3 race AI test 75 & 65 + picture data
As I'm not enjoying the game at the minute, I've decided to do some testing. Last year's game I started a My Team on AI 85 using Medium TC and anti-lock brakes. And had won drivers & constructors' ch...
Forum Discussion
Belgium race results 75 started 6th finished 6th, overtaking is so hard to do even though IRL it's easier. My total race time was 20.46
65 started 1st and finished 1st. But only because max DNF before the pit stops. Charles was only 9 seconds behind me even though I'd dropped the level and went 12 seconds faster myself, If I had done the same time as 75 I wouldn't have won???
My times in Belgium.
Best AI times in Belgium
TM times in Belgium.
Belgium Strategy
Belgium incidents
Somebody make it make sense
Thinking of doing more testing, not sure if I should do 3 different tracks, or knock the AI to 55 for Bahrain and Spain. open to suggestions
Some of the results are clearly affected by variation which is perfectly fine. If times wouldn't vary the same race run over and over again would give the same exact results. So if the pole time changes by 1.2 and teammate is faster by 0.7 it's fine (on average there is an improvement of around 1 second). The game has to simulate somehow a driver having a good day and a driver having a bad day.
In my opinion even the pole time has to vary a bit as it does. Even in real life there are sessions where a perfect lap is required for pole and others where a solid lap is more than enough to get P1.
Overall time is weird though but I wouldn't pay to much attention to it. There are so many factors that can affect overall time. I don't know if the grip stays the same. DRS trains and ERS management can affect it too.
I've been testing the AI since forever in 100% distance and what I'm seeing in terms of race pace usually makes sense. In short distances the results can be close. In a full distance race a 5/10 increment can be the difference between winning and not even fighting for the podium. In 25% cars never reach the clean air phase where you can truly judge pace.
Over the years I've seen content creators that put up the AI by 10 points and still get the same results. If I do the same I go from winning to getting 20/25 seconds...
@sirio42994Thanks for your reply. I know every race won't be the same and love the way the drivers are making mistakes and locking up, going wide....etc. In fact, the first time I saw it happen was when Ocon lost his car on lap 1 right in front of me and took my front wing. But I smiled because it happened. I know EA Codemasters are looking into AI straight line speed, but I just think it's more than that. And thought these might help in some way @EA_Kent @EA_Barry . Just Bahrain qualifying how my TM got closer to me even though I'd reduced the level by 10. I'm just trying to find a level to play my team at that's going to be enjoyable but not impossible to do. I remember back in 2020/21 game Codemasters released a video to show you how to get close racing. It involved a short qualifying where you do as many flying laps as you could in that session and be within 0.2 of your TM, so I've tried that on this year's game. I started at level 80 as Alonso. Using the default set up, which is preset 4. I have done a 1.31.252 while Ocon was 1.32.982. So according to that I need to knock the level up to 97!
Short qualifying AI 80 results. Even some AI times don't look right, I.E not in the right positions compared to car performance.
