Here's the text from the readme for the New Advanced Physics Patch...

The simracing community had great expectations on this new version, and while most of the people seem to agree that it's a great title, there are some
arguing about the authenticity of the physic data inserted, specially in terms of grip and aerodinamic stability. Critics point out how easy is to
mantain a drift without even modulating the gas pedal, and how difficult is to block the tires under braking. Other people believe that this is how the
cars should handle, but all seem to agree that there is a big difference in lap times between the real cars and the simulated ones. Top sim pilots seem
to be able to lap at over 7 seconds faster than the real cars. Considering that GTR's2 tracks are amongst the best ever produced for a sim, it's clear
that something is a bit off limits.

Taking all the above considerations into account, we started looking at the physics data of the sim, which for the first time in a simbin simulation,
have been left completely open to edit. We found lots of good things which are to our opinion very very accurate and precise but we also found bad things,
surelly designed so, to enable easier driving even from less experienced drivers, and probably a much easier online experience. For sure, the original physics of GTR2, permit close online racing with no frustration created from 600bhp beast. Some will argue that this is indeed a good thing, and we agree, but this is not our goal. What we seek is a more realistic behaviour, and a more authentic experienc even if that means So we decided to move on and change the physics with more realistic values.

In this readme text, we are going to explain what we changed, and why, trying also to prove why and how the original values have been maded "easier"
to handle. We hope that this patch won't be cause of flaming and trolling. It's a patch offered to the people that seek just a bit more, and are willing to pass the frustration of a racing driving. This doesn't means that whoever prefers the original physics is not able to drive well, or by any means not a simmer. Everybody is free to choose whatever he likes more and all opinions should be respected.
Thank you.

ISI engine physics discussion and values changed.
The ISI physics engine, (on which the Simbin titles are based), is one of the best physics engines actually in the market. Versatile, precise and accurate,
permits the simulation of various types of suspension, vehicles, aerodinamic configurations ecc ecc. Unfortunately, there is no such thing as the perfect
physics engine, and the ISI engine has it's own problems. One of it's biggest problems is the tires simulation, because it fails to combine well, lateral
and longitudinal forces. It's pretty obvious that actual realistic countersteer drifting has been almost impossible with the ISI engine, while for example
is very good on LFS (not that LFS is perfect, it has it's own problems too).

TYRES

DropoffFunction=1.0
With this new version of the ISI engine, there are some new values that permit to change the curves of slipangle and slipratio, depending on the load. This
is indeed handy, and permits good control of the car, over the limit. What most of the modders do, is put the above value at max (1.0) and indeed it gives
more control once the car is over its limits. Still, it doesn't provide a true slip and drift over the limit. Usually we keep the wheel straight on,
and we do a 4wheel drift with the gas pedal to the metal. This, we believe, is mainly due to some problems the ISI engine has to combine the forces.
We took the original tire curve of GTR2 physics, and we totally change it's design AFTER the peak slipangle point. We did lot's of testing and lots
of research work on SAE papers and we tried to simulated the behaviour of a tire over it's limits, with a curve that takes into account the
ISI engine problems in combining forces. We are very satisfied with the results, and we are even more satisfied from the fact, that we can use
different values on the dropoffFunction, more aggressive ones, that make the behavior of such large slick tires, more realistic and nervous.
Big improvement here is that while the tires are more nervous, they can be drifted with countersteer and gas, once over the limit, thus giving
full control of the car, even at high yaw angles.

LatPeak=(0.075, 0.235, 12500.0)
Here is another value that we focused on. This defines the maximum slipangle, depending on the load (in Sin) plus the maximum Load where this peak occurs.
The original values give a maximum slipangle at almost 14 degrees at 12500 of load. From our research it appears that this is a bit of a high value for
this kind of tires, so we adjusted this to a more realistic 8,5 degrees at maximum load. Amazingly enough this change, can be the cause of slower lap times
if the car is not driven correctly, but if the car is being driven up to the limit of adhesion and not beyong, this smaller slipangle gives even faster lap
times. But it takes some very fine driving.
We also changed accorgingly the longitudinal slipratio values too.

AERODYNAMICS.

One of the most strange things with ISI engine, is that the cars are more difficult to handle at low speeds, and as the speed goes up and the the negative
lift also raises, it becomes easier and easier to control the car, even at high yaw angles and amazing drifts. It's not really difficult to understand why,
if someone opens the aerodynamic physics files of various mods and games based on the ISI engine. Amazingly it seems that everybody is doing simple copy
paste of the sensibility values, and only changes the actual lift values accordingly to the car that is simulated. So we get the paradox of
a modern F1 car, that has the same aerodynamic sensibility to yaw and pitch, to an historic touring car with just a rear wing, meaning no aerodynamic
sensibility.
With the new tyre behaviour that permits better control at lower speeds, even with nervous big flat slicks, we tried to work on the aerodynamics of
the GTR2's cars to obtain a much more detailed and complicated behavior. We where able to not only make the car much more nervous (but not impossible
to control) and slower when it starts to drift at high speed, but also much more subject to setup choices in terms of aerodynamic efficiency. We are
very satisfied with the results.

FWLiftParams=(-0.28, -0.120, 0.00003)
RWLiftParams=(-0.372, -0.0280, 0.000025)
DiffuserBase=(-0.300, -2.33, 43.0)
Those are the lines that determine the maximum lift the cars receive from their wings, splitters and diffusers. Accordingly with our research,
we have lowered the lift of those values to more realistic ones. The modification has been made in terms of % for every value. In such a way we
haven't changed the actual relative car perfomances of the gtr2 grid, meaning a Maserati will remain faster than a Viper ecc...

FWLiftSideways=(0.0)
RWLiftSideways=(0.0)
DiffuserSideways=(0.3)
Now those are some very important values to us. They determine how the wing/diffuser/splitter, loses lift efficiency, when the yaw angle raises
(in other words setting the car in drift). We where amazed to see that in all the cars those values where identical and null (or very low for the
diffuser), but we weren't really surprised, because it's obvious that with no yaw aereo sensibility, a car handles easier and better while in drift.
Unfortunately this is not happening in reality. Aerodynamic efficiency is very sensible at yaw and lowers very fast usually in hyperbolic curves.
Just as an example, a CART single seater loses 15% of its aerodynamic load if it goes at just 5 degrees of drift! Of course GT cars are not that sensible,
but still suffer from these effects. There is a reason why modern pilots, never drift their cars. We changed these values accordingly and now you can feel
the car loosing aereo grip, when drifting at higher speeds. It's still controlable, thanks to the tires, but much much slower.

FWLiftHeight=(0.335)
DiffuserRake=(0.005, -5.0, 150.0)
DiffuserLimits=(0.010, 0.100, 0.055)
The above values determine the efficiency of the front splitter and the rear diffuser in regards of the height and rake of the car. We have found the
original values to be very forgiving, losing lift only at excessive setup settings. We have made a much more short range of work, for the splitter and
the diffuser. This means that while in the old GTR1 and original GTR2, players would opt for the softer possible spring rate, to achieve lots of mechanical
grip, but without losing aerodynamic efficiency due to pitch, roll and rake. With the patch values, pilots will be forced to find the correct compromise,
because softer springs will give better control at lower speeds, but will lose efficiency at higher speeds and might find unpredictable handling too. On
the other hand, with stiffer springs, they will get a more twitcy lower speed handling, but much better aereo efficiency at higher speeds, and probably
better lap times at fast tracks.

TRACK GRIP

RoadDryGrip = 1.00
The original value of 1.00, gives a perfect condition track, actually impossible to find in reality. Telemetry data and research, has convinced us to
decide for a slightly lower value of 0.97. Keep in mind that even the most slightly changes in the track grip, are immediately felt in the car grip and
behaviour.

FINAL WORDS.
We tried hard to not compromise the original grid performances. Each car, although different in behaviour from the original, should perform equally in
relationship to another, as in the original game. We where also happy to see that with our values, lap times are much closer to the real ones, and
sometimes even spot on. Only true aliens will be able to better them, but not by more than a couple of secs as their tests in beta testing have shown.

Once more, we hope that this patch won't bring flames and futile discussion in the communities. We encourage downloading it and test it, but if
you don't like it, just don't flame about it and please ignore it . We don't try to convince anyone about the authenticity of our patch, we just did it,
because we had fun doing it.
Thank you