Radar guided AAA (Oerlikon, Bofors, Skyguard, Flycatcher)

[LaBerthier]

Could we get a system like the Oerlikon GDF in combination with the Skyguard fire control system? Or the Flycatcher system with Bofors 40L70 guns? This would be absolutely fantastic!

https://en.wikipedia.org/wiki/Oerlikon_GDF

https://en.wikipedia.org/wiki/Flycatcher_(radar)

https://en.wikipedia.org/wiki/Bofors_40_mm_Automatic_Gun_L/70

Edited July 5, 2022 by LaBerthier

[Northstar98]

+9999999999

The Oerlikon GDF-002 would be perfect for the 1982 Falklands War, alongside the Skyguard system.

[LaBerthier]

1 hour ago, Northstar98 said:

+9999999999

The Oerlikon GDF-002 would be perfect for the 1982 Falklands War, alongside the Skyguard system.

I agree. That would be great!

[upyr1]

Flak 

[Andrew8604]

Towed KS-19 100mm and S-60 57mm AAA mounts linked with SON 9 "Fire Can" radar direction.  Vietnam era.

 

[Silver_Dragon]

Flak 

Flak was added on WW2 assets pack, and someone was added to the core. Ks19 and radar FC radar has planned by ED, but not have release date.

Enviado desde mi RNE-L21 mediante Tapatalk

[upyr1]

I was using flak in the generic sense 

[LaBerthier]

On 7/6/2022 at 3:40 AM, Andrew8604 said:

Towed KS-19 100mm and S-60 57mm AAA mounts linked with SON 9 "Fire Can" radar direction.  Vietnam era.

 

Would be nice to have something for both Red and Blue. Maybe the Bofors would be a nice start since the 40L60 is already in the game (if i'm not mistaken). Anyway radar guided AAA would be really cool and fill a void in the game...

[SUBS17]

14 hours ago, LaBerthier said:

Would be nice to have something for both Red and Blue. Maybe the Bofors would be a nice start since the 40L60 is already in the game (if i'm not mistaken). Anyway radar guided AAA would be really cool and fill a void in the game...

The latest RADAR GUIDED AMMUNITION never misses not a single bullet!

[Andrew8604]

18 hours ago, SUBS17 said:

The latest RADAR GUIDED AMMUNITION never misses not a single bullet!

I guess you're saying AAA is near perfect in performance in DCS, when in reality it wouldn't seem to be so.  I would agree with that.  Even optically aimed, 12.7mm or 14.5mm machine guns on the top of a turret of a T-55 tank in DCS seems to always be aimed with the exact amount of lead and elevation to hit any aircraft within range that's on a steady flight path...even looking into the sun or being rattled by nearby exploding munitions.  That does NOT seem very realistic.  When you look at the tracers coming at you from one of these, it is a nice, clean arc right into you.  Only if you are changing your flight path at the time do they miss.  Seems like in reality, that stream of tracers should be quite dispersed and mostly off target.  I have hand-aimed the M2, 50-cal machine gun mounted on a HMMWV in DCS and fired several thousand rounds at low-level aircraft making strafing runs nearby and have never been able to put even one round on an aircraft.  Maybe I just suck at gunnery, or maybe that's reality.  Or do the machine guns on the T-55 tank have precision, highly accurate range-finding, lead-computing sights?

It seems like AI in DCS has no "human error" and electro-mechanical imperfection/lack-of-precision factored into it.  When a Mk-82 detonates on the ground 100 meters away, and you have your upper body exposed, trying to aim and controllably fire a 12.7mm machine gun at a passing aircraft, do you think you might at least flinch a little and spoil your aim, if not be knocked aside by the pressure wave and forced to stop firing...and maybe not recover to fire any more for several seconds, if not minutes?  If doing the same thing and then cluster munitions start popping off 100 meters away, are you going to hold your concentration and sustain accurate lead and elevation as you fire the machine gun at an aircraft?  I think the reality is that 99% of people would stop firing and duck and cover.

I think this needs to be factored into AI air defense.  This would allow for effective air defense suppression.  You don't have to kill them, just make them duck for a moment while your fellow pilots make their bomb runs.

Now, radar-directed guns would be less likely to be "suppressed", unless the gun mounts are open and being hand loaded...in which case they should be forced to cease fire or reduce firing rate for a period of time.

But how to "code" this human error and mechanical error in DCS without bogging down the computing power...  Maybe the guns could have two levels of accuracy.  One for normal operation, and a greatly reduced accuracy for when munitions are hitting the ground in an effective vicinity.  And then a 3rd state where operation is suspended when damage or near damage is being taken by the gunners.

Also, could there be a diagram released illustrating the "as simulated" blast and shrapnel effects range rings of various munitions in DCS.  I'm not sure they're all realistic or consistent.  For instance, I recently dropped M117, 750-lb GP bombs from the F-86F (latest DCS release version) and they left a "visual" crater that looked more like a small BBQ fire pit than a bomb crater.  A single small "crater" from two bombs that hit together.  I do realize visual effects and simulated impact on ground assets are two different things.  But sometimes it seems like you can do more damage to your own plane with a dropped bomb than to an unarmored vehicle that's 10 times closer to the blast.

[upyr1]

4 hours ago, Andrew8604 said:

I guess you're saying AAA is near perfect in performance in DCS, when in reality it wouldn't seem to be so.  I would agree with that.  Even optically aimed, 12.7mm or 14.5mm machine guns on the top of a turret of a T-55 tank in DCS seems to always be aimed with the exact amount of lead and elevation to hit any aircraft within range that's on a steady flight path...even looking into the sun or being rattled by nearby exploding munitions.  That does NOT seem very realistic.  When you look at the tracers coming at you from one of these, it is a nice, clean arc right into you.  Only if you are changing your flight path at the time do they miss.  Seems like in reality, that stream of tracers should be quite dispersed and mostly off target.  I have hand-aimed the M2, 50-cal machine gun mounted on a HMMWV in DCS and fired several thousand rounds at low-level aircraft making strafing runs nearby and have never been able to put even one round on an aircraft.  Maybe I just suck at gunnery, or maybe that's reality.  Or do the machine guns on the T-55 tank have precision, highly accurate range-finding, lead-computing sights?

It seems like AI in DCS has no "human error" and electro-mechanical imperfection/lack-of-precision factored into it.  When a Mk-82 detonates on the ground 100 meters away, and you have your upper body exposed, trying to aim and controllably fire a 12.7mm machine gun at a passing aircraft, do you think you might at least flinch a little and spoil your aim, if not be knocked aside by the pressure wave and forced to stop firing...and maybe not recover to fire any more for several seconds, if not minutes?  If doing the same thing and then cluster munitions start popping off 100 meters away, are you going to hold your concentration and sustain accurate lead and elevation as you fire the machine gun at an aircraft?  I think the reality is that 99% of people would stop firing and duck and cover.

 

The human element is the hardest thing to simulate. You need to be able to factor in both Dori Miller and the guy that breaks down. Though tables for accuracy could help make things make guns less accurate and more realistic 

[okopanja]

Actually, natural match for Bofors guns is swedish radar Giraffe.

https://en.m.wikipedia.org/wiki/Giraffe_radar

 

[LaBerthier]

On 7/10/2022 at 2:35 AM, okopanja said:

Actually, natural match for Bofors guns is swedish radar Giraffe.

https://en.m.wikipedia.org/wiki/Giraffe_radar

 

I see your point. I have to say I don't like its long Giraffe neck. I think the flycatcher looks better. Now, in the real world that probably wouldn't be a really important argument, but since this is a game.......

https://nl.wikipedia.org/wiki/Flycatcher_(radar)

[okopanja]

1 hour ago, LaBerthier said:

I see your point. I have to say I don't like its long Giraffe neck. I think the flycatcher looks better. Now, in the real world that probably wouldn't be a really important argument, but since this is a game.......

https://nl.wikipedia.org/wiki/Flycatcher_(radar)

That neck offers through simple physics better detection of low flying objects. This is especially important on longer distances and with uneven terrain.

Just consider the countries of origin and their terrain.

Edited July 16, 2022 by okopanja

[SUBS17]

9 hours ago, okopanja said:

That neck offers through simple physics better detection of low flying objects. This is especially important on longer distances and with uneven terrain.

Just consider the countries of origin and their terrain. 

 

 

The latest guns Automatically level and adjust when on uneven terrain. They would calculate lead and auto aim automatically and fire bursts. It would not miss and the ammunition the latest type would detonate precisely for maximum damage. 1 gun 1 round 1 kill if the aircraft is within its WEZ!(WEAPON EFFECTIVENESS ZONE)  Tungaska and Shika have this. The guns mentioned in the title all have the same computer for that purpose as the Shilka and Tungaska and have the same capability. If it is manual mode then you would have error but that would be against how such weapon systems normally operate. On the move they all have the same capability 1 round 1 kill.(NO ESCAPE ZONE) 

[LaBerthier]

On 7/16/2022 at 3:45 PM, okopanja said:

That neck offers through simple physics better detection of low flying objects. This is especially important on longer distances and with uneven terrain.

Just consider the countries of origin and their terrain.

 

I understand its function. It's just ugly (Not an important argument in real life of course!) But I would be happy to have it in the game, none the less.

[Bozon]

On 7/10/2022 at 3:25 AM, upyr1 said:

The human element is the hardest thing to simulate. You need to be able to factor in both Dori Miller and the guy that breaks down. Though tables for accuracy could help make things make guns less accurate and more realistic 

The biggest error for human gunnery is most likely the range. The error on your estimation of range increase with the range - this quadratically affects the lead error against targets that do not fly directly towards you. Humans are also particularly weak at adjusting to a non constant range when the target has a velocity component towards/away from you.

There are other errors than increase with range such as bullet drop estimation and errors stemming from general angular accuracy, so the end result is that manually aimed AAA become statistical weapons pretty quickly with range - it is actually worse than a pure statistical errors because range estimation errors lead to bias not to scatter. This is unlike the WWII AAA in DCS that are able to snipe me out of the sky no matter the aspect and a weak function of range.

Radar guided weapons on the other hand solve the range pretty accurately and with an error that is largely independent of the distance, but generally have poor angular accuracy (vs optic aiming for example) - these are just the properties of radars. Angular aiming error create a miss distance that increase just linearly with distance. This is closer to a pure random aiming error around the true perfect solution.

Games usually calculate a perfect solution and then add a random angular error to simulate the inaccuracy - this is much more similar to the errors of radar guided AAA than to human gunners as I explained above and why in most games AI gunners are super humans. To simulate a human, a range-dependent error that is re-sampled at a low rate needs to be included in addition to angular errors.

[upyr1]

7 hours ago, Bozon said:

The biggest error for human gunnery is most likely the range. The error on your estimation of range increase with the range - this quadratically affects the lead error against targets that do not fly directly towards you. Humans are also particularly weak at adjusting to a non constant range when the target has a velocity component towards/away from you.

There are other errors than increase with range such as bullet drop estimation and errors stemming from general angular accuracy, so the end result is that manually aimed AAA become statistical weapons pretty quickly with range - it is actually worse than a pure statistical errors because range estimation errors lead to bias not to scatter. This is unlike the WWII AAA in DCS that are able to snipe me out of the sky no matter the aspect and a weak function of range.

Radar guided weapons on the other hand solve the range pretty accurately and with an error that is largely independent of the distance, but generally have poor angular accuracy (vs optic aiming for example) - these are just the properties of radars. Angular aiming error create a miss distance that increase just linearly with distance. This is closer to a pure random aiming error around the true perfect solution.

Games usually calculate a perfect solution and then add a random angular error to simulate the inaccuracy - this is much more similar to the errors of radar guided AAA than to human gunners as I explained above and why in most games AI gunners are super humans. To simulate a human, a range-dependent error that is re-sampled at a low rate needs to be included in addition to angular errors.

Not only do you need to figure out realistic errors, but you also need to factor in fear. A flak suppression run would be successful if the AA crews decide to take cover giving the attackers a window of time where they are not being shot at. 

[LaBerthier]

Any chance we can get an update from the devs on this request?

[LaBerthier]

Check out this site for some cool products that could be modelled:

https://rheinmetall-defence.com/en/rheinmetall_defence/systems_and_products/air_defence_systems/stationary_air_defence/index.php

[LaBerthier]

I saw there is already a Gepard that uses the right 35mm cannons. So modeling a system based on that cannon should be possible.

[LaBerthier]

I saw that Falcon 4.0 BMS has an Oerlikon GDF in combination with the Skyguard fire control system. So it should be possible. Would be really cool!