Jets in 1.4

[rPoXoTauJIo]

I might be wrong but don't the lazes only last for 10 seconds before requiring a refresh? That may have been the logic behind it...

24 seconds actually.

[DogACTUAL]

Some critique about stuff in the flight model that is annoying. I guess there is a good chance you already are working on these things.

Harriers take much more space to take of horizontaly than regular jets, even though they are suposed to at least have a short take off capability compared to them. Regular jets on the other hand should not be able to take off as quick as right now. Same for landing regular jets, they can land so quickly that you can land most of them on the amphipious assault ship, although it is difficult.

Su-22 and Su-25 flight model are too sluggish, landing Su-25 is very strange, it tends to bounce up in the air again/ doesn't wanna stay on the runway when landing with reasonable speed. You can also land it in extremely short space by just going really slow and letting it bounce down on the runway. Su-22 has too much drag imo. Both jets could use a speed increase, maybe A10 too.

New flight model is a big improvement upon the old one of course, just some stuff that needs to be ironed out!

[DogACTUAL]

Any way you could give Su-25 and maybe A10 a speed boost? Su-25 especially looks very cartoonish on a flyby with how slow it is.

[winject]

@devs any idea on how to replicate this bug?
https://www.youtube.com/watch?v=ciqx14P ... u.be&t=55s

[FFG]

@devs any idea on how to replicate this bug?
https://www.youtube.com/watch?v=ciqx14P ... u.be&t=55s

It was a redirect, it happens.

[AfterDune]

I once saw a video about a jet that got killed by a friendly missile that missed its target quite far away. Couldn't find it right now, but it's out there.

These situations are really rare, but it can happen.

[AlonTavor]

What I think happens is the missile collides with the jet, killing it, clients think the missile died, but the missile continues.

[winject]

It was a redirect, it happens.

yes I know we call it a "redirect" bug but we need to know what is causing it.

What I think happens is the missile collides with the jet, killing it, clients think the missile died, but the missile continues.

sync issue?

somehow this bug only occurs on PR, never saw it live on BF2 so I start to believe it has something to do either with the aa missile proprieties or the flares themselves.
maybe is it related to the collision group / material id of those objects?

I saw this comment on some v1.2 video


Source : https://www.youtube.com/watch?v=nDvo3pQ ... e=youtu.be

seems like the aa missile material used in PR are hardcoded by default to bounce on the flare material.

maybe playing with the following template settings could give another result :

Code: Select all

// Extracted from aim9.tweak

ObjectTemplate.hasMobilePhysics 1
ObjectTemplate.hasCollisionPhysics 1
ObjectTemplate.physicsType Point
ObjectTemplate.collisionGroups 8

the reason why I asked this is because I need as much intel as possible so I can get a chance to reproduce it on a local server, I'll see what I come up with but I need confirmation if these are the settings involved

[Suchar]

I once saw a video about a jet that got killed by a friendly missile that missed its target quite far away. Couldn't find it right now, but it's out there.

These situations are really rare, but it can happen.

[LimitJK]

i dont think it was even a bug. look at the minimap, jet and hind are really close to each other. AA has insane splash so it probably hit both.

[FFG]

What I think happens is the missile collides with the jet, killing it, clients think the missile died, but the missile continues.

no. The first missile was the one that killed the Hind. not the second one.

[Hunter291]

Working as intended

[AlonTavor]

no. The first missile was the one that killed the Hind. not the second one.

Check the map. It would have had to do a 150 degrees turn for that. If it did, that's a bug of its own.

[AlonTavor]

yes I know we call it a "redirect" bug but we need to know what is causing it.



sync issue?

somehow this bug only occurs on PR, never saw it live on BF2 so I start to believe it has something to do either with the aa missile proprieties or the flares themselves.
maybe is it related to the collision group / material id of those objects?

I saw this comment on some v1.2 video


Source : https://www.youtube.com/watch?v=nDvo3pQ ... e=youtu.be

seems like the aa missile material used in PR are hardcoded by default to bounce on the flare material.

maybe playing with the following template settings could give another result :

Code: Select all

// Extracted from aim9.tweak

ObjectTemplate.hasMobilePhysics 1
ObjectTemplate.hasCollisionPhysics 1
ObjectTemplate.physicsType Point
ObjectTemplate.collisionGroups 8

the reason why I asked this is because I need as much intel as possible so I can get a chance to reproduce it on a local server, I'll see what I come up with but I need confirmation if these are the settings involved

We have edited the detonateDistanceToTarget assembly. Vanilla used to check every 0.25s. we have changed it to check every tick. There might be some really weird side effects due to this, but I doubt. I don't remember if we also updated clients assembly, as this code doesn't run on clients. I'll check later.

[CAS_ual_TY]

Fix only 1 of these and the survivability is either completely OP or UP, fix both: Profit!

I also believe its a Client-Server thing. It kinda has to be

[winject]

We have edited the detonateDistanceToTarget assembly. Vanilla used to check every 0.25s. we have changed it to check every tick. There might be some really weird side effects due to this, but I doubt. I don't remember if we also updated clients assembly, as this code doesn't run on clients. I'll check later.

thanks, that is what I wanted to hear.
I don't believe either that this is a side effect to it but I think it is rather an unexpected effect from the object templates or the behaviour of the LPTarget component itself.

that assembly change was also applied in v1.4 ifrc but this issue was already up in v1.2 so it can't be assembly-related.

forgot to say as well that the redirect bug also occurs with 1 missile killing the initial locked target and immediately switching to another one and killing it as well. Strange that it doesn't go for a third target..

I also experienced the same problem when I was locking an enemy cow very far away with the us manpad and suddenly the missile headed straight for a nearby blackhawk and damaged it, can't recall whether the blackhawk was in the same path though.

the only way to really understand what is going on would be to record the missiles/flares trajectory and its behaviour by reversing it in the memory and drawing with directx, for example, the 3D position and the current target retrieved from the LPTargetComponent.

i'm going to run a few tests locally and see where this leads me.
not sure though if the local results will be accurate since dedicated servers behave slightly differently.

Edit :

Any way you could give Su-25 and maybe A10 a speed boost? Su-25 especially looks very cartoonish on a flyby with how slow it is.

If you could update this for 1.5.6

[AlonTavor]

Querying for all GenericProjectile objects every tick and doing getPosition on them in Python is very reliable.

As for "current target" you'll need to reverse engineer the projectile struct and find the comp there.

[Mats391]

As Alon already mentioned, redirect were and are an issue of bf2 checking AA missiles too rarely. We tried to fix that by making it check more often and right now it looks like the missile does explode more often on flares thanks to our changes. However the explosion does not seem to kill the missile, so it goes on and finds next target. We are still looking into it, but it is kind of a mess.

[DogACTUAL]

A long time ago i wrote Fastjack a PM asking him about his approach and ideas and he wrote me back in some detail.

To paraphrase his ideas:

1) explosion.coneAngle for AA missiles. Have AA missiles only deal damage towards a certain limited angle when exploding, like claymores do, so good manouvering can defeat them.

2)Change flares from being a Heat_Object to being an object with an invisible sphere that has ProjectileCollision. So flares would work less as decoys to misdirect the missile and more like obstacles in the way of the missile that have the potential to 'block it' from hitting the aircraft. Putting flares in the right spot in the path of the missile would defeat it.

Would both or one of these be feasible to do, to mitigate the current problems with AA?
And to make the AA system more fair, interesting and skill based for both parties (AA and pilot) and less 'random' and frustrating?

[CAS_ual_TY]

A long time ago i wrote Fastjack a PM asking him about his approach and ideas and he wrote me back in some detail.

To paraphrase his ideas:

1) explosion.coneAngle for AA missiles. Have AA missiles only deal damage towards a certain limited angle when exploding, like claymores do, so good manouvering can defeat them.

2)Change flares from being a Heat_Object to being an object with an invisible sphere that has ProjectileCollision. So flares would work less as decoys to misdirect the missile and more like obstacles in the way of the missile that have the potential to 'block it' from hitting the aircraft. Putting flares in the right spot in the path of the missile would defeat it.

Would both or one of these be feasible to do, to mitigate the current problems with AA?
And to make the AA system more fair, interesting and skill based for both parties (AA and pilot) and less 'random' and frustrating?

Can we get an official response to this? Especially the 2nd point sounds interesting.