Bing Ai

Asking Bing Ai to explain specific aspects of Ai behavior editing gives surprising results. It appears to read all forum posts and then compile a suitable response.

Example:

Explain how to edit the ai behavior in plr file for factor.


Bing response:

The AI behaviour in rFactor is controlled by a file called plr, which stands for player. This file contains various parameters that affect how the AI drivers behave on the track, such as aggression, overtaking, braking, cornering, etc. Editing the plr file can allow you to customize the AI behaviour to your liking or to suit different types of cars and tracks.

To edit the plr file, you need to locate it in your rFactor installation folder. The default location is C:\Program Files\rFactor\UserData\YourName\YourName.plr, where YourName is your profile name in rFactor. You can open the plr file with any text editor, such as Notepad or WordPad. However, it is recommended to use a more advanced editor that supports syntax highlighting and line numbers, such as Notepad++ or Sublime Text.

The plr file is divided into sections that start with a square bracket and end with a blank line. Each section contains several parameters that have a name and a value separated by an equal sign. For example, AI Driver Strength=(100) means that the AI driver strength is set to 100%. You can change the value of any parameter by editing the number or text after the equal sign. However, you should be careful not to change the name of the parameter or the format of the file, as this may cause errors or crashes in rFactor.

Some of the most important sections and parameters for editing the AI behaviour are:

[AI]
This section contains general settings for the AI drivers, such as how many drivers are allowed on track, how often they pit for fuel and tyres, how they react to yellow flags and safety cars, etc.

AI Driver Strength=(100)
This parameter controls how fast the AI drivers are compared to you. A higher value means that they are faster and more competitive, while a lower value means that they are slower and easier to beat. You can adjust this parameter to match your skill level or to create different levels of difficulty.

AI Limiter=(0)
This parameter controls how much the AI drivers are limited by their car performance and setup. A higher value means that they are more affected by factors such as weight, drag, tyre wear, fuel load, etc., while a lower value means that they are less affected and can drive faster than their car should allow. You can adjust this parameter to make the AI drivers more realistic or more challenging.

AI Mistakes=(0)
This parameter controls how often and how severe the AI drivers make mistakes on track, such as spinning out, crashing into walls or other cars, missing braking points, etc. A higher value means that they make more mistakes and are more prone to errors, while a lower value means that they make fewer mistakes and are more consistent. You can adjust this parameter to make the AI drivers more human-like or more perfect.

[AI Calibration]
This section contains settings for calibrating the AI drivers for each car and track combination. These settings affect how the AI drivers handle specific aspects of driving, such as acceleration, braking, cornering, overtaking, etc.

AI Brake Power Usage=(1.00000)
This parameter controls how much brake power the AI drivers use when braking for corners or slowing down. A higher value means that they use more brake power and brake harder and later, while a lower value means that they use less brake power and brake softer and earlier. You can adjust this parameter to make the AI drivers more aggressive or more cautious.

AI Corner Grip Usage=(1.00000)
This parameter controls how much grip the AI drivers use when cornering or turning. A higher value means that they use more grip and corner faster and sharper, while a lower value means that they use less grip and corner slower and wider. You can adjust this parameter to make the AI drivers more skilled or more sloppy.

AI Overtake Space=(1.00000)
This parameter controls how much space the AI drivers need when overtaking or passing other cars. A higher value means that they need more space and only overtake when there is a clear gap, while a lower value means that they need less space and overtake whenever there is an opportunity. You can adjust this parameter to make the AI drivers more respectful or more ruthless.

These are some of the most important parameters for editing the AI behaviour in rFactor. However, there are many more parameters that you can explore and experiment with in the plr file. You can also find online guides and tutorials that explain each parameter in detail and give tips on how to
 
This is fantastic! :) Thanks so much for sharing.

IMO this is just another reason that the site I've been working on over at https://sim-racing.gitlab.io/isimotor2-ai-docs/ can be useful. The more places the information is available, the better a large language model chatbot web crawler like Bing will be able to see it and digest the info.

It's also why reposting old forum posts and website content that you find can be useful - you can help archive that information for the future. This is why I recently dumped a lot of text from an old Image Space Incorporated website about rFactor 1 into a big thread here: https://www.racedepartment.com/thre...s-corner-a-long-lost-website-from-isi.256994/ By posting it another time, hopefully it has even less of a chance to be lost to the sands of time.
 
Last edited:
I signed up and started playing with chatgpt. Initially I had no interest in this, but it does appear to have some good uses in my type of sim racing. For example it wrote a VEH file and retrieved details of a car I specified, and put the values in.
It also wrote a talent file, made up the driver's name and gave him the type of strength I asked for. It wasn't perfect, but close enough that it would probably work.
 
Holy crap. That's wild! :D
In another experiment yesterday, I informed ChatGPT of its errors in the talent file. It said thank you and adjusted its output to match what I had given as an example.
I then asked it to make a set of talent files for all the drivers who competed in the 1988 Monaco Grand Prix, which it proceeded to do, with adjusted parameters for each driver and with their correct names and country.

The only issue is there is some limit to the output. Chat GPT will occasionally stop for no apparent reason. I then have to ask it to continue from the last line.
 

Latest News

Are you buying car setups?

  • Yes

  • No


Results are only viewable after voting.
Back
Top