Do bullet holes in AI planes affect the flight model?

[bephanten]

damaged ai anton, with bullet holes in the fuselage and canopy continues to fly with full throttle evasive maneuvers. i cant see a difference in behavior between an intact plane and swiss cheese plane. if the control surfaces are fine, no fluid leaking and no fire, it flies fine even if the plane has lots of holes in it. 

there is a whole lot of undefined area between damage model and flight model. and that, it hurts. like a damage model. but for players. single players.

 

[Gunfreak]

17 hours ago, bephanten said:

damaged ai anton, with bullet holes in the fuselage and canopy continues to fly with full throttle evasive maneuvers. i cant see a difference in behavior between an intact plane and swiss cheese plane. if the control surfaces are fine, no fluid leaking and no fire, it flies fine even if the plane has lots of holes in it. 

there is a whole lot of undefined area between damage model and flight model. and that, it hurts. like a damage model. but for players. single players.

 

If there is an effect. It's very very small.

[Cool-Hand]

I've noticed the same thing, I think the bullet holes are just for decoration.

[PawlaczGMD]

19 hours ago, bephanten said:

damaged ai anton, with bullet holes in the fuselage and canopy continues to fly with full throttle evasive maneuvers. i cant see a difference in behavior between an intact plane and swiss cheese plane. if the control surfaces are fine, no fluid leaking and no fire, it flies fine even if the plane has lots of holes in it. 

there is a whole lot of undefined area between damage model and flight model. and that, it hurts. like a damage model. but for players. single players.

 

This is true, but the Anton is very tough anyways. You need to hit the engine, pilot, or hope to start a fire. Firing low caliber machine guns into the wings or tail from 6 o clock is not a great way to take them down. Lead more in the turn to hit the front of the plane, or attack from below and hit below the engine. Good thing is the Anton is probably the worst-performance dogfighter we have.

[peachmonkey]

about 6 months ago I remember seeing nineline's comment either on reddit or in this forum, something about ED working on the next level of features for ww2, which is wing ammo explosions and the turbulence from the holes in the wings...  So, it sort of told me that currently we don't have any.

[Gunfreak]

14 hours ago, PawlaczGMD said:

This is true, but the Anton is very tough anyways. You need to hit the engine, pilot, or hope to start a fire. Firing low caliber machine guns into the wings or tail from 6 o clock is not a great way to take them down. Lead more in the turn to hit the front of the plane, or attack from below and hit below the engine. Good thing is the Anton is probably the worst-performance dogfighter we have.

In the end you reach a point were there is more air than aircraft left. You would simply have total structural collapse. That's not simulated. Neither is fuel and ammo detonation. Doesn't look like you can shoot away struts either.

[Michal]

I think it is a problem of damage model for AI planes only. Player controlled Anton (or Dora) behave differently with holes in wings, stalls more, turns worse, etc. There is no fuel tank in wings so the damaged wing means usually flat tire or not working flaps mechanism. Damaged fuselage sometimes ends in fuel tank leak or fire. I usually keep 50-100l in rear tank before switching to front tank. It helps in case of front tank leak and it is not fatal if rear tank will catch fire.

[Dragon1-1]

5 hours ago, Gunfreak said:

In the end you reach a point were there is more air than aircraft left. You would simply have total structural collapse. That's not simulated. Neither is fuel and ammo detonation. Doesn't look like you can shoot away struts either.

With .303s? Most of them will deflect off the pilot seat armor or even the skin itself, if they hit at a low angle, and if you hit a strut, forget about doing anything to it. These guns were not especially useful, especially by the end of the war. Many models of the Spitfire (just not the one we have in DCS) replaced the two .303s with a single .50 in each wing, which could do some actual damage. The main use I found for the .303s is checking my aim before firing a 20mm burst (they have tracers and ample ammo supply, the 20mm has neither). The other is for strafing infantry.

[Gunfreak]

46 minutes ago, Dragon1-1 said:

With .303s? Most of them will deflect off the pilot seat armor or even the skin itself, if they hit at a low angle, and if you hit a strut, forget about doing anything to it. These guns were not especially useful, especially by the end of the war. Many models of the Spitfire (just not the one we have in DCS) replaced the two .303s with a single .50 in each wing, which could do some actual damage. The main use I found for the .303s is checking my aim before firing a 20mm burst (they have tracers and ample ammo supply, the 20mm has neither). The other is for strafing infantry.

The dewild .303s combined with AP ammo was decently effective through most of the war. The E wing only showed up in numbers after Normandy invasion. 

But I was talking general. The .50  in the E wing would do very little in DCS given how the damage modeling is. 

All guns in DCS seem to under preform vs historical sources. The 20mm hispano does not seem to do the "frightening " damage it is described to have.  The German and American heavy machine guns are far less effective then they should be. A few good hits would usually take out an aircraft. Remember virtual pilots are on average far better shots then even real ww2 aces. We have infinite amount of hours to practise gunnery in games. While real pilots had just a fraction of that. We also get to fire off without the effects of Gs, cold/warm. And all other physical effects that are negative. Yet in DCS on average it's far harder to kill an enemy aircraft then what myriad of first hand accounts say. 

In the book tail end Charlie. One of the pilots fired a short long range burst with his P47. He got a few hits in the wing of a 109 and the wing just fell of. This is impossible to recreate in DCS, as there simply isn't any damage modeling there.

In a spitfire. Blasting the wing of a German fighters with 20mm ammo would be a perfectly good way of taking our. A few 20mm hits and the wing would either explode from the ammo in the wings going off. Or simply the struts holding the wing in pplace would be destroyed. 20mm Hispanos delivers 50 000 joule in kinetic energy. On top of that the high explosive shells add another 45 000 joules in chemical energy. Compare that to the 3000ish joules of the .303(or other rifle calibres) and the 15 000ish joules of the .50 BMG. In DCS you can fire all 500 rounds of 20mm on a 190 wing and nothing really would happen. In real life that wing would simply not be there anymore.

Yes there are extraordinary tales of planes surviving the most extreme punishment. But those are highly unlikely things that only happened a handful of times throughout the entire war. In DCS that seems to be the default. And quick efficient kills are the exceptions.

[kablamoman]

2 hours ago, Gunfreak said:

as there simply isn't any damage modeling there.

There's no question there's tons of room for improvement with the damage modeling, but ED have shown several times in the forums here that damage to wing struts is in fact modeled, and to quite a high fidelity indeed. Each strut has its own damage value, but it takes quite accurate and sustained fire from .50s to damage them to the point where they actually break. It does account for structural weakness and can fail if they become somewhat weakened by battle damage and then are subsequently put under load.

I wish I could find the forum posts, as they had some really cool screenshots of their internal development tools for debugging the damage modelling. What they have really is pretty impressive, and I wish they would add to and expound upon it a bit more.

Edited August 16, 2024 by kablamoman

[Gunfreak]

12 minutes ago, kablamoman said:

There's no question there's tons of room for improvement with the damage modeling, but ED have shown several times in the forums here that damage to wing struts is in fact modeled, and to quite a high fidelity indeed. Each strut has its own damage value, but it takes quite accurate and sustained fire from .50s to damage them to the point where they actually break. It does account for structural weakness and can fail if they become somewhat weakened by battle damage and then are subsequently put under load.

I wish I could find the forum posts, as they had some really cool screenshots of their internal development tools for debugging the damage modelling. What they have really is pretty impressive, and I wish they would add to and expound upon it a bit more.

 

I know they can break under G after damage. Never seen it ever happen without a lot of G load though. While sources show this to be quite common from relatively few hits. Wings breaking off appears to be just as common as engine damage, fire or explosions. Yet in DCS  it's is as rare as a unicorn. If they are in fact modeled then ED has either made them to powerful or the damage from the projectiles is too low(which appears to be true for the 20mm Hispanos at least) air combat in ww2 was quick and brutal most of the time. In DCS it feel more like battleship flinging ineffective shells at each other.

Much like AAA in DCS. DCS doesn't match the sources (Granted ED has acknowledged the problem with ground AI accuracy) 

I'm sure ED has gone over all the technical manual of weapons and aircraft. But if the results doesn't match primary sources then something is missing from the pure number crunching.

[kablamoman]

28 minutes ago, Gunfreak said:

While sources show this to be quite common from relatively few hits. Wings breaking off appears to be just as common as engine damage, fire or explosions. Yet in DCS  it's is as rare as a unicorn. If they are in fact modeled then ED has either made them to powerful or the damage from the projectiles is too low(which appears to be true for the 20mm Hispanos at least) air combat in ww2 was quick and brutal most of the time. In DCS it feel more like battleship flinging ineffective shells at each other.

Not sure if you missed the debate, but this has been brought up a lot, with 9L often taking issue with the assertion that wings detaching was a common outcome. You can look yourself at all the readily available guncam footage, and I have to say I tend to agree with him, as wings and structural elements are rarely separated unless there's some kind of ammo explosion (there are several guncams featuring FW-190s seemingly hit in the ammo storage, leading to wing-snapping explosions). Most guncams depict damage to oil and rad coolers, lines and hoses, with the resultant streaming fluids, or fuel tank/drop tank explosions and fires. It's relatively rare to see wings snapping -- go ahead and look for yourself.

Where I think they can improve the most would be more dynamic and varied fuel fires (with progressive damage or explosion), ammo explosions, and perhaps a review and QA check of their pilot damage calculations, as it seems really rare for in-game pilots to get shot up even if the cockpit area gets raked thoroughly with incoming fire.

Edited August 16, 2024 by kablamoman

[Dragon1-1]

3 hours ago, Gunfreak said:

But I was talking general. The .50  in the E wing would do very little in DCS given how the damage modeling is. 

On the P-51, they do plenty of damage. Yes, there are four more of them, but either way it doesn't take a whole lot of those bullets to make the target completely unfit to fight. It's also important to recognize when it happens and stop pumping bullets into it.

It does not immediately cause the plane to turn into a fireball, but that's not how most WWII kills went down, and neither did ones from Hispano cannon. Most of the time, you've got the bandit leaking all sorts of fluids, and he'd crash some time after, on landing or otherwise. That's the "terrifying" level of damage the 20mm does, a big hole torn out in the wing or in the fuselage. Unless you hit a control surface linkage, you won't see any drastic effect, because by itself, it doesn't cause a huge explosion. There were fuel fires and you could set off ammo explosions, but the vast majority went down without fireworks.

[Gunfreak]

14 minutes ago, Dragon1-1 said:

On the P-51, they do plenty of damage. Yes, there are four more of them, but either way it doesn't take a whole lot of those bullets to make the target completely unfit to fight. It's also important to recognize when it happens and stop pumping bullets into it.

It does not immediately cause the plane to turn into a fireball, but that's not how most WWII kills went down, and neither did ones from Hispano cannon. Most of the time, you've got the bandit leaking all sorts of fluids, and he'd crash some time after, on landing or otherwise. That's the "terrifying" level of damage the 20mm does, a big hole torn out in the wing or in the fuselage. Unless you hit a control surface linkage, you won't see any drastic effect, because by itself, it doesn't cause a huge explosion. There were fuel fires and you could set off ammo explosions, but the vast majority went down without fireworks.

Expansion were far more common than what happened to Bob Johnsons P47. 

In DCS Bob Johnson is the default. And explosions are impossible.

A few 20mm in the wing would often detonate the ammo, of goes half the wing and down the plane goes. A few 20mm or a good burst of .50 cal and they would dig into the aircraft and detonate the fuel tank. 

In the book clean sweep. You get descriptions of hundreds of engagements from the 8th fighter command. From 42 to 45. 

The number of times those .50 explode a 190 or 109 is quite common. 

Remember we are talking enormous amounts of energy from these projectiles are relatively small aircraft.

10 20mm Hispanos rounds has the same energy as 1 stick of dynamite.  Or the energy of a 1 ton car slamming into a wall at 100kmh. 

1 second burst of .50 cal from a p47 is 1.5 sticks of dynamite (in pure kinetic engey, not taking chemical energy from the incendiary rounds into account. ) 

Fighters aren't tanks. Saying the A8 is well armored doesn't mean it can shrug off millions of joules in energy deposited into it. 

I've been doing  a series that test the AI of the warbirds in a medium engagement and the silly amount of damage the A8 take is just wrong. You'll have 10 A8s riddled with holes and still fighting like nothing has happened. You'll have A8s flying around that are so damaged the aircraft would literally just have dissolved as they are 90% air at that point. Taken hundreds upon hundred of .50 cals. 

Can it happen? Once in a million perhaps, like with Bob Johnsons. In DCS that 1 in a million is the norm not the exception. 

Saying stuff like hitting from perfect 6 o'clock isn't effective. Because you have to hit the pilot or engine. Just isn't true, firing  from a perfect 6 o'clock in real life is perfectly(and the most common way ww2 aircraft got shot down) If you hit the back of the aircraft you'll chew off the tail in a fraction of a second or your bullets will dig into the aircraft and explode the fuel tank. Or if you hit the wings. They'll usually fall of after a few hits and they really aren't made to handle that kidna energy delivered into then.

Unlike human casualties were you roughly can count 1 dead for every 3 wounded. In air combat between fighters. Destroyed will often outnumber damaged. Because once a burst lands. That aircraft is often out of combat.

If you really whats under the red box.

I don't think anyone will say the 20mm on the Spitfire has an effect on aircraft that can be described as "frightening "

There are other sims were indeed the effect of 20mm hispano could be described as frightening.

In DCS you can fire and fire of 20mm and unless you hit some very specific places on the aircraft. You can just fire of all 500 and have little effect.

 

[kablamoman]

54 minutes ago, Gunfreak said:

Saying stuff like hitting from perfect 6 o'clock isn't effective. Because you have to hit the pilot or engine. Just isn't true, firing  from a perfect 6 o'clock in real life is perfectly(and the most common way ww2 aircraft got shot down)

Watch the video at timestamp.

[Dragon1-1]

4 hours ago, Gunfreak said:

In DCS Bob Johnson is the default. And explosions are impossible.

It's been already stated that ED is going to add ammo detonations eventually. This was indeed a serious problem, but it required hitting the wing in the specific place, in case of the 109 it was the right wing root, FW-190 had ammo boxes for wing guns, making this easier. Of course, Allied pilots would know to aim for those bits, so I guess it wasn't that rare, but remember that many engagements ended with the bandit bugging out and not being recorded as kill at all, at least until they captured the airfield and found the plane in question augured into a field not far from where the fight took place (or even crashed trying to land because pilot didn't realize he was missing a tire).

Did you verify those "hundreds of .50cal holes" in the post-mission log? Because that's very much not my experience. Sure, it happened that I was shooting at the bandit all day and couldn't kill him, but it usually meant I wasn't hitting him a whole lot. The guns are harmonized to converge at a specific distance, and if you're off that distance, you won't be hitting much.  Especially when shooting from directly behind, it's very easy to get too close and have the bullets just whizz by. When firing at convergence distance, raking a 190 with a short burst from Spitfire's Hispanos reliably sends it down in flames, and the P-51 doesn't need a whole lot of shooting, either. In DCS they're harmonized to a point, not a box like they were IRL, so it's harder to hit with them, but they hit hard.

[motoadve]

Quick kills happen a lot in DCS

 

[Hobel]

Am 15.8.2024 um 20:00 schrieb Gunfreak:

If there is an effect. It's very very small.

With the Bf109 the effect is very strong with the p51 it seems to go towards 0 but I have to test it again 

[Bowie]

On 8/16/2024 at 2:05 PM, Gunfreak said:

Saying stuff like hitting from perfect 6 o'clock isn't effective. Because you have to hit the pilot or engine. Just isn't true, firing  from a perfect 6 o'clock in real life is perfectly(and the most common way ww2 aircraft got shot down)

Range - makes a big difference.

Have gotten close behind Fw-190's in the P-51D and absolutely shredded them, to no avail.

Airplanes are mostly empty space, with a few critical components that, from the rear, are generally armored.

But back off until their wingspan is half the 70-MIL gunsight ring, where the 6x0.50Cal are concentrated into a trash can lid sized pattern, and it usually only takes a few 1/4 second bursts, as each contains ~ 20 rounds.

 

Bowie

[Bowie]

As to the OP, was having stall/snap-roll problems in a mid altitude fight w/ a Bf-109, which was unusual for the trimmed A/C.

Then noticed this:

So there is some effect.

 

Bowie