Project Reality Forums

Posted: **2015-12-08 04:23**

Back in Bf2 there was an extremely irritating glitch with jets where if someone is behind you locking on with Air-Air missiles if you turn and feather the reverse key the missiles would constantly overshoot. Like they would reach your planes engines then keep going rather than detonating, this was extremely easy thing to do with Delta-wing jets like the J10 and Eurofighter in particular.

Anyway today on Hades Peak I saw this in PR for the first time, I fired 4 consecutive missiles from a Su 27 both types of missile, and without any flares the Eurofighter just kept turning and the missiles passed his engines every time.

If this is the same exploit then it should definitely be looked into.

Posted: **2015-12-08 15:17**

Brozef wrote:-snip-

Yeah, it's pretty easy to exploit. I find that the IR missiles tend to impact at the edges of the wings, but since the profile of the Euro\J-10 are quite small and that they are agile, the combination of the two probably end up in more dodges than say a gigantic Su-30.

Posted: **2015-12-08 16:09**

You are saying you shot 4 missiles, so i guess you were using AA-10. Since you have 4 of them and additionally 2 AA-11.

AA-10 are long range missiles, if you were very close to him in a turn fight, its very well possible that the missiles overshot, due to their low turn radius. The AA-11 have much tighter turn angle for these purposes.

Posted: **2015-12-08 17:05**

I said both types, I shot 2 long range ones switched to short range and fired both. I'm telling you the same thing that happened in Bf2 happened here the missiles reached his engines and just kept going.

Posted: **2015-12-08 18:22**

i can confirm that i was piloting a J-10 doing a turn fight with a SU-27 in black gold he fired 2 missiles and i watch on the back camera that i was clearly hit by two missiles, one in the engine and one in the wing but did no damage and i could keep turn and run away.

maybe this is also related to the hit bug?

Posted: **2015-12-08 19:44**

The problem might be in two points.

First is overcomplicated and sophisticated problem of old engine and pr pushing it out of limits.
During tweaking jets physics and analyzing them via external tool i were sending raw data each server frame, which is somewhere around 0.03-0.04s on ~35-36 maximum server fps.

Position : (-356.0014953613281, 343.4421691894531, -808.2207641601562)
Rotation : (-87.0715103149414, -21.12311553955078, -170.7117156982422)
Timestamp : 868.883181307
Delta : 0.0339969496496

Position : (-350.5141906738281, 341.3143005371094, -808.541931152343
Rotation : (-87.07022094726562, -21.340808868408203, -170.6194610595703)
Timestamp : 868.849184357
Delta : 0.0551704688954

Delta between those two frame updates gives us vector distance of ~6m per server frame.
Now lets go into missile stuff.
I didn't measured their speeds directly in PR, but rather approximated their delta based on vbf2 values, and in PR we have missiles flying 3-6 times faster than jets, so their delta per server frame would be around 18-36 meters.
As some experienced modders already know, missiles ignores physics when being locked, only things they actually consider after lock is acceleration and speed on z-axis.

Now there's some complicated things come into leading. Bf2 uses tail-chase method, completely ineffective and non-realistic, but most simplier for programming sumulation. You simply calculating each game frame when you need to aim, and then give command to missile to aim at that point, that's where yaw and pitch limit values(like max deg per sec and so on) comes in play, but that's secondary. The major problem of this method is that you have to 'cheat' to make missiles hit by giving them insane acceleration and speed values.

Now there's tracking and proximility issues come in play.
As example, i'll be using AA-11\R-73.
Pretty much all the AA missiles in vbf2 and PR aswell has track time of 0.05s(for performance purposes), with approximated time to solve the server frame it would we like 0.06s - the double of server frame.
With the proximility radius of detonate of 35m it's giving us a hemisphere facing plane of around 17-18m.
Remember i talked about missiles speeds? As their delta move being around 18-36m, and with doubled time of track it simply could be that missiles just bypassed jet while solving their positions.

Another solution could be that you simply fired too close, R-73\AA-11 has arming time of 1.3 secs

Posted: **2015-12-09 02:07**

rPoXoTauJIo wrote:The problem might be in two points.

First is overcomplicated and sophisticated problem of old engine and pr pushing it out of limits.
During tweaking jets physics and analyzing them via external tool i were sending raw data each server frame, which is somewhere around 0.03-0.04s on ~35-36 maximum server fps.

Delta between those two frame updates gives us vector distance of ~6m per server frame.
Now lets go into missile stuff.
I didn't measured their speeds directly in PR, but rather approximated their delta based on vbf2 values, and in PR we have missiles flying 3-6 times faster than jets, so their delta per server frame would be around 18-36 meters.
As some experienced modders already know, missiles ignores physics when being locked, only things they actually consider after lock is acceleration and speed on z-axis.

Now there's some complicated things come into leading. Bf2 uses tail-chase method, completely ineffective and non-realistic, but most simplier for programming sumulation. You simply calculating each game frame when you need to aim, and then give command to missile to aim at that point, that's where yaw and pitch limit values(like max deg per sec and so on) comes in play, but that's secondary. The major problem of this method is that you have to 'cheat' to make missiles hit by giving them insane acceleration and speed values.

Now there's tracking and proximility issues come in play.
As example, i'll be using AA-11\R-73.
Pretty much all the AA missiles in vbf2 and PR aswell has track time of 0.05s(for performance purposes), with approximated time to solve the server frame it would we like 0.06s - the double of server frame.
With the proximility radius of detonate of 35m it's giving us a hemisphere facing plane of around 17-18m.
Remember i talked about missiles speeds? As their delta move being around 18-36m, and with doubled time of track it simply could be that missiles just bypassed jet while solving their positions.

Another solution could be that you simply fired too close, R-73\AA-11 has arming time of 1.3 secs

how about the actual hit and explosion not resulting in a kill/damage ?

Posted: **2015-12-09 06:25**

Totally could be.
R-73\AA-11 for some reason has no damage besides explosion one, and while you're seeing hit AND explosion due to interpolation, the server just registering missile collide BEFORE explosion, leading to no or tiny damage from collision mesh.

But anyway, i have yet seen clear video\demo(1.3.5

) nor specific steps to reproduce in 100% cases.

Project Reality Forums

Eurofighter dodges missiles

Eurofighter dodges missiles

Re: Eurofighter dodges missiles

Re: Eurofighter dodges missiles

Re: Eurofighter dodges missiles

Re: Eurofighter dodges missiles

Re: Eurofighter dodges missiles

Re: Eurofighter dodges missiles

Re: Eurofighter dodges missiles