The reason why AP is so powerfull on the game.

[Ehret]

41 minutes ago, VO101Kurfurst said:

Kinetic energy is pretty much irrelevant when it comes to HE shells capacity for destruction. Chemical Energy > Kinetic Energy.

 

Again "energy this, energy that" where it's about momentum and pressure waves. Twice the energy yield of HE will not increase effects significantly. You need 8-10x more for that - an increase (coincidentally?) offered by the MK108 30mm.

 

To interact with something with a mass efficiently you need some mass, too. Stuff with more weight at similar energy levels will have a greater impact. That is the consequence of inertia and air has very little of it.

 

Still - the tests and videos show that hit-boxes are too coarse - looks like too many AP (and bigger fragments) register as structural hits. This is not a problem with shells as such.

Edited August 6, 2018 by Ehret

[=EXPEND=13SchwarzeHand]

1 hour ago, Ehret said:

 

Again "energy this, energy that" where it's about momentum and pressure waves. Twice the energy yield of HE will not increase effects significantly. You need 8-10x more for that - an increase (coincidentally?) offered by the MK108 30mm

 

I don’t know why you keep bringing the momentum argument, when it has been stated repeatedly, especially in regards to your post, that the original test documents,  come to the same conclusion of gun power that Ivy does and they explicitly mention that they account for momentum. Please read the report, you will come to the conclusion that the in game representation is wrong regardless of whether a a mix of momentum and chemical energy (as in the original ) or exclusively energy terms (as Ivy has) are used to determine gun power.

 

 

First one is Shvak, last one is Mineng. as used in Ivy´s test

 

CARTRIDGE

TYPE

ROUND WEIGHT

MV M/SEC

PROJECTILE WEIGHT GM

% HEI CONTENT

DAMAGE

POWER

20x99R

API  /  HEI

183

750  /  790

96  /  95

2  /  6

86  /  120

11

20x82

API  /  HET  /  HE(M)

205  /  183

720  /  720  /  800

117  /  115  /  92

3.1  /  3.2  /  22

110  /  109  /  236

16

 

 

Quote:

"For all of these reasons muzzle energy (one half of the projectile weight multiplied by the square of the velocity) has not been used to calculate kinetic damage as this would overstate the importance of velocity. Instead, momentum (projectile weight multiplied by muzzle velocity) was used as an estimate of the kinetic damage inflicted by the projectile. It might be argued that even this overstates the importance of velocity in the case of HE shells, as noted above, but the effect of velocity in improving hit probability is one measure of effectiveness which needs acknowledging, so it is given equal weighting with projectile weight.

Chemical energy is generated by the high explosive or incendiary material carried by most WW2 air-fighting projectiles. First, there is the difference between HE and incendiary material, which were often mixed (in very varying proportions) in the same shell. HE delivers instant destruction by blast effect (plus possibly setting light to inflammable material within its blast radius), incendiaries burn on their passage through the target, setting light to anything inflammable they meet on the way. The relationship between the effectiveness of HE and incendiary material is difficult to assess. Bearing in mind that fire was the big plane-killer, there appears to be no reason to rate HE as more important, so they have been treated as equal."

 

You get your momentum accounted for in the italicized line

 

He derives power by averaging all different bullets in the standard belting divided by 10.

 

But it gets even worse when you look at the MG vs the Shvak HE, which I highlighted in red, because it relates to the ammunition that Ivy tested in game. MG almost deals doulbe the damage that Shvak does. Yet in game they are treated equally.

 

There is a constant call for scientific data, yet when it is presented it seems no one bothers to even take a look at it, but rather makes various claims suiting his opinion.

 

 

 

Edited August 6, 2018 by =EXPEND=SchwarzeDreizehn

[Brano]

1 hour ago, VO101Kurfurst said:

 

Kinetic energy is pretty much irrelevant when it comes to HE shells capacity for destruction...

Not true. As example, wartime standard ShVAK 20x99 ОЗ projectile with K-6M fuse (осколочно-зажигательньй=fragmentation-incendiary) had also AP properties and could penetrate 7mm of homogenous armor at 200m.

[MK_RED13]

We have ShVAK 20x99 O3 with K-6M fuse in this game? 

[unreasonable]

2 hours ago, VO101Kurfurst said:

 

Kinetic energy is pretty much irrelevant when it comes to HE shells capacity for destruction. The shell will self destruct into tiny fragments at impact and will have little chance to convert its kinetic energy into physical damage. There is also magnitudes of difference between the capacity of a relatively small round (or its even smaller fragments) to make damage after flying several hundred meters and between the capacity of the explosives contained in the shell itself to do the same. Take note that all kinetic energy of the round itself comes from the powder charge in the cartridge to start with, and that powder charge is far, far less powerful than the explosive charge in the shell - otherwise it would simply rip the cannon breach apart.

 

Chemical Energy > Kinetic Energy.

 

Both fragments (due to their very poor aerodynamic shape, low weight and high velocity they loose their kinetic energy very fast) and explosive pressure loose the capacity to make actual damage within very short distances, however explosive force (pressure) will effect every and quite possibly destroy all surface in its near vicity, while fragments maybe do or maybe not. Their effect is far more random.

The little fragments are accelerated by the explosive energy. They also retain whatever kinetic energy they had from the speed of the shell: it does not all disappear. Hence the fragments move out and forwards. So for a ShVAK or Hispano HE shell the kinetic energy is extremely important: the blast effect is almost immaterial.  

 

You can calculate the initial velocity of the splinters using this little calculator or the Gurney equations if you want to DIY.  https://www.un.org/disarmament/un-saferguard/gurney/ 

Probably a little over 2000 m/s for the ShVAK (although I am not sure what explosive to use for the Soviets).  If the shell is moving at 700m/s on impact, the total KE of the flyers is about 25% from impact speed and 75% from the explosion. So yes, the HE does contain more energy than the propellant, but as you can see a great deal of it goes into accelerating the flyers.

 

The fragments are unaerodynamic but will cause damage on impact an order of magnitude further away from the explosion than any blast effect. edit - I agree this is more random the further from the point of impact you go. 

Edited August 6, 2018 by unreasonable

[Raymondo77]

[ZachariasX]

1 hour ago, unreasonable said:

the blast effect is almost immaterial.  

Actually, I found the blast effect of the Oerlikon 20 mm impressive in the light of how snall the charge actually is. It is strong enough to blast a hand sized hole in a strong aluminum struchture, roughly 30 cm behind impact. Shrapnel travels forward mostly, giving it about a 30 to 40 degrees scatter cone that acts as shot to anything behind the blast. Both effects, blast and shrapnel, can be clearly identified as such.

 

 

[unreasonable]

5 minutes ago, ZachariasX said:

Actually, I found the blast effect of the Oerlikon 20 mm impressive in the light of how snall the charge actually is. It is strong enough to blast a hand sized hole in a strong aluminum struchture, roughly 30 cm behind impact. Shrapnel travels forward mostly, giving it about a 30 to 40 degrees scatter cone that acts as shot to anything behind the blast. Both effects, blast and shrapnel, can be clearly identified as such.

 

 

 Do not forget that that hole is also made by the fuze and nose cone that are projected forwards, which make up a fairly meaty fragment.

[ZachariasX]

14 minutes ago, unreasonable said:

 Do not forget that that hole is also made by the fuze and nose cone that are projected forwards, which make up a fairly meaty fragment.

Might well be.

 

I‘m still looking for one of those hit demonstators Oerlikon used to reassure the prospective and hopeless 20 mm gun crews. I should take some pics. But so far I didn‘t find it any the Army museums. Contacting the company for that might not be that straight forward. It is Rheinmetall now, not sure what they kept of the old junk.

[unreasonable]

34 minutes ago, Raymondo77 said:

 

Good to give the source when you post a document. Textbook? pdf?

 

I agree that the KE at impact is a function of the closing speed of firer and target as well as absolute speeds: head on shots should do more damage.

 

I have no problem with a KE+CE formula as an overall energy estimate, but it has the disadvantage that it does not accurately reflect the ratio of energy that actually causes damage. This is because part of the CE is converted into KE: the flyers (splinters, nose cone and base plate) have to be accelerated which uses CE as work, and given the typical speeds at which they are accelerated this increases the total KE of the flyers by 200-300%. 

 

So saying that a HE shell hits with only the KE of the non-explosive mass times (impact) velocity squared of the shell is incorrect. The KE is far more than that, if the angle allows the splinters to hit the target, since they are travelling much faster than the projectile. The CE available to damage the target is correspondingly less. 

[Raymondo77]

49 minutes ago, unreasonable said:

Good to give the source when you post a document. Textbook? pdf?

 

The scientific paper was writen by Assoc. Prof. Dipl. Eng. Lt. Col. Janosek M., University of Defence in Brno,  Air Force and Aircraft Technology Department. He is a distinguished expert in the fields of aerial gunnery and effectivity of air ordnance.

 

49 minutes ago, unreasonable said:

I have no problem with a KE+CE formula as an overall energy estimate, but it has the disadvantage that it does not accurately reflect the ratio of energy that actually causes damage. This is because part of the CE is converted into KE: the flyers (splinters, nose cone and base plate) have to be accelerated which uses CE as work, and given the typical speeds at which they are accelerated this increases the total KE of the flyers by 200-300%.

 

Please enlighten us. How would you mathematically and analytically determine the effectiveness of HEI damage?  Could you provide your formula and calculus?

 

49 minutes ago, unreasonable said:

So saying that a HE shell hits with only the KE of the non-explosive mass times (impact) velocity squared of the shell is incorrect. The KE is far more than that, if the angle allows the splinters to hit the target, since they are travelling much faster than the projectile. The CE available to damage the target is correspondingly less. 

 

Fine. Instead of criticize the man's work, please provide your POD.  Thanks!

Edited August 6, 2018 by Raymondo77

[unreasonable]

I am not arguing with the total energy calculation: simply that as I said before, part of the chemical energy is converted into kinetic energy.  The Gurney equations give fragment velocities for the shells in Ivy's other paper of about 2,000 m/s for a cylindrical charge. That is a lot of KE, and it has to come from somewhere.

 

So if you think that total energy is proportional to total damage a KE+CE formula is fine, and given the complexity of the problem is no doubt a fair way to start. So I am not out to criticize the man's work at all. If you were comparing shells of similar design, and did not care whether the damage was due to blast or splinters, then such a formula is enough.

 

But it does not help very much here if we want to differentiate between the type of damage caused by CE and the type caused by KE, and the conventional HE shell and the mineshell were designed to create damage in different ways under different circumstances.  In the case of the conventional HE shell much or most of the damage is from splinters: ie KE damage. In which case you have to account for the total KE of the material that hits the target, not only the part that is due to the shell's impact velocity, and similarly the CE that can directly affect structure is reduced by the same amount.  I am perfectly happy for someone more mathematical to work out the numbers.   

 

Personally I doubt that we can get a much more sophisticated damage model given the game's limitations, but it is still interesting to speculate.

 

 =EXPEND=SchwarzeDreizehn earlier linked to another attempt to answer this question: that survey used momentum not the usual KE equation as in your source to estimate the damage ratios. This gives lower weight to conventional HE relative to the mineshell than your paper's formula would do and I do not agree with that.  

[=EXPEND=13SchwarzeHand]

1 hour ago, unreasonable said:

Personally I doubt that we can get a much more sophisticated damage model given the game's limitations, but it is still interesting to speculate.

 

Right, so

a) pretty much every paper on this is wrong and MG isn’t that much better than normal HE

b) the Germans could have just swapped their MG for HE

 

I‘m sorry but a simplifying model has to produce output that fits simplified assumptions and not musings about conversion of CE into KE.

 

As Long as the DM is based on rather large hitboxed, detailed assessments of hypothetically different impacts of different types of ammunition is nonsense.

The only sensible way to include ammunition on a very general structure model is in turn a simplified general ammunition model.

Edited August 6, 2018 by =EXPEND=SchwarzeDreizehn

[Ehret]

2 hours ago, =EXPEND=SchwarzeDreizehn said:

 

I don’t know why you keep bringing the momentum argument, when it has been stated repeatedly, especially in regards to your post, that the original test documents,  come to the same conclusion of gun power that Ivy does and they explicitly mention that they account for momentum. Please read the report, you will come to the conclusion that the in game representation is wrong regardless of whether a a mix of momentum and chemical energy (as in the original ) or exclusively energy terms (as Ivy has) are used to determine gun power.

 

I don't disagree with most of the premises. It's just the number of shots to "wing-off" is kinda weird unit of measurement for many reasons. Neither is right to rationalize it using total energy disparities. I'm not sure how it can be connected to results from the report - there is nothing about "wing-offs". It's quite the jump to assume that stated AmmoPower has to translate to one type of results in the game or such thing (wing-offs like in the videos) would be repeatable in the reality.

 

Yet, the in game tests/videos show where the real problem is - it's quite evident that hit-boxes register too many hits as structural ones - they are too coarse. Why then insist with a superfluous argument? Buffing one type of shells will not fix this - to the contrary - it may break something instead.

[=EXPEND=13SchwarzeHand]

That is exactly the problem though. I agree  that in real life there is no connection. In game where the world is simplified however, there should be a connection to Wing offs, as in game this is the major damage driver and as you have already correctly mentioned, hitboxed are too coarse to produce any realistic damage for AP.

So if the game isn’t able to produce a finer damage representation it should also stick to more general measurements of damage dealt by ammunition.

Edited August 6, 2018 by =EXPEND=SchwarzeDreizehn

[Ehret]

Just now, =EXPEND=SchwarzeDreizehn said:

That is exactly the problem though. I agree  that in real life there is no connection. In game where the world is simplified however, there should be a connection to Wing offs, as in game this is the major damage driver and as you have already correctly mentioned, hitboxed are too coarse to produce any realistic damage for AP.

 

If fragments use similarly coarse hit-boxes then bigger but fewer fragments (like from ShVAK or Hispano) could have an extra advantage over the much lighter but numerous (like from M-shells) fragments.

[Raymondo77]

1 hour ago, unreasonable said:

 I am perfectly happy for someone more mathematical to work out the numbers. 

 

Sure, leave it to others. Until now, you haven't provided anything substantial to support your claims. All what you do is to arguing in a cicle, claiming that the author's scientific approach was "inaccurate", "incorrect" while later admitting: [it is still] "a fair way to start" and "[good] enough".

 

1 hour ago, unreasonable said:

that survey used momentum not the usual KE equation as in your source to estimate the damage ratios. This gives lower weight to conventional HE relative to the mineshell than your paper's formula would do and I do not agree with that.  

 

Nonsense. I deliberately left over the final formula and the author's conclusion from the next pages. How you came to that verdict is beyond me, as the author clearly adopted Anthony Williams methodical calculation for combat effectiveness of aircraft cannons.

 

Please, spare me with your further pseudo-knowledge.

Edited August 6, 2018 by Raymondo77

[unreasonable]

10 minutes ago, =EXPEND=SchwarzeDreizehn said:

 

Right, so

a) pretty much every paper on this is wrong and MG isn’t that much better than normal HE

b) the Germans could have just swapped their MG for HE

 

I‘m sorry but a simplifying model has to produce output that fits simplified assumptions and not musings about conversion of CE into KE 

 

So far I have only seen one reference to an empirical test by JtD although I have not seen the actual paper, and of course the Germans were saddled at the time with rather low velocity cannons.   The paper you cite above (Williams) is a model based on some assumptions, just as BoX's is a model.  In particular, by using momentum rather than KE as his index of kinetic destructiveness Williams ensure that KE will be understated, and his reasons for this are very unconvincing - mostly about the difficulty of hitting the target, which is irrelevant if you do hit, and anyway applies to all shell types.  

 

I agree that A simple model has to use simplified assumptions. Of course, BoX does exactly that, it just does not deliver the results you want.

 

 

[=EXPEND=13SchwarzeHand]

6 minutes ago, unreasonable said:

Of course, BoX does exactly that, it just does not deliver the results you want.

 

It´s not about what I want, but about what the majority of literature on these simplified assumptions has to say. Which is the opposite of what the game has implemented.

[unreasonable]

9 minutes ago, Raymondo77 said:

 

Sure, leave it to others. Until now, you haven't provided anything substantial to support your claims. All what you do is to arguing in a cicle, claiming that man's scientific approach was "inaccurate", "incorrect" while later admitting: [it is still] "a fair way to start" and "[good] enough".

 

 

Nonsense. I deliberately left over the final formula and the author's conclusion from the next pages. How you came to that verdict is beyond me, as the author clearly adopted Anthony Williams methodical calculation for combat effectiveness of aircraft cannon.

 

Please, spare me with your further pseudo-knowledge.

 

The paper you quote shows the formula  KE=1/2 mv^2 in different notation, which is also included in the composite formula.   

 

Williams says : " For all of these reasons muzzle energy (one half of the projectile weight multiplied by the square of the velocity) has not been used to calculate kinetic damage as this would overstate the importance of velocity. Instead, momentum (projectile weight multiplied by muzzle velocity) was used as an estimate of the kinetic damage inflicted by the projectile."

http://www.quarryhs.co.uk/WW2guneffect.htm

 

So unless you - or the secret last page - can show how 1/2 mv^2 becomes p=mv I will stick with what is on the page.

Edited August 6, 2018 by unreasonable

[Raymondo77]

5 minutes ago, unreasonable said:

 

The paper you quote shows the formula  KE=1/2 mv^2 in different notation, which is also included in the composite formula.   

 

Williams says : " For all of these reasons muzzle energy (one half of the projectile weight multiplied by the square of the velocity) has not been used to calculate kinetic damage as this would overstate the importance of velocity. Instead, momentum (projectile weight multiplied by muzzle velocity) was used as an estimate of the kinetic damage inflicted by the projectile."

http://www.quarryhs.co.uk/WW2guneffect.htm

 

So unless you - or the secret last page - can show how 1/2 mv^2 becomes p=mv I will stick with what is on the page.

 

Not going to happen. Pay me.

That should be sufficient:

 

[unreasonable]

22 minutes ago, Raymondo77 said:

 

Not going to happen. Pay me.

That should be sufficient:

 

 

Well since we are talking about WW2 guns and not the modern era I am happy to look at what Mr Williams says on his website on WW2 guns to which I have linked.  Perhaps he changed his mind before he wrote his book. 

 

Actually if you are indexing damage from KE and then using a multiplier it does not matter which you formula you use, the relative index numbers comparing HE and AP for the same gun will be almost the same using either method, unless you assume a different impact speed, so it is odd that Williams thought it necessary to state what he did.

 

  

 

 

Edited August 6, 2018 by unreasonable

[=EXPEND=13SchwarzeHand]

14 hours ago, unreasonable said:

The paper you cite above (Williams) is a model based on some assumptions, just as BoX's is a model.  In particular, by using momentum rather than KE as his index of kinetic destructiveness Williams ensure that KE will be understated, and his reasons for this are very unconvincing - mostly about the difficulty of hitting the target, which is irrelevant if you do hit, and anyway applies to all shell types. 

 

You do realize I, quoted this paper to show that there is literature using momentum, because someone said that it would yield different results? Fact is it doesn’t. The thing is Ivy‘s calculations are based solely on E and come to the same conclusion of destructiveness ranking as the one including momentum. So using any method will come to the conclusion that the game‘s representation is wrong, yet you try to argue over 2 pages that it is right, without giving any reasons why it might be so, or having any literature to back it up. Where is the math or literature that backs the current Il2 model? Maybe there is a point where it is best to give up...

Edited August 7, 2018 by =EXPEND=SchwarzeDreizehn

[=362nd_FS=Hiromachi]

18 hours ago, unreasonable said:

Good to give the source when you post a document. Textbook? pdf?

Here is the source if you are interested, I dont know why someone would be making of it secret when its available to general public:

http://aimt.unob.cz/articles/07_02/07_02 (3).pdf

 

Parts that were quoted above come from pages 37 and 38 (or 5 and 6 of the file).

[Ehret]

Just now, =EXPEND=SchwarzeDreizehn said:

You do realize I, quoted this paper to show that there is literature using momentum, because someone said that it would yield different results? Fact is it doesn’t.

 

Aside this topic... The thing about momentum means, when you have two projectiles at the same energy, the heavier one will have better penetration. That's because by adding weight you can derive more momentum than by increasing velocity at same energy budget. It's a direct consequence of the E = 0.5 * m * v ^ 2 - the momentum (m * v) doesn't care which variable is bigger but the energy does - the v is square when the m isn't.

(weight could help with sectional density and slows rate of deceleration which are helpful, too)

 

This applies to everything including fragments - at the same E budget heavier ones will be more destructive assuming most of them will hit. Now returning to the issue there (IMHO) come the hit-boxes problem - they are too wide and allow for too many hits.

[unreasonable]

6 hours ago, =EXPEND=SchwarzeDreizehn said:

 

You do realize I, quoted this paper to show that there is literature using momentum, because someone said that it would yield different results? Fact is it doesn’t. The thing is Ivy‘s calculations are based solely on E and come to the same conclusion of destructiveness ranking as the one including momentum. So using any method will come to the conclusion that the game‘s representation is wrong, yet you try to argue over 2 pages that it is right, without giving any reasons why it might be so, or having any literature to back it up. Where is the math or literature that backs the current Il2 model? Maybe there is a point where it is best to give up...

 

Firstly, the method used - either Williams' or Janosek's - does change the rankings of per shell energy: not between the ShVAK and the Mineshell, but between the Mineshell and the Hispano.

 

I add an extract from a spreadsheet using Ivy's input data to illustrate.  Ivy's method puts the Hispano well behind the mineshell, Williams' a little behind, but Jansek's puts it just in front.

The ShVAK always looks sucky.  

 

Your appeals to "the literature" are not useful: these are all just models, and both William's and Janosek are doing them so that they can compare a wide variety of weapons, taking into account a number of issues such as rate of fire, installation weight etc to determine measures of effectiveness using an aggregate index.  BoX is not doing that: it has to simulate individual shell hits.

 

I am not arguing that BoX always simulates hits convincingly: quite the contrary, it is clear that AP (and MG bullets in FC) should differentiate better between going straight through a hollow structure and hitting something that can absorb it's KE. I would like to see this fixed as much as you and the OP - I have been making the same complaint about RoF DM for years.  In that case we can mod the DM to some extent for SP, but I would still like to see this changed in both BoX and FC.

 

When it gets to the mineshell vs other HE debate things get much more murky: which is why I have raised objections to using Williams' webpage numbers as gospel, especially as his reasons for using P rather than KE are unconvincing.   I do not know why the ShVAK scores equally to the mineshell in Ivy's tests, but I assume this is because of the splinter modelling.  I would love to have more information about the DM but I do not have it.   

 

[=EXPEND=13SchwarzeHand]

16 hours ago, unreasonable said:

Your appeals to "the literature" are not useful: these are all just models, and both William's and Janosek are doing them so that they can compare a wide variety of weapons, taking into account a number of issues such as rate of fire, installation weight etc to determine measures of effectiveness using an aggregate index.  BoX is not doing that: it has to simulate individual shell hits.

 

You are wrong, look at the test again and you will find that the aggregate numbers are made up (as the name already implies) single shot values, again they all tell the same story...

I believe that your unfounded theories are far less useful than my references to literature on the subject

 

A sinulation, with a simple model should reproduce simple outcomes of literature and not your or any other layman‘s musings. So again, please, just point to ONE source in support of Il2‘s current DM.

Edited August 8, 2018 by =EXPEND=SchwarzeDreizehn

[unreasonable]

Williams' model is just a model -  for his single shot values he takes momentum, just as in my spreadsheet, and then multiplies it by an number for the contribution for CE - for which zero empirical evidence has been produced, BTW, since even if we know the theoretical CE as per Ivy's bottom up model, we do not know the efficiency with which that CE is converted into KE and transmitted to the target.    Same with Janosek's model except he uses KE - and as I said they do not always give the same answer, not do they give the same answer as Ivy's analysis. 

 

You can see how crude their one - shot analysis actually is by using their formulae to calculate the total energy of an unexploded HE shell sitting on your desk as a paperweight.

 

I have spent many years of my professional life making predictive and analytical models far more complex that these: I am extremely well placed to understand how they work and for what they are intended.  But you can believe what you want: TBH I an only assume that you do not understand the calculations.

 

 

 

[JtD]

Imho, any attempt to attach a specific general damage value to a certain type of projectile is good for an interesting read, but a certain failure in terms of accurate damage modelling.

 

Any reasonably accurate approach for damage modelling needs to take into account properties of projectile, target and hit. Now while I have no doubt that the German 20mm mine shell was more effective against general WW2 type aircraft structure, I can think of a dozen types of targets and hits where it gets beaten by a .50 AP round.

 

Thinking of adjusting damage values in accordance with one or another theory will just change the errors, not fix them. We need a somewhat more complex damage model, and this wouldn't necessarily require a lot more work. Adusting an increasing number of aircraft will make this task increasingly difficult, however.

[Kurfurst]

Very much agree JtD, I think an approach the differentiates hit boxes into separate categories like airframe skin, load bearing elements, ammo, coolant/fuel, and equipment would be a good start. Then, perhaps an empirical % reduction of damage could be applied to each ammo type for different structures - for example, with just some AP would do 10% of its "full" damage to skin hits but 50% to load bearing elements and so on. The % can be adjusted on the basis of known firing trials. It can be further refined with additional rules in the set - AP for example could have the trait of "passing through", so this 50% damage could apply to several load bearing elements or other equipment, skin etc. as it goes through the aircraft, while HE would only do local damage, mostly to the skin and effecting flight controllability, but not so much the structural integrity. 

 

 

Edited August 9, 2018 by VO101Kurfurst

[LLv34_Flanker]

S!

 

 HE blows off panels and skin of the aircraft. And does more as can be seen on the Spitfire wing/fuselage or Blenheim fuselage vs MK108 Minengeschoss. On some planes the skin of the wing can be a component bearing part of the load. So gaping holes do not do any good to it nor aerodynamical qualities. An AP can just pass thru a wing skin and spar punching a hole, but still not reducing the integrity or aerodynamical qualities as much as a gaping hole or missing skin. How to make it work in a game is an another matter..

[JV69badatflyski]

10 hours ago, JtD said:

Imho, any attempt to attach a specific general damage value to a certain type of projectile is good for an interesting read, but a certain failure in terms of accurate damage modelling.

 

Any reasonably accurate approach for damage modelling needs to take into account properties of projectile, target and hit. Now while I have no doubt that the German 20mm mine shell was more effective against general WW2 type aircraft structure, I can think of a dozen types of targets and hits where it gets beaten by a .50 AP round.

 

Thinking of adjusting damage values in accordance with one or another theory will just change the errors, not fix them. We need a somewhat more complex damage model, and this wouldn't necessarily require a lot more work. Adusting an increasing number of aircraft will make this task increasingly difficult, however.

 

8 hours ago, VO101Kurfurst said:

Very much agree JtD, I think an approach the differentiates hit boxes into separate categories like airframe skin, load bearing elements, ammo, coolant/fuel, and equipment would be a good start. Then, perhaps an empirical % reduction of damage could be applied to each ammo type for different structures - for example, with just some AP would do 10% of its "full" damage to skin hits but 50% to load bearing elements and so on. The % can be adjusted on the basis of known firing trials. It can be further refined with additional rules in the set - AP for example could have the trait of "passing through", so this 50% damage could apply to several load bearing elements or other equipment, skin etc. as it goes through the aircraft, while HE would only do local damage, mostly to the skin and effecting flight controllability, but not so much the structural integrity. 

 

 

 

6 hours ago, LLv34_Flanker said:

S!

 

 HE blows off panels and skin of the aircraft. And does more as can be seen on the Spitfire wing/fuselage or Blenheim fuselage vs MK108 Minengeschoss. On some planes the skin of the wing can be a component bearing part of the load. So gaping holes do not do any good to it nor aerodynamical qualities. An AP can just pass thru a wing skin and spar punching a hole, but still not reducing the integrity or aerodynamical qualities as much as a gaping hole or missing skin. How to make it work in a game is an another matter..

Won't get that. unless they rewrite the engine.
iF the dev's wrote the soft in spaghetti mode, it would be a nightmare to rewrite as all classes and libraries are inter-connected. one change having directly  consequences at multiple levels. A small change may create serious bugs in the whole calculation. then finding the issue becomes a nightmare.
iF the dev's wrote the soft in Lasagne mode, it would be doable but still a lot of work...and all the 3d models would have to be modified to comply with the new DM anyway.
DM is something you think of when you start coding as this is a main parameter in an aero combat game. It's better to see big, code longer and implement all functions from a WWII plane (mono or multi engined) and then not even use all the capacity of the DM at the game launch and incorporate a new DM function with each game release, than start with a limited DM model and trying to add new functions into it.
In this field, BoX seems to have very limited DM, like the one from original il2.
This point is much better handled in CloDo visually and "system'aticaly"..(structure and mechanical failures).

[Operatsiya_Ivy]

I just stumbled across this thread and i just wanted to clear something up.

 

 

I am not assuming that my test is perfect nor that it necessarily reflects/represents accurate real life data.

 

The result of the test was that the different shells "rank" in a very odd way and that there is no explanation as to why (or at least as of yet, someone has to find out). While it is completely correct to say that there are too many variables to accurately simulate the damage in the game, it still begs the questions why some shells seem to be favored by the system in place. 

 

In the end, i think one of the biggest mistakes (due to engine limits or whatever) was to not model the Minengeschoss. A round that is used by nearly 50% of all aircrafts in the game.

[JtD]

I don't know if I am the only one to be curious what exactly the changes in the DM did, in particular I was curious about

 

21. Aircraft primary structure won't erroneously receive double damage in case of hits on the internal components like fuel tanks, engine, radiators, etc.)

 

which I thought could have been the root of all evil around the overpowered AP destructiveness.

 

Well, today, I've been shooting up La-5's. Many of them. In AI dogfights, until the enemy was 100% out of action (typically, that involves a bit of overkill). Generally I fired low/medium angle deflection shots from behind. Of course, I counted hits (from the logfiles), so I'd be getting something out of it. One check up was about the current state of HE vs. AP. I used the ShVAK 20mm gun for comparison. Pleasant surprise, overall. On average, I needed

- 15 HE hits to bring down a La-5. Most frequent cause of loss were engine fires, followed by damaged controls.

- 25 AP hits to bring down a La-5. Fires were less common, and no real most frequent cause of loss became apparent. However, I shot the tail section off the La-5 twice, which I didn't manage with any HE round I tested so far.

 

The overall figures appear like a huge improvement to me and generally right in relation to each other and also in absolute number. Big thumbs up!

Details of structural damage modelling imho still leave room for improvement.

[303_Kwiatek]

Average 15 hits from 20mm HE or 25 20mm AP to shot down  fighter?   And these is right direction?  I suppose to shot down a fighter from 20mm cannons it would need average 5-6 hits.

[DD_Arthur]

49 minutes ago, 303_Kwiatek said:

Average 15 hits from 20mm HE or 25 20mm AP to shot down  fighter?   And these is right direction?  I suppose to shot down a fighter from 20mm cannons it would need average 5-6 hits.

 

Agree with Kwiatek.....yes, really 

[CUJO_1970]

7 hours ago, JtD said:

Pleasant surprise, overall. On average, I needed

- 15 HE hits to bring down a La-5. Most frequent cause of loss were engine fires, followed by damaged controls.

- 25 AP hits to bring down a La-5. Fires were less common, and no real most frequent cause of loss became apparent. However, I shot the tail section off the La-5 twice, which I didn't manage with any HE round I tested so far.

 

The overall figures appear like a huge improvement to me and generally right in relation to each other and also in absolute number. Big thumbs up!

 

15 HE or 25 AP 20MM rounds to take out a La-5 is a good thing? Am I missing something ironic here?

[BraveSirRobin]

4 hours ago, 303_Kwiatek said:

Average 15 hits from 20mm HE or 25 20mm AP to shot down  fighter?   And these is right direction?  I suppose to shot down a fighter from 20mm cannons it would need average 5-6 hits.

 

A single hit from a 20mm round probably results in multiple damage entries in the mission log files.

[JtD]

Like I said, the intention was to instantly shoot it down and that involves a portion of overkill. I don't fire one shot bursts, so I can't tell you if hit 1 or hit 6 from the final 6 hits burst killed the plane. Additionally, I kept firing if there were only oil and fuel leaks, even though they'd be deadly eventually. Also, I only stopped firing when the prop stopped, but that doesn't happen until several seconds after the engine received lethal damage. Just a couple of examples.

 

So yes, for the amount of damage I dealt, I feel the absolute number to be in the right ball park. And once structural damage is modelled properly, high side outliers will disappear and then the average might be spot on.

 

But more importantly the relation between HE and AP appears to have made huge progress, and that's why I posted it in this topic.

[IRRE_Centx]

9 hours ago, 303_Kwiatek said:

I suppose to shot down a fighter from 20mm cannons it would need average 5-6 hits.

 

In real life, where a pilot would instantly jump after few hits to save his life, yes.

In a video game, against a player or (even worse) an IA who doesn't care about his life, you need more ammo because you need to completely DESTROY your opponent.

 

 

Always the same, some things can't be simulated like the panic of a young pilot who takes 20mm shot in the wing and jump instantly because he thinks his plane is out.

Should the dev make planes more fragile to "compensate" this? No.

Edited December 30, 2018 by -IRRE-Centx