Nose-Tipping, Rollovers, and 'Broken' Handling of MR Cars: An Analysis of GT6's Center of Gravity

I agree with everything the OP said and I will throw my hat in on the subject also, The game is broken, I can say this is total honesty I have been in an Evo 6 IRL and we got some air and it was jolly exciting but at no point midflight did the car tip either forwards or backwards it remained level, Granted I know not every undulation on every road is exactly the same but my point is something badly wrong is causing the cars in the game to not behave as they actually would and should in the real world.
 
Cars in Gran Turismo have never displayed proper mass, so these issues dont surprise me. Whoever programs the physics is probably the same guy from GT1. If they fix one issue they create another while still leaving legacy problems from 15 years ago still in there, like awful collision and crash physics, which are likely due to poor simulation of car mass. Every new physics engine feels like a simple upgrade of the previous instead of a total overhaul.

If everyone at PD outside of the car designers were fired tomorrow, future games would probably be much better off. Thats how bad its gotten.
 
Is ridiculous to clip a curb and roll over like a bomb exploded under you. I think they overdid the rolling parameter compared to GT5 that was almost impossible to roll over.
 
Hi guys, I´ll trow my 2 cents here. I agree with the OP, and I´ve posted on many threads about that. (please excuse my english beforehand)

The problem isn´t a FIXED COG, but the whay COG transfer from back to front on braking.

Remember Enthusia? Remember the v-Box wich shows the way the COG moves when driving? Well, in GT6 that seems to be like an on-off COG movement.

And that´s why the problem is with MR cars and not with FR. When you´re braking hard on a FR car the COG transfer to the front, the tires get all the grip on the planet, and it´ll turn perfectly well. On the other hand, being the engine on the front causes a LOT or weight to be shifted to the front, and that´s causing the stoppies on FR, and not so much on MR cars.

With all that said, the lack of grip that we are experiencing on MR cars happens to be caused by the same COG transfer. When on FR cars you have a lot of grip on the front, wich is good, on MR the lack of grip on the back causes the snap oversteer that is happening.

BTW, I´m doing a lot of testing, and the problem ISN´T A HANDLING PROBLEM. It isn´t about "You suck at driving/You are the best driver on the world"... it happens when you brake, or when the car lifts on a bump.

The driving model is lovely. I LOVE GT6 to death, is just that, but the COG transfer has to be tuned down just a little. And that why if you add ballast to the front on MR cars the problem is someway mitigated.(the COG is already moved the front, so the transfer isn´t that drastical)

Don´t be apologist people. We ain´t here to bash GT6 like a "diehard forza fan". All of us have bought GT6 and will keep on playing for years to come. It´s something that has to be fixed. If we took this attitude of "NOTHING IS WRONG" we´ll never have a Spec II like we all had on GT5.

Peace :)
 
Last edited:
Just saw this in another thread. Note that the Deltawing has a profound rear COG bias.


It's incredible, really. Obviously a bad first turn on my part. I accept that. Didn't control it well. Then I accelerate, and suddenly I'm scraping the back of the car on the ground, unable to change course since the front tires are off the ground.
 
Is ridiculous to clip a curb and roll over like a bomb exploded under you. I think they overdid the rolling parameter compared to GT5 that was almost impossible to roll over.
Not only that, they made it impossible to roll on Trail Mountain's famous hump at the end!
 
Just saw this in another thread. Note that the Deltawing has a profound rear COG bias.


That's actually the least unrealistic video I've seen in this thread so far. The car has little front downforce, plenty of torque and went over a kerb straight on. I doubt it wouldn't take off.

The only things that are exaggerated are the stoppies (actually posted a video) and that some MR cars have exaggerated lift off oversteer. The RGT was getting lift off oversteer at like 60 km/h on tsukuba. I know it's rear engine'd, but I'm pretty sure the R8 LMS I drove at the demo would do the same.
 
Not that I've driven it in GT6, but in GT5, I found the R8 LMS very easy to drive. I never had any particular problems with any cars. One car I find odd in GT6 is the GTByCitroen Road car, in that when tuned to 800bhp+, anything more than minimal front downforce seemed to make the nose of the car dive towards the inside of a cambered corner (high speed ring for example).

This also proved problematic even on flat tracks. Not sure if this is realistic or not, but when rear downforce was full on all setups I tried, it's frustrating when any front downforce above minimal makes the car extremely difficult to drive.

I've only driven the GT by Citroen Road Car in stock form, but I simply cannot go through the first turn on High Speed Ring with decent speed without spinning to the inside. I was using Sports Soft first, then even put Racing Hards on … same result every time. And that with a car that has great handling and is easy to drive at Silverstone. Now, obviously I haven't driven that car in real life, but I simply cannot imagine why that car would behave that differently in a banked corner. I tried it with every imaginable throttle position, nothing worked. It's like you have no rear grip at all. The only thing that does work is going through 50km/h slower than with a stock Corvette and that can't be right.
 
Although I agree with the weird stuff, e.g. roll overs at will etc. I think this is the most 'natural' the physics have been in a Gt game. Maybe it's not actually realistic, but it feels like it is.

I would dislike it if they made changes now I think.
 
Last edited:
Although I agree with the weird stuff, e.g. roll overs at will etc. I think this is the most 'natural' the physics have been in a Gt game. Maybe it's not actually realistic, but it's feels like it is.

I would dislike it if they made changes now I think.

It completely depends on what the changes are. If the physics give an even better feel I am all for it :)
 
This was posted by @GENERALSAVANT in another thread.
It almost seems like gravity on the right side of the car is reversed after he hits the curb.



He is turning, probably already has a wheel off of the ground, he taps the curb, moves the right side up a little higher and inertia does it's thing and flips the car, it's realistic, but maybe a little too fast.
 
He is turning, probably already has a wheel off of the ground, he taps the curb, moves the right side up a little higher and inertia does it's thing and flips the car, it's realistic, but maybe a little too fast.

It seems to happen in two phases. First he hits the curb, then for a split second nothing much happens and then he suddenly flips over. My sense of forces, which I developed by living on earth for almost 34 years now tell me this is not natural... If it was one fluid movement I maybe could've lived with it.
It just seems like there is an invisible guy at the side of the road giving it an extra push or if he suddenly catches a huge burst of wind from the side.

This is what I see in many of the roll over videos, especially the ones that are on the edge of rolling over. They all suddenly get a push (indeed like a bomb explodes under the car) and flip over. In many cases the gravity should have pulled the car back to the road or at best should've made it a close call, but instead the car just gets this huge kick and starts flying.

If physics truly worked this way we would see a lot (a LOT) more cars flip over in real life.
 
I've only driven the GT by Citroen Road Car in stock form, but I simply cannot go through the first turn on High Speed Ring with decent speed without spinning to the inside. I was using Sports Soft first, then even put Racing Hards on … same result every time. And that with a car that has great handling and is easy to drive at Silverstone. Now, obviously I haven't driven that car in real life, but I simply cannot imagine why that car would behave that differently in a banked corner. I tried it with every imaginable throttle position, nothing worked. It's like you have no rear grip at all. The only thing that does work is going through 50km/h slower than with a stock Corvette and that can't be right.

To make the Citroen work, I had to fully increase rear downforce, decrease front downforce and fully raise the suspension. my knowledge of real life handling characteristics isn't fantastic but I'm pretty sure I shouldn't have to do this just make a car take a cambered corner without spinning immediately (I say immediately because it still wanted to dive towards the inside, I had to change my steering input to minimum for it work).
 
Take the R8 LMS Ultra per example, you can entirely prevent the waving and snapping while braking above 300km/h by setting up a 135kg ballast up front. Or, and this is the interesting part, you can achieve the same result by setting the car height to 70-55 or 80-60 and suddenly the rear stays somewhat straight under light and heavy braking. This shows that either the rear lighten (lift) up a lot more than it should or that the center of mass is programmed higher than in reality.

The Honda Raybrig NSX '06 with almost the same weight distribution (46:54) as the R8 LMS Ultra(45:55) does not exhibit any waving or snapping under extremely hard braking, which even the BMW Z4 GT3 with a weight distribution closer to the front does have. Furthermore, if you tune the suspension of the NSX to exactly that of the R8, from ride height to toe-outs, same weight and distribution, a 350-600 downforce and racing hard tires; so technically from the game's perspective almost the exact same car except center of mass and wheelbase, where the R8 would twitch under the slightest touch of the brakes and flip if you did brake above 350km/h, you can jump both feet hard on the brakes of the NSX and it will remain straight and composed. I've tested both at equal speeds on LeMans 2013, so something is wrong with the R8.

That said, the weight of the car seems to have a tendency to pivot (pitch) around the point of friction beyond where the weight transfer would push the wheel(s) into a slide, like we witness, to an extreme degree, with AIs. Is the center of mass too high? Are the physics to blame? Combination of both? I don't know, but if the R8 LMS Ultra was as tricky to keep in a straight line under braking at high speed in real life as in GT6, I doubt anyone would dare race it.
 
Not only that, they made it impossible to roll on Trail Mountain's famous hump at the end!

I managed to do that last night though...

Although I agree with the weird stuff, e.g. roll overs at will etc. I think this is the most 'natural' the physics have been in a Gt game. Maybe it's not actually realistic, but it feels like it is.

^ This is pretty much my thoughts on it; but I think that some minor tweaks are really all that is needed with the physics right now.
 
It seems to happen in two phases. First he hits the curb, then for a split second nothing much happens and then he suddenly flips over. My sense of forces, which I developed by living on earth for almost 34 years now tell me this is not natural... If it was one fluid movement I maybe could've lived with it.
It just seems like there is an invisible guy at the side of the road giving it an extra push or if he suddenly catches a huge burst of wind from the side.

This is what I see in many of the roll over videos, especially the ones that are on the edge of rolling over. They all suddenly get a push (indeed like a bomb explodes under the car) and flip over. In many cases the gravity should have pulled the car back to the road or at best should've made it a close call, but instead the car just gets this huge kick and starts flying.

If physics truly worked this way we would see a lot (a LOT) more cars flip over in real life.

The original posters video was fluid though :)
 
Hahaha, just had to say it. As soon as I edited it I could foresee way too many jokes about being American, thus geographically stupid.
Meh take a map of the US and ask the average European to point out Nebraska or Pennsylvania and he wouldn't have a clue neither; a bit the same concept as with Europe and it's countries.
 
Back