The AI’s discrepancy between practice and race shows how much “rubber banding” there might be. The Road America race seemed to be the most evident of this. Since I was running a maxed out 962c against the AI’s “stock” PI cars, I was able to get them 4 seconds + faster than their practice times during the race. I was averaging 10 seconds ahead of the AI, then around the 3/4 point of the race, they kept running faster lap times to try and keep up.
Wouldn’t that basically be the same thing as rubberbanding? Sure sounds like it to me.
1 Like
I did the race on level 8, I noticed the AI had trouble overtaking the GTX, but they never caught up to me. If there was rubber banding, it clearly wasn’t very effective.
Thats pretty much the exactly want it is. Funny though because in another one of your informed posts you mentioned that FM7 did not have rubberbanding when it actually l had a very well known AI HP boost. So which one is it?
After actually playing and not AFK’ing I believe that the difference in AI performance is the tyre wear multiplier. We run at 3x and they remain at 1x. That’s my unproven opinion but seems to make some sense.
I think this was their fix to the complaints that the lead cars broke away from the field. Essentially, they’ve artificially slowed them down if they get ahead so that the field closes back up again. As such, if you are some distance behind them, they will drive relatively slowly before going back to their true pace when you get close/ahead.
I’d probably describe that as partial rubber-banding. What they don’t seem to do is use speed boosts etc to artificially stay close to you. For example, at the Road America race, I’ve used three cars with massively different PI. I’ve ran best laps in the 44s, 46s and 48s. In all three races the AI times have remained consistent - with a best lap between 47.9 and 48.3.
This was my experience in this race as well. After a few laps I took the lead and held a 1-2 second lead until we started lapping the GTX cars. The AI had no trouble passing GTX cars in the straights but in the curvy bits they really struggled. One thing I did notice was that around lap 17 the AI did pickup the pace and even though we were hitting traffic on every lap they were doing 1-2 seconds faster (or at least that is what their faster lap suggested).
I did do a 20 lap Maple Valley race with back markers ghosting turned on and noticed something strange. Late in the race I caught up to a group that was the 13-15 placed cars and as we came down the back hill I watched as the 15th place car drove through the 14th and 13th placed cars to take over 13th place. I don’t know if this only happened because I was near them or if this is how the AI races with ghosting turned on all the time.
I know ghosting happens when AI is pitting. It’s possible the two cars in front were heading for the pits, and the one behind just plowed through the g-g-g-ghosts.
I said FM7 had little or minimal rubberbanding. But in any game, the rubberbanding is relative to the players. And faster players are less affected by rubberbanding than slower players. As evidenced by the conversations here and in the rubberbanding thread. The experiences of the slower players are all different from the faster players. My experience in game play is much different from AdmPatate or Jezza14.
And I’m pretty sure, we were in an MP race together, those who have issues with rubberbanding would be much closer in competition on my level than those who don’t notice the rubberbanding. Those are just my observations and opinions.
By the 4th lap on the first Sebring race they were doing laps around 8 seconds faster than their practice laps.
That’s why I was convinced there’s some kind of rubber banding going on.
It’s the only time I’ve noticed such a discrepancy though, so I’m not sure what to think of it.
2 Likes
This was just as we started down the hill between the 1st and 2nd turn so they were not yet near where the ghosting for the pit normally happens. Or at least normally the ghosting doesn’t happen that far from the pit entrance.
The career endurance racing in FM7 has the worst rubberbanding I’ve ever seen in a racing game.
I’d hate to quote to people out of context but here is a literal copy and paste from your post.
‘AI in FM7 didn’t have rubberbanding’
I haven’t played MP in months, that’s a different story. So not sure why you’ve brought that up. I have played a lot of career though.
Hey siri how can I backtrack while trying to appear I know what I’m talking about.
Then don’t quote people “out of context” show the whole quote or get the f out of my face and stop trolling me.
My statement “IN CONTEXT” was; “AI in FM7 didn’t have rubberbanding. If it did, it wasn’t very noticeable or it was vrey minimal.”
Now bugger off…
I also had my sound go out for me towards the end of the first endurance race at Sebring. Kinda funky bugs
What type of times NT? The RA race seems a good one to test of any rubberbanding, as the AI always has the same 11 cars/PI.
Looking forward to the Forza Touring Cars and whether or not I can extend the life of the Subaru tires, the only car I ever want to use in that class.
Well, I’m not running level 8 difficulty (level 6 is my norm), nor am I as fast as you 
. The pole AI had “run” a 1:51.xx which is about the same as my practice time. I ended up running a 1:47.xx in race about 3/4 of the way through (still had decent tires). Every faster lap I ran, the ran the same amount faster (in difference, not over all time). I would run .9 sec faster, they would run .9 faster than their best. 3 laps straight I ran faster up until my 1:47, they (2nd-4th) would run faster as well, they never got to 1:47 though but were very close with low 1:48’s. The key is my PI was above theirs, by quite a few.
Okay, so normally I wouldn’t bite in a situation like this but I feel like a should help you. In the spirit of the forum and all.
Sub context A context within another context; a context that is subordinate to another .
Either way bro just chill with the pretending you’re the voice of the forum and a SME because you defo are not.
Interesting - the best lap in my three attempts has been 1.48.1, 1.48.3 and 1.47.9, each time by the PI969 XJR-9. Suggests my theory that they have a “genuine” pace and won’t go beyond that, but rubberband up to it (seemingly even on lower levels) is correct.
A shame this car doesn’t go higher than 981, as it’s an amazing car. Drove into a wall near the end on worn tyres so lost about 20 seconds limping home. AI still in the 1.48s.
In a way, but a bit more acceptable/believable if it means they don’t actually exceed realistic capabilities.
For instance, sandbagging might be driving at 80% when the player is significantly behind but 100% when they close up (happens in real life to hide performance from officials/competitors until the last second), rubber-banding might be 80% when ahead and 120% when behind. Yes it is splitting hairs a bit and also relies on having a clear definition of what 100% capability is (and whether that is realistic or not).