Gran Turismo 6 general Physics Discussion(as well as video)

[Ridox2JZGTE]

A video of me doing some slalom with the 370Z :

​

Real life Manji Slalom :

 

[Raitziger]

Physics and tire model seems improved. FFB was too low even on 10/10 setting with my turbo S wheel. I have to adjust new settings to my wheel to compensate the FFB strength loss today when I test more.

 

[Tenacious D]

I think discussing the GT6 demo in the context of other games should be completely valid, especially PC sims now that GT6 is offering physics of such good quality. This shouldn't be a reason to argue or get hurt feelings. And if you do, move along right now because I'm going to discuss a certain other racer as well as PC sims.

Well the inconsistency in laptimes is not down to an inconsistency in physics. It could just be a result of the greater variety of vehicle behavior in Forza, but to be playing for a long time and still have the issues is odd.

After digging up my PC racing rig and 360 Leet back in February I think for a Saturday race off, I concluded that both Forza 4 and GT5 were both so good that PC sims really didn't have anything major on either game. This is before rFactor 2 and Project CARS release of course, and they could change everything, but I still think F4 and GT5 have reached a point that you don't need a PC racing rig to have very satisfying and realistic racing.

Why some people can dive right into Forza, and others like me have issues with it, I'm not sure. I think it's because of a combination of having to use chase cam which isn't my favorite view but works best for me, and Turn 10s 3D perspective they developed which is similar to Need For Speed's. The physics they developed give the game life and dynamics which I only see in Live For Speed, but no other racing game. Well, but now much of that in the GT6 demo. While Forza is moving to much more powerful next gen hardware and GT6 isn't, it does seem that Kazunori and the team have still infused many aspects of real world physics that GT5 was lacking. I'll be very interested to see how GT6 compares to Forza 5, rFactor 2 and P CARS. But right now, this demo looks to bode well for GT6 standing pretty tall in comparison to this very stiff competition.

 

[CoolColJ]

Re the controller thing - I found Richard Burns Rally unplayable with a pad, with 1:1 setting. I had to use the in game options to change the curve to a non-linear controller mapping etc.
So while the physics didn't change the game responce sure changed. And my car control improved dramaticly
Same for Netkar pro

With my original comment, regarding the RC controller, it was with a real car I meant. Trying to do a salom in a real car with a RC controller would be impossible I feel

 

[TorqueHappenS08]

Still looking forward to that hardcore mode, which appears not to be in the Demo.

 

[WRT IVORBIGUN]

As im so out of touch with GT5/6 could anyone advise with some settings.

I noticed I could change my steering settings from Amateur, Pro, Sim. What is the difference and how does this actually affect the physics with my G25 wheel. Also is everyone who is using the wheel keeping the steering sensitivity on default?

Appreciate the advice.

 

[Griffith500]

Hm... Just had a thought. I want to put slick tires on relatively wide cars in GT6 final and see whether the suspension practically rips off in response.

Hopefully the bump stops are properly modeled...

In theory, hitting the bump stops would cause the car to corner worse, because it's a sudden spike in spring rate, meaning a momentary loss of grip at that wheel (depending on the vertical loading; hitting the bumps stops in a dip in the road is different, but the spring rate thing applies).

It would in any case make it very spiky to drive at the limit, more so than an "appropriately sprung" car on slicks tends to be already. Should be interesting indeed.

 

[Johnnypenso]

As im so out of touch with GT5/6 could anyone advise with some settings.

I noticed I could change my steering settings from Amateur, Pro, Sim. What is the difference and how does this actually affect the physics with my G25 wheel. Also is everyone who is using the wheel keeping the steering sensitivity on default?

Appreciate the advice.

Doesn't work with the G25. Only works with the wheels that are animated when you change the settings and the Logitech wheels aren't included. I don't believe the sensitivity has any effect either.

Regarding the "physics" of Samus' slalom run, how can anyone know how a real life car will behave if it was driven with a PS3 controller? I'm sure lots of things are possible if you could steer your real car with a controller instead of the wheel. Take the same car and put it on CS or CM's and I wonder if it would be as easy to control. How about producing a video of a real car driven with a DS3 that can't do what was done in the video as proof the physics are off?

 

[Goruk]

Physics

 

[AKps3]

edit: to late

 

[Lock2Lock]

Doesn't work with the G25. Only works with the wheels that are animated when you change the settings and the Logitech wheels aren't included. I don't believe the sensitivity has any effect either.

Yep...
The amateur, pro, and simulation settings don't have a affect on the wheels that are not highlighted.

But the "power assisted steering" and the FFB adjustment does have a effect.

 

[townnet]

Physics

What's your point? Even the best of racing SIM in this world also have bug.

 

[Exorcet]

If you steer quickly enough you need never run out of lock either because you control the momentum of the transition, but with a wheel that can be difficult. Easier with a pad (wasn't that the point of the argument?).

That's not necessarily true. There are inputs which lead to unrecoverable behavior. You can see them on Nyquist and Root Locus plots.

http://en.wikipedia.org/wiki/Nyquist_stability_criterion

http://en.wikipedia.org/wiki/Root_locus

 

[Ills1999]

Tornado I stated my opinion which was honest, I never pointed out a certain game that Gran Turismo stomps on, but then he comes(Exorcet) and sais Forza has better physics, so who started it..? I do feel that GT6 is the next generation sim, and I do feel that Forza games are simplistic, I use a steering wheel in GT and I own car, I know I am right because my experience tells me so. I said that the overall physics package in Gran Turismo is the best I have ever seen. I don't care much about other sim games the way I do about Gran Turismo.

So the obvious logical retort is that you're likely a PD/Sony agent.

 

[Samus]

What's your point? Even the best of racing SIM in this world also have bug.

The kerb bug isn't the physics issue, it's the weird floaty car physics once it get's airborne. The car seems to lose all weight and 'dance'.

 

[Cheshyrkat]

The kerb bug isn't the physics issue, it's the weird floaty car physics once it get's airborne. The car seems to lose all weight and 'dance'.

Looks like some programing is in place to assist the car in righting itself back onto its wheels. Kind of like a bottom heavy top or loaded die. Still looks weird and I would rather have the car just roll and if applicable slide on its roof.

 

[sinbad]

That's not necessarily true. There are inputs which lead to unrecoverable behavior. You can see them on Nyquist and Root Locus plots.

http://en.wikipedia.org/wiki/Nyquist_stability_criterion

http://en.wikipedia.org/wiki/Root_locus

Well if you try to spin a car then no matter how hard it "snaps" back the other way, it's only doing that because you kept steering it that way at the crucial moment. If you're simply saying that it should be possible to throw the car further than the steering lock allows recovery from, then I agree. My point was that impeccable and lightning fast steering would, of course, never allow that to happen in the first place.

 

[Lock2Lock]

Well if you try to spin a car then no matter how hard it "snaps" back the other way, it's only doing that because you kept steering it that way at the crucial moment. If you're simply saying that it should be possible to throw the car further than the steering lock allows recovery from, then I agree.

Great example of this here... https://www.youtube.com/watch?v=Ov-vzjtlDxU&feature=youtube_gdata_player
Skip to 1:20

 

[Wolfe]

If you're simply saying that it should be possible to throw the car further than the steering lock allows recovery from, then I agree.

That's it. Sorry if I wasn't clear; I was replying to your point that you have to be able to steer far enough, not ignoring it.

The fact that GT/Forza won't do this illustrates that there's a degree of handholding when you're oversteering. It makes the whole game easier, and if you're used to drifting in the real world or a more accurate simulator, it interferes with your technique. It's annoying, IMO.

 

[Exorcet]

My point was that impeccable and lightning fast steering would, of course, never allow that to happen in the first place.

No, there are situations where not even being perfect with the steering wheel would let you recover because any action would lead to an unstable motion.

Once the car reaches that point, it is gone.

 

[sinbad]

No, there are situations where not even being perfect with the steering wheel would let you recover because any action would lead to an unstable motion.

Once the car reaches that point, it is gone.

I don't know what that "point" is, or what that "unstable motion" is tbh. Even an intentional 360 spin could be stopped at the point where you choose to initiate it, and even stopped immediately after that point. No matter how violently you throw the car from one extreme angle to the other, with said uber-steering you could control it. If you steer the other way before the car even reaches straight, you reduce the inertia with which it "snaps" back massively. Controlling the inertia of recovery is ultimately how all slides are corrected.

There is no slide on a flat road with consistent grip that, if caught before you run out of lock, cannot be recovered completely.

 

[Exorcet]

I don't know what that "point" is, or what that "unstable motion" is tbh.
It is the point where a pole on the Root Locus crosses over to the right plane. Once an input strong enough to trigger that has occurred, it is physically impossible to recover from the slide. The car will depart from its direction of travel and then skid to a stop.

Even an intentional 360 spin could be stopped at the point where you choose to initiate it, and even stopped immediately after that point.

Before the spin sure, after it might not be possible to reverse. This isn't to say that the spin will go on forever because it can't. However you wouldn't just be able to make the nose of the car go right back to where it was originally.

No matter how violently you throw the car from one extreme angle to the other, with said uber-steering you could control it. If you steer the other way before the car even reaches straight, you reduce the inertia with which it "snaps" back massively. Controlling the inertia of recovery is ultimately how all slides are corrected.

But your tires can only do so much. There is no guarantee of reversing direction or being able to catch the slide after you do. If an input strong enough is applied, you get a right plane pole and the car is gone.

There is no slide on a flat road with consistent grip that, if caught before you run out of lock, cannot be recovered completely.

Even with "infinite lock" there are slides that can't be caught.

 

[Wolfe]

Many of the slides in Samus' video should not have been caught at all. They should have continued into spins or half-spins of various kinds, regardless of what the front wheels are doing.

It's not even just this one maneuver. The way the car pivots and rotates as he performs donuts and handbrake 180s, it appears almost weightless. Visually, the suspension modelling is the only thing conveying any sense of heft. It's nothing new to the demo -- Gran Turismo has always felt peculiarly "artificial" to me -- but I think the more Polyphony refine and upgrade the physics model they're using, the more it shows its age.

 

[arora]

I have not played the gta yet but based on what I am seeing and reading, this sounds eerily similar to the transformation of physics between gt3 and gt4.

Would make perfect sense in PD land of logic to step the second game on the same console back to an easier playability

 

[JerseyDriverSS]

New drinking game for any drinkers here. Take a shot everytime Ridox2JZGTE mentions no ABS. You are bound to get hammered.

 

[mister dog]

I've noticed this too in FM4; the slides get interrupted so to speak and the car doesn't have the tendency to throw itself back violently if you countersteer to abrupt.

Altough Forza is the better game in many points, i realized now going back to GT with the demo here that GT's physics are still the better of the 2. I once had an accident with my real car like this, countered the oversteer to much and the car slapped to the other side and ended up in the wall...

GT6 physics simulate these movements very good i must say, i had the same thing happening to me a couple of times already in the demo, i'm definitly a fan of the new physics 👍

You should actually play the game in the first place before attempting to comment on it in detail...

I've been playing FM4 for a year now, level 133 and 60 hours total driven on track (good thing we have a "my stats" section eh).
Anything to add still? Which level are you at?

 

[Samus]

New drinking game for any drinkers here. Take a shot everytime Ridox2JZGTE mentions no ABS. You are bound to get hammered.

I've been doing that for months now, the only time I stop is the for the AA meeting each month.

 

[sinbad]

your post

It's not a question of grip, you are asking very little from your front tyres except to roll where you point them. The only reason running out of lock results in a spin/half spin is that you can no longer sufficiently reduce the resistance of the front tyres enough for them to go where you want them to go. That's why f1 drivers have famously frequently locked the front wheels to achieve the same result, but obviously the freely rolling wheel offers less resistance than a locked sideways wheel. So yes, infinite lock would give infinite control of this and whilst you would be able to throw the car to wild angles, you would never lose control and so never spin unless you wanted to, no matter how rough you were.

Anyway, whilst I don't think the videos of weaving look particularly realistic, I don't think it's because the car should just spin out, the car is never recovering from beyond the lock stops with some weird srf type force or anything weird like that, and with such quick reacting steering there is no reason why the car should end up pointing backwards.

 

[GTPorsche]

New drinking game for any drinkers here. Take a shot everytime Ridox2JZGTE mentions no ABS. You are bound to get hammered.

I'll play the sobering up version. I'll take a shot every time he doesn't.

 

[Wolfe]

...whilst I don't think the videos of weaving look particularly realistic, I don't think it's because the car should just spin out, the car is never recovering from beyond the lock stops with some weird srf type force or anything weird like that, and with such quick reacting steering there is no reason why the car should end up pointing backwards.

There is some weird SRF type force. It's doesn't kick in after you've spun beyond the lock stops, it seems to help prevent the car from even getting to that point in the first place. Quick steering or not, you should not be able to whip a 3300lbs. car back and forth like a 3.3oz feather duster. Where is the momentum going?

You said yourself that the video doesn't look particularly realistic. It's not only a lack of spins we're talking about.