Jump to content

Wing shedding flight tracks thread


JG1_Hotlead_J10
 Share

Recommended Posts

JG1_Hotlead_J10

https://www.dropbox.com/s/k6z611ic05nvv03/Camel Rips Wings off too Easy - 2021-10-12.zip?dl=0 

(TRK file included now; thanks @Cynic_Al!)

 

In the effort to support the devs' research in the format they asked for, here is a track from multiplayer tonight where I bounced a Camel. After firing a few shots into him, he did a moderate turn and his wings peeled off. Look starting at 6:19 for the action's start. The engagement lasted 3 seconds.

 

Any other flight tracks to help research are welcome to post below! Please keep comments only to post new tracks. Thanks! 🙂

Edited by JG1_Hotlead_J10
  • Like 3
  • Thanks 1
  • Upvote 5
Link to comment
Share on other sites

https://drive.protonmail.com/urls/0CKWKA9EA8#vgyXqKnO8iQr

Albatross D5. PvP in training server. Few bullets into right wing. Entered a slight spiral dive. Wing fell off. Timestamp 1:01.

 

https://drive.protonmail.com/urls/PAAZ3GK80G#2NFAjMLSPP6A

Albatross D5. PvP in training server. Two bullets to upper left wing. Attempted pull up. Wing fell off.

 

https://drive.protonmail.com/urls/GD612S86M4#dh4xnEGdMKFA

Camel. Multiplayer training server. Few bullets to cockpit and wing. Slow spiral turn. Timestamp :22

 

https://drive.protonmail.com/urls/YEB20TZS5C#6fols4ID9tDI

Camel. Multiplayer training server. Few bullets to right wing. Attempt pull out of dive. Timestamp :09

 

https://drive.protonmail.com/urls/9VX5526H40#dWMOUqdaKBif

Camel. Multiplayer training server. Several bullets to upper wing. Turn and wing off. 

 

https://drive.protonmail.com/urls/Q5NJ19T55G#chh3VImthV9e

Halb CL2. PvP Multiplayer training server. Few bullets to right wing. Spiral dive and wing off.

 

https://drive.protonmail.com/urls/492B5X905C#ummDxGSFcJCS

Pfalz D12. PvP Multiplayer training server. Couple bullets to upper wing. Gentle dive and wing off. Timestamp 1:22.

 

https://drive.protonmail.com/urls/WF3S01PP5G#pt8IBOwjEanL

SE5a. Multiplayer training server. Several bullets left wing. Spiral dive and wingoff.

 

https://drive.protonmail.com/urls/EW18GRMPR0#0tzSL9wuSq3m

Spad XIII. Multiplayer training server. Couple bullets lower left wing. Pull out of dive and wing off. Timestamp :12

 

https://drive.protonmail.com/urls/6SWC2YXR1G#zvEn9gdGskFh

Spad XIII. Multiplayer training server. ONE BULLET to lower right wing. Pull up from dive and wing off. Time stamp 2:21

 

 

Thank you

 

 

Edited by US93_Low
  • Thanks 1
  • Upvote 3
Link to comment
Share on other sites

JG1_Hotlead_J10
3 hours ago, Cynic_Al said:

 

That zip does not appear to contain a .trk file, which would reside in the parent folder.

Thanks for the heads-up! I'm still figuring out the nuts/bolts with track sharing. 😉 Should be fixed now! 👍

Link to comment
Share on other sites

Helloooooooooooooooo where are you, 1C? You wanted tracks!! There they are.

 

TWO YEARS

  • Haha 1
Link to comment
Share on other sites

18 minutes ago, US93_Low said:

Helloooooooooooooooo where are you, 1C? You wanted tracks!! There they are.

 

TWO YEARS

Regardless as to whether or not you have a point, the way you're going about it is shockingly rude and disrespectful. If you came into my place of work acting like that, and talked to my employees in that manner, I'd sling you out. 

  • Upvote 8
Link to comment
Share on other sites

4 minutes ago, Diggun said:

Regardless as to whether or not you have a point, the way you're going about it is shockingly rude and disrespectful. If you came into my place of work acting like that, and talked to my employees in that manner, I'd sling you out. 


You’re free to sit there and do nothing while being shocked. Meanwhile I’ll do my best to get the game fixed for everyone. I’m not here to be your friend or Jason’s friend

  • Haha 1
  • Sad 1
Link to comment
Share on other sites

I'd say that throwing tantrums and sarcasm around like a spoiled child isn't the best way to go about it - if you look at what others have done in the past - for instance around the .50 cal update recently, you'll see other approaches are available. 

 

But whatever. You've clearly decided on this charming approach. I'm intimidated by your assertive manliness.

tenor.png

  • Haha 1
Link to comment
Share on other sites

JG1_Hotlead_J10
3 hours ago, US93_Low said:

Helloooooooooooooooo where are you, 1C? You wanted tracks!! There they are.

 

TWO YEARS

Low, I appreciate your desire to have this fixed. I share it too! But please respect my original post and keep replies to tracks only. Beating the devs over the head with rudeness isn’t going to fix anything, even if we’ve been waiting a long time for this to be fixed. This process will take time, which is probably why the devs haven’t responded right away.  If you have personal commentary, please feel free to post it on a new thread. Thank you. 🙂

Edited by JG1_Hotlead_J10
  • Upvote 7
Link to comment
Share on other sites

  • 1CGS
Jason_Williams
4 hours ago, US93_Low said:

Helloooooooooooooooo where are you, 1C? You wanted tracks!! There they are.

 

TWO YEARS

 

You really need to drop the damn attitude. Really, really fed up with it. I'll take a look at your tracks, but be warned. Just because you say something is not right, does not necessarily mean we will agree. Requires research and I have zero ETA on when if anything will be done. This is but a first step in an investigation. 

 

Jason 

  • Thanks 3
  • Sad 1
  • Upvote 4
Link to comment
Share on other sites

On 10/13/2021 at 12:07 AM, JG1_Hotlead_J10 said:

Thanks for the heads-up! I'm still figuring out the nuts/bolts with track sharing. 😉 Should be fixed now! 👍

 

As far as I can tell from the animation and sounds, you fired about 27 rounds per gun.  What effect that may have had is entirely subjective.

Link to comment
Share on other sites

US213_Talbot
44 minutes ago, Cynic_Al said:

 

As far as I can tell from the animation and sounds, you fired about 27 rounds per gun.  What effect that may have had is entirely subjective.

 

Looking at his parser for this sortie, he had roughly 30% accuracy which includes other Englishmen before breakfast. Using that figure it puts about 16 rounds into the target.

  • Upvote 2
Link to comment
Share on other sites

unreasonable

I am puzzled by what anyone posting tracks here thinks they are achieving. We already know that a wing hit box can be shot off after a few hits - experience and AnP's original graphs showed that clearly enough, and his graphs showed - with a bit of analysis - that some wing hit boxes were more prone to do that than others. 

 

It is a bit like people showing videos of people only over 6ft 6 inches walking around and claiming - "look, people are too tall".

 

From an analytic point of view, the only sensible metric would be the number of times a wing section is hit by n bullets and breaks, compared to the total number of times that wing section is hit by n bullets. You have to show all your tracks, not just the ones you cherry pick because you are unhappy with the outcome. Otherwise this whole exercise is completely misleading, except perhaps as a placebo to shut you all up while Jason gets on with something worthwhile.

Edited by unreasonable
  • Sad 1
Link to comment
Share on other sites

US93_Larner
1 hour ago, unreasonable said:

I am puzzled by what anyone posting tracks here thinks they are achieving. We already know that a wing hit box can be shot off after a few hits - experience and AnP's original graphs showed that clearly enough, and his graphs showed - with a bit of analysis - that some wing hit boxes were more prone to do that than others....*snip*



From Jason: 
 

Quote

I'll reiterate again what we need to "fix" this. We need tracks of the alleged issue happening and on what planes? Is it different on each plane? Not videos or anecdotes, but tracks please. I can't say this enough. Tracks are what we use to solve problems.


Big J says he can get a better understanding of whatever issue there might be with the DM through examining tracks of wingoffs - so now we're posting wingoff tracks. I'm guessing the Dev team has some kind of tool they use to analyse the tracks past just visually watching them (hence why Jason says videos are no good) 

Edited by US93_Larner
  • Thanks 2
  • Upvote 2
Link to comment
Share on other sites

unreasonable

To OP - sorry for discussion apart from posting tracks. (Cannot help myself when I see an interesting topic). Mod delete if you like. 

 

 

Edited by unreasonable
Link to comment
Share on other sites

JG1_Hotlead_J10
28 minutes ago, unreasonable said:

To OP - sorry for discussion apart from posting tracks. (Cannot help myself when I see an interesting topic). Mod delete if you like. 

 

 

It’s ok @unreasonable . 😌 I usually don’t speak up and post on the forums about issues like these unless I believe I’m achieving something. As Larner mentioned, Jason asked for tracks, so I am providing what he asked for. 💁‍♂️ I’m actually finishing off an article collecting historical data about how WW1 aircraft were most often destroyed in combat. I’ve been collecting data from various sources for a few months in my spare time. Hopefully this will prove useful to the devs once it’s published. 🙂 But in the meantime, I wanted to comply with Jason’s request. Hence the tracks. 👍

  • Upvote 8
Link to comment
Share on other sites

  • 2 weeks later...

All tracks completed in singleplayer quick mission with Camel, 30% / 50 L fuel, Aldis, Expanded Window. Some against Albatross Dva, some against D7F.

 

  1. Control test. Flat Turn. 90Mph. 4.5gs. Wings fine. https://drive.protonmail.com/urls/FMQAN1W604#g0hwNDqcaUOw

  2. Control Test. Dive and pull up with back stick. 140Mph 4.2g @ :27, 150mph 5g @ :52 https://drive.protonmail.com/urls/9SYPRX0K64#b1FKRcLlWRqn

  3. Control Test. Left corkscrew dive and pull up with back left stick- 140Mph pull up @ :11; Right corkscrew dive with pull up with back right stick – 150mph @ :30. Wings break. https://drive.protonmail.com/urls/470142VK3C#MnpiFkS49Uqn

  4. Wing damage occurs at :26. Flat turn reaches 90mph 4g at 1:00, no issue. Dive and back stick pull up 140mph 4.2g @ 1:24. Wings break. https://drive.protonmail.com/urls/DKA7ZJ6BMR#vran5iprZndt

  5. Wing damage occurs at :17. Dive and back stick pull up 120mph 3-4g @ :32. Wings break. https://drive.protonmail.com/urls/B02R0QT8S0#jE3owgSl6fbP

  6. Wing damage occurs at :17. Dive and back stick pull up 120mph 3g @ :25. Wings ok. Further 120mph attempts at :33 , :43, and :53. Wings ok. 130Mph 3-4g @ 1:06. Wings break. https://drive.protonmail.com/urls/FHBRWT3Y3R#K33cpKKQih8z

  7. Wing damage occurs at :18 including broken strut. Starting at :29 several gentle left-hand corkscrew pull-ups attempted at 120mph. Wings ok. Wings break at 1:19. https://drive.protonmail.com/urls/FCVMP5GXVR#WwoB18MMK4eF

 

My quick conclusions for the devs:

 

Undamaged camel wings can withstand up to 7g (not shown) in a vertical, straight back stick, pull up. And undamaged camel wings can withstand up to about 5g (150mph) in a corkscrew pull up. Damaged camel wings (whether light or heavy damage) can only withstand about 3-4g at 120-130mph in both the vertical straight back pull up, as well as the corkscrew pull-up. Any wing damage at all (when received at oblique angle) gives a standard reduction in wing loading gforce tolerance of anywhere from 20-50%.

 

The issue here is that this severe reduction in wing tolerance occurs with either a couple bullets, or so many that a strut is blown in half. The damage model isn't distinguishing severity. If the wing is hit with dozens and dozens of rounds causing a strut to be blown apart, it would be reasonable to assume that the wing will lose 20-50% wing loading capability. However, it is not reasonable to assume that a couple of 30 caliber bullets would always cause the same reduction (or any reduction at all). The practical result of this DM is that planes like Spads which were known for heavy endurance in a dive become 50% more likely to fall apart in exiting a dive after being hit with only a couple 30 cal bullets, yet the pilot accounts reported surviving steep dives while riddled with bullets to the point that the wings were barely attached. This also flies in the face of several aces teaching to primarily fire guns close range and at either the engine or the pilot to ensure a kill. This mentality would not be necessary if you could hit a wing with a wild spray of bullets and watch the enemy break up as they attempted to evade.

 

As has been stated repeatedly this is neither historically realistic nor is it fun from a gameplay perspective. You get neither realism nor general enjoyment out of this. The DM severely impacts the Spad 13, Camel, SE5a, Dolphin, Albatross D5, Halb CL2, and Pfalz D12. It also impacts other planes but seemingly not as severely.

 

Thank you.......

 

EDIT: I forgot to include this anomaly in my assessment - Damaged wings in a FLAT turn can withstand over 4gs. Damaged wings in a dive cannot (or very rarely) withstand 4gs in a pull-up. Perhaps the issue is in sustained vs instantaneous G forces, but in the pull-ups I used a smooth pull back. Not a jerk back.

Edited by US93_Low
  • Upvote 6
Link to comment
Share on other sites

@US93_Low A further point of discussion. Wing strength reduction due to battle damage seems to be inconsistent, unpredictable, and unrelated to the number of hits received or damage visible. I regularly fly camels, SE5as, and other planes well-known for their horrendous in-game wingshedding habits. In some cases they have been able to survive extremely aggressive combat maneuvers without losing their wings despite visibly heavy damage, while in other cases wings were lost with no player-identifiable damage at all. I submit that it is not enough to provide tracks of "early" or "wrong" wing failures. We should also endeavor to provide tracks where the wings didn't fail but should have.

 

I have started recording tracks much more often than before. Unfortunately I often forget to start recording before battles :(.

Link to comment
Share on other sites

US28_Baer

Apart from the frequency and inconsistency, i find the  gap to the unaffected planes exceeds belief.

 

Perhaps it would be informative to have some tracks of the D7, Pfalz D3a, Dr1s (and D8?) taking a huge amount of wing hits and still pulling high Gs, and flying unaffected despite being very badly shotup.

  • Upvote 1
Link to comment
Share on other sites

I posted tracks as a control showing in-game airframe limits without damage, and then showed a predictable result from any wing damage. 
 

Everyone is welcome to replicate. Do a singleplayer quick mission. Pick 1v1 , camel vs ace D7F, and start at say 2000m. Offset your approach to the D7F. As he closes in, lower your wing to face him and use opposite rudder to maintain straight flight. He will shoot your wing. You can rock wings to limit bullets hitting you. 
 

Now hold blip and enter a near vertical dive. Watch speed. At 120-130 mph pull back on stick with sufficient force to reach 3.5-4.5gs. Should break most of the time.

 

I posit that the variation we think we’re seeing (with camel) is simply being in combat at different angles, yet still reaching a specific g limit set by the damage model. 


I believe every plane I listed  can be predictably duplicated.

 

For instance, fly a spad 13 with balloon guns vs a cl2. Set it up as you on his 6. Shoot him in top wing. He’ll most likely explode 

 

As an aside, I can’t recall seeing a situation where wings should have failed. Maybe when a hun plane that is less affected by the DM like a D3 collides with another? Because his wings should break but I often see them intact. 

11 minutes ago, US28_Baer said:

Apart from the frequency and inconsistency, i find the  gap to the unaffected planes exceeds belief.

 

Perhaps it would be informative to have some tracks of the D7, Pfalz D3a, Dr1s (and D8?) taking a huge amount of wing hits and still pulling high Gs, and flying unaffected despite being very badly shotup.

Yep. I’ll try this next. 

Edited by US93_Low
  • Like 1
Link to comment
Share on other sites

8 hours ago, US28_Baer said:

Apart from the frequency and inconsistency, i find the  gap to the unaffected planes exceeds belief.

 

Perhaps it would be informative to have some tracks of the D7, Pfalz D3a, Dr1s (and D8?) taking a huge amount of wing hits and still pulling high Gs, and flying unaffected despite being very badly shotup.

I think these planes actually have the correct damage model for wings. Little tiny holes through fabric wings aren't going to do much. You need to hit pilots or the engine.

  • Like 1
Link to comment
Share on other sites

unreasonable
7 hours ago, kotori87 said:

@US93_Low A further point of discussion. Wing strength reduction due to battle damage seems to be inconsistent, unpredictable, and unrelated to the number of hits received or damage visible. I regularly fly camels, SE5as, and other planes well-known for their horrendous in-game wingshedding habits. In some cases they have been able to survive extremely aggressive combat maneuvers without losing their wings despite visibly heavy damage, while in other cases wings were lost with no player-identifiable damage at all. I submit that it is not enough to provide tracks of "early" or "wrong" wing failures. We should also endeavor to provide tracks where the wings didn't fail but should have.

 

I have started recording tracks much more often than before. Unfortunately I often forget to start recording before battles :(.

 

It is inconsistent and unpredictable because damage effects involve the application of a random number in addition to the fixed data about hits and target in each case. This is certain: the developers have even given us the graphs showing the probabilities in certain cases. See AnP's DM thread. 

 

The mechanic is that a hit to a wing hit box rolls an RNG to see if it hits the spar, based on the geometry of the box and spar and the angle: if it does, the max g-load of that hit box is reduced, (in some way that probably is not a simple x points per hit calculation. There are other things that happen as well). 

 

That means that any testing is only showing you the effects of those particular RNG rolls. To understand the overall results you have to do large numbers of identical tests: AnP's graphs were of 10,000 trials of firing at each wing hit box of a Camel, until it broke, from one particular angle. We obviously cannot do that, but it makes it very easy for the devs to say that small sample tests are worthless.

 

Having said that I do think that the devs may be underestimating how often wing breaks are happening because they are looking at the wrong average. I have tried explaining why this may be happening before, but it is a bit technical and I might even be wrong about that, but as always with no feedback it is hard to say.

Edited by unreasonable
  • Like 1
Link to comment
Share on other sites

If their test is “shooting until the wing breaks” I don’t think it’s going to be useful here considering the issue is reduced g tolerance from any amount of bullet fire.
 

The issue mostly isn’t wings breaking from one bullet, or from sustained fire. The issue is one or many bullets causing the same severe reduction in wing strength that then results in wingoffs during normal combat maneuvers.

 

But I find it curious that they tested shooting at wings with “10,000” trials but couldn’t find the CL2 exploding from a burst of balloon guns.

 

At any rate, what we all experience is wings breaking when the plane makes a maneuver after being shot. I’ve shown tracks of this happening predictably from multiple angles with multiple planes suffering multiple levels of bullet hits and multiple degrees of visible damage.

Link to comment
Share on other sites

JG1_Hotlead_J10

Salute all! After months of work, I've finished my report on how WW1 aircraft took combat damage. While this does not include flight track footage, hopefully the combined mass of reports (580 instances of aircraft sustaining critical damage in combat) will prove to be a reliable yardstick to measure the sim against as the developers review the damage model when they have time:

https://climbinghigher.wixsite.com/climbinghigher/post/pick-your-poison-an-exploration-of-how-ww1-aircraft-were-most-often-destroyed-in-combat

  • Like 1
  • Thanks 1
  • Upvote 2
Link to comment
Share on other sites

unreasonable
Just now, US93_Low said:

If their test is “shooting until the wing breaks” I don’t think it’s going to be useful here considering the issue is reduced g tolerance from any amount of bullet fire.
 

The issue mostly isn’t wings breaking from one bullet, or from sustained fire. The issue is one or many bullets causing the same severe reduction in wing strength that then results in wingoffs during normal combat maneuvers.

 

But I find it curious that they tested shooting at wings with “10,000” trials but couldn’t find the CL2 exploding from a burst of balloon guns.

 

At any rate, what we all experience is wings breaking when the plane makes a maneuver after being shot. I’ve shown tracks of this happening predictably from multiple angles with multiple planes suffering multiple levels of bullet hits and multiple degrees of visible damage.

 

There has already been a great deal of discussion about this: you really should read AnP's whole thread when the the new DM was introduced. They know perfectly well already that wings can come off under g loads more easily when damaged. That is by design. The issue is if it is happening too often, or at least more often than the developers think it should be happening, and if so why.

 

What AnP did, I think, is to carry out his 10,000 case test to destruction on each hit box, then average the results, probably by weighting by area or length. The problem is that the graph shows (as I have argued elsewhere) that some hit boxes have much lower tolerance to damage before they reach their break point. That means that the average number of hits taken by aircraft actually shot down with wing breakages is potentially much lower than the average in the test graphs, because the weakest box often fails first, so the potential additional hits in the stronger boxes do not get into play.   

 

 

 

77198584_AnPcamelbehind.thumb.JPG.e4f8e940bee420a2a27b10fd374daa9b.JPG

Just now, JG1_Hotlead_J10 said:

Salute all! After months of work, I've finished my report on how WW1 aircraft took combat damage. While this does not include flight track footage, hopefully the combined mass of reports (580 instances of aircraft sustaining critical damage in combat) will prove to be a reliable yardstick to measure the sim against as the developers review the damage model when they have time:

https://climbinghigher.wixsite.com/climbinghigher/post/pick-your-poison-an-exploration-of-how-ww1-aircraft-were-most-often-destroyed-in-combat

 

Excellent work and a very fair assessment. A couple of points:

 

Another reason for bumping up the preferred % of wing losses from the raw data as you did is that many - possibly most - of the OOC claims in your denominator were not actual kills at all, if we are to make any sense of the claims vs loss data.  Overclaiming, especially over hostile territory, must be taken into account when looking at combat reports with no crash site confirmation. 

 

The other point is that the probability of hitting a spar if you hit a wing box is purely a matter of simple geometry; I would be surprised if they have got that wrong; although stuff like that has happened. 

Link to comment
Share on other sites

JG1_Hotlead_J10
34 minutes ago, unreasonable said:

Excellent work and a very fair assessment. A couple of points:

 

Another reason for bumping up the preferred % of wing losses from the raw data as you did is that many - possibly most - of the OOC claims in your denominator were not actual kills at all, if we are to make any sense of the claims vs loss data.  Overclaiming, especially over hostile territory, must be taken into account when looking at combat reports with no crash site confirmation. 

 

The other point is that the probability of hitting a spar if you hit a wing box is purely a matter of simple geometry; I would be surprised if they have got that wrong; although stuff like that has happened. 

 

Thank you! 😊 I completely agree about overclaiming. I've added a line about this in the conclusion my article to reflect this.

 

As far as wing spars go, it's anybody's guess. I certainly know very little about programming! 😅 However it gets done, I would just appreciate seeing the DM match the history.

  • Like 1
  • Upvote 1
Link to comment
Share on other sites

I know what the issue is and I’ve demonstrated it repeatedly in tracks. I can produce predictable results. 
 

It’s not an issue having a damage model that reduces wing g tolerance PER SE as it would be reasonable to assume a flapping wing missing two struts wouldn’t survive the same g load.

 

The issue is that the DM doesn’t distinguish between a couple bullet holes and the broken struts. Regardless of level of damage, regardless of where the wing was hit, regardless of the angle it was hit from, the camel wing falls apart pulling vertically out of a dive with roughly 4gs. The camel can withstand at least 7g (perhaps more) in that same pull up with zero damage.

 

So like I said previously the practical implication is that in gameplay the affected planes get hit with a couple bullets of inaccurate long-range mg fire and lose the ability to conduct combat maneuvers.

 

This doesn’t match historical documentation or pilot accounts, and it’s not fun.

 

If you have tracks to the contrary please post them. If you want to assist in resolving wing shedding issue please conduct some QMB tests as I laid out above with ai shooting your wing. 

  • Upvote 1
Link to comment
Share on other sites

unreasonable

Here you go:  Tracks.zip

 

Two Camel tracks - using a MGAA to take some light wing damage then doing some pull ups. Did over 4g pull ups repeatedly on one going by the g-meter. On the other went over 7g. No wing failure.

 

Not sure how to make the instrument panel show on a track replay however - the time bar at the bottom seems to take over the  GUI.  

 

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...