Please do so.
Normally I wouldn’t respond to you because I’ve you trolling with the blind fanboy persona in multiple threads, but I’ll bite.
You want me to go and video myself doing circuit races where I maintain similar lap times throughout, but the AI speed is dramatically different?
Just yesterday I had a race where the average AI 2nd lap time was around 1:30 then jumped to 1:15 on the last lap.
Don’t be so naive. Rubberbanding has been a staple in the Forza franchise, they’re not going to take it out randomly for this one.
1 Like
I’m waiting. Just lemme know on Xbox when you’ve done it, and I’ll tell you why you’re wrong.
Given that it’s been over 24 hours, I’m guessing you couldn’t find any evidence supporting your statement. Thought so.
there’s no rubberbanding , drivatars have the track as their no1 priority even before the opponent , so with more leniency to sacrifice speed , they get faster , more clean corners than you do, thus saving more acceleration time once they exit , so when you’re at 1st and drop half a second on a corner , an obstacle or trying to dodge a spin out for example , their seconds start to add up ,so at the end you’d notice they’re too close for comfort , but with better cornering and track knowledge they tend to stay withing the gap you already created.
drivatars have received a noticeable cornering buff in FH4 , small circuits on unbeatable are a complete nightmare.
1 Like
A fair few people have said similar things, so their may be something to it. It might also be difficulty related I race regularly on the higher difficulty’s pro/unbeatable and the ai is consistent through out
I love complaints about drivatars on pro/expert/unbeatabe difficulty. Even if there is some sort of rubberbanding (which there is, I’ve tested it multiple times on Goliath) it’s for the purpose of difficulty. You don’t want the drivatars catching up to you if you make the smallest mistake? Lower the difficulty. I expect unbeatable drivatars to rubber band as well as have amazing cornering speed and what not. It makes it fun and makes you a better racer by making you drive perfect.
1 Like
Yes, like others peoples saying here, if you don’t support to be beat by IA, lower the difficulty, simple as that ! For me it’s a very good point to have good IA on Expert/Pro/Unbeatable, it’s push me to be better and it’s a very good test to see if my setup for my car is good or not ! And the weather and the nature of the ground is an important thing to take in consideration, in the past it was possible to win in the mud with a car not setup for that (with difficulty sometimes but it was possible in past FH) but now it’s very difficult and it’s important to prepare his car in the good way. It’s take time for sure but it will help you to be better and of course, a good learning of the race help too. I don’t want an easy game just to gratify my ego and I’m sure no ones here want that.
Sorry if I look a bit too agressive or other, english is not my first language 
There is rubber banding / unfair A.I/ wonky physics for A.I, whatever you want to call it, 100% in my mind.
Of all the races I have done on FH 4 so far, I have come in second place twice.
Do I run the perfect race every time ?, No.
Are there a point (Easy on a sprint race), where i know if i’m not at a certain position in the grid at a certain point in the race, I cant win, yep.
The first part of the race, goes one of two ways but the end result is always the same;
Good start, go outside of A.I cars, group of 2/4 cars always pull away,
Poor start, race through middle of grid easy and midway through race be up against the pack of 2/4 cars that have pulled away.
Midway to final part;
Work you way through leading pack, till there is just one out in front.
Final part:
Over take the final car, and then the A.I will stay close but never really threaten and you win.
Or there is also an alternative “blueprint”/ “template” that races go;
Good start, outside A.i pass most, carry way too much speed into the first corner, use cars in 3rd and 4th to slow down, very quickly catch and over take 1st and 2nd, have 1st sit close but never really threatening the rest of the race.
These three story lines are repeated in every single race I have had on FH4, I never really feel my driving abilities are 100% the decider in winning a race.
I started on Highly skilled and was asked to move up and up, etc, now drive on unbeatable.
1 Like
I’ve noticed some oddities.
Last night, I set up 20 laps of one of the short tracks in Edinburgh. I was within three tenths on all of my laps, and was a second up on the fastest lap in the chasing pack. All my lap times were at least a second faster.
They finished six seconds behind me. Maths is not my strongest point, but that strikes me as odd!
I was very much in the ‘there’s no rubberbanding in FH4’ camp and it’s definitely not like it was in FH3, but in some occasions I have seen the AI change their driving behaviour based on the player. Two races showed it more clearly than most.
Firstly a 15 lap race on one of the Edinburgh road courses, anything goes in a standard BMW M1 on unbeatable. I was in the lead by lap 3 and raced more or less flawlessly for the rest of the race, by the end I had a lead of around 20 seconds… until the last lap. Upon crossing the finish line it went to the results screen where I could see the running order and best laps, I didn’t have the outright best lap but it was marginal, then the AI crossed the finish line to complete the race and suddenly I had the slowest ‘best lap’ of the race with one car that finished in 7th setting a best lap more than 5 seconds faster than mine. The AI finished only 16 -20 seconds behind, meaning the 11 other cars finished within 4 seconds of each other. It may no difference to the outcome and the fact remained that I’d won the race comfortably, but it was interesting nonetheless that the game seemed to boost the AI in the final lap to make it look less of a walkover.
The biggest adjustment I saw though was on the quarry cross country course, here I setup an anything goes 30 lapper in heavy rain on unbeatable. I was in a standard Jeep Trackhawk and got a very lucky grid as there were no off-road vehicles amongst the AI line-up. I was in first by the first hairpin and was a good 4-5 seconds ahead by the end of the first lap. Again I made no mistakes and was expecting to have the AI lapped several times over the 29 remaining laps (the AI was really struggling for grip and hopeless over the jumps running cars like standard Lotus Exiges for example). However by lap 5 I was about 20 seconds ahead and the AI seemingly just shook off all their problems and instantly matched my pace, they did this for the entire rest of the race, all tightly bunched together without any further problems. This was as clear a blatant adjustment of their driving ability that I’ve seen, again it made no difference to the outcome and I suspect had I had a big crash, once they closed in on me they’d have gone back to complete incompetence almost immediately.
In Co-op however I’ve noticed that this doesn’t work the same, I’ve joined a couple of co-op events with multiple laps and ended up lapping the majority of the AI (almost like I was considered an outlier and they were only concerned with being matched to the host of the event). I’ve also not really seen any dodgy behaviour as mentioned by others in short races or sprints, it’s just dependant on how I drive and what car they’re in. If I drive perfectly then I win 9 times out of 10 with the other time being a lead drivatar that got a good start and had a great car for the event.
It’s nice to see others have noticed the poor AI as well. I raced a blueprint race (Moneyball @ 50 laps) on ‘inexperienced’ difficulty and throughout the entire race had 2 drivatars within 2 secs of me. On lap 48/50 I missed a checkpoint, respawned in 11th place and finished the race in 6th. Kind of defeats the whole point of having difficulty settings.
In response to all of this, you can also sometimes defending by blocking them if they do catch up, typically the A.I. will slow down if you get in their way.
But as other suggested you may have to do some tuning if you play unbeatable or pro.
I thought i was the only one. The AI magically gets better near the end of the race.
Yeah I get what people are saying, they might enjoy the challenge of artificial ‘on-rails’ racing with arbitrary modifiers. I would prefer something that felt more realistic and believable. It’s like when you go to a baseball game, you go to watch teams play against eachother, not to watch a batter play against a pitching machine.
1 Like
This should be the main takeaway: Racing against the AI in Forza Horizon games FEELS artificial, win or lose.
This is bad game design.
AI will never equal human opponents, but players should still have a sense of fairness when racing against the AI which is completely absent from the Forza Horizon series and looks to get even worse as time (the series) goes on.
I have accepted this will be my last FH game (most likely) because not only has the game itself been dumbed down (stripped of meaningful base content that deals directly with racing), but I don’t expect PG to actually fix the AI because it’s a cheap system that works for consoles limited hardware. Even if the PS5, and XB2 are equal to mid-to-high-end gaming PCs next gen, that is even more reason for them NOT to invest in fixing the AI because it’s all about cutting corners, ironically. They gotta put more budget into incorrect car models, botched sound design and most of all up the polygon and shader count because that’s what sells games, not the actual quality game play anymore, unfortunately.
P.S. – What would would help SuperHornet’s case if he would just admit their is rubberbanding instead of flat out denying it.
It doesn’t matter if he can beat it. I can beat it on lower difficulties, and so can a 12-year old. That’s not the point. He doesn’t seem to (or want to?) understand the majority of us don’t like racing against AI that feels artificial and that many, many have proven time and time again is blatantly unfair & inconsistent. That’s what a vast majority of racers want, at all levels: Fair and consistent AI that is FUN to race against, win or lose. Race outcomes that are dependent on driver SKILL (and tuning) rather than pre-scripted algorithms. This isn’t happening and people instinctively KNOW it when they come to the forms and question the AI and gasp provide proof of their experiences. To blatantly ignore, or gaslight them doesn’t help the problem and it certainly doesn’t tell PG about a major part of their game that needs redesigning, re-balancing if possible?
2 Likes
If there’s no rubberbanding, there’s still definitely something odd about the AI. On the Goliath, for example, you’ll find that all the AI cars will sit back at almost exactly the same distance throughout the race, and occasionally one or two will break free and catch up stupidly quickly. You then throw a couple blocks, and they almost kind of retreat back into the main pack. It’s completely bizarre. So if it’s not some weirdo implementation of rubberbanding, it’s something else entirely.
As I discussed in this thread: https://forums.forza.net/turn10_postst135276_Rubber-Banding-Is-Ridiculous.aspx
I think SuperHornetA51 might be right. I ran the Derwent Lakeside Sprint in a stock Peel P50 and won in Pro difficulty with no sign of rubber banding. In FH3 when I ran a Sprint using the slowest car in the game completely stock, I had a different result. In FH3, the cars pulled away, I caught them on corners because in FH3 they cornered terribly, then when I was WAY out in front, the 2nd place car got an HP boost and would have passed me were it not for my blocking. The cars were identical and AI should not have caught me like that.
As I mentioned in the other thread, not only did the other stock Peel P50s not pull away, I slowly caught them with smoother driving using a tighter line. I passed the lead car on the first off two turns where I had to ease off the throttle. That was about 65% of the race and after staying very smooth, they did not catch me. Unlike FH3, there was no sudden burst of power to catch me.
I’ve proposed a way to prove/disprove rubberbanding on this thread:
https://forums.forza.net/turn10_postst137726_Experiment-to-Prove-Disprove-Rubber-Banding.aspx