Driver ratings

[Matt Yeager]

Hi!

I would like to modify drivers ratings in some carsets that are way off or even don't have proper ratings at all.

I assume these figures work the same way as NR2/99.

DINFO  9 0 1 1 450 550 300 500 450 550 500 750 Scott_Dixon Scott INDYCAR ICR ICR Chip_Ganassi_Racing

I presume these are:
- POWER (min/max ; greater values = better power)
- TRACTION (min/max ; greater values = better traction)
- DRAG (min/max ; lower values = less drag and better performance)
- AGRESSION (min/max ; greater values = more agressive behaviour)

Is this correct?

[Pavel 69]

Yes, that's right. You need to modify drivers2.txt, drivers.txt is like backup.
Also there was good topic about testing all this AI drivers settings
https://icr2.net/forum/showthread.php?tid=1467

[samsepi0l]

My opinion -
"Aggression"
lower numbers - more likely to move up in a race
higher numbers - more likely to stay or fall back

[Matt Yeager]

My opinion -
"Aggression" 
lower numbers - more likely to move up in a race
higher numbers - more likely to stay or fall back

Thank you. This is what I understood as well after I read the thread Pavel has just shared the link to.

[Matt Yeager]

Yes, that's right. You need to modify drivers2.txt, drivers.txt is like backup.
Also there was good topic about testing all this AI drivers settings
https://icr2.net/forum/showthread.php?tid=1467

Thank you!

Regarding the carsets you've shared on your Google Drive, did you rate them every year according to each driver's performance, or did you apply more generic ratings? Just asking because I am a bit of a nitpicker and I won't be able to refrain from doing my own ratings if yours are not season-accurate. 

[Pavel 69]

Regarding the carsets you've shared on your Google Drive, did you rate them every year according to each driver's performance, or did you apply more generic ratings? Just asking because I am a bit of a nitpicker and I won't be able to refrain from doing my own ratings if yours are not season-accurate. 

It's not season-accurate rating. Because it never was my goal to have real life results.

[Matt Yeager]

Regarding the carsets you've shared on your Google Drive, did you rate them every year according to each driver's performance, or did you apply more generic ratings? Just asking because I am a bit of a nitpicker and I won't be able to refrain from doing my own ratings if yours are not season-accurate. 

It's not season-accurate rating. Because it never was my goal to have real life results.

Well, I will not replicate what I do with the NASCAR games (less races per year, more part-time drivers, more mechanical failures...) but at least I will try to get the "big boys" running up front, valorous outsiders getting some success and notorious backmarkers not sweeping the championship. If I can have this with your ratings, that's perfect. If not, I will tune them a little bit. 

On the other hand, I really like nooRok's 2018-2020 IndyCar carsets, but all the drivers are rated the same... I applied basic ratings, so that I can be satisfied when I watch Dixon, Power and Newgarden battling for the first spots, instead of having them randomly dominated by Max Chilton. 

[JR82]

I have done a few of these, not 100% happy with what i have as yet but fairly close. Have done 1990-2001 & 2018-2024 as far as i can remember.

Is it possible to "build" a car which is faster on Ovals & less so on road/street tracks for example? Would the Drag setting achieve this? I have had a muck around with those settings but am yet to find the solution i'm looking for.
Aggression am still trying to get best result so that cars overtake each other in the races, incl not sitting behind backmarkers for a whole stint

[samsepi0l]

Aggression am still trying to get best result so that cars overtake each other in the races, incl not sitting behind backmarkers for a whole stint

I think we understand aggression pretty poorly. In another parallel universe I am smarter and I reverse engineered the EXE with Ghidra to better understand how the EXE using this number value- but I can't get smart-me to pass the information over inaccessible dimensions of the space-time continuum to dumb-me.

However as dumb-me, I do know one thing: Your Pass1 and Pass2 LP files will do a lot more CC Cars getting around backmarkers easier. I learned this when I was working with SK to make LPs for his Detroit Downtown circuit. I moved pass1 and pass2 to a greater distance away from race and it made a HUGE difference. Total game changer.