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

Discussion in 'Gran Turismo 6' started by TorqueHappenS08, May 17, 2013.

  1. nangu

    nangu

    Messages:
    19
    Well, I used to own a 207 GTi (a French hot hatchback) and the thing is heavy understeer prone, you need to use the handbrake or throw it really hard (the way you know you will loose a lot of lap time) on turn in order to get the tail loose.

    Modern FF mini hot hatchs are very understeery.
     
  2. LazyJK

    LazyJK

    Messages:
    12
    This is surprising since I was under impression (based on my experience) that almost anything with 4 wheels can oversteer without using such brutal methods like handbrake. Examples: Mitsubishi Space Star (a minivan) has nice, gentle, progressive lift off oversteer, a rental Skoda Suberb I spent 3 hours in on a wet skidpad was actually much more nervous and snappy than my old BMW 320i. In my experience all that you really need is some speed so that inertia starts to work, a slightly abrupt lift of the throttle and also a slightly snappy steering input and the back end starts to move around: not drift mind you, but definately helps to turn in. I do agree that if you mees up anything can plow straight on and completely fail to turn but it is rare and you have to specifically screw up in a major way.
    I will go back and run the sway bar/spring sweep yet again tomorrow, again on the Delta since this is a car that I guess everyone has won by now and that seems to be really hopeless handling wise.
     
    FussyFez and poumpoum like this.
  3. nangu

    nangu

    Messages:
    19
    The tires used is a major element too. Stickier tires, and you need to brute force the lift off / late braking oversteer

    Also, weight distribution and traction on FWD cars are the main factors on why they are understeery. On my example on the 207 GTi, have into account the PSA chassis are very very neutral too. In the other hand, a Clio RS is more easy to get oversteer by lift off, in example.
     
  4. OdeFinn

    OdeFinn

    Messages:
    2,643
    Location:
    Finland
    @LazyJK, how low ride have you put it, and how stiff LSD you have, which tires? You can pretty much eliminate turning of most cars just using one of those "incorrectly".
    FF cars liftoff oversteer starts working soon as you take ABS off.
     
  5. LazyJK

    LazyJK

    Messages:
    12
    In general I am trying to keep to real world tuning principles: front lower than rear, spring/roll bar stiffness either equal or with rear bias, shock rebound one click higher than compression (unless I'm trying to tune weight shift behaviour), LSD preload/power/brake 10/12-14/5. I prefer to keep cars on the slightly higher & softer end of scale since I believe suspension must be able to do its job. The grippier the tire the stiffer I usually go, but I am wary of lowering the car too much and I dislike the rock-hard 'wheel barrow' setups.
     
  6. OdeFinn

    OdeFinn

    Messages:
    2,643
    Location:
    Finland
    @LazyJK, ok. Your LSD is pretty "open". Maybe fixing harder lock on brake and power sides or just putting near zero camber rear and decent camber on front? (High camber on rear helps only during high speed cornering, but on low speeds not)
    Just checked (not yet tested) @Ridox2JZGTE 's tunes, there is many good LSD settings what can help a lot. Maybe he can give you some advices also.
    Will get back to this when home + time.
     
  7. TorqueHappenS08

    TorqueHappenS08

    Messages:
    2,960
    Location:
    United States
    speaking of tuning the tuning for camber...is all over the place..
     
  8. FussyFez

    FussyFez

    Messages:
    1,460
    Location:
    United Kingdom
    Have you only just noticed?

    Extensive threads debating this since launch lol.
     
  9. OdeFinn

    OdeFinn

    Messages:
    2,643
    Location:
    Finland
    @TorqueHappenS08, Camber values are bit out of real use yes, on certain type of tune still helps a lot to use camber.
    Camber it self works, but tire modelling is relying on "bricks" as tires, bricks what have small (too small) rubber cover on it, and when camber is higher than soft layer on tires it stops helping/working on driving.

    @FussyFez, really awkward thing this camber, just praying for proper update/fix for suspension/tire model what will solve this.
     
    super_gt and FussyFez like this.
  10. super_gt

    super_gt

    Messages:
    4,339
    Location:
    Bulgaria
    Last edited: Mar 3, 2014
    Tenacious D and FussyFez like this.
  11. FussyFez

    FussyFez

    Messages:
    1,460
    Location:
    United Kingdom
    Even LFS type deformation would do.

    In my opinion, GT needs tyre flex, castor adjustment, and tyre pressure adjustment. Then wheel width and offset options.

    Hopefully gt7 will see all of these.
     
  12. OdeFinn

    OdeFinn

    Messages:
    2,643
    Location:
    Finland
    @super_gt, even diving in ground would be enough for graphical way, just hoping at they fix that tire flex on somehow, current situation is just a bad joke to The Yokohama Rubber Co. Ltd.
     
    super_gt likes this.
  13. TorqueHappenS08

    TorqueHappenS08

    Messages:
    2,960
    Location:
    United States
    Yes, after working on my civic...it became alarmingly clear..didn't really notice it too much before.
     
  14. LS Chiou

    LS Chiou

    Messages:
    368
    I'd like to put it here as the update threads are flooded in no time.

    Does anyone find/feel the camber effect is changed after 1.07 ?

    I admit I've ignored camber for quite a long time. In trial runs of the latest BMW VGT, in addition to turning down the big camber angle in default setting and getting a better grip, I kind of feel something more.

    With some camber, say, under 2.0 / 1.0, it feels smoother going up and down the curbs, also smoother in the transition between coner and straight. Seems reasonable, so far so good.

    But with such mild camber, the grip is reduced significantly. At the end of straight, it's much easier to lock the tire under braking compared to 0 camber. And the cornering limit is lowered somewhat. It feels more slippery, progressive, though.

    So, I suppose the camber bug is not 'solved' yet. I mean, any street cars IRL can have such degree of cambers with slight mod and/or lowered suspension. And it shouldn't be like the situations in GT -- less grip in the corner. Practically, it's the opposite.

    Ah, this is probably nothing to do with the latest update.

    Any thoughts?
     
  15. Johnnypenso

    Johnnypenso Premium

    Messages:
    28,465
    Location:
    Canada
    You mean you want the guys who can't get camber to work, to figure out how to incorporate tire deformation into the game?[​IMG]

    Next you'll be asking them to do this[​IMG] :

     
    Hyland, super_gt and Motor City Hami like this.
  16. Griffith500

    Griffith500

    Messages:
    8,631
    Once you get visual tyre deformation working, the visual representation of a puncture is trivial. The physical aspects are very different, unless the visual deformation is based on a physical model, as is the case with e.g. LFS.

    I'd be surprised, though, if the camber issue didn't turn out to be a tyre model problem. I think they should focus on the physical aspects before worrying about the visual stuff, especially since their physical model is unlikely to be compatible.
    There's a chance it's the new suspension model, either by itself, or through interaction with the tyre model.

    I think this branch of the physics engine is a dead-end anyway, and PD should be working on a totally new system.
     
  17. Imari

    Imari

    Messages:
    11,508
    Location:
    Australia
    That or they should just license a high-end system to use as a backend for their data and visuals. There seems little point reinventing the wheel (ha!) when there's good solutions out there already.

    For example, I think Game Stock Car has proved that isiMotor2 has a lot of life in it still, and a lot of the older games based on it still hold up remarkably well physically against newer titles. Licensing a newer version like the rF2 engine would just be gravy.

    http://en.wikipedia.org/wiki/ISImotor
     
    super_gt likes this.
  18. Griffith500

    Griffith500

    Messages:
    8,631
    The time it'd take to get a handle on the idiosyncrasies of any third party engine would already have allowed the creation of a new engine that PD can tailor to their own needs. It's not a question of capability, just executing the changes at the right time. PD focus on feel, no matter the physical complexity. Most PC sims are very stale by comparison, even if they are "more accurate". I'd prefer PD went into the same level of detail on their own terms, just like the old NKP team (now Kunos, i.e. Assetto Corsa devs) did.

    Plus, I really hate the feel of ISI-powered games, spoiled as I was at a young age by Papyrus sims.
     
    Zer0 and Tenacious D like this.
  19. Imari

    Imari

    Messages:
    11,508
    Location:
    Australia
    See, I disagree. PD have not demonstrated that they have the capability to create a physics simulation even on par with the best of a decade ago. It may feel as good, but it's a hack. Whatever it's doing is not modelling physics on the same level as demonstrated by things like the camber bug.

    A lot of the time, a hack may be all that's required, but it certainly lacks the depth of response that more complex sims offer.

    And that's not a simulator. That's pretty much the definition of an arcade style racing game, one that focuses on the feel over the absolute physical simulation. :(

    It's just that the feel they're going for is something that suggests realistic driving instead of OMGAWESOMEDRIFTBRO, and that makes some people confuse it for a simulator.

    Maybe so, but I think it's more a matter of gameplay and FFB rather than physics. Would GT6 really be a worse game if the physics were based on [insert your favourite PC sim] instead of PD's proprietary system? They could still tune for the same "feel" (there are many FFB mods on PC and people tend to have individual preferences), but the underlying physics would be more complex and given the correct numbers, hopefully more accurate. Maybe they'd need to tone down the physics to fit on the console, but I sort of doubt it unless they go with one of the very recent and complex tyre models.

    Certainly. I don't think GT6 even reaches the levels of physical simulation that NKP did. You can't run before you can walk. The GT6 tyre model is rudimentary at best, even by the standards of ten years ago.

    I'll admit to liking Papyrus sims better as well. I picked ISI because it's well known to be in a bunch of stuff, but it's not the best engine available. I have rF2 and I almost never play it.

    I imagine Kunos would license the AC engine for the right amount of money, and there's probably other choices out there if you have the industry stature of Kaz. I imagine most companies would be willing to at least have the discussion with him.

    Take it as an example to demonstrate the point, not that I think that isiMotor would actually be the best thing to use.
     
    Cote Dazur and Johnnypenso like this.
  20. Motor City Hami

    Motor City Hami

    Messages:
    4,803
    Location:
    United States
    It is a bit mysterious to me that the seven post rig has not shown up in the game yet. It seemed like such a big part of development in the marketing of the game last fall. So why didn't it show up in the game early on? I think it is because the seven post rig data started to show the flaws in the physics model. Adding data to the mix should lead to adjustments to be made. But, when those adjustments produce random results, the tool becomes useless.
     
    Nato_777, Johnnypenso and donpost like this.
  21. MustangManiac

    MustangManiac Premium

    Messages:
    2,110
    Location:
    United States
    I don't recall for sure, but I thought they meant that they were using the 7 post rig for the tire modeling/physics in the game, not that it would be included for us to use. As for the reason why my guess would be that 99% of the people that play the game are not race engineers and wouldn't have a clue as how to use a rig. Here is an example of the data generated, does any of this make sense to you and do you see how it could be used in the very limited settings we have in the game?


    7-post_tools_640x302.jpg

    I am an engineer (electronics) and have been a car nut for over 40 years. I have built my share of hot rods, raced on both 2 and 4 wheels, restored a few cars and this doesn't make sense to me,
     
    Johnnypenso, donpost and poumpoum like this.
  22. Griffith500

    Griffith500

    Messages:
    8,631
    That's your opinion - I agree that their simulation isn't as physically forward (in terms of the calculations) as the brute-force platform (PC) typically allows, but I think you underestimate what a good hack can do. The level of sophistication and system-level thinking that has gone into the "hacks" demonstrates a degree of understanding that I can't actually begin to appreciate properly. But then, I don't know any mathematicians, which might help things.

    PD are perfectly capable, they have just chosen to keep a certain level of user-friendliness and to use what they have until they absolutely have to / can make the big change (which they will be working towards). The new suspension model is a statement of intent, currently hamstrung by (my perception of) the existing solver framework (which is why I say they need to re-evaluate that).

    It's what happens with every game, no matter the pretence. At the end of the day, you sit a big lump of jelly in front of it and say "how does it feel?". You can paint a veneer of "accuracy" over it, too, and everyone will be happy. Look at pCARS.

    "Feel" is about more than just FFB; keyboard controls have "feel". Systems. You're being one-dimensional in your analysis, over-simplifying again.

    NKP was made by one person, in terms of the major accomplishment of the feel of the game, which is the unique tyre model. Not to diminish the achievement (which is only possible by completely ignoring the status quo and doing your own thing - more of that, please), but I think PD have the muscle to pull in any expert they might want if they don't have their own already.

    And it's totally irrelevant, because it would still make more sense for PD to stick with their own models, which they are clearly continuously working on (and have a long-term roadmap for, just like everything else; e.g. sounds). It would cost less and give better results in less time. It would make more sense for someone like me to license an engine; I'd still rather make my own.
     
    Zer0, Firedragon and Tenacious D like this.
  23. Johnnypenso

    Johnnypenso Premium

    Messages:
    28,465
    Location:
    Canada
    I could not agree more with this, a hundred likes would not be enough. :lol: I used to work in the video business and our software was outdated. I found a company that had a ready made proven software package that met all of our needs for about $50k. The owner decided he was going to develop his own software so we spent $250k over the next two years upgrading the piece of crap software we already had in place and after two years...we had an underperforming but updated piece of crap. By that time I was out on my own and had purchased the very same software I recommended to my previous employer for my own stores and it was like night and day.

    Point is, you don't need to reinvent the wheel out of some misplaced pride. The money is there, the computing grunt should be there with the PS4, swallow or lease the software from AC or something similar and get on with it. Other than arcade games, I'd have to say that GT has the least sophisticated physics model of all the games I've played, even those that are several years old.
     
    Cote Dazur, super_gt and Imari like this.
  24. Imari

    Imari

    Messages:
    11,508
    Location:
    Australia
    Obviously this is something that we're unlikely to agree on, it's almost entirely opinion based. However I don't believe that a wrong calculation that spits out the right answer is ultimately as useful as the right calculation to give the right answer.

    Yes and no. If you take it as far as stuff that is developed purely as a simulator like the F1 systems, I doubt they would sacrifice any accuracy for feel. They need both.

    The way you mention painting a veneer of accuracy points to the difference I feel in some games. Some games are developed for feel, and then reverse engineered to respond to real-world adjustments in a correct way. Some games are engineered around real world constants from the beginning, and the "feel" is adjusted taking those constants as an absolute.

    The difference being how the systems handle edge cases. Simulation systems are developed because once in place all the behaviours are present "automatically", as it were. A system that is a black box designed for feel needs to have every behaviour explicitly placed in it.



    That you refer to pCARS as an arcade game developed for feel with a veneer of accuracy painted over the top makes me severely doubt that you're a member of WMD, and therefore probably have no idea what you're talking about.


    It certainly is, the physics behind it matter a great deal also. What I was trying to say is that FFB and gameplay matter a great deal, more than most people give them credit for.

    To go back to pCARS, it's completely possible to make it feel awful or awesome by tweaking the FFB, on the same physics. The same is true of any control scheme, how the inputs and outputs (if there are any) are handled is a big part of how the user perceives the physics.


    Uh, isn't that what I'm saying? That they probably should? Is there a significant difference between buying someone's system, and buying someone's time to come and make your system look like theirs?

    Again, we're unlikely to agree on this. I see no evidence that you could possible use to make the judgement which method of development would be cheaper, faster or more effective. Their current methods have resulted in a physics system which feels nice, but is demonstrably at odds with how it behaves with respect to real world values and tuning techniques.

    It begs the question of what they're trying to do. Because a simulation at least attempts to conform to known physical results, even though the perfect simulation will always be just out of reach.

    You may enjoy making your own engine more, but it's business. It's not about what's more fun as a developer, it's what results in a better product. See @Johnnypenso 's example above.
     
  25. Griffith500

    Griffith500

    Messages:
    8,631
    And what, pray tell, is the "right" calculation? If you think no-one else is using "hacks", I think you need to have a word with yourself and think about the depth of your knowledge on this issue.

    It's funny, because GT is designed forward from first principles, according to Akihiko Tan. So there's clearly more to it than what you're describing. As I said, you're over-simplifying. And there is no difference between the "black box" and the "simulation" as far as behaviours are concerned; that is, if you want certain behaviours, you need to consciously make sure they're there through either method.
    I never said any such thing. And I am a "member" of WMD. Sure is a lot of smoke and mirrors, handwavey nonsense over there, though - not from everyone, but it's there in a systemic sense. At least they have Ben Collins on hand to tell us it's accurate; he gets that impression from the driving feel, of course. But this is off-topic.

    The physics contribute to the feel, by interaction with the "inputs and outputs", including the player. So do the graphics and sound. System.

    You don't take something off the shelf and try to bend all the other components to fit. You design something as a complete package (system) and evolve the components, like you said with "feel". What takes longer: tweak an existing package, or redesign the whole package to accommodate a single totally new (in terms of data flow and so on) component?

    If all you want are "results", then the superficial "here's a thing that does what we want" method can work. It depends. Business practice dictates that a reskin is all you need anyway, so I don't know why we even care.

    As a counter example, iRacing wasted a lot of time and money trying to get a sound system like everyone else's, because that's what everyone was telling them they should do. They hired a guy to do just that, to make their system look like everyone else's. It was an abject failure, because they couldn't get the "off the shelf" product to play ball with what they already had, and it wouldn't do what they wanted (roadmap of improvements: e.g. drivetrain simulation) without significant modification anyway. They went back to developing their own in-house engine a year later.

    It's up to you, but I think maybe you shouldn't think of your own entertainment as "just business", and I wasn't talking about "fun", either.
     
    Zer0, Firedragon and Tenacious D like this.
  26. Imari

    Imari

    Messages:
    11,508
    Location:
    Australia
    Of course people are using hacks, the computer power doesn't exist to do everything at the limit of current physics knowledge.

    Perhaps I'm explaining badly. You can model a spring, a damper, suspension linkages, and a wheel separately. You can combine them and have a system that will describe the motion of a wheel attached to a car (mostly, I'm aware there's more to it in most cars but I'm trying to keep this short).

    Or you can just make a lookup table for every position and force that the wheel might encounter.

    The difference being that the first setup takes a few relatively simple parameters and produces an output. It's doing it's best to replicate reality, as far as the computing power will allow. Simplifications are made for complex stuff like say, fluid dynamics, but a simulator will always have some objective link back to reality.

    The second requires everything being precalculated. If you're precalculating from larger sim, fine. If you're putting numbers in from experimental data, fine, until you get outside your experimental range. If you're just making numbers up that feel good...you're not simulating squat.

    I think of it as the difference between a photograph of a mountain, a painting of a mountain, and an imaginary painting of a mountain.

    -The photograph is what the developers aspire to create, a highly accurate rendition of the real object.
    -The painting is what a simulator ends up being. Because of restrictions on available techniques and materials, the picture has to be produced by a painter with paint and canvas, which isn't ideal. It's a representation of the real mountain, but filtered through the artist's perception and limited by the way he can paint it. Accuracy may vary.
    -The imaginary painting may appear to be just as real, but it's not based on a real mountain. It uses the artist's experience with mountains to produce something that appears to be a representation of a mountain, but because it's not related to any real mountain it may include features that couldn't be possible, but enhance the picture artistically.

    There's a difference in how one approaches the design. One method holds real world data as an absolute, the simulation must use that data in the form it's given to produce the final result, no matter what transforms need to take place on the way.

    The other has no absolutes and will use any method available to get the correct result. Which would be fine, if games weren't so complex that it's essentially impossible to exhaustively test every possible result of the physics system. That's why something that models the behaviour from a small set of variables is preferable to attempting to describe every possible output.


    I'm a chemist. I create models of molecular and physical interactions. I could define every possible interaction by simply testing them all. I'm not given the time to do that, it would be hugely expensive for the knowledge that would be gained.

    Instead I define a small set of experiments that elucidate the mechanisms behind the reactions and the constants that govern them. Then I create my model, and create more experiments to double check that the model is in fact behaving in accordance with the reality. The amount of work to do this is orders of magnitude less than testing every interaction, even to a very low degree of accuracy.

    What's more, a good model should show edge cases and odd interactions that might have otherwise gone unidentified.

    This is pretty standard scientific procedure. I don't see why it would be preferable when creating a model of a physical system to go and test everything, rather than creating a model of the system and allowing that to generate your data for you.

    Maybe they did, and just omitted the final step of double checking their model against reality again. Which is a pretty big whoops.

    When words and actions disagree, believe actions.

    A system designed from first principles does not omit camber effects.
     
  27. Griffith500

    Griffith500

    Messages:
    8,631
    Yeah, when I said PD focus on feel, I didn't by any stretch mean they say "it should feel like this" and reverse engineer a solution - you brought that claim, and the one does not automatically imply the other. I think you need to offer proof that that's what they're doing, or at least brush up on your physics (continuum mechanics) to see how that just wouldn't work for the behaviour we're seeing in the game.


    If all you want to do is simulate feel, then I can't see a problem with doing some kind of empirical solution (most of engineering is still empirically-based to a first approximation), but GT has broader goals than that, and where those goals intersect, you're left with building a vehicle dynamics engine. And everyone cheats, in precisely the manner you described for whatever molecular models you might be building.

    Except that vehicle dynamics involves a few more influences, the physical underpinning to which is not understood to the degree that, say, electron orbitals or "spin" are. There's also a difference of range of scale, in that molecular physics is constrained to the molecular scale (for chemistry's purpose), but the things that matter on a car exist over several orders of magnitude of time, space and number (discretisation). Then there's still the lump of jelly to consider, and "entertainment".


    A system designed from first principles can do anything you want it to do. You're ignoring your own prescription regarding throwing things away, simplifying. Doing that has system-wide consequences. Imagine you decided to change the "length" (bear with me) of a single kind of inter-atomic bond (obviously, you'd have to go "lower level" than that, and the consequences would probably be "unpredictable"). What effect would that have on the structure of the universe? If you, personally, were wanting to do that for whatever reason (I don't know, cheaper catalysts), I think your attention would stay at the molecular level.

    For a game, your attention, never mind if it's at the flow resistance in damper valves, or the thermo-kinetic dissipation of the turbulent wake on the back of the car, or the fluctuation in the torque cycles of the engine (I wish), always comes back to the lump of jelly - the player. Or at least, it should. That says nothing about how you build the individual parts of the overall system, though, which is where the differences lie. It's basically that origami metaphor from Kaz.


    When making a game, you're building a system, not distilling behaviours into compartments - classical science is not good at systems because it always seeks to simplify and consider in isolation (thanks to formal logic, and to its credit, don't get me wrong). Games are about the interactions between parts yielding some complex whole, so you need to focus just as much on the interactions as on the parts (although the parts determine the interactions to a degree, it's the "connections" / "mappings" that matter more). The interactions do not exist in isolation.

    Many people have an intuitive knack for it, such that it can, in practice, be an art (it is, of course, also formalised out of the branches that followed robotics and cybernetics / systems science). When there are lumps of jelly involved, there tends to be a lot of artfulness required. And I say that as a ("former") scientist and model builder myself. I'd love GT to be the best simulation of suspension linkage flex, but it's the human interpretation of that complexity that matters, ultimately. "Realism is boring", but only if you fail to communicate that realism by neglecting to include the player in your analysis of the feedback loop that constitutes the game in motion.


    What's promising with GT is that that authentic feel has always been there, even when the physics were far more primitive, and, more importantly, it remains as PD add complexity. I am surprised that, for a mass-market game like GT, PD continue to improve the physics in the specific way that they are ("simulate everything" - eventually). Even if the small changes to the parts result in weird changes to the whole in the short term, probably partly because those (new) parts are intended for a different whole...
     
    FussyFez and Tenacious D like this.
  28. Imari

    Imari

    Messages:
    11,508
    Location:
    Australia
    Then I misunderstood. Could you clarify what you did mean?
     
  29. Zer0

    Zer0 (Banned)

    Messages:
    1,624
    "In the virtual driving universe," he says, "drifting is a problem."

    The issue is complexity. Unlike acceleration, cornering, and braking - which rely on basic Newtonian formulas - drifting is an esoteric confluence of g-forces, coefficients of friction, and vectors of velocity.

    "On paper, the physics don't make sense," Kazunori says. "So our engineers couldn't understand it enough to make a formula." In theory, a car can no more drift around a corner than a bumblebee can fly. And yet in the real world, it happens.

    It's easy to simulate drifting with a hack -- a slide-sideways subroutine, activated with a jiggle of the wheel and a pump on the gas, as seen in rival games like Capcom's Auto Modellista and Microsoft's Project Gotham. But to Kazunori, that's cheating.

    "We cannot ignore physics," he says. "If GT has a realistic enough physics engine, a player will be able to accurately drift a car, whether the engineers understand why it slides or not. It is the goal and the proof." When drifting happens in the game, Kazunori's engineers call it "the miracle."

    http://archive.wired.com/wired/archive/11.10/drifting_pr.html
     
    Streeto and Tenacious D like this.
  30. Imari

    Imari

    Messages:
    11,508
    Location:
    Australia
    Do you really think nobody understands how drifting works? You don't think anyone can sit down and map out the forces and equations required for the car to perform that particular manoeuvre?

    The bumblebee thing is crap, by the way. It has never been the case that any serious scientist thought that a bumblebee couldn't fly, even in theory. You have to give the bumblebee only the most basic of aerodynamic modelling treatments to actually come up with a result that says it can't fly.

    People who believe the bumblebee story are suddenly much less reliable when it comes to anything else scientific they might have to say.

    Kazunori is either ignorant (which I seriously doubt, the man may be many things but he's not an idiot) or he's spewing marketing BS, which casts doubt on that whole quoted passage.

    I don't know what you think those passages are proof of. That they don't use shortcuts? We know they do, and as @Griffith500 and I have gone over it's largely impossible to make a simulation running in real time without using at least some simplifications.

    No one ever claimed they were scripting drifting. I claim that their tyre physics are not based entirely on physical principles, and possibly their suspension physics aren't based entirely on real suspension either. Both of these things are the simplest explanations I can think of for the lack of camber and torque response in the cars. Possibly similar effects could be observed by making the body of the car respond incorrectly to applied forces, but it gets a lot more complicated.