Okay, so here is what is happening.
If you look at the distant target engagement, the missile trajectory has enough horizontal space available to get the missile on a parabolic arc slightly above the target and then dive onto it:
When engaging the close-in target, the short horizontal range was less sufficient for this. Additionally there was in the code a "cap" present which limited the ceiling of the pseudo-ballistic trajectory, essentially preventing the missile from reaching the (more favorable in terms of drag) higher altitude. You can observe this here:
(When engaging an air-breathing target with a boost-coast weapon, the choice of mid-course altitude is a delicate tightrope between
"too high and I may not be able to get down on terminal dive in time, especially if he maneuvers; and the increased look-down angle may also complicate terminal guidance" and
"too low and he'll use the increased atmospheric drag to outrun me, or I may even stall before I get there".)
I was able to correct this issue in two ways:
* Added a slight upward bias to the loft trajectory arc during the boost phase. While the "pure" pseudo-ballistic trajectory is broadly the most energy-efficient one, it overlooks the detrimental effect of increased drag at lower altitude. "Nudging" the trajectory upwards has the effect of getting the missile up to more favorable (drag-wise) environment faster, allowing for a more efficient ballistic curve post-burnout.
(This BTW is also why in real-world artillery shots the maximum-range elevation angle is 47-49 degrees rather than the theoretical 45)
* Removed the "cap" on the trajectory arc. This will generally result in a slightly higher apogee, and thus reduced speed bleed due to drag in the mid-course phase.
As a way of comparison, with the original trajectory the missile typically arrived at the target with a ~900-950 knot speed, and the resulting PH was 50-55%. With the improved trajectory the missile arrives at 1200-1250 knots and the PH increases to 72-76%.
These tweaks will be included on the next update release.
Thanks very much for the clear, concise example provided and giving us the benefit of the doubt.
