CM spawn point on capship models

[Compellor]

It does sometimes seem like I see the missile detonation animation when a missile passes through a chaff, but I've made a missile with long lifetime and low acceleration which not only passed through the chaff and kept going, but I was able to ram myself into the missile after it passed through the chaff, and I took damage as normal. The weirdest thing from that testing session was that at one point the missile hit me prior to the chaff, and kept going! I was unable to confirm whether it was able to double-hit, there might have just been some lag effects causing the missile to continue being displayed.

[Broodwich]

thats interesting, but I'm sure in game you have seen or had happen to you a missile hit you or someone and the damage to be taken, then in a half second taken back. I would say test it a few more times before concluding anything. Especially because i know i have been hit and have hit people with spoofed missles

[Jimen]

Factoid's analysis explicitly states that chaff has HP as well as a time-to-life and can be destroyed before that time runs out. On top of that, it suggests that even if the missile resists the chaff, you'll still be safe if you manage to get the chaff cloud directly into the missile's path. Given both of those, I think it's worth assuming that missiles do explode when they hit chaff, but given their short lifetime and the number of limitations on their effects, it's not something that's often seen.

[Compellor]

QUOTE (Jimen @ Nov 19 2009, 05:40 PM) Factoid's analysis explicitly states

Yes, yes it does. I'm disagreeing with it because I have yet to see any evidence of it. If I can make a missile that will drift back and forth over the center of a custom chaff cloud for thirty seconds, and will continue to do damage if it hits me, that very much suggests that either factoid is wrong, or that he is right for normal missiles/chaffs and I have created an edge case. The only thing I'm still wondering about is what Brood brought up, which is that sometimes it seems like you get hit by a missile, and then the server changes its mind or something, and I have no idea whether that's related to chaff at all. Also, the fact that sometimes I see a explosion animation when the missile passes through the chaff is interesting, but it doesn't seem to affect the missile's capacity to continue drifting back and forth across the center of the chaff cloud.

[Broodwich]

the same thing happens with bullets, especially with high pings for me to eu and recently with solap

[TurkeyXIII]

QUOTE (TurkeyXIII @ Nov 18 2009, 05:42 PM) change .cvh format to include a chaff dispenser, both of which are beyond me

I've made a liar of myself







I figured out howcopy-pasted from the cockpit code to take a cvh joint labeled "chaff" and dispense from that position. It still travels 'upwards' at 5 m/s. Ships without a the joint dispense from the center of the model as per R5.

For the two ships I was testing with I edited the .cvh manually in notepad. The Attack Carrier dispenses from 4 'units' to the left, while the dest dispenses from 0.9 'units' upwards.
I don't know how to use modeling software to create cvhs so somewhere down the line, somebody's going to have to make sure the spawn points it uses actually correspond to the location of the joints, and I haven't just fluked this result.
Also there's a comment in missileigc.h:
QUOTE //Assume m_tImpact does not change (not unreasonable given that the only reason to call SetTarget
//is when the missile is spoofed and then the chaff is on top of the target.[/quote]I'm not sure what it's referring to here, but if the chaff is not on top of the target when SetTarget is called, is that going to make the missile behave funky? I have no idea.

[Compellor]

I'm not much of a coder, but since m_tImpact is the estimated time for the missile to impact with the target, and it is used in the equation for determining which direction the missile should turn, then what's going to happen is that on the next update cycle the missile is going to calculate the new time to impact based on the relative position of the chaff, the relative velocity of the chaff, the missile's acceleration, and the old time to impact the original target ship. The missile then decides which way to turn based in part on the new (incorrect?) time to impact. The time to impact is then decreased by the amount of time that has passed since the last update cycle (or since the missile activated, if that's more recent). So whether there's a problem depends on the function which takes in the old time to impact and outputs a new time to impact.

I don't have the inclination or even enough background to go through the "solve" function to determine what exactly the effect would be, but I suspect the error would diminish with each update or even disappear immediately, since I think the old time to impact is mostly just used as a starting point, to minimize the number of calculations necessary each cycle.

If the chaff originates at the center of the target, then when SetTarget is called the relative position of the target doesn't change immediately. If the chaff originates 200 meters closer or farther away, then that could affect things greatly... best way to find out is to test. If it turns out that there's a problem, then I guess you'll have to add some time to impact calculations to missileigc.h... good luck figuring out how to do those.

Also, without being a modeler, you can still determine if the chaff is dispensing from where it's supposed to. Just copy the joint location from a joint with an obvious position, like a weapon mount.

[TurkeyXIII]

Testing with missiles that live long enough to alt-tab between clients reveals that a displaced cm spawn doesn't seem to bother the missile behavior at all. It'll still turn plenty sharp when its new target appears. 'yp! I wasn't looking forward to sifting through those solve functions.
Some missiles with lower turn rates will still hit the ship even with a greatly displaced CM position, only because the missile can't maneuver enough to avoid such a big target. This is exactly how I'd expect it to work, so I ain't changing it.

Copy-pasting a weapon joint and renaming it as chaff results in the bullets and counters coming from the same point. Looks like I was all worried over nothing.

Here's the patch file, if anyone wants to point out the flaws in my coding style.