I'll add that in my experience the massive slowdown happens once ballistic missiles are fired, during the boost phase, and ends as soon as the missiles become RVs.
Edit: And that this is a problem I have only noticed recently as well, across multiple powerful computers. This certainly seems to be a genuine bug as opposed to being a matter of complex calculations naturally slowing things down. Either way it renders certain scenarios terribly slow and tedious to play.
For me it continues regardless of the state of the missile. I tested this problem again a couple days ago, and there was a significant performance drop until the RVs impacted their targets. However interestingly, the slow down was less compared to the other test I used. When using SRBMs over a short distance (between China and Okinawa) 1 second in game took about 2 or 3 seconds while using ICBMs over an intercontinental distance (between China and the US) 1 second in game took over a minute. Just to make it clear, this is all during ABM engagements.
In the case with Okinawa, the Patriots continued firing ABMs at the warheads right up until impact because one of the batteries is incorporated into the airfield itself and thus the warheads remained in the DLZ. Also, during the boost phase, the game worked completely fine. It was only after the Patriots started firing that the slow down occurred.
I would also like to note that the first time I reported a similar issue, which was temporarily fixed, I was playing Salvo from CoW, which had 3000+ units. However my short range Okinawa test also had a high number of units, 2000+. So range seems to play a role in the slow down more than unit count. When the issue occurred during Salvo, ABM engagements were underway across intermediate missile ranges.
I hope I didn't say anything that is obvious to the devs who have access to the actual code and whatnot, I just want to provide as much info as possible.
Formerly known as Project2035, TyeeBanzai, and FlyForLenin