I was running Silverstone on 95%AI strenght 25% aggression and found that everytime I ran a practice session the fastest AI will get about 1:45 laptimes. This will carry on to qualy and then racing with laptimes around that mark. However, if I were NOT to run a practice session and go straight to qualy, the fastest AI will run around 1:40 and average laptimes will be 1:43. Race times for fastest AI can go below 1:40. Can someone tell me what's wrong and is there something I can tweak in the AI settings to make the AI more consistent?
Like real drivers, the AI like the road to get rubbered in. Start off with a green or lightly rubbered track and there's no rubber build-up if you skip sessions.
Hi, this makes sense if AI is faster when I run a practice session, but what I am experiencing is the other way round - AI is faster if I do not run a practice session and go straight to qualy. Very weird to me.
No, first post says 1:45 with a practice session and 1:40 without. * @Senght I would help if I could, but I've basically never run against AI so don't know anything about its behaviour.
"...but what I am experiencing is the other way round - AI is faster if I do not run a practice session.. " He's clearly contradicting himself
I don't follow, but it's important we at least clear this hurdle so the issue can be perhaps solved/helped. To paraphrase that quote: AI is faster with no practice session. First post, says AI do 1:45 with a practice session, 1:40 without a practice session. So AI is faster with no practice session. I'm a little worried I'm somehow reading this wrong, but I'll persist...
Thanks very much, you clearly understood my post. For the benefit of those who are confused about my initial post, Lazza has very kindly helped me summarise the issue I’m facing in his post.
There, in Bold for you Either way it's not something I've noticed, @Senght first things first (well a bit late!) What content are you using?
I... still don't get it. What you're bolding agrees with what I said, and how I read the first post 12 hours ago when I first read it. Emery I think replied to the title rather than the first post content. Or I'm completely wrong, I don't know.
Hahaha oh my god i need coffee, i do apologise I've been reading the Numbers completely backwards #Facepalm , sorry chaps that's a very tired brain there. As you were
Marc, you need to go and have a sleep mate I'm as old as dinosaurs and even I understood what he was saying. I think I read somewhere that the AI behave oddly if you don't run a full quali, they also behave differently if you speed quali up with the time lapse (called something else in game but can't remember what....dinosaur...) button. Maybe a similar thing happens if you don't run practice??? Might be black magic. Sorry can't help.
Well I'm relieved, I was just about ready to have myself committed @JimmyT I know that like many games, the times recorded for a session will differ depending on whether you skip to the end of it or run through it in accelerated time (since the second one makes the AI actually do the driving, while the first just generates some times based on some parameters), but I don't know if I've seen anything about subsequent AI speed changing based on whether a session is skipped or not. If anything you'd expect the opposite of the OP, either through AI learning or track rubbering, so it seems a bit strange. But here endeth my knowledge of AI, so...
Yeah I've not been well plus lack of morning coffee = massive brain fart! As above (and i know this is right!) If you Skip Session then the times are generated automatically which won't represent the actual times the AI would get if you let the qualify session run normally. @Senght Also make sure you're running the same Real Road profile and that each session isn't different, i.e. Load one or start Green for practice then Naturally Progressing for Qual & Race. I've seen people get this mixed up before and have Green for race (for example) and they wonder why times are slower!
IF the rcd files truly work, they act as a filter to slow down the ultimate pace of a car. Anything less than 100(on a scale) is to the detriment of the actual pace. So skipping practice or accelerating practice, or other sessions, probably also bypasses the filter of the rcd file. Thus the driver with the lousy qualifying rating doesn't pass on that fact to the computer and he's assigned a nice average pace that the CAR is capable of. If you are looking for a logical reason, that is the best I can come up with. The interaction between various components may be absent when sessions are skipped. And it appears, that once items like the rcd are ignored, even for practice, they stay ignored for the remainder of your time at that track.
I've also read that you should never interrupt a qualifying session because RF2 will generate Q times for the AI rather than using their actual track times. The suggestion I've read is to speed up the session with Ctrl X and let the time expire.