It seems like sometime during the past year, the ETK's vehicle configuration for the Downhill Rally run was changed from having the "Stage 2 Turbocharger" in earlier versions of the game to only having the "Stage 1 Turbocharger" in the current version. It's a difference in 10psi of boost and nearly halves the power of the ETK Rally(Gravel) setup from what I'm able to tell changing configurations around in freeroam(400+bhp vs 220bhp), at least with the current version of BeamNG. The real difference is that it pulls through corners at speed since it doesn't bog when you try for a higher wheel speed in a higher gear.

I really hate to be "that" guy, and the new 1st place run is incredibly solid otherwise, but I don't see a way for anyone to get a time faster than that with the newest version of the game. I'd honestly rather the devs changed it back to the previous configuration since the stage certainly feels a lot better with the bump in power, but that's neither here nor there.

Regardless, thanks for the consideration and your time in reading this.

The scenarios timer stops when both laps have been completed, and the current 1st place is only for one of the two laps.

Should the WR be the total scenario time or just your best lap out of 2? If it's for the total time, the current run in 1st place should be changed to 4m 04s 93ms, and not the 2m 01s 169ms it currently is.

Thank you for your time in reading this, it's a little nitpicky but I thought that it should be brought up anyways.

There are a couple of runs that rely on the AI behaving wrongly to achieve the fastest times, ways that might be/become impossible to get as of today since the devs are always making AI improvements amongst other various bug fixes.

Examples of this would be in the scenarios "Bank Robbery" and "Police Patrol".

In both of the current 1st place speedruns, the AI bugs out and causes the scenario to end in a success state earlier than what it's supposed to. Also in "Bank Robbery", the 2nd/3rd place runs use a path that is no longer possible and results in a failure state of the scenario. "Police Patrol" is better in the fact that the scenario hasn't been altered much by updates, but the AI occasionally just faults immediately which causes an early success state. I wasn't able to reproduce the AI behavior in my attempted runs, but even then, I can't say for sure that the AI was fixed or not since the first place run took place.

I also found that the scenario "Canyon Jump" has been changed since some of the runs, altering the starting area from the dirt back further to the parking lot. It seems to have made achieving a sub 20 second time just about impossible. Also in doing so, it makes the current 1st place run obsolete, since it starts already well into the dirt before the first jump.

IMO if the AI bugs out before the player has a chance to interact with them during a run, it probably shouldn't be counted. It leaves quite a lot to RNG/randomness in a game that already relies on a lot of good luck with AI.

Even though I believe that this may introduce other aspects to verifying speedruns as well, such as if the AI behaved sub-optimally at the start of a scenario and/or didn't move for the first couple of seconds after the beginning of a speedrun for that scenario and whether it should be counted based on that.

This is not to be confused with the pathing that an AI takes during any "chase" scenario, which is dependent upon where the player is located, which can be manipulated/altered in-game for faster runs without relying on just pure RNG.

That's all I have to say, thank you for taking the time to read all of this and may fortune be on your side for faster speedruns.