isotaan Posted February 7, 2022 Posted February 7, 2022 This is a problem with many of the AI aircraft; they cheat by not bleeding the speed they should when they turn and their acceleration figures don't account for gravity, so they can actually accelerate faster in the vertical than a human-flown aircraft even if the human plane has a better TWR. That's why you'll often hear of AI F-5s and others as having "UFO engines." This cheating of physics is more apparent at the higher difficulty levels (since Ace AI planes will pull harder). The F-5 has it bad, but the Korean War jets have it *really* bad. They're supposed to be roughly equivalent, but try taking a F-86 out against a Mig-15 and watch as they demonstrate their superior UFO physics modeling, especially in the vertical. 5
TLTeo Posted February 7, 2022 Posted February 7, 2022 Yeah, some AIs are notoriously messed up. It would be worth re-doing the test with the F-5/Mig-15 set at Normal (and maybe a Tomcat or Sabre set at high or very high?) to check how the results change. 3
LowRider88 Posted February 10, 2022 Author Posted February 10, 2022 From my other tests, the worse UFO behavior seems to come from the F-5, the Viggen, and th AV-8. However, I don't have flight manuals for the others. But Viggen pulls long number of seconds over 9Gs, and AV-8 out turn rates both MiG-19 and F-5.
Tiger-II Posted March 21, 2022 Posted March 21, 2022 Prior to the files being locked, I found just halving the values for the F-5 AI afterburner was sufficient to make it more reasonable. Motorola 68000 | 1 Mb | Debug port "When performing a forced landing, fly the aircraft as far into the crash as possible." - Bob Hoover. The JF-17 is not better than the F-16; it's different. It's how you fly that counts. "An average aircraft with a skilled pilot, will out-perform the superior aircraft with an average pilot."
Recommended Posts