Report: How WW1 aircraft most often took combat damage

[JGr2/J5_Hotlead]

Salute all! After months of work, I've finished a report on how WW1 aircraft took combat damage. You might want to pour yourself a glass of your favorite beverage while you read — it's a long one! Hopefully this helps by establishing a historical precedent as FC's damage model faces review soon. I know it certainly was eye-opening for me as I researched these past few months!

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

 

Thank you to @JG1_Dudley, @US213_Talbot, and @US93_Larner for assisting me and compiling some of the source results for me. I sincerely appreciate it! ?

Edited October 27, 2021 by JG1_Hotlead_J10

[US41_Low]

Very good article. Thanks for putting the work into it.

Also I agree that FC’s pilot damage system is very good. My experience with it makes me believe that most of the historical “OOC” claims were in fact a pilot kill or pilot wound.

[=IRFC=kotori87]

Very fascinating read. Thanks for putting in the effort! I think it's worth pointing out that not all FC planes have implemented the full control cable fix yet. The chances of control cable hits have been corrected for all planes. However, only some of the planes have the more accurate loosely-flopping control surfaces, while others still have the old rigidly-jammed controls. They have stated their intent to add loosely-flopping control surfaces to all FC planes, but it has not been fully implemented yet.

Edited October 28, 2021 by kotori87

[JGr2/J5_Hotlead]

43 minutes ago, kotori87 said:

Very fascinating read. Thanks for putting in the effort! I think it's worth pointing out that not all FC planes have implemented the full control cable fix yet. The chances of control cable hits have been corrected for all planes. However, only some of the planes have the more accurate loosely-flopping control surfaces, while others still have the old rigidly-jammed controls. They have stated their intent to add loosely-flopping control surfaces to all FC planes, but it has not been fully implemented yet.

Thanks! I’m glad you enjoyed it. ? And you’re certainly right; not all control cables have been fully fixed. But, since the devs told us they would be, I wrote the article from that standpoint in good faith.

[No.23_Triggers]

Thoroughly enjoyed the article, and think you've made a good analysis on the FC Damage Model. 

The CL.II wing update today shows promise...haven't really put it through its paces yet, but so far it seems to be quite notably sturdier. I haven't tried the DeHav or D.VII out yet, but I'm hearing good things. Can't wait for a "weak" Scout update (D.XII, Dolphin, Camel, etc), to really put the changes through their paces...! 

Edited October 28, 2021 by US93_Larner

[ZachariasX]

Out of reactions, but great read @JG1_Hotlead_J10!

 

One thing to compare wing failures in the real world vs. in game is that one should somehow normalize flying styles. I think this could be done.

 

Of the people flying multiplayer, has there ever been a telemetry plot of the g-forces during a dogfight? It would be interesting to see what the commonly reached peaks in g-forces are. We do have such a plot for a mock dogfight with the rreal aircraft back then. We know how the DM in FC works when AnP presented it and made us guess on the "how many bullets it takes to sever a spar vs g loads". For the sake of the argument, If we in game reach on average a 5 g peak in maneuvers and not like 3 g on that plot, you will see an according increase of wing failures. Now, if you then took the incidence of wing failures in FC at 5 g (as we have it in this example) and go back to AnP's model of wing damage, you could project the number of wing failures when you assume we would only reach 3 g in most maneuvers. How many wing failures would FC produce then? One should compare that number to your data.

[1PL-Husar-1Esk]

46 minutes ago, ZachariasX said:

Out of reactions, but great read @JG1_Hotlead_J10!

 

One thing to compare wing failures in the real world vs. in game is that one should somehow normalize flying styles. I think this could be done.

 

Of the people flying multiplayer, has there ever been a telemetry plot of the g-forces during a dogfight? It would be interesting to see what the commonly reached peaks in g-forces are. We do have such a plot for a mock dogfight with the rreal aircraft back then. We know how the DM in FC works when AnP presented it and made us guess on the "how many bullets it takes to sever a spar vs g loads". For the sake of the argument, If we in game reach on average a 5 g peak in maneuvers and not like 3 g on that plot, you will see an according increase of wing failures. Now, if you then took the incidence of wing failures in FC at 5 g (as we have it in this example) and go back to AnP's model of wing damage, you could project the number of wing failures when you assume we would only reach 3 g in most maneuvers. How many wing failures would FC produce then? One should compare that number to your data.

I did analyze G load charts during wing failure but only during single player and damaged wing  by personal weapons.  I would be good to do it during multiplayer and with multiple planes and damage levels. 

[SYN_Vander]

Quote

It would be interesting to see what the commonly reached peaks in g-forces are.

 

I agree this is a very important factor. Most players will not use force feedback sticks (where have they gone?!?)  and it's very easy to pull G-forces in the sim no pilot in WW1 would dare to do.

 

But thanks for this insightful compilation and analysis!

Edited October 28, 2021 by SYN_Vander

[BMA_FlyingShark]

Thank you for this, very interesting.

Have a nice day.

 

[J2_Bidu]

Still reading. Great job, Hotlead!

 

British and American aviators tend to use the "out of control" language a lot while German aviators (at least in this source) tend to use the "forced down" or "forced to land" nomenclature frequently.

 

This may be related to the fact that germans fought more often on their own side of the lines and so were more inclined to go low and check, while also being more likely to receive feedback from their ground troops.

[Holtzauge]

Fascinating read and very good compilation of data @JG1_Hotlead_J10! It was especially interesting to see more sources and how MvR’s often quoted victory statistics compare to others and that it contains a higher proportion of wings off occurrences than the other data so thanks again for putting all this data together and doing the comprehensive analysis.

 

I think what stands out is that the reports of wings falling off are (with the exceptions of MvR’s claims) not that common and that the overwhelming majority of the victories claimed are in the “Out of control” category. One can of course argue why the planes were out of control in the first place but I think the most likely explanation is actually the pilot being incapacitated not that a control wire, wing or control surface had been damaged to the point of the pilot losing control. Two reasons for this: First of all from the behind sector (where probably most of the firing was done from) the pilot is a big and vulnerable target and secondly, without getting into morbid details, as any hunter will tell you, game on adrenaline will most often show very small reactions to being hit mortally unless hit in the central nervous system which is usually the exception and not the rule. Consequently we could (just as we see) expect a low number of reports that pilots were killed outright. More likely then is that the numerously reported out of control dives are actually caused by an incapacitated pilot, not structural damage or loss of control.

 

Regarding the wings off statistics, another way of looking at it is that we don’t know at what point in time the wings came off: In the previous threads in the forum these occurrences have by many been taken as proof that the structural integrity of the planes has been compromised to the point that the wings folded. Another explanation could however be that these actually started off in the “out of control” category and while the victor is watching his victim go down the wings come off due to the combination of high speed and high g-loads in an out of control dive with an incapacitated pilot.

 

Summing up: Based on what we know about how these planes were constructed and to what level they could withstand loads undamaged, I would think (based on the statistics you have provided) that just a few percent actually lost their wings due to battle damage and the rest simply lost their wings due to structural overloads in dives with incapacitated pilots or due to the pilots panicking and jerking the stick when being shot at. As long as you have a bit of speed (as in an out of control dive) there is not much it takes to pull the wings off an even basically structurally undamaged plane.

[unreasonable]

Excellent article, wanted to expand on my comment in the other thread.

 

No doubt some proportion of the OOC claims were due to dead or incapacitated pilots, but spinning out of combat was also a widely used escape method, or simply a panic reaction.  A lightly wounded pilot might well spin OOC and then head home once clear of immediate danger. This would not be too hard in a large multi plane battle where following an aircraft down could be very dangerous, although I doubt it works in FC MP. These were certainly victories tactically, but did not represent an aircraft loss. 

 

The various air forces tried to sort out the good claims from the bad, but there seems little doubt that they erred on the side of optimism, if only to boost morale. Often unable to match up claims with crash sites, for understandable reasons, even the supposedly severe German confirmation requirements seem to have been fudged in a good proportion of cases, according to the Franks  books, especially of the "other Huns". Looking at claims and losses for WW1 (not easy) it seems that there may have been a ratio of about 2:1 between confirmed claims in aerial combat and aircraft actually destroyed in those fights. 

 

I have taken the liberty of stating the numbers from the article in a table, simplified by omitting the (few) double cause entries and the POW numbers, which give interesting  insight but which are of a different type. Given the comments above, if some false claims occurred in the sample - a virtual certainty - they are most likely to be in the "OOC uncertain fate" or "Other Destroyed" categories. So it is also worth looking at the data excluding these entries. It actually does not change the summary results that much. 

 

I agree with your conclusions and recommendations in general: (especially for more fires!).  The main remaining game design problem, however, is that if people playing the game do not act much like pilots in WW1, they will not get similar results even if the simulation is perfect.  

 

 

 

Edited October 28, 2021 by unreasonable

[US41_Low]

Faking ones own death and spinning down is a real world tactic that works in FC MP. Just FYI. 

[JGr2/J5_Hotlead]

16 hours ago, US93_Larner said:

Thoroughly enjoyed the article, and think you've made a good analysis on the FC Damage Model. 

The CL.II wing update today shows promise...haven't really put it through its paces yet, but so far it seems to be quite notably sturdier. I haven't tried the DeHav or D.VII out yet, but I'm hearing good things. Can't wait for a "weak" Scout update (D.XII, Dolphin, Camel, etc), to really put the changes through their paces...! 

 

Thanks Larner! The recent updates are really positive. First the control cables fixed and now some of the wing shedding issues addressed? I feel quite optimistic about the future if this trend continues! ?

 

 

12 hours ago, ZachariasX said:

Out of reactions, but great read @JG1_Hotlead_J10!

 

One thing to compare wing failures in the real world vs. in game is that one should somehow normalize flying styles. I think this could be done.

 

Of the people flying multiplayer, has there ever been a telemetry plot of the g-forces during a dogfight? It would be interesting to see what the commonly reached peaks in g-forces are. We do have such a plot for a mock dogfight with the rreal aircraft back then. We know how the DM in FC works when AnP presented it and made us guess on the "how many bullets it takes to sever a spar vs g loads". For the sake of the argument, If we in game reach on average a 5 g peak in maneuvers and not like 3 g on that plot, you will see an according increase of wing failures. Now, if you then took the incidence of wing failures in FC at 5 g (as we have it in this example) and go back to AnP's model of wing damage, you could project the number of wing failures when you assume we would only reach 3 g in most maneuvers. How many wing failures would FC produce then? One should compare that number to your data.

 

10 hours ago, SYN_Vander said:

 

I agree this is a very important factor. Most players will not use force feedback sticks (where have they gone?!?)  and it's very easy to pull G-forces in the sim no pilot in WW1 would dare to do.

 

But thanks for this insightful compilation and analysis!

 

Thanks ZachariasX and Vander! Yes, I definitely think pilots in FC feel free to push far more punishing Gs than real life WW1 pilots could have stomached. Probably if we virtual pilots actually could feel the Gs either pressing us into our seats or trying to rip us out of them, we'd fly a tad more conservatively! (Force feedback sticks are a good compromise in this regard though... sad they've become all but extinct! ?)

 

Nonetheless, I do feel that for many of the aircraft (less now, thanks to yesterday's update ?) pulling even reasonable Gs after receiving a few hits would pull wings off in an unrealistic manner. For instance, there is this instance when I was dogfighting in the N.28. I had pulled multiple reversals to the right and the left before extending. Nothing too crazy, mind you, but still moderately-aggressive. Then I extended and my opponent placed a few bullet holes into my wing. Judging from the sounds and the parser, hardly any damage was laid. (Look for the two entries at 2:11:42. Prior to those two hits, I had been able to maneuver just fine.) After those two hits, I used rudder to go into a right turn and lost wings. To me, that doesn't seem like I was pulling too many Gs for my damaged airframe. ?‍♂️

 

9 hours ago, J2_Bidu said:

Still reading. Great job, Hotlead!

 

British and American aviators tend to use the "out of control" language a lot while German aviators (at least in this source) tend to use the "forced down" or "forced to land" nomenclature frequently.

 

This may be related to the fact that germans fought more often on their own side of the lines and so were more inclined to go low and check, while also being more likely to receive feedback from their ground troops.

 

Thanks Bidu! I certainly agree with this. This probably is also the reason why so many American pilots (as shown in the POW reports) went down with engine damage to become captured.

 

9 hours ago, Holtzauge said:

Fascinating read and very good compilation of data @JG1_Hotlead_J10! It was especially interesting to see more sources and how MvR’s often quoted victory statistics compare to others and that it contains a higher proportion of wings off occurrences than the other data so thanks again for putting all this data together and doing the comprehensive analysis.

 

I think what stands out is that the reports of wings falling off are (with the exceptions of MvR’s claims) not that common and that the overwhelming majority of the victories claimed are in the “Out of control” category. One can of course argue why the planes were out of control in the first place but I think the most likely explanation is actually the pilot being incapacitated not that a control wire, wing or control surface had been damaged to the point of the pilot losing control. Two reasons for this: First of all from the behind sector (where probably most of the firing was done from) the pilot is a big and vulnerable target and secondly, without getting into morbid details, as any hunter will tell you, game on adrenaline will most often show very small reactions to being hit mortally unless hit in the central nervous system which is usually the exception and not the rule. Consequently we could (just as we see) expect a low number of reports that pilots were killed outright. More likely then is that the numerously reported out of control dives are actually caused by an incapacitated pilot, not structural damage or loss of control.

 

Regarding the wings off statistics, another way of looking at it is that we don’t know at what point in time the wings came off: In the previous threads in the forum these occurrences have by many been taken as proof that the structural integrity of the planes has been compromised to the point that the wings folded. Another explanation could however be that these actually started off in the “out of control” category and while the victor is watching his victim go down the wings come off due to the combination of high speed and high g-loads in an out of control dive with an incapacitated pilot.

 

Summing up: Based on what we know about how these planes were constructed and to what level they could withstand loads undamaged, I would think (based on the statistics you have provided) that just a few percent actually lost their wings due to battle damage and the rest simply lost their wings due to structural overloads in dives with incapacitated pilots or due to the pilots panicking and jerking the stick when being shot at. As long as you have a bit of speed (as in an out of control dive) there is not much it takes to pull the wings off an even basically structurally undamaged plane.

 

Thanks Holtzauge! Those are some great thoughts. I would say pilot hits certainly contributed a great deal to out of control cases, especially given how vulnerable of a target pilots were in WW1. Or, as unreasonable mentioned below, they could have been wounded (or just plain panicked) and diving out to live another day. 

 

As far as wings-off goes, that is also quite true! I'd say a good 2/3 of my wing off results are cases where bullets directly caused wing failure. The other cases are not so clear. Fire may have caused weakening, G forces in the out of control dive may have caused some as well, like you said. 

 

Really good thoughts! Thanks again for sharing! ?

 

4 hours ago, unreasonable said:

Excellent article, wanted to expand on my comment in the other thread.

 

No doubt some proportion of the OOC claims were due to dead or incapacitated pilots, but spinning out of combat was also a widely used escape method, or simply a panic reaction.  A lightly wounded pilot might well spin OOC and then head home once clear of immediate danger. This would not be too hard in a large multi plane battle where following an aircraft down could be very dangerous, although I doubt it works in FC MP. These were certainly victories tactically, but did not represent an aircraft loss. 

 

The various air forces tried to sort out the good claims from the bad, but there seems little doubt that they erred on the side of optimism, if only to boost morale. Often unable to match up claims with crash sites, for understandable reasons, even the supposedly severe German confirmation requirements seem to have been fudged in a good proportion of cases, according to the Franks  books, especially of the "other Huns". Looking at claims and losses for WW1 (not easy) it seems that there may have been a ratio of about 2:1 between confirmed claims in aerial combat and aircraft actually destroyed in those fights. 

 

I have taken the liberty of stating the numbers from the article in a table, simplified by omitting the (few) double cause entries and the POW numbers, which give interesting  insight but which are of a different type. Given the comments above, if some false claims occurred in the sample - a virtual certainty - they are most likely to be in the "OOC uncertain fate" or "Other Destroyed" categories. So it is also worth looking at the data excluding these entries. It actually does not change the summary results that much. 

 

I agree with your conclusions and recommendations in general: (especially for more fires!).  The main remaining game design problem, however, is that if people playing the game do not act much like pilots in WW1, they will not get similar results even if the simulation is perfect.  

 

 

 

 

Thanks unreasonable! I think this table of yours is another excellent way to view the data! Probably the truth lies somewhere between the raw interpretation of the data where all out of control cases are taken as "destroyed" and where none of the "out of control, fate unsure" and "destroyed, cause unknown" results are counted. Overclaiming definitely occurred and there were even cases of "confirmed" victories that never actually were destroyed, as I saw while going over MvR and the "other Huns'" claims in the Franks books. Also, many pilots did indeed escape by faking their own death — a trend I observed in nearly all sources I examined. My gut tells me that probably, about 1/2 - 2/3 (at most) of the "out of control, fate unsure" cases lived to fight another day.

 

Nonetheless, it is highly-interesting that even when you take all of the "out of control, fate unsure" and "destroyed, cause unknown" results away from the data, the percentages remain roughly the same. I think that is telling for sure!

 

 

Thank you to all who've commented! I appreciate the feedback and hope this article can be a useful resource for a while to come. ?

Edited October 28, 2021 by JG1_Hotlead_J10

[US103_Baer]

Comments regarding G forces used by sim pilots are kinda missing the point as it relates to the DM issue in FC.

 

The point is the disparity between planes that was introduced in 4.006. 

You can keep pulling the 'unrealistic' Gs in some planes after damage,  but others you can't. 

 

If it were possible to get details of victories/losses from FC in the way Hotlead has done from historical reports, I'm quite sure that the D7, Dr1, D8, PD3s and Bristol would be much closer to historical numbers for wing offs than all the other planes. 

Edited October 29, 2021 by US28_Baer

[JG1_Vonrd]

On 10/28/2021 at 12:20 AM, 1PL-Husar-1Esk said:

I did analyze G load charts during wing failure but only during single player and damaged wing  by personal weapons.  I would be good to do it during multiplayer and with multiple planes and damage levels. 

Tacview records and will display several flight parameters, including "G", Alpha, airspeed and altitude in all planes within the recording distance (I'm not sure what the actual radius is). You can also see the in-cockpit view of any plane recorded. I'm not sure if Flugpark allows Tacview. There are some servers that don't allow it. I'd take a look but don't have the time at the moment.

[=IRFC=Gascan]

Pretty sure some of the parameters are recorded, while others (like G) are calculated by Tacview based on the data given. The recording doesn't record data at the same rate that the simulation runs at, so you only get a glimpse of what happened. If the wing reaches a failure point, the G forces calculated by Tacview probably won't reflect the forces the game saw at the failure point. You can really see this when you are looking at planes at the edge of the recording area: they seem to jump around a bunch. They weren't jumping in game, its just the recording has fewer data points to work with.