coldburst

Symplectic Euler is often a good compromise. It's as simple as forward Euler but has better energy conservation.

Suppressors only reduce the loudness of the muzzle blast without impacting the projectile's ability to break the sonic barrier. In fact, suppressors slightly increase projectile speed because they effectively add barrel length.

Having a kill switch "just in case" really is a strategically logical move. Consider the case of Iran: In the 1960s and 1970s, the US delivered several US-built aircraft to Iran, including F-4s, F-5s, and F-14s, some of the most advanced fighters of that era. After the 1979 revolution, US-Iran relations collapsed. Despite the US cutting off spare parts and maintenance, Iran has been able to keep the aircraft operational to this day (likely through effortful reverse engineering). So now there is an adversary that is armed with equipment you provided. The obvious way to avoid that is to install a kill switch.

For "boosted fission" and early thermonuclear bomb designs yes, but not so for the more evolved thermonuclear designs. Navajo (Operation Redwing, 1956) was 95% fusion, Tsar bomba (1961) was 97% fusion, Housatonic (Operation Dominic, 1962) was 99.9% fusion.

Would that be much different/better than TCAS?