SpitIX FM and DM bugs.

[Holtzauge]

This guys has also lost quite a bit of wing and is still flying.

 

BTW about the fast spinning Fw-190: It's a big difference if you lose the wing while flying straight or pulling g's like the 190 was in the gun cam: In the latter case the rolling moment is huge due to the high wing lift which will spin you fast if you lose a piece.

Edited June 9, 2018 by Holtzauge

[Poochnboo]

30 minutes ago, AndyJWest said:

But then I'd be surprised to see the SBD flying too.

Normally I'd have bowed out of a thread, by now. But this is a fun little discussion, so I'll stick with it. 

I'm more inclined to understand the SBD and TBF as surviving that kind of damage, for this reason. They are bombers, not fighters. They both have big, wide wings with thick airfoils. That's meant to give lots of lift to carry heavy loads and to make the airplanes more stable for getting them aboard aircraft carriers. You could fall asleep at the controls of one of those airplanes and they'd go right on flying without you. 

A fighter is designed to be unstable, to a certain degree. A fighter can't be as docile as a Cessna 150 or it could not perform the way a fighter is supposed to. So, it takes less damage to a fighter aircraft to make it unflyable then a bomber. 

That's why the only examples anyone has of planes surving such a hit, are bombers. Haven't seen a fighter, yet. As rugged as a P-47 was, take 2/3's of it's left wing off, and the plane is going down 

[DD_Arthur]

17 minutes ago, Poochnboo said:

 As rugged as a P-47 was, take 2/3's of it's left wing off, and the plane is going down 

 

 That might be true,  I have no idea but this one managed to get home after doing this to it's right wing.

O

Ah...quick google search for the left wing.....

 

[JtD]

The upper one is missing maybe 10% of the right wing, the lower one maybe a third of the left wing span, quarter of the left wing area. Same is true for the Avenger or basically any picture I've seen of damaged WW2 aircraft. All these pictures look spectacular, but are far from "half a wing", let alone two thirds. It's mostly the wing tip, size and definition of tip varying.

 

The impressive thing is the pilot, because with damaged ailerons, you get all sorts of unwanted behaviour. That makes the aircraft extremely hard to control. Not everybody can do that.

[Poochnboo]

Well, I'm the first one to say I was wrong. But it's not common enough, I think, for it to be built into the plane's FM in a flight sim, I think. What I mean is, the devs can't make it so that  it

18 minutes ago, DD_Arthur said:

Ah...quick google search for the left wing.....

 

happens all the time. It would have to be random, some how. You can't be bringing a P-47 home with no wings all the time. I think the Luftwaffe guys might complain.

[VesseL]

Half is more than 1/3. But could it be so sensitive thing that if the cutting is clean or formed the way that some shred sticks out up or down , meaning it will push the wing up or down, that would be enough to decide the outcome what will happen. Survive or not.

[DD_Arthur]

I don't think this sort of thing happened very often at all in real life.

 

  I have also lost part of the left wing of the Spit mk. IX in this game.  My 'plane flew on - very poorly - for a short while but eventually there was nothing I could do to stop it slide-slipping into the ground.

 

Haven't had that happen on the mk V yet either.

 

Does that make anything 'wrong' as opposed to 'right'?  No idea.  I would imagine it does mean they have different flight models that reflect different characteristics.

[Holtzauge]

In the OP video you can see that the Hispano and a bit of the wing outside that is still there: That means that missing wing part (See Mk9 drawing here) is basically the part with the aileron so again: There is no problem to generate zero/slight negative lift on this piece of the wing to counteract the missing part on the other side.

[AndyJWest]

An observation:

 

We have a single video showing a BoX aircraft apparently modelled as flying when damaged to extent that appears (visually) to be questionable. Since this is a single sample, why should we assume that it is a problem specific to the Spit IX at all? It could be. On the other hand, it could be something that can occur on any aircraft? It presumably isn't common, but that's no reason to assume it is more likely to occur on one model than another.

 

We need more data,  because without that, we have no way of knowing whether this is a 'Spitfire IX FM/DM' issue, rather than a general one.

[LLv34_adexu]

4 minutes ago, AndyJWest said:

An observation:

 

We have a single video showing a BoX aircraft apparently modelled as flying when damaged to extent that appears (visually) to be questionable. Since this is a single sample, why should we assume that it is a problem specific to the Spit IX at all? It could be. On the other hand, it could be something that can occur on any aircraft? It presumably isn't common, but that's no reason to assume it is more likely to occur on one model than another.

 

We need more data,  because without that, we have no way of knowing whether this is a 'Spitfire IX FM/DM' issue, rather than a general one.

I can check if I have SpitV tests, which I made yesterday. DM model was pretty much same, only we did short test with Mk108 only, we did not take russian cannons too like we did with spit9. Wing broke in spit5 after 3 rd hit. Right after it started to spin. There was no way to be able to fly it.

 

I would appreciate if ppl could test different planes with different guns and post videos about test results.

[AndyJWest]

Another photo of an aircraft apparently flying with an implausible amount of wing missing:

 

Note the up aileron on the intact wing.

 

[Poochnboo]

 

 

13 minutes ago, AndyJWest said:

Another photo of an aircraft apparently flying with an implausible amount of wing missing:

Yes, that's pilot officer Kanichi Kashimura. He collided with a Chinese fighter. 1937. That's an early version of the Claude. Several reasons it was still flying, I think. Nice clean break. Not the type of damage you get with an explosion. He still had more than half his wing. And the A5M had long ailerons. Long enough that he still had some aileron control from the damaged wing. He was lucky.

His luck, as for most Japanese pilots during WW2, didn't last. He was shot down and killed by Wildcats over the Solomons in March of 1943.

 

Edited June 9, 2018 by Poochnboo

[AndyJWest]

Possibly relevant. A video posted by Wandalen back in 2014 showing a BoS Bf 109 dogfighting and successfully landing after losing part of a wing:

 

Wing damage occurs at around 2:30. Clearly there is less wing missing than the Spit IX example, but being able to dogfight in this condition might be evidence of a more general limitation with the 'damaged' FM.

[=EXPEND=Tripwire]

4 hours ago, AndyJWest said:

Possibly relevant. A video posted by Wandalen back in 2014 showing a BoS Bf 109 dogfighting and successfully landing after losing part of a wing:

 

Wing damage occurs at around 2:30. Clearly there is less wing missing than the Spit IX example, but being able to dogfight in this condition might be evidence of a more general limitation with the 'damaged' FM.

 

The FM was revised after that video was created to ensure that planes with partial wing loss became uncontrollable. 

 

Prior to that change it was quite common to see online what you have shown in that really old video clip. 

 

[Legioneod]

One instance of this happening is not really enough to make a judgement. Take a wing off in a dogfight and the spit will be out of control rather easily.

The problem with the video is that you are flying straight and level, you're not puling any G or making maneuvers so when you do lose the wing it's less likely to lose control.

 

Imo this doesn't happen often enough to warrant a change, if a change is even needed since flying with wings heavily damage or even partially missing is possible irl. 

If you were to conduct the test in more realistic conditions then I'm willing to bet that the aircraft will lose control instantly or very quickly.

Edited June 10, 2018 by Legioneod

[LLv34_adexu]

6 hours ago, Legioneod said:

One instance of this happening is not really enough to make a judgement. Take a wing off in a dogfight and the spit will be out of control rather easily.

The problem with the video is that you are flying straight and level, you're not puling any G or making maneuvers so when you do lose the wing it's less likely to lose control.

 

Imo this doesn't happen often enough to warrant a change, if a change is even needed since flying with wings heavily damage or even partially missing is possible irl. 

If you were to conduct the test in more realistic conditions then I'm willing to bet that the aircraft will lose control instantly or very quickly.

? Good morning laugh. 

The problem is with FM. Not with maneuvering. This is test situation, not combat. I am 100% sure no plane would survive my 109 from those angles. I would empty whole clip in to it if needed.

[Legioneod]

56 minutes ago, LLv34_adexu said:

? Good morning laugh. 

The problem is with FM. Not with maneuvering. This is test situation, not combat. I am 100% sure no plane would survive my 109 from those angles. I would empty whole clip in to it if needed.

Your not understanding what I'm saying though. Aircraft are capable of flying with extensive wing damage, just because that spit was able to keep flying doesn't mean it's wrong.

My point was that if that spit were to do any sort of significant maneuver it would lose control.

[Poochnboo]

11 hours ago, Legioneod said:

The problem with the video is that you are flying straight and level, you're not puling any G or making maneuvers so when you do lose the wing it's less likely to lose control.

 

Have to disagree, here. The sheer violence of the impact would cause the airplane to go out of control. You can't possibly believe that an airplane could take such a catastrophic hit and continue to go on flying in a straight line. Even with examples that have been shown, you can be sure that those pilots lost control of those planes but were able to regain them in time to save their lives. The Spitfire in question, however, goes on flying as though it had barely been grazed. 

[Mauf]

16 minutes ago, Poochnboo said:

Have to disagree, here. The sheer violence of the impact would cause the airplane to go out of control. You can't possibly believe that an airplane could take such a catastrophic hit and continue to go on flying in a straight line. Even with examples that have been shown, you can be sure that those pilots lost control of those planes but were able to regain them in time to save their lives. The Spitfire in question, however, goes on flying as though it had barely been grazed. 

 

There were wing losses through air collisions and the plane was still landed on a wing stub. You don't get more violent than a mid-air collision and yet those kept in the air.

[AndyJWest]

I decided to see if I could reproduce the issue with the Spit IX. The first time, the aircraft spiralled in straight away, but on my second attempt I got the same result as the initial video - the Spit applied aileron to counteract the roll, and flew on under control. It crashed before landing though, and I suspect this was because it ran out of control authority as it slowed down.

 

I'll run some more tests, but from this there definitely seems to be some sort of issue.

 

Has anyone been running tests on other aircraft? At the moment we still don't know for sure if this is a Spit IX only issue.

[Raptorattacker]

On 6/9/2018 at 4:45 PM, AndyJWest said:

TBD flying after loss of part of one wing:

 

Has lost less wing % than the Spit, obviously. 

 

I bet they had to give the seats a damn good clean after they got home?!!!

[unreasonable]

Is the OP's video - and other tests - of damage to AI controlled Spitfires?  You would expect the AI to maintain control better than 99% of human pilots, unless it has been programmed to panic and make errors: it makes errors like flying into the ground, but those are errors of failing to plan ahead, not of maintaining control.

 

If so, what you are looking at here may be more of an artifact of the AI rather than the FM itself.  The FM says if you make such and such control movements the plane - for a given damage state - will do such and such. If the AI can make control adjustments almost instantaneously and without error, in situations where control is possible but difficult for a human pilot, you are going to see more of these situations than seems plausible.  

Edited June 10, 2018 by unreasonable

[Poochnboo]

1 hour ago, Mauf said:

There were wing losses through air collisions and the plane was still landed on a wing stub. You don't get more violent than a mid-air collision and yet those kept in the air.

Mauf, please reread what I wrote.

[AndyJWest]

I assume the initial video was against AI - otherwise there would have been comments about how easy it was to control the Spit. I've been testing against AI myself, but finding it difficult. You can't really run the test against an AI Spit 'enemy', as it will turn fight, and getting the sort of shot in required is difficult (or it is for me, anyway). I've tried setting up the QMB with the Spit as the 'second flight' against an AI enemy (an unarmed JU 52, to keep things simple), which makes it a little easier, but I can't figure out a way to persuade the Spit to fly straight so I can get in consistent shots. I think a test proper really needs setting up in the mission builder, but I've not used that myself. Does anyone have any suggestions as how to best do tests?

 

BTW, I've also seen the Spit have an engine quit twice, for no obvious reason. I wonder if there is an issue with air starts?

[Poochnboo]

One way that could possibly fix it, if it's possible, is to make it so that if the airplane takes a certain percentage of damage, the pilot is killed. That might not be acceptable to some.But it means that if a wing is blown off, the pilot is dead and can't maintain control of the airplane. If the nose is blown clean off of the plane, the fuselage is cut in half, the plane is a ball of flames....etc. The AI pilot dies.

[LLv34_adexu]

13 minutes ago, AndyJWest said:

I assume the initial video was against AI - otherwise there would have been comments about how easy it was to control the Spit. I've been testing against AI myself, but finding it difficult. You can't really run the test against an AI Spit 'enemy', as it will turn fight, and getting the sort of shot in required is difficult (or it is for me, anyway). I've tried setting up the QMB with the Spit as the 'second flight' against an AI enemy (an unarmed JU 52, to keep things simple), which makes it a little easier, but I can't figure out a way to persuade the Spit to fly straight so I can get in consistent shots. I think a test proper really needs setting up in the mission builder, but I've not used that myself. Does anyone have any suggestions as how to best do tests?

 

BTW, I've also seen the Spit have an engine quit twice, for no obvious reason. I wonder if there is an issue with air starts?

? Check General thread, about K4... Temuri is commenting there. 

[Legioneod]

3 hours ago, Poochnboo said:

One way that could possibly fix it, if it's possible, is to make it so that if the airplane takes a certain percentage of damage, the pilot is killed. That might not be acceptable to some.But it means that if a wing is blown off, the pilot is dead and can't maintain control of the airplane. If the nose is blown clean off of the plane, the fuselage is cut in half, the plane is a ball of flames....etc. The AI pilot dies.

No. Why would we want this, Just because some people don't like the fact that it still can fly with part of it's wing missing? Killing the pilot just doesn't make any sense.

[AndyJWest]

Test results:

 

I tested shooting a half wing off 5 times, and the AI successfully flew it afterwards 4 times - until it slowed for landing, when lack of aileron authority caused it to spiral in.

 

Video of typical test:

 

This test was done with stock cannon for the Bf 109 - I don't think it makes any difference how the wing is shot off, as long as it is hit further than about halfway out. I think it is reasonable to report this as a bug at this point, though in practice since the AI can't land it (and I'd be very surprised if a human player could) it will count as a kill anyway.

 

[BraveSirRobin]

Great catch finding this bug.  Not sure if the genre could have survived with Spitfires barely staying flyable, and then crashing, after having half the wing shot off.  

[Venturi]

The only two questions that should be asked here are these:

 

1. Can the aircraft sustain enough lift with the remaining wing area to stay aloft, and can it sustain enough speed to maintain the lift needed?

2. Is the remaining aileron able to counteract the asymmetrical lift from the lost wing area?

 

If the two answers above are yes, then what you're seeing in the video is accurate, and changing it to fit assumptions is creating inaccuracy...

[ZachariasX]

There are some anectotes of a Spit „flying“ with half a wing missing, a prominent one is mentioned by Johnnie Johnson in his book. In a collision, a pilot of his flight cut of the tail of a sqadmate, losing the wing from aileron outwards.

 

That guy could hold the plane like that, but even at cruise speed, it took a lot of force to hold the stick. Some minutes later he was so exhausted that he couldn‘t even use the radio anymore to respond to R/T. Before he coukd cross the Channel, he lost strenght and he gradually rolled over and crashed into the sea. His canpoy was blocked after the collision and he couldn‘t open it anymore, hence his desperated try to make it to British soil.

 

If there is a bug, it is more about the AI (or the comfy armchair pilot) not exhausting its strenght over time.

 

[unreasonable]

17 hours ago, AndyJWest said:

I assume the initial video was against AI - otherwise there would have been comments about how easy it was to control the Spit. I've been testing against AI myself, but finding it difficult. You can't really run the test against an AI Spit 'enemy', as it will turn fight, and getting the sort of shot in required is difficult (or it is for me, anyway). I've tried setting up the QMB with the Spit as the 'second flight' against an AI enemy (an unarmed JU 52, to keep things simple), which makes it a little easier, but I can't figure out a way to persuade the Spit to fly straight so I can get in consistent shots. I think a test proper really needs setting up in the mission builder, but I've not used that myself. Does anyone have any suggestions as how to best do tests?

 

BTW, I've also seen the Spit have an engine quit twice, for no obvious reason. I wonder if there is an issue with air starts?

 

If you have not already worked it out - the best way to do a test like this in the Editor is to set up the target plane with a waypoint so that it flies in a straight line, and a speed which makes it fairly easy to catch: then set your own plane to the same nationality as the target, which you can do for any plane.  The AI plane will then ignore you.  

 

IMHO the evidence is that the lift is sufficient, the control is theoretically possible - in fact documented for a variety of similar cases, albeit temporarily.  So there is no FM bug: it is just the AI doing what we cannot, just as it flies dead straight and level: which we cannot.  

[AndyJWest]

1 hour ago, unreasonable said:

 

If you have not already worked it out - the best way to do a test like this in the Editor is to set up the target plane with a waypoint so that it flies in a straight line, and a speed which makes it fairly easy to catch: then set your own plane to the same nationality as the target, which you can do for any plane.  The AI plane will then ignore you.  

 

IMHO the evidence is that the lift is sufficient, the control is theoretically possible - in fact documented for a variety of similar cases, albeit temporarily.  So there is no FM bug: it is just the AI doing what we cannot, just as it flies dead straight and level: which we cannot.  

 

I'd already figured out what I'd need to do if I used the mission editor: it's just that I've never got around to using it, given the learning curve.

 

As for whether the FM is bugged, there really isn't a yes or no answer. As I wrote earlier, everything involved here is an approximation, and there are lots of variables. Not least of which is whether the remaining aileron is going to work at all after the damage. 

 

Regarding the AI doing things we can't, I've now learned that the original test which started the thread was done with a human pilot - see this other thread: 

https://forum.il2sturmovik.com/topic/37134-bf-109-k4-information/?page=2

[HR_Zunzun]

"Landing" is still possible after losing such (visual) amount of wing.

Being hit in berloga while flying straight (not checking my six ?). Flipped over but managed to regain control. Started recording at this point. Flying with full right aileron and almost full right rudder. While turning applying full elevator to keep it from falling off the sky. Full rpm/throttle.

Managed to land/crash that possibly would account as a full landing in wol/taw. I do not think I would have been able to bring to a base having started farther apart from it (possibly damaging the engine),

Not saying is right or wrong. Just happened in Berloga after reading the post and wondered if landing was possible. Well, technically is still possible.

 

 

Edited June 11, 2018 by HR_Zunzun

[MK_RED13]

What is it? Another Spitfire "tuning"? This Spit followed my FW at 580 km / h on the deck a few minutes… 

[Ehret]

44 minutes ago, MK_RED13 said:

What is it? Another Spitfire "tuning"? This Spit followed my FW at 580 km / h on the deck a few minutes… 

 

The left wing radiator box is shared with the oil cooler - perhaps that is why it opened more.

[LLv34_adexu]

12 hours ago, HR_Zunzun said:

"Landing" is still possible after losing such (visual) amount of wing.

Being hit in berloga while flying straight (not checking my six ?). Flipped over but managed to regain control. Started recording at this point. Flying with full right aileron and almost full right rudder. While turning applying full elevator to keep it from falling off the sky. Full rpm/throttle.

Managed to land/crash that possibly would account as a full landing in wol/taw. I do not think I would have been able to bring to a base having started farther apart from it (possibly damaging the engine),

Not saying is right or wrong. Just happened in Berloga after reading the post and wondered if landing was possible. Well, technically is still possible.

 

 

I think that clipped wing spit is harder to fly with broken wing. Thank you for vid. Proves that it’s not one time thing. Also clipped wing!

Edited June 12, 2018 by LLv34_adexu

[LLv34_Temuri]

On 6/10/2018 at 8:06 PM, LLv34_adexu said:

? Check General thread, about K4... Temuri is commenting there. 

From the discussion in the general thread:

"Yes, rolling is possible with wing shot off, but can't really start maneuvering.

 

IIRC, we actually did a bit of dogfighting after I had received one 30 mm to the wing (wing not cut off yet), but that's not in the video. Wing didn't come off."

 

Basically the autopilot was able to keep the plane leveled, any manouvering in the OP video is done by me.

[unreasonable]

1 hour ago, LLv34_Temuri said:

From the discussion in the general thread:

"Yes, rolling is possible with wing shot off, but can't really start maneuvering.

 

IIRC, we actually did a bit of dogfighting after I had received one 30 mm to the wing (wing not cut off yet), but that's not in the video. Wing didn't come off."

 

Basically the autopilot was able to keep the plane leveled, any manouvering in the OP video is done by me.

 

Fair enough - we know that the AI can stay leveled if there is a control setting that can achieve that, just as it can fly perfectly straight and level in an undamaged plane when a human pilot cannot.  So the only reason to think that there is an FM bug would be if what the AI was doing was impossible. This is at best not proven, as the remaining wing loading shows, and planes that RTB successfully with only on aileron also suggest. 

 

So no bug: at most it is just the AI, plus the fact that human pilots had to exert considerable force to maintain level flight in this kind of situation and we, with our little HOTAS units, even with FFB, do not.  Again, absolutely nothing to do with the FM as such, more the integration of FM with the pilot strength model.  Currently you can keep your HOTAS control at one extreme more or less forever - or until you need a piss.  So the only way to deal with this would be through a model of AI and human pilot strength that was dynamic - ie changed over time by incorporating fatigue.

 

I would have no problem with that but whenever it has been raised in the past it has been met with howls of outrage from the MP set, especially those without rudder trim.

 

 

 

[MK_RED13]

10 hours ago, Ehret said:

 

The left wing radiator box is shared with the oil cooler - perhaps that is why it opened more.

Wow.. nice! Thx for explanation!