Was browsing reddit and came across this video:
It seems it was recorded during actual playing, so it cancels any replay issues, correct me if I'm wrong.
I think it's quite obvious there is something wrong, this video is probably one of the best showcasing a deep down issue in the physics ( which may not be related at all with what is being currently talked ). Not even on tarmac with heated slicks you would be able to do this such of a change in orientation coming from a slide like this lol.
I haven't done enough testing nor just played enough of the game to have a clear mind on what's wrong specifically, also I'm not playing the game with a wheel anyway, so it's hard for me to compare with the other sims I've played with my wheel, but even on controller, I tend to easily notice unrealistic behaviors, I'll not speak for now because I did see a big improvement with the 1.06 update and thus I need to play it more in its current state, specifically with one drift car I had done, which fully went from having myself restarting the race multiple times to doing runs in first try just after the update, so clearly something has changed drastically under the hood, at least from what I can tell ( quick note that imo it's the traction loss curve that changed, it seems in certain scenarios the rear tires can lose as much grip as before, but it does not happen nearly as fast and thus as often as it did before, so it results in having the rear tires now having more grip while sliding and getting the front wheels leading them much more than it did before in most scenarios of drifting, before it felt like the rear tires would snap into "extremely low grip" mode and go on its own, making the pedal dance and steering input extremely important and precise to keep the rear in its place and not have it fly away, not impossible by any mean but way more difficult than in real life and other sims imo, in theory this should reduce snap oversteer, as the loss of traction is more gradual on the rear tires and they seem to fly less on their own as fast as it actually did, but I need to play more of the game to have a clear mind out of this, and test more cars ).
To be honest, I'm actually a bit impressed by the physics on a personal level, all the physics flaw don't come at me as a surprise, I've always came up with these kind of issues on GT honestly. Not that it should be accepted for a game that targets "sim" handling, but I wasn't expecting to suddenly see it gone with gt7, so I'm not disappointed personally.
Also my 2 cents but this thread is a bit of a mess, and I think it will be for ever without proper testings of the physics, unless everyone suddenly admits that something is wrong or the opposite perhaps. Showing off videos of real life slides, or you sliding X car, doesn't prove anything. What we're all experiencing for now is a feeling, until someone come up with a proper understanding of what is happening and show it with actual tests.
It creates a lot of confusion to what is actually being talked about too. I think proper testings need to be done with more or less predictable car movements and actually properly analyse what's happening. Ofc it's easier said than done but I've been reading for quite some time and the discussion so far is turning in circles with different people coming in and arguing over and over with each other.
I think it needs to be clarified that from what I felt and what I've read, most people talk about the rear snapping, stepping away from the car forward direction too fast ( which is why everyone is talking about snap oversteer, not just oversteering in itself ), but be careful on the understanding of this, people aren't saying cars should NOT oversteer, but specifically that cars shouldn't have their back going away this fast during an oversteering scenario ( hence why most people talk about this on/off feeling of oversteer, when cars tend to slide they just suddenly slide instead of gradually losing its grip more and more ). Problem is when comparing to IRL videos is none to useless as the rear can totally go away on its own if you don't have the skill to correct with your steering and throttle, perhaps comparing with something impossible to manage in GT7 would be smarter imo, but comparing car spinnings are worthless, as both results on their own are absolutely correct, cars do spin in real life and so do in gran turismo.
I think most people get this oversteer part wrong and immediately assume that people are just referring to oversteer in general, which is wrong, cars have to oversteer obviously, even low powered rwd cars in low speed, tend to oversteer if they're asked to do so with a proper diff, it's just that right now, to me and for others it seems, the current loss of traction is completely messed up, or at least was, cause again on my side I need more testing with this last patch. So showing any car oversteering at any track under X circumstance doesn't prove anything, as it's obvious cars oversteer, but it's rather the behavior during that oversteer that has to be looked at, but throwing just random videos is none to useless really.
Also just to say that differentials make a huge difference irl with cars behavior on the limit, open diff will make it harder to slide than a welded diff, even with all driving aids off. This actually makes me think that I wonder if there even would be a possibility of recreating this in gt7 ? Definitely would like to see the car's behavior with an open diff in gt7.
