@Herve_Versabeau You write "I think they had more fuel at the start of the race." When you have fixed Setups set properly, they use the same Amount of Fuel that you use too, if you talk about the same Car that you use. So i have to ask if the BMW M8 GTE AI had 24 seconds longer Pistops than you. And did you do the full Pitstop without to reduce the refuel Amount or skip something? And does it happen with all AI of the same Class? If you have set fixed Setups, it only works for the Car Classes, for that the assigned Setup File can be found in the Settings->Location Folder. I maybe have to add, that i never race against AI, so it maybe is really bugged. But as far as i know, from Statements of offline Drivers that race against the AI, different Pit Stop Times normally have a Reason, like Damage Repair for Example. But with fixed Setups set up properly, the "Basis" for all Cars, for that an assigned Setup is in the Settings Folder, is the same.
My test today with "Fixed AI Setups":true and "Fixed Setups":true with my setup assigned to IA SPA grand prix 1H race with 29 IA only on GTE and I am on BMW M8 GTE for this race, you need about 102 L of fuel for 1H00 pit times for me and the AIs come from leaderboad secondmonitor when I stop in the pits, I add 12L of fuel and change all 4 tires practice : 30 min with accelerated times no qualification when i stopped in the pits, there was a car on my spot. I pushed it, but it cost me about 5 seconds more. the AIs took between 16 and 18 seconds longer than me to stop the BMW M8 GTEs took between 16 and 18 seconds longer than me to stop Maybe the practice had a slight impact on pit-stop times for AIs
@Herve_Versabeau have you checked the logs (manually or something like log analyzer) to check AI fuel levels?
@Herve_Versabeau I think the Problem is this: when I stop in the pits, I add 12L of fuel and change all 4 tires. Did you set that in the Setup or do you reduce it in the Pit Stop Menu? If you don't set it in the assigned Setup already, the AI will refuel more (i guess), which of course lasts longer.
My test this afternoon with "Fixed AI Setups":true and "Fixed Setups":true with my setup assigned to IA SPA grand prix 1H race with 29 IA only on GTE and I am on BMW M8 GTE for this race, you need about 102 L of fuel for 1H00 pit times for me and the AIs come from leaderboad secondmonitor when I stop in the pits, I add 12L of fuel and change all 4 tires practice : 30 min with accelerated times no qualification with no pit strategy weird, when I asked for a pit stop the pit menu told me 80L for refuelling I thought it would give me 0 as I didn't have a pit strategy the AIs took between 26 and 28 seconds longer than me to stop I continue my tests with practice = 30 min pit strategy = one pit stop with10L for refuelling
I found this https://forum.studio-397.com/index.php?threads/ai-running-out-of-fuel.64490/page-2#post-1046194 So I did a lap race with 25 laps + a pit stop strategy with 10L for refuelling. SPA grand prix 1H race with 29 IA only on GTE and I am on BMW M8 GTE pit times for me and the AIs come from leaderboad secondmonitor when I stop in the pits, I add 10L of fuel and change all 4 tires practice : 30 min with accelerated times no qualification pit strategy = 1 pit stop with 10L for refuelling when I asked for a pit stop the pit menu told me 10L for refuelling....as expected the AIs took between 6 and 10 seconds longer than me to stop it seems to be confirmed that for time races, refuelling does not work properly.
@Herve_Versabeau If the Cars have to refuel, you have to use a Pit Strategy to get the Refueling working properly for you and the AI. So set the Pit Strategy, set the Amount of reloaded Fuel for Stop 1 and 2, save the Setup, assign it and put it in the Server Install. If you don't do that and the AI has to refuel, the default Values from the HDV File get used (80 L as it seems). And don't Skip or select anything in the Pit Menu.
As written in my Tutorial, the assigned Setup File (*.svm) has to be put into the Servers ..\UserData\player\Settings\TRACK Folder. Where TRACK is the Name of the Track for that the Setup File is. And the FavoriteAndFixedSetups.gal with the correct Entry for the assigned Setup, has to be put into the Servers ..\UserData\player Folder. If you use the integrated Server, of the Installation with that you play, it is not necessary.
So just in case you can't get the strategies to do what you want (or until you do get them working for AI), yes, this log shows that the AI all fill up their fuel tanks completely at pitstops. So to have a competitive race against them you'd have to do the same. Pit timing itself is identical between you and AI cars (ie tyre changes, and refuelling, take the same time; there's no separate AI pit timing settings). And for info: that file is a text file. You could open it in notepad or similar. All the cars' laptimes and some other info (fuel level, tyre wear/compound) are listed at the bottom.
The ai do take longer to complete a pitstop. A ai car can enter its pit stall before me and I take on full fuel and 4 tires and I always leave before the ai. Other sims I've played, you have to be perfect entering the pits or the ai will pass you.
Normally, if you use the default Setup and don't do anything when you make a pit Stop, there should not be any Difference, in the Time that a Pit Stop lasts, between you and the AI, except if some Damage gets repaired.
The ai always fill the fuel tank and change tires when they pit. So if I fill the fuel tank and change tires also why would changing the player setup prior to the race effect pit time. The only way pit time should be effected is if I don't completely fill the fuel tank or don't change tires, making pit stop shorter, or changing the wing setting or other changes making the pit stop longer.
There's no reason for there to be any difference, as pit timing is in the .hdv so is the same for player and AI. There are multiple reasons they could end up taking longer as above, but there's no underlying setting or code that would make the AI take longer in pitstops. I'd go so far as saying a measured difference in careful testing eliminating other variables would justify a bug report.