Arcade and online are the full physics package. Contrary to what I have read from a couple of people in this thread, the cars handle much more realistically in those areas.
In "A spec" I am finding the handling model waaay too forgiving. Almost like "standard" physics from GT5P. this also includes "Practice".
Well, I can tell you for that, at least in the case of certain RR cars, the offline physics are much more accurate in terms of the dynamics of how the cars handle. The basics of RR driving in real life translate very well to offline physics, but extremely poorly to online physics.
However, I do feel you are absolutely correct in that the offline physics are too forgiving. The thing is, at least for those cars I can properly judge, it is a bit too easy but it is at least accurate. I know maybe I'm not making a clear distinction there, but I don't know how else to put it. Let's say in real life I'm at an autocross, one of those one's where it is really just a track with a few cones and one car on track at a time. Speeds in excess of 100mph are not all that uncommon at some of these). Let's say I'm in the 1980 911sc (Import from Germany, about 200hp, compared to 180 for the US model). Gobs of torque. Completely stock suspension. I'm on good performance street tires, quality but nothing specific to racing. Probably Sports Hards in GT5 lingo. I enter the corner too hot! Oops, understeer! I let the right foot come up a bit, front settles, often I even let the rear begin to rotate on purpose. How far and how much depends on the corner. I counter-steer and then, when I'm approaching where I want to be, squeeze the pedal back down and get the car to settle *exactly* where it should. Viola, weight transfers back to rear and away I go. Those are some of the basics of performance Porsche driving. I use this technique around almost every hairpin also, and in sweepers. Constant juggling of over/understeer with throttle. This all WORKS offline in appropriate cars. Yes, it works *too easily*, but it *works*. Online, it doesn't. It just doesn't work in certain RR cars (also, I feel that the MR Stratos road car should behave in a similar manner, but I'll leave final judgement to real-life MR drivers in that case). It doesn't work even at lower speeds (watch the Alpine video, the car is all over the place clear down around 60mph in some places). And that's plain wrong. Now, I won't try to judge FR cars, or AWD cars, or FF cars, because I don't do motorsports in those very often, but I've driven RR cars my entire driving life (going on 30 years), and used them in amateur motor-sports for 5 years now. I'm also one of those people who truly feels that the proper place for an engine is in the back. I've also studied and read copious amounts of racing guides over these years. I'm hoping to make you understand that my opinion in this case is based in theoretical (reading) and practical real life experience.
So I'm intimately familiar with the handling of RR cars and I'm bloody good at it! I really don't like to toot my horn, there is way too much arrogance on the internet as it is, with too many of us wanting to be experts in things we really know nothing about. But there is value in experience so I'm going to say this once and you can either take it for what it is worth or you can dismiss my experience as useless: For the first season I stunk. I was godawful, terrible. I understood very well how the cars handled overall, but I simply could not fully grasp of how easily most cars push into understeer, thus I was continually overcooking it into corners. The second season I was passable, and for the last 3 seasons it is the very rare event where I don't take 1st in my class (and I often run in multiple-classes in multiple-cars, so I might actually race in 2 classes per weekend, both on a Saturday and a Sunday, and walk away with 3 1st and 1 2nd place finish in 2 different classes against different cars and drivers both days). With one of the clubs I race with I recently bested a former national SCCA H-stock champion. I also do ice racing in one of the Porsches, where I'm equally good (this is solo ice-racing, on frozen lakes, on DOT legal street tires. No studs! Wheeeeeeeeeee. It's a lot of fun and boy does it teach car-control).
I don't mind the difficulty online at all, in fact. I think it would be good if offline physics were a bit more challenging. What I do mind, very much, is when the online physics are just plain wrong. If you haven't watched the Alpine video I posted above, please do so. The problem is clear as day. (I also think it's very problematic to claim that online physics are more realistic when you can't turn a car over online. That's a serious contradiction!) So while I think that the *difficulty* online is more accurate overall, the *physics* seem to be terribly lacking, at least for the cars I can judge.
Could I be wrong? Of course. I've been wrong before, it'll happen again. Probably tomorrow. Maybe the Alpines won't come back at all once the rear starts to go. Maybe, despite sharing most of their chassis and suspension with my real life motorsports vehicles, the BTR and Yellowbird also won't come back around. Maybe these cars do have terrible snap-back even at low speeds like I've never encountered in real life. And when someone here comes along as says "I've raced my Alpine 1600s for the last 20 years and it actually doesn't handle like other RR cars", well then I'll believe that person. (And in the end, even if I'm 100% wrong, it *still* doesn't excuse PD for using two different models for vehicle dynamics).
) but I do have one that I felt was completly diferent from offline wich was the caterham 7 fireblade 02 - with 250 hp and about 350 -400 kg wich can be even better to do the test.
or should it be done in a 4wd car ???
