Cpt_Cool Posted March 7, 2018 Posted March 7, 2018 The best of course would be a way to test as JaffaCake suggests, although I think it is a long shot we will get such a feature. Until then, I think Unresonable's test is as sound (statistically) as we can hope to achieve in game. I am assuming there is no way to tinker with the AAA ammo belts, which is honestly what we would need to explore HE vs AP effectiveness with that setup. Alternatively, could it be possible to set up a repeatable test where a parked bomber (perhaps a tank would work as well) shoots at parked aircraft with a turret? If we know what the ammo belts are made up of (easy in the tank), it could be possible to demonstrate the difference between HE and AP this way, although it would be extremely tedious, especially if there is a RNG involved with the DM. I don't understand how anyone can say "there is no evidence to suggest that player in that MP video had lag so we have to trust those results" and then "the AA ballistics might be simplified compared to player fired ballistics, so we should disregard these hundreds of test runs" Even if we could show that the AP is significantly more effective than HE, the next challenge would be to prove this is historically inaccurate, and figure out how to quantify how "off" it is in the sim.
Ehret Posted March 7, 2018 Posted March 7, 2018 Just now, Cpt_Cool said: Even if we could show that the AP is significantly more effective than HE, the next challenge would be to prove this is historically inaccurate, and figure out how to quantify how "off" it is in the sim. There are more questions - what type of ammunition against what type of target? A small frame with closely packed internals is probably very vulnerable to AP hits. For bigger frames, like many allied planes, area effect HE rounds should be more effective than AP, because odds piercing something important are lower.
Cpt_Cool Posted March 7, 2018 Posted March 7, 2018 36 minutes ago, Ehret said: There are more questions - what type of ammunition against what type of target? A small frame with closely packed internals is probably very vulnerable to AP hits. For bigger frames, like many allied planes, area effect HE rounds should be more effective than AP, because odds piercing something important are lower. From "can't miss" range, shoot each section of the 109 with HE only (record number of hits until the section breaks off). Hopefully (for times sake not for realism sake) it takes the same number of hits to remove the same component when shot in the same spot. If there is a randomness to the number of hits, do enough trials to get a good average. Then do the same test on each section of the 109 with AP only, and compare the results. THEN, repeat the entire process with the Yak et cetera, et cetera. I said it was going to be tedious. If these tests show a significant difference in the number of shots required to knock off pieces of the airplane (regardless of what aircraft, or what aircraft section). You cold make the argument that that ammo type is more powerful than the other (with regards to the structural integrity). Then, sort out if it should be or shouldn't be (this would probably be the tough part).
=RvE=Windmills Posted March 7, 2018 Posted March 7, 2018 Using the BK 3.7 on the G2 is a good way of more easily testing the effectiveness of HE vs AP. Just start shooting at friendly bombers and see how they differ.
unreasonable Posted March 8, 2018 Posted March 8, 2018 9 hours ago, Cpt_Cool said: The best of course would be a way to test as JaffaCake suggests, although I think it is a long shot we will get such a feature. Until then, I think Unresonable's test is as sound (statistically) as we can hope to achieve in game. I am assuming there is no way to tinker with the AAA ammo belts, which is honestly what we would need to explore HE vs AP effectiveness with that setup. Alternatively, could it be possible to set up a repeatable test where a parked bomber (perhaps a tank would work as well) shoots at parked aircraft with a turret? If we know what the ammo belts are made up of (easy in the tank), it could be possible to demonstrate the difference between HE and AP this way, although it would be extremely tedious, especially if there is a RNG involved with the DM. I don't understand how anyone can say "there is no evidence to suggest that player in that MP video had lag so we have to trust those results" and then "the AA ballistics might be simplified compared to player fired ballistics, so we should disregard these hundreds of test runs" Even if we could show that the AP is significantly more effective than HE, the next challenge would be to prove this is historically inaccurate, and figure out how to quantify how "off" it is in the sim. In reference to the bold issue - I seem to recall JtD doing some testing in this manner a while back. How many hits to remove a tail etc. You might find it if you search his threads - not sure if changes in the game since that time have rendered his results obsolete or not and I do not recall the conclusions. It may be the case that AP is too effective against structures compared to HE while correctly doing what it should for a low deflection shot. The whole point of AP is to increase the lethality of a certain kind of hit - one with a trajectory intercepting the engine or pilot in particular! As for the LAA round mix - I did not see any way to change it in the ME. I assumed that it was all HE actually since this seems the most sensible RL approach, and the 37mm shells have a self destruct/time fuze (although 20mm does not I find it unlikely that AP would be used in this context).
=EXPEND=13SchwarzeHand Posted March 8, 2018 Posted March 8, 2018 (edited) While all this is interesting, the main points have already been made by JtD On 5/4/2017 at 6:50 AM, JtD said: Thank you Jason for the long explanation. It's really appreciated on many levels. I've conducted a lot of detailed testing in the last couple of days in order to better understand the damage modelling. From this, I however would say that you've missed the main technical point. The main point would be Please look into the effectiveness of AP vs. HE ammunitions. I've tested several calibres now and the behaviour is slightly different with the different projectiles, but I think a generalization is possible. The AP round and the HE round do very similar amounts of damage against everything they hit. That excludes hits to critical points, just general aircraft structure. They shouldn't. A 20mm AP round will make a 20mm hole and that's it. A 20mm HE round easily destroys ten times as much structure in a single hit, which has been tested in real life. In addition to doing too much damage when hitting structures, after hitting the first structure, the AP round will continue straight on, and hit further structures in its way, It will damage four structures in total, disappearing after the fourth, dealing high damage to each of those. The HE round is typically gone after the first hit, though 23mm and 37mm HE rounds have some AP capabilities. In effect, the AP round does four times the damage of a HE round if it hits a complex structure or penetrates tail, fuselage and wing in an attack from the rear. With standard armament, the Yak-1 and LaGG-3 use a belting consisting of AP rounds almost exclusively, the La-5 can use AP rounds exclusively. The Germans are stuck with a HE mostly belting, and selecting HE only for the La-5 is just waste. As a result, in my tests, the Yak-1 was twice as hard hitting as a Bf109G and twice as hard hitting as a La-5 with HE only, even if the La-5 actually has twice the guns of a Yak-1. This needs to be addressed with highest priority in damage modelling - according to my test the AP deals way too much general damage, whereas the HE could probably use a bit more punch. Further points: HE round do damage not only by splinters, but also by gas pressure. In particular the German Minen rounds with a high HE content. Changing splinter velocity is not going to address this. Wood, even Delta wood, is much worse at dealing with damage than Aluminium is. Spar, rib and panel construction and monocoque construction have their own particular vulnerabilities, less spars don't necessarily mean less strength. In game there are actual bugs with some details (nothing critical imho & so far). You don't need to shoot 20mm round at WW2 aircraft to know some of that. Much of it is just physics or even common sense. I'll probably make a few longer posts about my findings in the near future, to show what's good, and where improvement's possible. Should help to reduce unqualified bitching on the forums, but maybe it can also help you by pointing you in the right direction. After this excellent post (and maybe as a result), a few days later, Devs posted this in Update 2.010 10. German 20 mm shell fragments have a higher initial speed because of their higher explosive mass to total mass ratio so they cause more damage; 11. Bf 109, Fw 190 and some other planes durability is fixed (it was undeliberately lowered because of the many earlier changes in these aircraft). Now twin longeron Soviet fighters with delta-wood wings are roughly 20% more durable than Bf 109 fighters which have single longeron wings, while Fw 190s with triple longeron wings are roughly 20% more durable than Soviet fighters. P-40 is even more durable thanks to its five longeron wings. Pe-2 and Bf 110 planes are roughly twice more durable than Soviet fighters while IL-2 fits somewhere in between Pe-2 and LaGG in terms of combat durability; There are three main points to take away from this: 1) Problems with the structural damage that AP inflict have not been covered. 2) They increased the initial speed of the shell fragments of the Minengeschoss. If you read JtD´s post closely you will see that this is not the main critique. The main point is that there will be a lot more structural damage due to gas pressure. IMO this has not been addressed either. 3) Apparently the Devs got some numbers confused, which does happen, if you have ever worked with xls, csv, you name it... shit happens (I know this might shock some of you... yes even to the Devs make mistakes), however the 110 being as stronk as the Pe-2??? Really??? Has anyone done any tests for that (that would be an unreasonably useful test ) IMO this sums up the main problem when talking about the D Edited March 8, 2018 by =EXPEND=SchwarzeDreizehn 2
JaffaCake Posted March 8, 2018 Posted March 8, 2018 To add to the above - while the german shrapnel from HE shells has an increased starting speed, it still has lower MASS, thus dealing less damage. This is not even considering the shockwave damage that would be much larger from the minen shell than the equivalent soviet one. Indeed AP damage and HE damage is still unresolved and is in awful state as currently implemented. Just wonder what it takes to get it to be fixed? The issue has been raised too many times not to be noticed. 1
Holtzauge Posted March 8, 2018 Posted March 8, 2018 Another thing that would be good to verify with the in-game AP is how do they behave when they hit? I have a nagging suspicion that they go straight, i.e. even if they hit at a very small angle they will drill a hole straight through the plane in the angle they hit ticking of damage points in the DM along that line. But I may have that backwards since it's just a subjective impression connected to the lethality of AP fired from the cowl guns when I fly Russian planes . Anyway my limited time in the army as a conscript and my somewhat greater experience as a hunter tells me that IRL bullets tend to be deflected when they hit things and while an FMJ and an AP round of course are not directly comparable I'm sure that an AP round would change its trajectory as well when it hits something. AFAIK this was also something that happened when they tested and evaluated battle damage IRL : Stuff aft of the back armour could make a round from the rear tumble and that would even make it hard for an AP to penetrate armour. I'm aware that you can in rare conditions in-game see tracers ricochet off but I'm unsure of if that is an indication if changes in bullet trajectories are modeled or if that's more of an eye-candy effect? 1
Ehret Posted March 8, 2018 Posted March 8, 2018 I wonder how exactly the HE is modeled? Is there a distinction between air-bursts and fragmentation? Maybe it's done just by spreading few dozens of shrapnel - does the frame model support the effect of volume damage in it?
JtD Posted March 8, 2018 Posted March 8, 2018 1 hour ago, Holtzauge said: Another thing that would be good to verify with the in-game AP is how do they behave when they hit? I have a nagging suspicion that they go straight, i.e. even if they hit at a very small angle they will drill a hole straight through the plane in the angle they hit ticking of damage points in the DM along that line. But I may have that backwards since it's just a subjective impression connected to the lethality of AP fired from the cowl guns when I fly Russian planes . Anyway my limited time in the army as a conscript and my somewhat greater experience as a hunter tells me that IRL bullets tend to be deflected when they hit things and while an FMJ and an AP round of course are not directly comparable I'm sure that an AP round would change its trajectory as well when it hits something. AFAIK this was also something that happened when they tested and evaluated battle damage IRL : Stuff aft of the back armour could make a round from the rear tumble and that would even make it hard for an AP to penetrate armour. I'm aware that you can in rare conditions in-game see tracers ricochet off but I'm unsure of if that is an indication if changes in bullet trajectories are modeled or if that's more of an eye-candy effect? According to my testing, AP bullets lose speed when penetrating are occasionally are completely deflected when hitting at shallow angles (like a wing from behind), which is visible as a ricochet. A projectile still damaged the surface which deflected it. What does not appear to happen are systematic disturbances of the flight path, such as tumbling or small changes in the flight path after low angle hits. I had a wall several hundred meters behind my targets, and while I measured a lower velocity after penetration, I never saw a changed or unstable flight path. So it appears to be either complete and straight penetration (with little loss of speed), or projectile break up/ricochet.
Holtzauge Posted March 8, 2018 Posted March 8, 2018 (edited) 1 hour ago, JtD said: According to my testing, AP bullets lose speed when penetrating are occasionally are completely deflected when hitting at shallow angles (like a wing from behind), which is visible as a ricochet. A projectile still damaged the surface which deflected it. What does not appear to happen are systematic disturbances of the flight path, such as tumbling or small changes in the flight path after low angle hits. I had a wall several hundred meters behind my targets, and while I measured a lower velocity after penetration, I never saw a changed or unstable flight path. So it appears to be either complete and straight penetration (with little loss of speed), or projectile break up/ricochet. OK, thanks for the clarification and that's good to know. So in some sense the DM is quite advanced then but still, if it's sort of digital between going straight or ricocheting I think the addition of a small randomly added angular disturbance at each obstacle would be a nice feature. In addition, I don't see it would take that much in terms of implementation effort either if a reduction of the velocity is done in the code already. Would be much more realistic IMHO, especially for those shot taken from dead astern. Edit: Come to think of it: If I understand the results of your testing correctly, the code for ricochet and speed reduction is already in there so why not add a branch to the ricochet code so one option is the complete ricochet with a low random probability and an "else" option in the code where you get the reduction in speed and a small angular disturbance added. Edited March 8, 2018 by Holtzauge
1CGS LukeFF Posted April 8 1CGS Posted April 8 Don't bump up old topics like this if you aren't contributing anything new and useful for the team to review.
Recommended Posts