AI LMP Cars Very Mistake Prone

[Peter Kerényi]

Hello!

I would like to spend more time with rFactor 2, mostly in single player. I like to set up GTE/LMP2/DPi practice sessions. From my experience, LMP2 cars are extremely prone to making errors, spinning, when close to a slower class. I have AI Mistakes at 0 in the player.json file. I do most of my sessions on official tracks with official cars (came here from Sebring). Aggression about 25, limiter at 5. It is a pretty fresh install of rFactor 2.

I guess enabling AI learning would not solve this, right? As it is more racing related than driving related. Any tips are appreciated... Thanks.

Edit: Decreasing the limiter to 1 seemed to improve it? Have to test it more.

 

[Peter Kerényi]

@Durge Driven Increased aggression to 40 with the exact same race I set up yesterday and the result was better. Less yellows, not caused by clumsy GTs. Subsequent testing at Road Atlanta also showed improvements. This time, however, it is clearly some overly eager paydriver LMP2s causing trouble. A little bit of aggression reduction would provide optimal results I think, of course in the bounds of what we were given. Thank you for the useful advice. Bit counterintuitive, but logical and works at the end of the day. Of course, needs more testing. Portland was also problematic, that will be my next destination in the sim.

 

[Ace King]

I literally reinstalled rFactor 2 like... 4-5 days ago. I will check what the track state is/was. In Sebring they tend to spin around e.g. turn 11 trying to overtake the GT cars around the outside, clipping the grass.

Regarding very odd ai behaviours can sometimes be corrected by deleting your player.json as MarcG stated but reinstalling rF2 will "Not" delete your player.jason so manually delete it and see if it helps.

 

[Peter Kerényi]

Regarding very odd ai behaviours can sometimes be corrected by deleting your player.json as MarcG stated but reinstalling rF2 will "Not" delete your player.jason so manually delete it and see if it helps.

The whole folder was manually deleted before that (after the last uninstall). But thanks.

 

[Durge Driven]

Sorry 115% was another mod
I used 100% for this
Did 90 minutes session so far

 

[Peter Kerényi]

Sorry 115% was another mod
I used 100% for this
Did 90 minutes session so far

No problem. I did not increase strength (I was running 103%, wouldn't be able to go much higher), only tweaked the aggression.

 

[Durge Driven]

Yes higher aggression works well with some mod and track combos
F3 Eve at Isle of Man I run 115% and 50% and reduced crashes heaps
More agro stops them lifting so much in fast turns and under steering

 

[Squalido]

I am new to r factor 2 and I have been mainly running lmp1/lmp2/gte multiclass races using the GEM. I did a 2 hours race at Sebring this week (split in 3 sessions), not using the AI calibration, and also noticed the protos spinning from time to time. Another thing that I noticed is that after my last resume from replay (starting the 3rd session) there were 5 minutes of chaos with many cars retiring. Maybe a side effect of the race restarting with cold tires or not rubbered? In the end 9 cars from a total of 31 retired, mainly lmps. Any tips for using the resume from replay feature?

I did enjoy a lot that race, and running from a GTE the AI from the faster classes behaved correctly at the moment of lapping me.

 

[davehenrie]

Run a long practice session and then save the rubbered in track. I haven't done a replay race in years, but you should be able to load the track surface. Also there is an AI mistake prone value in the player json. the closer the number is to zero, the fewer mistakes. Look for these lines:

"AI Mistakes":0,
"AI Mistakes#":"a range of (intentional) AI mistakes from 0.0 (none) to 1.0 (sometimes). Anything above 1.0 multiplies the frequency",

 

[Durge Driven]

Also if you not bothered with track evolution use the real road you save in all sessions and set real road to "static" which lowers system usage

Personally I think it's useful for consistency with garage and AI baselines

Like if you do a long "offline online" session swapping 10 GT3 cars in garage using evolving grip the cars you run last will get advantage