Ride Height Glitch Returns in GT6

Yeah, I guess GT6 over exaggerate the effects :) I think PD should reduce the effect, I tested on Stratos, even 2mm difference can be felt :eek:, it should take at least 20mm for the same 2mm effect.
 
in THEORY setting up a neutral car with level ride height, setting ballast at 50 and increasing the ballast by set amount till we get the same turn in as the RHG would give us the % factor it has in the game. Same with toe or any other setting that induces oversteer.

@Motor City Hami to answer your previous query about the Elise being brought up, if it was bad programming and the physics where truly reversed than using the same theory just in reverse High rear low Front would induce understeer in a natural oversteering car like the Elise thus giving a RHG effect for the rest of the cars in the game not just the understeering one, as tests and previous seasonal are showing this is not the case and something specific to the front ride height being raised. *is my understanding of whats being shown and discussed so far in here.
 
donpost - I've never really thought about the point of rotation before. So playing with the ride height moves this point? I wonder if "point of rotation" and balance are linked - oversteer = rotation point near front of car and understeer = rotation point near back of car? Not sure.

I did a quick calculation: On a car with a 50/50 weight distribution, a wheel base of 2m and a CoG height of 75cm, raising the rear ride height by 100mm would only move the CoG forward by 37mm. This would change the weight distribution to 50.85F/49.15R.

I actually wrote that paragraph above expecting it to show that ride height doesn't really affect the CoG much but actually it has (although both the adjustment of 100mm and a CoG height of 75cm are large, possible overestimates)

So raising rear ride height = move CoG forward = more understeer/less oversteer and raising front ride hide does the opposite. So maybe things are not back to front after all, and its just a CoG moving thing - is that what you have been getting at all along and I've only just twigged on?



This theory makes the most sense by far and would be the easiest to programme so i'm assuming this is whats going on, I'm almost certain of it.
 
It used to actually say in the GT tuning section ( in GT3/4 ? ) that uneven ride height changes weight distribution.

I think PD exaggerated the effect to allow some weight distribution change, before ballast became available. And the effect is still there now.
 
It works to a degree but it would be nice if they fiddled with the sums a little so the effects were less pronounced but it depends how likely that would be to mess up other things that use the same algorithm for data. There seems to be no real negative side effect apart from on top speed tunes maybe where it would mess up the air resistance, but then again why would you worry about the cornering on a SSRX car.
 
Hi guys, another test/theory which I'd really appreciate some opinions on.

For this test I used the Eunos Roadster J limited '91 (premium). FC suspension installed. (I happened to set all dampers and ARB to 1 but I don't think it's of concern for this test). Camber and toe stock, then set her up with max front/min rear RH. Start arcade mode (I used Tsukuba, the back straight is flat), immediately park, and turn the steering full lock. Stop, go to replay, and close up photo the "inside" front wheel. Use the portrait view so the edges of the frame give you a good vertical guide. You will see a lot of camber inherited from the steering at full lock.

Now do the same, but with min front/max rear RH. I thought I saw a difference, in that the low front setup showed less camber at full lock. However I'm not sure if there was discrepancy in how I positioned the camera, because sometimes if you don't picture the wheel exactly head-on, it's hard to ascertain the true camber. Also, there is the risk of accidentally moving the camera off from it's initial vertical alignment. It'd be really great if anyone else could prove/disprove this.

If there is a difference, is the rake somehow changing caster/steering axis? I'm imagining the situation with a 4 legged table (your car and suspension struts). If you cut 2 legs shorter, not only will the top of the table be leaning (rake), but your table legs will no longer be vertical. This does assume that the top of the suspension is modelled at right angles to the car body, like a table top and table legs.

I'm aware that I'm taking the leap of faith of using the graphics model to infer an effect on the physics. I've also not thought through whether this would give the backwards effects on handling that people have seen, I literally was just mucking about with race photos, and thought something looked weird.

Will post pics if I get time today, mega busy :(
 
bread82
Hi guys, another test/theory which I'd really appreciate some opinions on.

For this test I used the Eunos Roadster J limited '91 (premium). FC suspension installed. (I happened to set all dampers and ARB to 1 but I don't think it's of concern for this test). Camber and toe stock, then set her up with max front/min rear RH. Start arcade mode (I used Tsukuba, the back straight is flat), immediately park, and turn the steering full lock. Stop, go to replay, and close up photo the "inside" front wheel. Use the portrait view so the edges of the frame give you a good vertical guide. You will see a lot of camber inherited from the steering at full lock.

Now do the same, but with min front/max rear RH. I thought I saw a difference, in that the low front setup showed less camber at full lock. However I'm not sure if there was discrepancy in how I positioned the camera, because sometimes if you don't picture the wheel exactly head-on, it's hard to ascertain the true camber. Also, there is the risk of accidentally moving the camera off from it's initial vertical alignment. It'd be really great if anyone else could prove/disprove this.

If there is a difference, is the rake somehow changing caster/steering axis? I'm imagining the situation with a 4 legged table (your car and suspension struts). If you cut 2 legs shorter, not only will the top of the table be leaning (rake), but your table legs will no longer be vertical. This does assume that the top of the suspension is modelled at right angles to the car body, like a table top and table legs.

I'm aware that I'm taking the leap of faith of using the graphics model to infer an effect on the physics. I've also not thought through whether this would give the backwards effects on handling that people have seen, I literally was just mucking about with race photos, and thought something looked weird.

Will post pics if I get time today, mega busy :(
Click to expand...
This might have some substance.

High front low rear would give more castor, meaning more negative camber on the outside front tyre at full lock.


But camber is... Broken?

No?

What?

:dunce:
 
I don't think castor is modelled either, it makes real world sense but possibly not so much within the game
 
This was debated in the old camber forum. Camber is broken therefore caster and suspension geometry is also out of whack.
 
DolHaus
I don't think castor is modelled either, it makes real world sense but possibly not so much within the game
Click to expand...
killerjimbag
This was debated in the old camber forum. Camber is broken therefore caster and suspension geometry is also out of whack.
Click to expand...
Yes, I agree completely.

But think about his table analogy.

Now imagine that the rear of the table has the shorter legs.

The suspension (wheels) in gt6 move up and down, in a linear motion. The axis on which they turn (when steering) , can be presumed to be rotating around the same line that the suspension moves along, this is the front legs of our table(stay with me)

So if we have any amount of castor, even 0 (which is dead upright, like our table legs) at a level ride height, then dropping the rear, will lean the top of the leg backwards, therefore giving castor to the front (and rear) struts.


I hope that made sense.


The fact we can't adjust castor, doesn't mean is doesn't exist, it just implies that the struts are vertical when at a level ride height. Adding any rake will put the struts off vertical.

That makes more sense.
 
FussyFez
Yes, I agree completely.

But think about his table analogy.

Now imagine that the rear of the table has the shorter legs.

The suspension (wheels) in gt6 move up and down, in a linear motion. The axis on which they turn (when steering) , can be presumed to be rotating around the same line that the suspension moves along, this is the front legs of our table(stay with me)

So if we have any amount of castor, even 0 (which is dead upright, like our table legs) at a level ride height, then dropping the rear, will lean the top of the leg backwards, therefore giving castor to the front (and rear) struts.


I hope that made sense.


The fact we can't adjust castor, doesn't mean is doesn't exist, it just implies that the struts are vertical when at a level ride height. Adding any rake will put the struts off vertical.

That makes more sense.
Click to expand...
It makes very good sense but my thinking is if castor is not modelled and therefore not considered an input factor, would it have any effect on the overall suspension model? It would change the angle for sure, but would that be recognised as a change by the programming?
 
Well in therory, yes it would make sense. But if camber is broken, which the consensus is yes,then caster is irrelevant.It doesn't work. Don't bang your head against a wall trying to theorize what PD programmed into this game. Nobody knows. I can send some T3's if you need them.
 
Back in 1999 NASCAR ran high fronts and low rears in Daytona 500 qualifying. It's use was banned shortly after. I recall it may have been to reduce drag on the rear spoiler. Is there anything about that theory that is relevant to this discussion?

 
Motor City Hami
Spec Miata '97 SR-Limited

100/80 ride height had good corner rotation
120/80 better
130/80 best
140/80 back toward understeer
150/80 closer to stock understeer
160/80 worse than even ride height

There seems to be a bell curve involved.
Click to expand...
This would make sense with shifting the centre of gravity, there's definitely a sweet spot before a decline. When I've seen it applied the distance between the heights has been increased in correlation with the nose weight bias or natural understeer tendencies.
 
kapnk006
Back in 1999 NASCAR ran high fronts and low rears in Daytona 500 qualifying. It's use was banned shortly after. I recall it may have been to reduce drag on the rear spoiler. Is there anything about that theory that is relevant to this discussion?

Click to expand...

I don't think it is to do with aero in this case but interesting to know, would be interesting to see if any max speed advantage could be gained on a Nascar by running a higher nose?
 
DolHaus
I don't think it is to do with aero in this case but interesting to know, would be interesting to see if any max speed advantage could be gained on a Nascar by running a higher nose?[/quote]

Not sure about GT6, but in GT5 that was indeed the case. My RH settings for my NASCAR cars was 15/12 for Daytona Superspeedway, similar for Indy and Motegi. I couldn't go any lower with the rear otherwise it would bottom out in the corners. I haven't messed with the NASCAR cars much, if at all, in GT6 so I'm not sure if it still holds true.
Click to expand...
 
That was 1999 different aero packages. Not saying I'm pretty good at NASCAR, but if you need a setup or any questions, well PM me.
 
DolHaus
Interesting, did you notice where the extra speed was being generated (straight line/cornering speed)?
Click to expand...

Both, but mostly in the corners. From mid corner on out, I was roughly 2mph faster with those settings as opposed to any other RH settings I tried. I continued picking up speed going down the backstretch also.

@killerjimbag Aero had nothing to do with it, at Daytona anyway, it was purely a RH thing. That's why I was so secretive with my Daytona settings back then, not a lot of people knew about setting up the car that way. Not to rehash old memories, but you know how good I was with a NASCAR car in GT5. If you have selective memory :lol:, I got a lot of other people to back up my claim. ;)
 
Last edited:
In GT6 adding ballast for day and night makes a difference. If you gentlemen would like to try a race, I'm sure myself and I know Cargo can tune NASCAR would be willing to help anyone out. Not as easy as it looks.
 
DolHaus
It makes very good sense but my thinking is if castor is not modelled and therefore not considered an input factor, would it have any effect on the overall suspension model? It would change the angle for sure, but would that be recognised as a change by the programming?
Click to expand...
That's my point.

Even if it's not adjustable, the sheer fact that it's a car, with suspension that moves up and down along the path of the strut.

The steering must have an axis on which to turn, even if this axis is dead plumb vertical (at level ride height) it is still there. This is castor, whatever angle it is at, and whether we can adjust it or not, it's still there. Therefore adjusting the rake will affect the steering axis.

I suspect that if castor was adjustable, top tuners would be adjusting it in conjunction with rake to balance any rake adjustments, and retain the desired castor.

I agree entirely that it's a long shot. Because camber doesn't seem to be working. But I think it is indisputable that changing the rake, will change the angle of steering axis, and therefore castor.



But I agree. I don't know what's being taken into account by the physics etc.
 
A small mention i would like to add from real world exp concerning how ride height on mcpherson strut vehicles actually have a sweet spot. I actually had custom made autocross springs made for my car. Which means the company spun a set of coils, installed them, tracked the car, gathered telemetry, and made adjustments. EVERY other spring made for performance applications dropped the vehicle 2+ inches. Conventional wisdom would suggest that lower COG would keep the weight from shifting and result in a better handling car.

Surprisingly, the end result was a ride hight that set my control arms parrallel to the ground. Around 1.7" of drop. When the car was lowered further, the leverage from the grip of the car will be transfered increasingly into the strut, increasing body roll and instability. Being a fwd car, the best set-up was to get the control arms parrallel so that forces would be transfered at the bushing of the control arm, the lowest point of the vehicle.

As much as i know gt is not very good at simulating accurate suspension geometry and the physics behind them, im inclined to think that this could be somthing that was incorporated accidentally while testing various vehicles. For the most part, a gigantic chunk of gt cars are mcpherson strut, however even the double a-arm suspension will follow the same trend, just less extreme.

If you think about it, race cars are usually engineered with ride height, tire size, suspension travel and COG usually worked out before suspension locations are plotted. There is an ideal range of travel for any given car and outside of it weight gets transfered due to leverage higher up the body, causing roll and instability. Just think about lifted trucks, after a certain point, you must lower the entire suspension to be within an acceptable range.

I realise that considering castor isnt even adjustable that the suspension model is rudementary. However, its my theory that a general model was created to replicate as close as possible to real world results in the same car. It would stand to reason that, if the model is designed to replicate real world performance, the oversight would be a suspension model that reacts to changes like in the real world. However, in comparison, they would use a variety of tunable cars to see what results change various aspects of the vehicle and have accidentally mistaken the suspension leverage with over/understeer as it does in real life. It just doesnt carry over to each and every car, as every measurment of the geometry makes each car unique to another.

Back to my real life expierence, i drove an identical car to mine, with the exception of springs. Same front bar, same dampers, same rims and very similar tires. The alien car had springs about 3/4 of an inch lower than mine, however it wasnt nearly as predictable in mid speed corners and weight transfer. His control arms were \/ (exaggeration) and was the only difference i could pinpoint as the source of the change. This is the pivot point of the body roll.

Anyways, enough for me. Just my .02 cents
 
This third line will cross the vertical centerline of the car. The place where this third line comes back and intersects the centerline is called the “Instant Center”, or roll center, in chassis design terms. This is the theoretical pivot point of the front suspension during the beginning of body roll. Now simply measure the dimension from the pavement up to the instant center (you have been drawing all this to scale, haven’t you!). In a further analysis, this point is then compared to the instant center of the rear suspension to determine the roll axis of the car, but that is another chapter. If you do all these diagrams, even if you just roughly sketch them, you can see how the angle of the upper and lower control arms can really affect the location of the intersecting lines, and therefore the instant center. In general, basic handling rules would like the front instant center to be low. If both upper and lower control arms are going downhill towards the center of the car, it is possible to have the instant roll center so low that it is below the pavement. Since it is only a theoretical point, that is OK, but it really does not help the overall roll of the suspension. When compared to the rear instant center it will generate a very steep roll axis, which will understeer excessively. Again, this for another chapter.
From this page
http://www.maxchevy.com/columns/hottopic/ii_2-heidt-1.html
 
Last edited:
weston19
A small mention i would like to add from real world exp concerning how ride height on mcpherson strut vehicles actually have a sweet spot. I actually had custom made autocross springs made for my car. Which means the company spun a set of coils, installed them, tracked the car, gathered telemetry, and made adjustments. EVERY other spring made for performance applications dropped the vehicle 2+ inches. Conventional wisdom would suggest that lower COG would keep the weight from shifting and result in a better handling car.

Surprisingly, the end result was a ride hight that set my control arms parrallel to the ground. Around 1.7" of drop. When the car was lowered further, the leverage from the grip of the car will be transfered increasingly into the strut, increasing body roll and instability. Being a fwd car, the best set-up was to get the control arms parrallel so that forces would be transfered at the bushing of the control arm, the lowest point of the vehicle.

As much as i know gt is not very good at simulating accurate suspension geometry and the physics behind them, im inclined to think that this could be somthing that was incorporated accidentally while testing various vehicles. For the most part, a gigantic chunk of gt cars are mcpherson strut, however even the double a-arm suspension will follow the same trend, just less extreme.

If you think about it, race cars are usually engineered with ride height, tire size, suspension travel and COG usually worked out before suspension locations are plotted. There is an ideal range of travel for any given car and outside of it weight gets transfered due to leverage higher up the body, causing roll and instability. Just think about lifted trucks, after a certain point, you must lower the entire suspension to be within an acceptable range.

I realise that considering castor isnt even adjustable that the suspension model is rudementary. However, its my theory that a general model was created to replicate as close as possible to real world results in the same car. It would stand to reason that, if the model is designed to replicate real world performance, the oversight would be a suspension model that reacts to changes like in the real world. However, in comparison, they would use a variety of tunable cars to see what results change various aspects of the vehicle and have accidentally mistaken the suspension leverage with over/understeer as it does in real life. It just doesnt carry over to each and every car, as every measurment of the geometry makes each car unique to another.

Back to my real life expierence, i drove an identical car to mine, with the exception of springs. Same front bar, same dampers, same rims and very similar tires. The alien car had springs about 3/4 of an inch lower than mine, however it wasnt nearly as predictable in mid speed corners and weight transfer. His control arms were \/ (exaggeration) and was the only difference i could pinpoint as the source of the change. This is the pivot point of the body roll.

Anyways, enough for me. Just my .02 cents
Click to expand...
Very interesting, nice to have some real life comparison and observation.
 
HTR Tuning
so when im tuning a car what should i do with the ride height ?
Click to expand...

Theories aside, try front higher than rear on cars with understeer. I am finding that max/min is not as fast as something more balanced like 125/80. I have not tried the reverse on nasty handling MR cars with oversteer. Someone should test that.
 
Hami I have to agree. Theories are great with real world physics. Have tuned some cars higher front lower rear, seems to work pretty good. Prime example, go look at the Triumph Spitfire in the dealership. Self explanatory. Maybe PD got that right.
 
Motor City Hami
Theories aside, try front higher than rear on cars with understeer. I am finding that max/min is not as fast as something more balanced like 125/80. I have not tried the reverse on nasty handling MR cars with oversteer. Someone should test that.
Click to expand...

I've been driving some of the Stratos tunes in the Difficult Car thread and low/high definitely decreases oversteer. It's the similar with well balanced FR like the S2000, makes it understeer.

As to the theories, they are fascinating to read but I'm always drawn back to two things that say PD have screwed up ride height.

1. They had the same issue in GT5, tried to correct it, failed and just neutralised the effect of rake altogether.
2. The in-game description says raising rear ride height will induce oversteer but it clearly doesn't work in they way they intended. So, if anyone comes up with a plausible real world theory for PD's settings, it's no more than coincidence.
 

Latest Posts

Back