[FIXED] Build 1566 - Strange missile performance

[Transient]

Please see attached.

In the scenario the incoming CJ-10s are unable to be intercepted at all by the Aster 30 SAMs, after the update.

Pre-update there was a fairly good chance of intercept.

The problem is that now, none of the ASter 30s even get anywhere near the CJ-10s for missile intercept calculations to occur.

Is this WAD?

[blu3s]

Thanks for the report,

there's something weird, we'll investigate it.

Logged 16553

[thewood1]

If you upgrade to db 510 they seem to function well.

[erick33]

Thanks for the report,

there's something weird, we'll investigate it.

Logged 16553

Based on my own tests, the main cause seems to be that the rocket boost time for missiles in the database is too short. Currently, the boost time for the Aster rocket in DB 510 is 19 seconds. Because of this, the boost phase ends shortly after launch, causing the missile's speed to drop and preventing it from utilizing its full range. The SM-6 also has a short boost time, making it difficult to reach its maximum range.

[Dimitris]

Thanks for the save. This has been looked into, and an improvement made for the next update.

Some comments from the dev who looked into this:

There are a few things going on here.

I am seeing a different SAM battery fire depending on sim speed (fidelity). The reason the SAMPT battery in direct line of the missiles often does not shoot at 0.1 sim speed is because of the WRA shooter limit -- at 0.1s sim speed the off-axis SAMPT battery shoots first and this prevents the battery 'in line' from shooting because WRA shooter quantity has been fulfilled. 1s vs. 0.1s sim speed changes the timing of which SAM battery creates a salvo first.

A lot of shots miss because the SAMs are not lofting at all -- this is due to the high speed of the incoming cruise missiles triggering the intercept into a [direct-flight instead of loft] situation. But the cruise missiles are terrain following and the terrain is quite hilly, causing the interceptors to do a lot of climbing and diving to keep from hitting the ground, which slows them down and can leave them at unexpected pitch angles as they go active. At 1s sim speed there is less of this up-and-down than at 0.1s sim speed.

[........]

The root cause is something odd with the saved game -- it has CJ-100 missiles flying at 9m altitude, as if they were terrain following. If I create a sample scenario the CJ-100s loft to cruise altitude and then dive on target normally. However, this save file did expose a couple of issues with the weapon code. (NOTE: The CJ-100 IRL is reportedly a fast & high-flying missile).

The first is the lack of lofting I mentioned above. I fixed this by adding a target altitude and range check to the 'is the target in terminal dive' check, so the SAMs will continue to loft vs. distant and low targets like cruise missiles, even if the [target missile] is in terminal dive mode.

I also made an improvement of the initial intercept / terminal point calculation for boost-coast missiles. The existing check was a bit optimistic and the missiles were thus arriving too late to even detect the CJ-100s in most cases.

These two changes do improve the behavior, but it is still very difficult to hit the very fast, very low CJ-100s with 'blind' shots from the SAMs (where the SAM system does not have the CJ-100s on its own sensors.) Even with the SAMs getting into a decent position the SAM terminal sensor takes a few seconds to detect a CJ-100 and by then it's often too late for a successful intercept.