straight line speed, it only reduces your cornering grip.

And in any case, setting front ride high in real life would create more drag as it has the effect of increasing frontal area.

I still refuse to have my cars set up retarded like that using glitch advantages it kills the tune relying on glitches. Can you Tune or do you just know how to take advantage of the various glitches.....

It's not a glitch if you simply see the ride height settings as transposed... and if you see them like that they work exactly as they have in previous versions of the game and as they should do in real life.

In this case I suspect it's simply a programming error.

 

[praiano63]

I've moved this to the Tuning Forum as the topic is a very important one for the wider GT5 community and a lot of people are either unaware of it or are not fully understanding of it.

So I ask new posters to:

A) Read the last 4 pages. They contain content that demonstrates tuning and racing acumen that we should acknowledge. If you cannot add constructively to the discussion (this means posting statements that are not substantiated with fact/ analysis/ experimentation results), do not post.

B) If you have read the last 4 pages and still have questions, do not be afraid to ask . We are all here to help each other and the current contributers are the most knowledgable people on this topic.

Ok i didn't read the 2&3 pages first ,english is not my linguage, lazy. I have no questions, i 've just take my mercedes 300 sl on speed ring // 380 hp 1077 kg // comfort soft tires // stock tune.

1) ride height F+50 R-25 = OVERSTEER A LOT impossible to drive

2) ride height F-25 R+50 = STABILIZED NO OVERSTEER

So it's totaly right. This people spend so much time on this game programing physics, probably a long time drinking coffee with cookies .
"Bravo" to them.

All the rest seems OK to me first look. I tune tryng and tryng again settings ,empirical way. I don't know enough about cars tuning to compare with the real life tunes ,just the basic like most of us.
If anybody can do a summary.....

Sorry for that. (But my pics of the security rail are so goods ....can't stop watching)

 

[Johnnypenso]

This "talk" is all well and good, but what I'd like to see is some empirical testing. Somebody, anybody, who believes in this "theory" grab a car, tune it up the ordinary way or grab a decent tune for it from a tuning garage, run some laps, then reverse the f/r ride heights and run some more laps and then compare. These are repeatable and testable by all members since we can all grab the same car, same tune, same track and go nuts. Anyone willing to post their tune and their on track results? I'd be more than happy to run the same car, same test, and post my results as I'm sure would several others.

 

[Stotty]

Err, my settings in the NASCAR are in the 1st post in this thread (I'm happy to share gearbox and LSD settings if wanted too, though if you leave these stock you shouod still be able to feel the effects of suspension changes)... go test them against the reverse... car is any NASCAR on race mediums, track is Indy Road.

 

[Merquise]

Ok i didn't read the 2&3 pages first ,english is not my linguage, lazy. I have no questions, i 've just take my mercedes 300 sl on speed ring // 380 hp 1077 kg // comfort soft tires // stock tune.

1) ride height F+50 R-25 = OVERSTEER A LOT impossible to drive

2) ride height F-25 R+50 = STABILIZED NO OVERSTEER

So it's totaly right. This people spend so much time on this game programing physics, probably a long time drinking coffee with cookies .
"Bravo" to them.

All the rest seems OK to me first look. I tune tryng and tryng again settings ,empirical way. I don't know enough about cars tuning to compare with the real life tunes ,just the basic like most of us.
If anybody can do a summary.....

Sorry for that. (But my pics of the security rail are so goods ....can't stop watching)

1) ride height F+50 R-25 = OVERSTEER A LOT impossible to drive
This should understeer like hell.

2) ride height F-25 R+50 = STABILIZED NO OVERSTEER
This should oversteer like hell.

Ergo: graphix are correct, physics are wrong.


I have a question: is this glitch online AND offline?

 

[No_OBsT33R]

Drag isn't programmed in GT5... setting downforce at minium doesn't give you any increase in straight line speed, it only reduces your cornering grip.

And in any case, setting front ride high in real life would create more drag as it has the effect of increasing frontal area.

It's not a glitch if you simply see the ride height settings as transposed... and if you see them like that they work exactly as they have in previous versions of the game and as they should do in real life.

In this case I suspect it's simply a programming error.

Have you tested and confirmed drag is not present in the physics? I notice slower acceleration with higher DF.

I also have a way to test your theory.

This is a little formula I came up with to calculate the rolling radius (used for other calculations) the trick is to run the test as slow as possible because the rolling resistance and drag increase progressively with speed build up. So the slower you run the test the closer you get to the good numbers.

Gear your in (2nd) * FD / 0.00595 / rpm / Speed = rolling radius

2nd gear = 1.479
FD = 4.764
RPM = 4000
Speed = 54

1.479 * 4.764 / 0.00595 / (4000/54) =15.98

Rolling radius = 15.98

It also accounts for the rolling resistance & drag resistance, this will provide the results for our test.

Get in the car, get into the gear you selected to run the test, drive at the rpm you chose for the test, and note the speed your driving at that rpm in that gear.

If you want to truly see if drag is in the physics run the test at a slow speed, then a high speed, adjust only the downforce and re-run the test and if there is a difference drag is there, if it's exactly the same, there is no drag.

I'm not at home to run the test, but I will when I get home.

 

[Stotty]

Haven't tried it like that, but if drag was modelled correctly you'd lose top speed proportionally to how much downforce you used... and top speed remains the same no matter what you set downforce to.

Much like softer tyres lasting longer than hards, this is bloody irritating. It means you may as well just max the aero as it only has a positive effect on the car.

 

[No_OBsT33R]

Haven't tried it like that, but if drag was modelled correctly you'd lose top speed proportionally to how much downforce you used... and top speed remains the same no matter what you set downforce to.

Could be up to an extent, maybe keeping top speed but loosing rate of acceleration.

 

[Stotty]

Could be up to an extent, maybe keeping top speed but loosing acceleration.

Which would be equally as silly as a high front ride height reducing understeer

Drag from downforce increases disproportionally with speed... so it should effect top speed much more than acceleration in the lower gears!

 

[No_OBsT33R]

Which would be equally as silly as a high front ride height reducing understeer

Drag from downforce increases disproportionally with speed... so it should effect top speed much more than acceleration in the lower gears!

It is as simple as running the test and seeing for sure, it wouldn't surprise me as PD seems to do things a lil odd from time to time.

Anybody in front of their system with GT5 running have 5 min to spare?

 

[yanaran]

This "talk" is all well and good, but what I'd like to see is some empirical testing. Somebody, anybody, who believes in this "theory" grab a car, tune it up the ordinary way or grab a decent tune for it from a tuning garage, run some laps, then reverse the f/r ride heights and run some more laps and then compare. These are repeatable and testable by all members since we can all grab the same car, same tune, same track and go nuts. Anyone willing to post their tune and their on track results? I'd be more than happy to run the same car, same test, and post my results as I'm sure would several others.

I ran the Lamborghini Time Trial 12 on Monza in the Gallardo with one of the tunes in Adrenaline's "all tunes" thread. With the posted ride height settings (I think -15 front and read) my best lap in many many hours of trying was a 1'46.4xx which put me solidly in the top50. With every other tuning option exactly the same, but front ride height set to maximum, I beat that time on the first lap. I proceeded to bring my best lap down to 1'45.9xx in and that put me top10 at the time (I've since been bumped down some). That's a .5 improvement and it all came from the almost complete removal of understeer, and even created some oversteer in some situations.

 

[No_OBsT33R]

The same test can be used on ride height, adjusting one side (raise or lower) and see variation on result for effect.

 

[budious]

Problem is nobody is doing controlled tests which means setting up a car as neutral as possible and testing one variable at a time and delivering specific observations about changes in handling. So far all I have seen is arbitrary setups with highly conjectured reviews as concrete evidence.

You can also have differences in cold lap best times and hot lap best times. Setting a car up with certain characteristics such as com > ext or com = ext usually results in slower cold laps with gradual lap time increases. A car with ext > com on the other hand will post faster cold lap times with gradual decreases in lap time as tires warm. I don't see any elimination of other factors in any of your arguments either.