Forums

The overall grip level is in the TXT file at the lines TCFF and TCFR, which means probably tire compound friction, and friction=grip.
The first 4 numbers are the grip levels of each compound (soft, medium, hard, rain) at dry conditions, the last 4 at wet conditions. Setting the first 3 values equal to the last value, you get for all 3 accessible compounds the grip level of the rain tire in wet conditions, which simulates a wet track. As I mentioned earlier, the AI suffering also from the reduced grip level, and go off the track, because they still try to achieve the LP speed, which is impossible at the reduced grip level.

-----------

To be more scientific, let's think about, which parameters make the car physics. Maybe then it's easier to find the right values.
If you know more, you can extend the list.
- drag
- downforce
- tire grip
- tire rolling resistance
- car weight
- center of gravity
- wheelbase
- track width
- G-force
- engine power
- brake power

(12-19-2021, 06:34 AM)Dennis Wrote: [ -> ]The overall grip level is in the TXT file at the lines TCFF and TCFR, which means probably tire compound friction, and friction=grip.
The first 4 numbers are the grip levels of each compound (soft, medium, hard, rain) at dry conditions, the last 4 at wet conditions. Setting the first 3 values equal to the last value, you get for all 3 accessible compounds the grip level of the rain tire in wet conditions, which simulates a wet track. As I mentioned earlier, the AI suffering also from the reduced grip level, and go off the track, because they still try to achieve the LP speed, which is impossible at the reduced grip level.

-----------

To be more scientific, let's think about, which parameters make the car physics. Maybe then it's easier to find the right values.
If you know more, you can extend the list.
- drag
- downforce
- tire grip
- tire rolling resistance
- car weight
- load transfer
- G-force
- engine power
- brake power
- wheelbase
- track width

Dennis

You are right about the data contained in the track.txt.

When we choose a certain type of tire (Soft, medium or hard) some addresses change simultaneously and these values 65535 appear.

I noticed that some tire data does not work for all types of tracks (Street courses values are not compatible for inclined ovals, for example) for this reason I did not defunme in these parameters.

Regarding the list you quoted, some data I shared impact scans directly on some of them, such as drag, tire grip, engine power, and brake power.

Certain addresses jointly impact these items.

Now that the data has been shared, I'm sure others will go deeper and find even more things that will allow us to further enhance our game.

Unfortunately I'm not a programmer, but if we have a few in the group, we may even have a program that allows you to parameterize the data so that the player and AI behave differently from the standard.

I'm even imagining the folks creating versions of old machines to make each season even more realistic.

As I said earlier, I was able to reproduce the behavior of the cars of 1987, so if you choose to ride with the Lola-Chevrolet Mario andretti, you will have a fast car, but very subject to breaks, now if you choose the Lola-Cosworth Rahal, you will have a car not so fast, but much more reliable. The same goes for March-Cosworth of Michael Andretti and Roberto Guerreiro.

If you opt for the Penske-Chevrolet of the Emerson Fittipaldi, you can be sure that you will have few moments of joys and many abandonments, if you run with March-Judd of Geoff Brabham, you will have to deal with competitors much faster, at least you will have a car with good reliability, especially when compared to the Lola-Chevrolet or Penske-Chevrolet.

(12-19-2021, 08:49 AM)sandro_nunes Wrote: [ -> ]

(12-19-2021, 06:34 AM)Dennis Wrote: [ -> ]The overall grip level is in the TXT file at the lines TCFF and TCFR, which means probably tire compound friction, and friction=grip.
The first 4 numbers are the grip levels of each compound (soft, medium, hard, rain) at dry conditions, the last 4 at wet conditions. Setting the first 3 values equal to the last value, you get for all 3 accessible compounds the grip level of the rain tire in wet conditions, which simulates a wet track. As I mentioned earlier, the AI suffering also from the reduced grip level, and go off the track, because they still try to achieve the LP speed, which is impossible at the reduced grip level.

-----------

To be more scientific, let's think about, which parameters make the car physics. Maybe then it's easier to find the right values.
If you know more, you can extend the list.
- drag
- downforce
- tire grip
- tire rolling resistance
- car weight
- load transfer
- G-force
- engine power
- brake power
- wheelbase
- track width

Dennis

You are right about the data contained in the track.txt.

When we choose a certain type of tire (Soft, medium or hard) some addresses change simultaneously and these values 65535 appear.

I noticed that some tire data does not work for all types of tracks (Street courses values are not compatible for inclined ovals, for example) for this reason I did not defunme in these parameters.

Regarding the list you quoted, some data I shared impact scans directly on some of them, such as drag, tire grip, engine power, and brake power.

Certain addresses jointly impact these items.

Now that the data has been shared, I'm sure others will go deeper and find even more things that will allow us to further enhance our game.

Unfortunately I'm not a programmer, but if we have a few in the group, we may even have a program that allows you to parameterize the data so that the player and AI behave differently from the standard.

I'm even imagining the folks creating versions of old machines to make each season even more realistic.

As I said earlier, I was able to reproduce the behavior of the cars of 1987, so if you choose to ride with the Lola-Chevrolet Mario andretti, you will have a fast car, but very subject to breaks, now if you choose the Lola-Cosworth Rahal, you will have a car not so fast, but much more reliable. The same goes for March-Cosworth of Michael Andretti and Roberto Guerreiro.

If you opt for the Penske-Chevrolet of the Emerson Fittipaldi, you can be sure that you will have few moments of joys and many abandonments, if you run with March-Judd of Geoff Brabham, you will have to deal with competitors much faster, at least you will have a car with good reliability, especially when compared to the Lola-Chevrolet or Penske-Chevrolet.

This is a real game changer

Friends,

Below is a short explanation of how to change the addresses of our game.

1) Open the cheat engine program;
2) Open the windy.exe file
3) Select a process and choose “Cart Racing”
4) Go to “Add address manually” and enter the address you want to change
5) Go to “memory view”
6) In the menu, choose the option “Tools”
7) Go to Auto assemble or change the chosen address.

The magic addresses were already mentioned at the beginning of this post.
It is possible to increase the AI random breakdowns rate by changing the address 43BEA6=> MOV ECX, 1B7740.
If we change the value 1B7740 to 9000, there will be several abandonments, both from the AI and from the player himself.
It is possible to dose the abandonments also by changing the address 43BEAF => CMP EDX,64, if we change it to CMP EDX,10 for example there is a variation in the abandonments of the player and the AI.
I won't go into the details of how to use the Cheat Engine because there are thousands of videos on youtube with several tutorials that will explain it better than I could explain.

Guys,

Has anyone been able to test the parameters?

I'm curious to know the results.

(03-06-2022, 10:57 AM)sandro_nunes Wrote: [ -> ]Guys,

Has anyone been able to test the parameters? 

I'm curious to know the results.

I use indycar.exe with dosbox ece, so no... I'm sorry. (not windy)

It is a pity that these parameters changes are not the same as indycar.exe.

The changes in parameters produce fantastic results.

The difference in performance and driveability are clear.

(03-06-2022, 10:57 AM)sandro_nunes Wrote: [ -> ]Guys,

Has anyone been able to test the parameters? 

I'm curious to know the results.

I haven't done any more work on this since my previous post when I verified that several of your discoveries work for me. Just been prioritizing my free time to work on Detroit. However, from what I saw, I think this is absolutely worth continuing to explore and I hope others can try and see for themselves.

I think these parameters will allow us to create different types of cars and new contexts with more accidents and abandonments by both the player and the AI.

I just picked up Cheat Engine again and did a quick scan to see if it is possible to get the car position. So far I barely know what I am doing, but I have found a couple of memory addresses that change when the car moves:

023EF7F8, 023EF80C = DLONG

Edit: I couldn't find those values again using the same addresses in a different session of Windy. I will need to learn more about this but at least I know it should be possible to get DLONG from memory somehow.