Tires and propellers damage model?

[Wolferl_1791]

EDIT: Nice pics and MP test later in this thread:

 

Here's a scenario you can test for yourself. I jumped into a bomber/attacker (anything with an aft gunner) career mission, started from the runway.

 

While on the runway, I jumped into the gunner position and started firing at the other friendly planes behind me.

 

- wheels (tires): almost no effect. No "friendly fire" warning. No deflation. The planes eventually took off without problems. One time I did aim at the carriage and the plane stopped his engines and the tire disappeared.

 

- moving propeller: sparks. Lots of sparks. No "friendly fire"warning. Some bullets went through and I could see dust rising behind the plane. I aimed at parts of the plane within the propeller radius. Again, lots of sparks on the propeller, some bullets went and hit various parts, eventually shutting down the engine.

 

- stopped propeller: sparks. Only sparks. No holes, nothing. Almost perfect collision detection on the individual blades.

 

I've repeated my test in flight, in close formation with the AI. I had to hit carefully but again I could see only sparks on the propeller, but without any effect on the engine, unless the bullets went through the blades. Would you guys mind testing this scenario in multiplayer? I don't have a wing to test it on

 

It feels strange that the propeller would act as a Star Trek shield for the plane. Yes a small bullet wouldn't do much damage to it, but I'd assume 1000 bullets would eventually induce enough mass difference in one blade to create massive vibrations that would damage SOMETHING. I can't find any info about enemy damage to propellers (most google searches end up with the stupid WW1 synchronization gear). But I doubt that people told bomber gunners "don't fire at the front of the enemy, because they have this magic propeller that ricochets 99% of your bullets".

Seems I was wrong on this part, see below

 

 

As for the landing gear, there are quite a few planes with exposed tires. Normally, one single bullet would mean a busted tire and a not so pleasant landing, usually without any warning.

 

 

Edited March 14, 2018 by Wolferl_1791

[Bilbo_Baggins]

Interesting topic. 

 

I've certainly seen wheels missing many times, but I'm not sure how susceptible the wheel or tyre is to damage.

 

I never even knew the moving propeller could be hit and I actually find that amazing. I thought it would be far too intensive to calculate the extremely high speeds of both a moving propeller and a bullet moving through it's path. Will have to see that for myself!

 

I've also previously tried searches on google for the effects of propeller damage from projectiles but ended up with WW1 synchronization gear as you did yourself. I imagine it could cause tremendous vibration and RPM fluctuation?

Edited March 11, 2018 by Mcdaddy

[CrazyDuck]

[Wolferl_1791]

If it damages the prop, yes. I'd expect most of it to bounce off do to the blade's shape when firing from the front. But when hit from behind, the shape and the rotation actually increases the chance of damage. Small bullets is all I could test it with and I'd expect those to just shatter (but not on a stopped propeller). But a cannon round should ruin the propeller completely, although that's really hard to test in single player consistently. Anyway, at least now I know not to spend bullets at the prop circle whenever I'm in a gunner position.

 

@CrazyDuck

I've seen that as well, notice that the shot came from behind. He must've been in the air when he got hit. Any detail on how he got home? Did he have to reduce RPM? Did his engine stop? Also, the hole is closer to the center, less speed in that area.

Edited March 12, 2018 by Wolferl_1791

[CrazyDuck]

12 minutes ago, Wolferl_1791 said:

I've seen that as well, notice that the shot came from behind. He must've been in the air when he got hit. Any detail on how he got home? Did he have to reduce RPM? Did his engine stop? Also, the hole is closer to the center, less speed in that area.

 

No idea for certain - but according to internet sources, pilot didn't even realize anything odd and only discovered the damage after landing!

I'd imagine a hole like this would produce considerable vibrations. Or maybe rip the rest of the blade off!

[Wolferl_1791]

Most of the metal is still there, just peeled out. Maybe the slight loss of mass was counteracted by the gain in drag. He was one lucky dude.

 

Anyway, I did one more test with an I16. I turned around and came close to a wingman, nose to nose. I started firing and I noticed that none of my bullets were hitting his prop. Then a little birdy reminded me that since we both had the engine at idle, we probably both had the same RPM. So I increased throttle slightly to raise RPM a bit and suddenly my bullets began hitting his propeller since now they were no longer in sync That's some amazing collision detection.

 

I guess the current behavior might explain why FW190 pilots complain about being insta-engine-damaged by gunners. Radial engines have a much bigger target area in the middle of the prop circle than inline engines. It would also explain why my Yak can sit all day long at the aft of a Ju87 till the gunner wastes his ammo, while my I16 gets engine damage almost instantly.

[Bilbo_Baggins]

Wolfer- that's amazing. Good find. Had no idea that kind of detail was possible. Thought individual propeller hit detection was something still a long way away. That is tremendous collision detection.

 

I've always wondered what the "engine damaged" message meant when you get hit but without smoke or any problematic engine symptoms apart from a very subtle fluctuating in RPM- this is probably what it is, bullet damage on the propeller blade.

 

Yet another surprise from this sim.

Edited March 12, 2018 by Mcdaddy

[Finkeren]

33 minutes ago, Mcdaddy said:

Wolfer- that's amazing. Good find. Had no idea that kind of detail was possible. Thought individual propeller hit detection was something still a long way away. That is tremendous collision detection.

 

It’s a carry-over from Rise Of Flight, which was the first sim to properly model synchronization gear.

[Bilbo_Baggins]

Outstanding.

 

I'd like to arrange a meet up with someone on a server to check out the effects of these propeller hits online. Wolfer or someone- are you keen?

Edited March 12, 2018 by Mcdaddy

[Wolferl_1791]

1 hour ago, Mcdaddy said:

Outstanding.

 

I'd like to arrange a meet up with someone on a server to check out the effects of these propeller hits online. Wolfer or someone- are you keen?

 

Yep, the 72AG Training server looks perfect for this. I'll sit there for a while, waiting for people.

Edited March 12, 2018 by Wolferl_1791

[[APAF]VR_Spartan85]

Prop damage looks to be interesting.  I can imagine what he must have thought when he landed and saw that humbugs hole.... I would like to see more gear damage, flat tires, blown out tires, loose or wobbly tires.... 

i has made a post a few days ago, story about my wingman that bumped his gear off a hanger on takeoff, they jammed going up, Then us troubleshooting and fixing them using gravity... most fun I’ve had..

 

ive seen tires get shot off mid air but I think it would be more realistic if they were to get shot and blown out/deflated...

 

[Wolferl_1791]

1 hour ago, spartan85 said:

i has made a post a few days ago, story about my wingman that bumped his gear off a hanger on takeoff, they jammed going up, Then us troubleshooting and fixing them using gravity... most fun I’ve had..

I've seen your post. Cool one.

 

As for the props, I've found a few more interesting images. Unfortunately they're pinterest copyrighted so I can't link them

- One shows a japanese propeller blade littered with small holes (probably from 50 cals) but without any catastrophic failure.

- another one shows a Do17 that skimmed the water, broke both of its propellers (all 6 blades) in half, with really random and ugly edges, but still managed to somehow fly back home.

- then there's this gem:

 

Long story short, not only is a swiss cheese prop enough to get you home, you can even reuse it after good old O'Neal uses some superglue and a trusty hammer (exaggeration). Note that O'Neal might be out of his league on the US duralumin high precision blades used on more powerful engines (https://www.youtube.com/watch?v=PY8oQ3XNreY).

 

So for the purpose of the game, it seems that you'd have to catastrophically hit a blade (and nothing else) to demand any effect, otherwise the enemy should realistically get back home without even knowing he was hit. Then it would be the engineer's job to fix it. It seems our Star Trek shields really work as intended.

 

I've been trying to figure out why, and it seems the answer sits in the airfoil nature of the blade and it's overall low weight (the hub, especially on variable pitch props weighs A LOT more). Just as your wing can continue to fly even with heavy damage, so does the air screw, especially at low pitch. And the extra vibrations are cushioned by the bearings. From what I read, it seems that pilots around the Channel were more scared about water splashes that would damage the prop.

 

So what about WW1, with Flying Circus on the way and a bunch of pushers coming along eventually? And why did they invest in syncro gear? My guess is:

- why damage your plane yourself? Let the enemy do it.

- why waste bullets on yourself? Keep them for the enemy.

- that blade will eventually sheer off someday. Maybe. Why risk?

 

Edited March 12, 2018 by Wolferl_1791

[Wolferl_1791]

I've managed to do another test, in a custom mission, so that I could force my target to fly straight. My targets were a Ju52 and a He111. The Ju52 has 2 bladed props so hits were very rare. On the He111 they were more often, but still sparse, most bullets got through. Mg fire did nothing, even though I've spent at least 10 reloads (infinite ammo) on it. My guess is 10% of the bullets made sparks so that's an entire ammo load on one propeller, in a relatively consistent area towards the edge. Chances to get such a scenario in the game are of course zero, but in real life I'd expect that prop to be visibly cut after such bombardment. Cannon fire with HE rounds produced explosions which damaged the engine and exposed the landing gear. One engine stopped and I could target one of the stopped blades with consistent hits. No visual damage on the prop.

 

However, this test also allowed me to Taran my target. I used slow time and really came in close. With a soft contact I shredded his elevator with low damage to my engine. When I tried it again, my propeller received progressive damage. I noticed the propeller animation being different (the RPM was fluctuating due to engine damage as well). When I tried it a third time, finally my engine died. No visible damage on my stopped propeller. I'm 100% certain that my engine itself never came in contact with the enemy, so it was due to prop contact alone.

 

[JaffaCake]

This is super interesting. It could also explain why the BF110 MG81Z is so ineffective against russian aircraft.

 

But then while I may understand, given the examples, that prop actually can take some damage - it would be interesting to see how many bullets does it actually absorb in IL2, as we may have a case where large caliber guns, such as USB are modelled to go through the prop, while LMGs are absorbed by a prop in disproportionate manner, and thus make the engine "invulnerable" from the front.

[Wolferl_1791]

10 minutes ago, JaffaCake said:

This is super interesting. It could also explain why the BF110 MG81Z is so ineffective against russian aircraft.

 

But then while I may understand, given the examples, that prop actually can take some damage - it would be interesting to see how many bullets does it actually absorb in IL2, as we may have a case where large caliber guns, such as USB are modelled to go through the prop, while LMGs are absorbed by a prop in disproportionate manner, and thus make the engine "invulnerable" from the front.

 

Not quite. The propeller seems to absorb all damage, but it does trigger the explosion effect of the HE rounds that hit it, which do splash damage all around. You can fire at a stopped propeller all day long with AP, probably even from a tank, it won't do anything to the enemy.

Edited March 12, 2018 by Wolferl_1791

[Mauf]

9 minutes ago, JaffaCake said:

This is super interesting. It could also explain why the BF110 MG81Z is so ineffective against russian aircraft.

 

But then while I may understand, given the examples, that prop actually can take some damage - it would be interesting to see how many bullets does it actually absorb in IL2, as we may have a case where large caliber guns, such as USB are modelled to go through the prop, while LMGs are absorbed by a prop in disproportionate manner, and thus make the engine "invulnerable" from the front.

 

I know examples of the contrary. The tailgunner of the 110 does damage and eventually kill the engine. Being on the recieving end of said examples, I can tell you the damage does happen, but the engine rarely seizes outright from the hits. So it's not an instant showstopper and that's usually all that's needed to ruin your day in the 110. So there's a component of perception bias at play too. It probably comes down to the damage model applied to the engine and the blades. If indeed the blades are invulnerable on all accounts, I hope they remedy this.

[JaffaCake]

21 minutes ago, Wolferl_1791 said:

 

Not quite. The propeller seems to absorb all damage, but it does trigger the explosion effect of the HE rounds that hit it, which do splash damage all around. You can fire at a stopped propeller all day long with AP, probably even from a tank, it won't do anything to the enemy.

 

 

I am mostly considering LMG and HMG - cannon HE is a bit out of the league here. Also Il2 does NOT model splashdamage of HE. HE is modelled purely by creating spray of random fragments at the point of explosion. Its a known issue that caused many arguments with LW HE being "weaker" than it should be due to higher HE filler. Devs attempted to address it by making LW HE shells generate more shrapnel with faster speed, but they also made it lighter (so understandably less damage dealt). However shockwave/splash is still not modelled.

 

When you say fire at the prop - do you mean fire at the blade with engine behind it? So that prop stops ALL AP rounds with no damage dealt to the engine behind? (and as proxy all LMG/HMG rounds?).

Have you tested it with Pe2 gunner by any chance? Pe2 87 has 12.7mm USB heavy machine gun - would be interesting to see if the prop blade stops it dead too.

 

 

12 minutes ago, Mauf said:

 

I know examples of the contrary. The tailgunner of the 110 does damage and eventually kill the engine. Being on the recieving end of said examples, I can tell you the damage does happen, but the engine rarely seizes outright from the hits. So it's not an instant showstopper and that's usually all that's needed to ruin your day in the 110. So there's a component of perception bias at play too. It probably comes down to the damage model applied to the engine and the blades. If indeed the blades are invulnerable on all accounts, I hope they remedy this.

 

 

Of course some damage is dealt. But if you test with 110 gunner firing at the engine from the side / behind vs. from the front through the prop, you'll generally get the engine seize in less than a second of fire from sides, but it can take 2-4 times that from the front. (I don't believe there is much armour on yak engine to explain the difference?)

Edited March 12, 2018 by JaffaCake

[Wolferl_1791]

You wanna test it now? I'm up.

My test was only firing at the propeller with nothing behind it. A difficult thing to do. I was dead six on the enemy plane and below it so that rounds wouldn't accidentally hit the engine bay itself. But on the runway I did it with a gunner to test what happens when bullets do get through the blades. Engine damage as expected.

Edited March 12, 2018 by Wolferl_1791

[Mauf]

12 minutes ago, JaffaCake said:

Of course some damage is dealt. But if you test with 110 gunner firing at the engine from the side / behind vs. from the front through the prop, you'll generally get the engine seize in less than a second of fire from sides, but it can take 2-4 times that from the front.

 

Definitely interesting if it indeed was tested that way. Again, we would need to know the damage model details. Are you hitting different components of the engine from the side than front? From the front, are you hitting the prop nose which would constitute quite a bit of buffer compared to the naked side? Remember there's a cannon situated right behind it which isn't a lethal component of the engine.

 

Notice that from a front view, the Prop nose make up about 40-50% of the area which would contain engine components. Since I don't know how the damage model would behave here, there could be quite a bit of buffer between Prop and critical engine components, not to mention the prop blades "sweeping" the bullets. And I remember losing guns to tailgunner fire, so it might be just that effect.

 

Edited March 12, 2018 by Mauf

[Wolferl_1791]

8 minutes ago, JaffaCake said:

When you say fire at the prop - do you mean fire at the blade with engine behind it? So what you are saying is that prop stops ALL AP rounds? (and as proxy all LMG/HMG rounds?).

 

Wait wait, no that's not what I'm saying...

Imagine that each blade is 3d modeled with an accurate collision box. Also, it's invulnerable. Also, it triggers the HE round if it hits.

 

As the propeller spins, some rounds will get through, between the blades. If there was something behind them, it will act normally and do damage or fly into oblivion. Easy to understand for AP. Your prop blades act as a shield. You want them to be hit. Chances increase if :

- you have more blades

- you have big blades

- you have high RPM

- the bullet hits closer to the hub, or wherever the blade is thickest)

- relative speed between bullet and blade is low (so, unfortunately, not in head to head or against enemy gunners)

It seems that's realistic, because not only would a real blade push the bullet away, even if it gets chipped or punctured, it won't affect the engine as much as a direct hit to a more important part.

 

But HE is another story, if it hits, it will do some damage. I set engines on fire without firing at the actual engine, just hitting the propeller with sky behind it. Even hits to the stopped propeller did additional damage to the wing area.

 

Also, the prop hub is a very heavy and thick part. It houses the RPM governor (or whatever that cillinder is called). But, if you fly a radial.... especially if it gets hit by HE... ouch.

 

 

 

 

 

 

[JaffaCake]

23 minutes ago, Wolferl_1791 said:

You wanna test it now? I'm up.

My test was only firing at the propeller with nothing behind it. A difficult thing to do. I was dead six on the enemy plane and below it so that rounds wouldn't accidentally hit the engine bay itself. But on the runway I did it with a gunner to test what happens when bullets do get through the blades. Engine damage as expected.

 

 

No, I'm afraid not the best time for me to test right now. If you happen to be able to though, would be good to know if prop acts as an absolute shield, or selective to certain calibre of ammunition.

 

 

To answer your second post - if AP rounds are blocked by invulnerable prop blades and NOTHING takes damage behind the blade, we may have an issue. As while the blades are fast, many fighter aircraft used wooden blades, and it should reduce AP damage, but not completely block it. A bit of my own searching got me to this, which of obviously not realistic setting, but curious nevertheless : 

 

 

16 minutes ago, Mauf said:

 

Definitely interesting if it indeed was tested that way. Again, we would need to know the damage model details. Are you hitting different components of the engine from the side than front? From the front, are you hitting the prop nose which would constitute quite a bit of buffer compared to the naked side? Remember there's a cannon situated right behind it which isn't a lethal component of the engine.

 

 

Notice that from a front view, the Prop nose make up about 40-50% of the area which would contain engine components. Since I don't know how the damage model would behave here, there could be quite a bit of buffer between Prop and critical engine components, not to mention the prop blades "sweeping" the bullets. And I remember losing guns to tailgunner fire, so it might be just that effect.

 

 

 

Prop hub is a good point, but hits to the prop hub should be detrimental to the operation of the engine, as that is where bearings / pitch control is located. However I do not know if prop hub is armoured in fighter aircraft (and which ones? and is it modelled?).

 

 

As it is right now, the performance of LMGs as rear gunner defensive just begs to ask the question why would they even bother sitting a guy there during the war, if he had no chance to bring the aircraft down or damage it seriously, unless the enemy literally sat on the 6 sipping tea for a few seconds. Versus using the gunner weight towards  speed / performance / etc. of the aircraft.

 

 

Edited March 12, 2018 by JaffaCake

[Wolferl_1791]

No idea. But I've notice in some old videos how heavy that cover seems to be, at least compared to a blade (even a big one).

 

But the video you posted is with a cheap propeller. The wooden propellers they used in WW2 were quite tough. Plus, they're hitting that plywood from point blank range. Yes, it would go through but I doubt it would be able to retain any serious energy after that.

[Mauf]

10 minutes ago, JaffaCake said:

As it is right now, the performance of LMGs as rear gunner defensive just begs to ask the question why would they even bother sitting a guy there during the war, if he had no chance to bring the aircraft down or damage it seriously, unless the enemy literally sat on the 6 sipping tea for a few seconds. Versus using the gunner weight towards  speed / performance / etc. of the aircraft.

 

 

 

I think you misjudge the psychological effect these gunners had. In the game right now, we don't really give alot about getting shot at. In the real thing, having loads of tracer flying your way will make you think twice or thrice whether you want to keep your plane pointed at the enemy. Even if it only made you flinch and mess up your aim, it would already be worth it. Not to mention the "felt" security for the crew of the plane having the gunner.

 

Components that mattered in the real thing that don't really translate to the game.

 

If indeed the prop blades simply "eat" AP bullets, that should be looked at.

Edited March 12, 2018 by Mauf

[Cpt_Cool]

37 minutes ago, JaffaCake said:

Also Il2 does NOT model splashdamage of HE. HE is modelled purely by creating spray of random fragments at the point of explosion.

 

This is NOT a contradiction. It is a contradiction because you made a firm claim and then said the opposite in the next sentence.

[JaffaCake]

8 minutes ago, Cpt_Cool said:

 

This is NOT a contradiction. It is a contradiction because you made a firm claim and then said the opposite in the next sentence.

 

 

Does not model splashdamage

Models only fragmentation of the HE shell.

 

fragmentation of the HE shell is NOT splashdamage that HE shell deals. Splashdamage is a shockwave that is mostly uniform in all directions, unless it is met with obstructions, which it damages. Fragmentation damage is damage caused by splinters or parts of the original entity, where the fragments follow some distribution pattern based on the circumstances.

 

HE shell should deal BOTH splash damage and fragmentation damage. As it is now only the fragmentation damage is modelled. Given the design of the HE shells LW minen shells deal more splashdamage than soviet equivalent, but less fragmentation (as the fragments are much smaller)

Edited March 12, 2018 by JaffaCake

[Wolferl_1791]

29 minutes ago, JaffaCake said:

 

To answer your second post - if AP rounds are blocked by invulnerable prop blades and NOTHING takes damage behind the blade, we may have an issue.
 

 

Ok, good news. I got really lucky with an airfield test. A couple of HE111 behind me, me firing from gunner position. I could fire through the right prop of the closest one, into the right prop of the further one. I could clearly see sparks coming of from the two propellers and damage on the second plane. He got spooked, and ran into the plane behind me. They both shut their engines, but still alive. So now I could really shoot at a fixed propeller with the engine behind it.

- after the 2nd ammo clip, visual damage apeared

- after the 7th ammo clip, engine came on fire

 

Just for lolz, I tried the same with the planes on the other side of the runway, one clip of direct fire to the engine was all it took to shut it down.

It's really difficult to test without the pilot in those planes telling me what damage he's seeing. The visual damage model doesn't really.... work right.

 

It seems we're getting the best scenario out if it, I'm impressed.

 

Oh , and just so we're clear, if I didn't make it clear before, at least on the runway, at idle RPM, maybe 10% of the bullets actually hit the blades, the rest went between them.

Edited March 12, 2018 by Wolferl_1791

[JaffaCake]

1 minute ago, Wolferl_1791 said:

 

Ok, good news. I got really lucky with an airfield test. A couple of HE111 behind me, me firing from gunner position. I could fire through the right prop of the closest one, into the right prop of the further one. I could clearly see sparks coming of from the two propellers and damage on the second plane. He got spooked, and ran into the plane behind me. They both shut their engines, but still alive. So now I could really shoot at a fixed propeller with the engine behind it.

- after the 2nd ammo clip, visual damage apeared

- after the 7th ammo clip, engine came on fire

 

Just for lolz, I tried the same with the planes on the other side of the runway, one clip of direct fire to the engine was all it took to shut it down.

It's really difficult to test without the pilot in those planes telling me what damage he's seeing. The visual damage model doesn't really.... work right.

 

It seems we're getting the best scenario out if it, I'm impressed.

 

 

LMG? HMG? Cannon fire?

 

What we really need to verify is :

1) Prop blades "Absorb" AP bullets

2) Prop blades reduce AP bullet damage

3) Prop blades create pretty sparks but let AP through with no other impact

4) Prop blades vs AP of different caliber

5) Prop at full RPM chance of "stopping" a bullet

 

What we are certain of so far? :

6) Prop blades trigger HE

 

I'll try testing this myself as soon as I am able.

[Wolferl_1791]

6 minutes ago, JaffaCake said:

 

 

LMG? HMG? Cannon fire?

 

What we really need to verify is :

1) Prop blades "Absorb" AP bullets

2) Prop blades reduce AP bullet damage

3) Prop blades create pretty sparks but let AP through with no other impact

4) Prop blades vs AP of different caliber

5) Prop at full RPM chance of "stopping" a bullet

 

What we are certain of so far? :

6) Prop blades trigger HE

 

I'll try testing this myself as soon as I am able.

 

I don't know what the ammo clip on the top rear gunner of the HE111 uses. Previous in flight tests were done with a standard loaded 109G-2, if that helps.

 

Excellent questions. I'd have better answers to them if I had a friend to kill repeatedly Hug me.

Edited March 12, 2018 by Wolferl_1791

[JaffaCake]

3 minutes ago, Wolferl_1791 said:

 

I don't know what the ammo clip on the top rear gunner of the HE111 uses. Previous in flight tests were done with a standard loaded 109G-2, if that helps.

 

Excellent questions. I'd have better answers to them if I had a friend to kill repeatedly Hug me.

 

 

I am sure getting on TS / Discord would likely net you someone masochistic enough to let you perform your inhumane experiments on  Otherwise give it time, people will test.

 

HE111 - H6 or H16? H16 is HMG, H6 is LMG.

Edited March 12, 2018 by JaffaCake

[Cpt_Cool]

39 minutes ago, JaffaCake said:

 

 

Does not model splashdamage

Models only fragmentation of the HE shell.

 

fragmentation of the HE shell is NOT splashdamage that HE shell deals. Splashdamage is a shockwave that is mostly uniform in all directions, unless it is met with obstructions, which it damages. Fragmentation damage is damage caused by splinters or parts of the original entity, where the fragments follow some distribution pattern based on the circumstances.

 

HE shell should deal BOTH splash damage and fragmentation damage. As it is now only the fragmentation damage is modelled. Given the design of the HE shells LW minen shells deal more splashdamage than soviet equivalent, but less fragmentation (as the fragments are much smaller)

 

My whole point is that they model the HE shell using fragmentation, but there is no reason the chemical energy from the explosion can not be accounted for with that model. 

 

Explosive pressure damage is way too complex to directly model, especially given the limitations of our DM (structurally speaking, which is already the best in the biz). You would essentially need a soft body physics damage model, with stretching, tearing in the skin etc. All of this even before you start to open the can of worms that is pressure expansion and fluid mechanics (difference between a shell going off inside a mostly enclosed structure like a wing or on the outside).

 

So what is one to do to simplify the problem? Easy. Model the explosion as lots of little pieces of shrapnel that explode in a relatively even distribution. Then fine tune the kinetic energy the fragments have until you arrive at a believable result. I do agree that the HE is probably lacking, but if is to be addressed, it will be to make the fragments more effective. 

[JaffaCake]

32 minutes ago, Cpt_Cool said:

 

My whole point is that they model the HE shell using fragmentation, but there is no reason the chemical energy from the explosion can not be accounted for with that model. 

 

Explosive pressure damage is way too complex to directly model, especially given the limitations of our DM (structurally speaking, which is already the best in the biz). You would essentially need a soft body physics damage model, with stretching, tearing in the skin etc. All of this even before you start to open the can of worms that is pressure expansion and fluid mechanics (difference between a shell going off inside a mostly enclosed structure like a wing or on the outside).

 

So what is one to do to simplify the problem? Easy. Model the explosion as lots of little pieces of shrapnel that explode in a relatively even distribution. Then fine tune the kinetic energy the fragments have until you arrive at a believable result. I do agree that the HE is probably lacking, but if is to be addressed, it will be to make the fragments more effective. 

 

 

Or even simpler! Take everything within a certain radius of the explosion and apply root-distance damage to it, no need to even model the fragmentation as a proxy for splash.

 

I'd believe you if they  modelled HE with a good number of fragments, but unfortunately, as far as I recall, they only use 20-25 fragments at most per shell.  That is laughably small to be even more accurate than a simple AOE splashdamage model. Edit: they also use fewer more damaging fragments for soviet shells vs. lw shells, which goes contrary to modelling splash damage of the shockwave (where you'd at least keep number of fragments consistent)

 

HE shells need both - fragments and splash. At the moment we do not have splash, just the fragments that are not in large enough quantity to model splash.

Edited March 12, 2018 by JaffaCake

[Bilbo_Baggins]

It's a long shot but I was wondering if anyone is willing and available to test the effects of propeller damage with some simple on-ground experiments on a server right now? 72AG would be good.

 

I'll be checking the thread as PM alerts seem to arrive delayed for me. 

 

Cheers

Edited March 13, 2018 by Mcdaddy

[BlitzPig_EL]

A small pedantic point...

 

With rare exception, machine guns are fed by a magazine, not a clip.  A clip is a device that is typically used to load a magazine.  The magazine holds the rounds in, or feeds them to, the weapon.

 

The US M1 Garand rifle is however loaded with a clip, one of the few exceptions to this general rule.

 

[JaffaCake]

34 minutes ago, BlitzPig_EL said:

....

 

 

Thank you for an exceptionally useful contribution to the thread topic!

[Ehret]

20 hours ago, JaffaCake said:

Or even simpler! Take everything within a certain radius of the explosion and apply root-distance damage to it, no need to even model the fragmentation as a proxy for splash.

 

From the point of the shock-wave it may be that simple, but not necessarily for receiving plane's elements. Probably, their DM can only handle single AP like hits at time.

 

If the ballistic are computed on clients' computers then the simplest way to improve HE handling could be increasing number of fragments.

[JaffaCake]

15 minutes ago, Ehret said:

 

From the point of the shock-wave it may be that simple, but not necessarily for receiving plane's elements. Probably, their DM can only handle single AP like hits at time.

 

If the ballistic are computed on clients' computers then the simplest way to improve HE handling could be increasing number of fragments.

 

 

Damage is handled client-side as far as I know. (Game is mostly client side anyway and there were cases of people hacking to get invulnerable fighters dropping 1-ton bombs).

 

Again, as far as I know, the damage model is basically a selection of boxes with healthbars - AP or HE damage is dealt to the box as a whole, which takes a certain hitpoint damage. AOE splash would work just fine with this model. 

 

Remember that the original claim that fragmentation is sufficient to simulate splash is false. The fragments vary between shell types and have very different characteristics (LW fragments are more numerous and lighter, while VVS fragments are less numerous and are heavier) and are too few to be of use for shockwave.

 

Using fragmentation as a proxy to splashdamage is an extremely flawed design for multiple reasons :

- Fragmentation has a chance to avoid small structural parts that would otherwise be always damaged by the splash

- Fragmentation fails to "go around" parts, as shockwave usually would, whether through conduction through the said part, or go around through the enclosure.

- Fragmentation is fairly expensive to calculate vs. simple AOE, while it does not add any realism/accuracy to the damage dealt.

 

 

Ideally one would separate the aircraft into isolated "volumes" that take damage if HE is to explode in the said volume and parts outside of that volume "group" would be ignored. But fragmentation modelled splash is the worst way to go about it.

Edited March 13, 2018 by JaffaCake

[Ehret]

Dunno - what you want from splash-modelling demands very sophisticated computations. You would need to implement a complex material science just to able determine if an element should deflect or absorb the shock-wave.

 

IMHO, today, even a budget CPU should be able to track hundreds, if not thousands fragments on the client. The numerical approach is what is used for serious stuff, too - why it wouldn't work for us?

[JaffaCake]

You misunderstand my point completely I'm afraid.

 

No matter how large of a fragmentation cloud you can generate it is still better to use a simple AOE splash than to use fragmentation.  Afterall, we don't use complex material simulation for AP rounds - they just go through with some random deflection.

 

And by no means at all am I proposing to simulate physically-correct shockwave. That requires solvers that take hours to run a sim frame. However what I have suggested is a fairly computationally-light way to have slightly better shockwave sim than a simple AOE.

 

To summarise :

 

Fragmentation is the worst

AOE splash is OK, but can "bleed" to parts of the aircraft that shouldn't have been damaged

AOE splash with volumes is still OK and prevents "bleed"

..

..

..

Scientifically-accurate solver that has been experimentally verified to work within % error. (no thanks)

Edited March 13, 2018 by JaffaCake

[Ehret]

Just now, JaffaCake said:

You misunderstand my point completely I'm afraid.

 

No matter how large of a fragmentation cloud you can generate it is still better to use a simple AOE splash than to use fragmentation.  Afterall, we don't use complex material simulation for AP rounds - they just go through with some random deflection.

 

Not true - with large enough number of fragments It will track shapes well enough. It will handle deflections inside current planes models well enough, too. What you are proposing is the layer of hit-elements and tracking hits but in a reverse order - from the to be damaged element to the HE source. Not a bad idea, but how you will determine amount of health-bars loss then?

 

Objects of different compositions, orientations and shapes may react very differently to the same shock-wave. It's not just how distant from HE something is - the shock-wave will spread accordingly to the environment it meets - the plane's fuselage will act like a guide for it and so it will affect the result.

 

You said the shock-wave may go through an element being conducted by it - but how do you set the correct amount without the material modelling, then? Manually fiddling with parameters maybe?

[JaffaCake]

14 minutes ago, Ehret said:

 

Not true - with large enough number of fragments It will track shapes well enough. It will handle deflections inside current planes models well enough, too.

 

Or you could just use AOE model without having to spend all of those CPU cycles. Handling deflections is an order of magnitude more complex too, and is not done currently for HE fragments as far as I'm aware.

 

 

 

Quote

 

What you are proposing is the layer of hit-elements and tracking hits but in a reverse order - from the to be damaged element to the HE source. Not a bad idea, but how you will determine amount of health-bars loss then?

 

Otherwise known as splash damage, or AOE damage. The amount of damage is left to the complexity of your model and how much work you want to spend. I'd take a simple surface area / volume, but that is just a guess.

 

 

Quote

Objects of different compositions, orientations and shapes may react very differently to the same shock-wave. It's not just how distant from HE something is - the shock-wave will spread accordingly to the environment it meets - the plane's fuselage will act like a guide for it and so it will affect the result.

 

Which is why you could do a volume-based AOE. Everything within volume takes damage, but nothing outside of it. Tune the damage values until they are reasonable. Lower volume compartments take more HE damage. Again, depending on how complex you want to make your AOE model. Still better than fragmentation.

 

 

 

Quote

You said the shock-wave may go through an element being conducted by it - but how do you set the correct amount without the material modelling, then? Manually fiddling with parameters maybe?

 

Sure, just like everything else is done in this game - manual fiddling until looks right. As long as you use a model that actually somewhat represents the reality.

 

 

P..S. Can we get back on topic of this thread? If you are so keen on proving that current HE modelling works - there are plenty of threads in the dev suggestions forum describing the contrary.

Edited March 13, 2018 by JaffaCake