Internet and opinions exaggerate things. Always. Sometimes we need to take things into perspective and by "we" I don't intend only us devs, but also "we" as a whole community.
A couple of things to keep in mind.
- Vehicle physics, tyre behavior, aerodynamics are all an ongoing research for every single sim developer out there and also for the same manufacturers. Lot's of patents, lot's of secrecy, lot's of magic sauce. On top of that, put personal sensations to the loop and yeah...
- Internet gave us the golden age of game development. In simracing in particular, we now get physics updates every 3 to 6 month, and I'm talking about substantial ones. 20 years ago, GPL was the undiscussed king of driving simulation. It might still be fun and exciting to drive nowadays (which is an honour to the driving model that Sir Kaemmer build back in the days) but in terms of physics accuracy, complexity and detail is simply laughable now. We had to wait 3 years until the release of Nascar Racing 4 with a slightly updated GPL physics engine... 3 years! Now every simulator gets physics updates at least every 6 months.
- Generally we tend to forget the past and look at it with nostalgic feelings. Take AI for example. GPL AI is still thought as amazingly good. Yet people tend to forget that in order to avoid you, the AI would magically move left or right during racing, completely ignoring of any physics calculations. We tend to ignore that GTLegends and GTR AI would be extremely slow in the opening lap, especially if you had mixed classes in the game and that you could overtake the whole grid in the first 2 corners. The trick to have good AI races was to wait the pack to open up, then start racing them. I'm not saying that AC AI is good, I'm just saying that AI in general has problems and it's difficult... it's a universal issue.
So let me get back to the AC physics. Our philosophy regarding vehicle handling and physics has always been the same, based also on our not so small real life experience that we have had and still are lucky to have. Cars are engineered with million of hours of development work by very skilful engineers and every single one of them has just one thing in mind. To create a car that does NOT kill its driver. To be fast yet easy to push close to the limit and safe.
If you try the AC tech preview and now the new v1.5 you will be certainly capable of understand this philosophy in the physics of each car in AC. It is through this philosophy that we try to create the cars and when a car is unstable then we know that something is not perfectly good.
We also develop our physics engine step by step. If a car should be more difficult but to do so we need to make it unstable under the limit, then we won't. We will make it stable under the limit and then try to understand what makes it more difficult over the limit.
If you compare older versions of AC to newer ones, you'll see this evolution.
Stable cars with a bit too easy over the limit handling.
Stable cars with a bit less easy over the limit handling
Stable cars with a bit harder over the limit handling and so on.
From what I can recall there was just one situation where the handling of AC became unstable, I think around v1.2 and that was because we had found a bug in the core and by eliminating it, it created an overall instability especially on coasting. Hard work, data collection and verification, put things back to normal again in the next version.
Also, from what I recall there was also just one situation when we took a jump instead of a single step and that was with the introduction of V5/V6 tyres. The amount of work needed to redo all the tyres was simply over my capabilities for such a small amount of time. We had to do the job we did in about 2 years before the first early access public release of AC, but in just 3 months and for the triple amount of cars. As a result the v1.4 of AC had up an downs and was generally not optimized enough. We tried hard and what you have now with v1.5 is a much more mature physics engine with attention to detail. On top of that, many extra physics features.
So where's that "completely different experience" between the various versions of AC? Well how about there isn't any. Yes there are differences of course, but there isn't night and day.
Here are some objective numbers, you can't argue with numbers right?
The difference in grip while sliding from the v1.4 to v1.5 is in percentage: 3% for street tyres, 1% for semislicks, 0% (yes zero) for slick GT tyres. How you arrive at that part of the grip curve has changed in speed. It is a tiny bit faster for street tyres, a bit faster for semislicks and faster for GT tyres.
Aero yaw you say? Rear diffusers depending on car and configuration can lose from 1.25% to 4% of downforce at... 30°...yes, THIRTY degrees of yaw angle (drift)
If I've had announce to you those crude numbers I'm pretty sure most of you would say "hey that's no difference at all, you're playing with numbers that will give no change to actual driving!". But when you drive it you feel the difference and you might like it a lot or somebody might not like it. When you have no actual data to deal with or you know your audience doesn't have or understand the actual data, then you need to express an opinion and opinions are often sensationalistic.
Let me take as an example Empty Box' video.
Please believe me I don't want to point a finger to Empty Box nor to offend him, I think he does entertaining videos and a good service to the community. I apologize to you and to him in advance if my words might sound offensive in anyway, I really have no intention.
So at 4:50 of his video, he starts talking about physics and he says that the "driving is basically entirely different!". Well it's not. The sensations are different, not entirely if I may add, but yeah I can understand why he says that. He then says that there is a "wider margin of balance handling" and that is "less twitchy and less sensitive" but "however at the same time if you go out of that window, things do get a bit worse"
Again he is right but again that's an opinion based on his personal sensations. I might agree, you might agree, but being that an opinion there is no other way to emphasize on that, than to exaggerate.
This is typical also with professional drivers. You have certainly hear professional drivers complaining about "no grip" situations if they can't heat up the tyres. Actually no, let me show you how they usually express themselves in such situations: "NO GRIP! NO GRIP! NO GRIP AT ALL, IMPOSSIBLE TO DRIVE THE DAMN THING!"
I know, we have our offices in the race track...
You want to know the difference in grip between a cold tyre and a properly heated up tyre? 2-4%.
So as you can see, we developers are trying to figure out changes and situations in a range of 0.1% to 1% for micro differences and 1% to 4% for macro differences. And we have thousands of numbers that can influence this general grip feel. So it's a big juggling of available real data, unknown data, personal feel, drivers feedback, players (yes why not) feedback and experience... And I'm pretty sure we as KUNOS simulazioni are not the only developers that we have to deal with that, but I'm 99.999% sure that the same headaches happen inside iRacing, ISI, Reiza, Sector3, WMD and so on (no particular preference in this order... nowadays we have to point out that too... bah)
So yeah if you've read all my wall of text and still do not understand what I'm talking about, heres my conclusion.
"All journalists swear that the latest Porsche 911 has none of the understeer of the previous model!" for the last... 50 years.
Opinions are emphatic and usually exaggerated, take them always with a grain of salt and have perspective.
If you think that in 3 days you've managed to understand exactly what a simulated racing car does because you tried your last setup and you've been faster/slower for whatever reason, then think again. Devs usually need years to decide on a model and then months to go from one evolution to the other. If you don't understand what has changed or why, ask a developer. At least in our forum it's very easy to do so and we think we are friendly enough. If you are polite and respectful you get extra bonus points meaning that we are more willing to reply and understand your point of view and hey... you might be good enough to show us a new bug or better data and we will always thank you for that
das klingt jetzt so als ob ich auf alles eine Gegenüberstellung habe. Nicht meine Absicht 
Dazu eben die Farben, die das angepasste Arbeitsfenster wiedergeben
