Hi guys,So I’ve recently started a new MyTeam career, this time doing 50% races (previously did 25%).Have noticed at certain tracks the AI race pace is just too fast - Miami so far. I just can’t keep up. My car is 2nd to last in development and I’m racing 90 AI.I spend a fair amount of time in practice doing the programs to ensure I pass them, and in the case of Miami, I passed all the practice programs fairly comfortably. My lap times are comfortably better than where my car development is during practice.I was at least 0.3s above the race practice program (with ERS being used minimally, how I would in the race, no DRS) and was able to maintain this for all 5 laps. The issue is compared to the AI race pace, the practice programs give me no warning as to how fast the AI will be, and also I can’t change the AI level on race day. Anyone else noticed this happen to them? I’m also interested in specific tracks where this might happen, so I can have a chance of reducing AI before the weekend starts.Any comments appreciated 🖖🏽

I'm gonna be as honest and direct as possible (and it's ok if people disagree with me), BUT..The AIs in F1 22 are just completely bad.I mean.. their practice, quali and race paces are literally all over the place - not to mention their behavior when fighting for positions (attacking/defending) and all of that together is why there is such a big discrepancy in terms of performance between different sessions and it's even worse (more noticeable) in some tracks, like Miami, Baku, Singapore, among a few others..It's also related to how they were coded - including, for example, how they 'manage' their tyre temps (always keeping them at 92 ºC, no matter how hard they push), which helps them to make their tyres last longer and keep and amazing pace throughout their whole stints (including getting their fastest laps at the very end of each stint - sometimes going up to 1 or even 2 secs faster than their regular pace during the stint) + their unrealistic and most of the time impossible to keep up with acceleration after the corners - especially the slower corners (like the one before the long back straight in Miami, for example) + super top speed and that's why and how they gain so much on us, players. at that particular sector..I keep complaining about the AIs in this game and I apologize for sounding like a broken record, but.. this is why things like these happen: the AIs are just badly coded in this game - change my mind! 😅

All,I’ve set myself up a little spreadsheet and re-started my career.I will keep the AI at 90 throughout and farm for data; Will note down all targets and my performance in practice programs, qualy and in races, compare this against car development as well as teammate performance.Once I’ve got data for decent mix of tracks (maybe around halfway in the season), will report back.🖖🏽cc: @dancrodrigues @SexyBrigadeiro @sc31245 @Bauersome @ScarDuck14

@Monzstar83 Same as you I gain my time in the programs on that straight. I thought last years game had balancing problems but compared to this year It was great.,. This year needs a Thanos intervention 😂

@Monzstar83Hey mate,I always use my race pratice program (RPP from now on) to both calculate tyre and fuel for the race and the AI level - so I do them a lot. Like, a real lot hahaWhat I feel doing this since F1 2019 is that the output depends strongly on the track's length and average speed. For example:Tracks with regular length and average speed (Bahrain, Spain) or fast short tracks (Brazil, Austria) if I let the delta to be +0.3s on the RPP them the race would be ok. Long fast tracks (Spa) this delta needed to be raised to ca. +0.5s. Slow tracks (Monaco, Singapore) I could let this time fall to +0.1s and I'd still do alright.On F1 2021 I did it meticulously plotting tracks' length x AVG speed, so I kept the AI fixed throughout an entire season and felt if it was easy, ok or hard. That's how I came to those deltas. It's tiring and one likely wastes a season doing so but it's rewarding from there on. Comparing throught 19-22, I noticed that I could keep the deltas though the AI absolute values would change.This graph is lost somewhere in GDrive, if I find it I bring it to you.Bottom line, I feel it depends heavily on expected cornering efficiency (braking and exiting) and its cumulative effect on the track layout (ammount of corners and their speed and tracks' length).So, I know I don't directly approach what you asked but I hope it brings some insight into it.Cheers,Danedit: @SexyBrigadeiro, I added small info since your reply.

AI race pace discrepancy compared to Practice Programs

37 Replies

ScarDuck14
Legend
3 years ago
@Monzstar83 I don’t like Miami so it’s a track I tack all new parts and poodle around at the back. Always find it’s the long straight that they are killer.

The first straight I find you have to nail the last corner by not scrubbing to much speech into it.

Singapore practice programs still way to demanding

But for the most part they aren’t as bad as they were
Monzstar83
Seasoned Ace
3 years ago
@ScarDuck14 You’re right about that back straight - it just kills me.

A few things I can’t square in my particular situation;

1) In the race practice program, that back straight is where I gain the most delta against target, which suggests my straight line speed is good - going into that straight my delta against target is positive and doesn’t drop too much (maybe 0.1s on the corner exit), I make up something like 0.5s against delta on that back straight.
2) my speed trap value (also at the end of that back straight) is like top 5 when compared to the rest of the grid

They just overtake with such ease on that straight and they can always overtake me if they have DRS by the middle of it, even if I’ve kept ERS on all the way from straightening on corner exit. In the roles reversed scenario, I can barely keep up even with DRS and ERS being applied the length of that straight - just cannot catch them.

I think a lot of it has to do with their exit speed out of the previous corner, their straight line speed and some disparity between race pace and practice program that the targets don’t inform you of.

I would be OK if I was failing my practice programs because then I could try to address issues with setup or reduce AI.

I think either the practice programs need to set targets indicative of race pace or allow us to change the AI level at any time during practice/race.

I don’t think on any other game release (21,20,19 etc), I’ve had this issue.
ScarDuck14
Legend
3 years ago
@Monzstar83 Same as you I gain my time in the programs on that straight. I thought last years game had balancing problems but compared to this year It was great.,. This year needs a Thanos intervention 😂
Monzstar83
Seasoned Ace
3 years ago
@ScarDuck14 wrote:
Thanos intervention 😂
😎
dancrodrigues
3 years ago
@Monzstar83Hey mate,

I always use my race pratice program (RPP from now on) to both calculate tyre and fuel for the race and the AI level - so I do them a lot. Like, a real lot haha

What I feel doing this since F1 2019 is that the output depends strongly on the track's length and average speed. For example:

Tracks with regular length and average speed (Bahrain, Spain) or fast short tracks (Brazil, Austria) if I let the delta to be +0.3s on the RPP them the race would be ok. Long fast tracks (Spa) this delta needed to be raised to ca. +0.5s. Slow tracks (Monaco, Singapore) I could let this time fall to +0.1s and I'd still do alright.
On F1 2021 I did it meticulously plotting tracks' length x AVG speed, so I kept the AI fixed throughout an entire season and felt if it was easy, ok or hard. That's how I came to those deltas. It's tiring and one likely wastes a season doing so but it's rewarding from there on. Comparing throught 19-22, I noticed that I could keep the deltas though the AI absolute values would change.
This graph is lost somewhere in GDrive, if I find it I bring it to you.

Bottom line, I feel it depends heavily on expected cornering efficiency (braking and exiting) and its cumulative effect on the track layout (ammount of corners and their speed and tracks' length).

So, I know I don't directly approach what you asked but I hope it brings some insight into it.

Cheers,
Dan

edit: @SexyBrigadeiro, I added small info since your reply.
dancrodrigues
3 years ago
Found it.
SexyBrigadeiro
Seasoned Ace
3 years ago
I'm gonna be as honest and direct as possible (and it's ok if people disagree with me), BUT..
The AIs in F1 22 are just completely bad.
I mean.. their practice, quali and race paces are literally all over the place - not to mention their behavior when fighting for positions (attacking/defending) and all of that together is why there is such a big discrepancy in terms of performance between different sessions and it's even worse (more noticeable) in some tracks, like Miami, Baku, Singapore, among a few others..
It's also related to how they were coded - including, for example, how they 'manage' their tyre temps (always keeping them at 92 ºC, no matter how hard they push), which helps them to make their tyres last longer and keep and amazing pace throughout their whole stints (including getting their fastest laps at the very end of each stint - sometimes going up to 1 or even 2 secs faster than their regular pace during the stint) + their unrealistic and most of the time impossible to keep up with acceleration after the corners - especially the slower corners (like the one before the long back straight in Miami, for example) + super top speed and that's why and how they gain so much on us, players. at that particular sector..
I keep complaining about the AIs in this game and I apologize for sounding like a broken record, but.. this is why things like these happen: the AIs are just badly coded in this game - change my mind! 😅
Bauersome
Rising Scout
3 years ago
@Monzstar83 Baku - the AI are broken on that track. I can't figure out if they have better grip or down force or speed but it is impossible to race against them - they disappear into the future.
sc31245
3 years ago
@SexyBrigadeiro"...their practice, quali and race paces are literally all over the place..."
Yes, definitely.
The AI's pace can vary significantly from one track to another. Eg Bahrain and Japan -- I really like racing both circuits, I don't feel like I'm better at one than the other, and yet I have to set the AI 10 clicks lower at Bahrain than Japan to get a comparably competitive race.
And the pace required to pass practice programs, especially the race strategy program, is not indicative of the race pace you'll encounter. 90 - 92 is a competitive setting for me at Spa, yet I can pass the race strategy program by a couple of seconds each lap. Meanwhile at Singapore at roughly 84-85 I struggle mightily to pass any programs (and I usually don't pass any).
So here's how I figure my AI settings -- I keep notes on every race and make tweaks to the AI setting accordingly. I did a season in normal career driver mode with Gasly as a teammate and used him as a benchmark, and then continued to tweak season after season.
Tracks that I find easier than average (I have to increase the AI setting) are Zandvoort, Spa, Japan, Great Britain, Jeddah, and (to a less degree) Abu Dhabi and Monza.
Tracks that are tougher than average are Bahrain, Hungary, COTA, Imola, Singapore and Canada. I know the "tougher than average" isn't all on the AI (except for Bahrain and maybe Hungary). 100% Singapore with sim damage and no flashblacks is tough regardless of AI level, I absolutely stink at Canada, Imola is a crash waiting to happen and I have yet to hit an apex at COTA.
Monzstar83
Seasoned Ace
3 years ago
@dancrodrigues wrote:

Tracks with regular length and average speed (Bahrain, Spain) or fast short tracks (Brazil, Mexico) if I let the delta to be +0.3s on the RPP them the race would be ok. Long fast tracks (Spa) this delta needed to be raised to ca. +0.5s. Slow tracks (Monaco, Singapore) I could let this time fall to +0.1s and I'd still do alright.
Hey @dancrodrigues,
Thank you for your input and providing your research.
Just so I am under no confusion, if in RPP, for regular length and speed tracks, if you are between 0.0-0.3 below delta (so delta is in green), then you’d be able to compete in the race ok?
The issue I have in Miami is that I am below delta (so delta in green), by at least 0.3 but I cannot even get close in the race, which leads me to believe the RPP needed to increase the delta time.
Your chart is interesting, as it is very likely there is a direct correlation between length and average speed for AI performance - I remember reading that 1 AI point equites to roughly 0.1s in lap time.
Linking the two (and using avg speed and length tracks as a base), one would have to increase AI for longer tracks with higher speed and vice versa for shorter tracks with lower speed?
I will check my general performances in races to see if the above holds true (I keep my AI at 90). There may be slight differences in my findings due to ability on certain tracks, but will be interesting to see.
At my last race (Monaco), I think I was about 0.3 below delta, I qualified 11th and finished 10th in an uneventful race (no crashes etc that would skew my finishing position).
🖖🏽