And Kevin, happy to report that the ASR 1992 Hungaroring track is very competitive. In the 1988 version ai was doing like 1:30-1:32. Now look at this. Amazing. Even Patrese (ai 90) improved and went to 6th in the end 1:28:208.
Still unfortunately the easiest track to win even with a half-assed effort. Maybe it's the track shape that AC ai struggles with. Cars are slow on races. But now they're at least doing QF quite well.
Too bad you can't set the ai level (for QF and race) individually for each race on the champs files. That would be huge improvement because the ai performance varies so much between the tracks.
So as i am a non-CM user i set up my championshiops through the vanilla launcher.
When i create a champonship for the first time i set the difficulty for example on 94% and name it like - "f1 1999 94%."
So every AI driver is on 94% difficulty and i have to rename the AI driver with real ones. First thing to know is at this point - there are fast and slower AI drivers by default. From experience i know the default drivers pretty well so i know approximately who i should replace with wich real driver.
Thats the base, but it´s not realy the key.
Then i learn from wikipedia the standings of the real season and add ballast accordingly to the real life results.
For example f1 1999 again:
Häkkinen gets 0 ballast
Schumacher 5 (yes he was 5 but only due the injury)
Irvine 10
Frentzen, Coulthardt, Ralf S. each 20
and so on.... the slowest drivers should not get more than 40-50 ballast.
It´s also possible to additionally slow down the slowest teams with the restrictor by adding 2% or something.
From my experience over 3000h with about 60 diverse championships .... it´s working pretty well, however not perfect, because it´s sometimes also depending on the given cars, mods
The Pro:
You have just more variety with the results. It´s not gonna be Häkkinen or SChumi all the time on the podium and you beating half of the grid with easy but beeing maybe unable to compete with the best.
It can sometime happen at a GP a guy like Panis or Wurz will win and the favorites will struggle, BUT by the end of the day (season) The ending results should be pretty realistic.
Another good thing is, most of the series will provide tighter quali lap times, the gaps will be not that big as using diverse AI strenghts.
The contra:
It may not work with any series and any season. Sometimes it will happen a favorit drive will struggle and do not dominate.
If you hate the fact a guy like Luca Badoer can reach the podium, and its obviously unrealistic - yeah ths may piss u off.
Semi: Often the AI quality on diverse track will vary. If you gonna beat them on 94% with ease on Track X, it could happen you`ll have not a chance on the next Track Y
But in fact it also brings variety and you never know what gonna happen, so it can be seen positive as well as negative.
If i am fe. done with a championship and struggeled to win it on 94%, i start the new season with another car/team on 92%.
If i win the championship it`s obvious i increase the AI strenghts ....
I think thats it for now.
SORRY FOR OFF Topic