Jump to content
Sign in to follow this  
Venturi

P-47 Toughness Analysis

Recommended Posts

I fly mostly online and can almost swear that when I'm flying in certain aircraft I can get hit once and the blooming thing feels like its made of tinfoil and just fails apart. Invariably I get taken down an awful lot from being 1 shot pilot killed (particularly flying in the 190). However, when I'm the one shooting back, I can convince myself that the enemy aircraft (Allied or Lufty) is soaking up tons of damage

 

With regard to the P-47, having flown in it and also shot at it a fair few times purely from a personal view point in seems to soak up MG fire fairly well but it can't take too many Mk 108 hits and indeed should it, or any of the current aircraft, considering it's a bomber killer round?

 

I personally think that the most effective gun platform currently is the P-51. it just seems to swat enemy aircraft down with relative ease.

Share this post


Link to post
Share on other sites

Most of those hits were pretty solid hits, tho if it was only 20mm hits then they do preform quite well.

With the new pilot damage and endurance model, ive noticed that pilots from all sides are dying like flies... Was this like so in the war, too? 

Share this post


Link to post
Share on other sites
1 hour ago, [Pb]Slegawsky said:

P-51 damage model is a different story but I just hope Razorback will be much harder to disassemble.

 

 

Dude in those videos that P47 is getting absolutely hammered. I don't see any issue whatsoever with it coming apart if it's hit like that. You don't see German wings coming off much in the Western maps because it's mostly being hit by 50BMG. The Tempest has taken my wings off many times though. Think about it, with the late German planes you are being hit by 3 fast firing cannons, and most of the time if it's a 109, one of those is a 30mm. When I'm in the A8 I take 2x30, 2x20, and 2x13 cannons and sometimes I still need a second pass to drop a P47. Mustangs, Tempests, and P38s come apart nicely in one pass though.

Edited by III./JG7-MarkWilhelmsson

Share this post


Link to post
Share on other sites
4 hours ago, LukeFF said:

 

You may want to dial back the hyperbole just a tad. 

 

Probably: grumpy old man syndrome. 

 

 

 

 

   

Share this post


Link to post
Share on other sites

I've been focusing on flying the 47 lately.  I gotta say, I've gotten beat up quite a bit, but haven't experienced the fragility that was seemingly so evident in prior patches.  The only time I broke apart was from having my tail fair-and-square broken off by a ferocious cannon volley and a wing broke off after trying to pull heavy G with critical damage.  There was no being playfully slapped on the shoulder and insta-break.

The main thing I've experienced was instant engine death.  From what I can tell all were due to turbocharger damage and if I had unlinked the turbo in time, dropped it to zero and got the exhaust bypass outlet valves opened, it's quite possible I could have saved it.  It's hard to tell for sure though.  

 

At least that's my experiences with it so far.   

Edited by Mobile_BBQ
  • Like 1

Share this post


Link to post
Share on other sites

Hoping back and forth between the P-47 and the P-51 it really does feel like I can take hits in the P-51 that will send the P-47 home in a box. Just ran a set of QMB, all vs Fw-190A-5 with MG-FF cannons in both the P-47 and P-51.

 

First one in the P-47 I tunnelled on a 190 and got my tail shot off. 

 

Second one in the P-47 I tunnelled, took a hit and the engine died not long after. 

 

Third run in the P-51, also tunnelled and got raked early and went into a spin but recover a and shot down six of the Fw-190's.

 

That's pretty much the way it goes. 

Share this post


Link to post
Share on other sites
5 hours ago, Gambit21 said:

 

Tunneled?

Tunnel vision; locked onto a single thing and neglected to check six or my periphery. 

Share this post


Link to post
Share on other sites

Looking at real damage on photos and what we got here. 
If you overheat the oil it stop within 30 sec. 

same happen after a bullet in one cylinder. 
there are pictures and written statements that it could fly with one cylinder/ piston shot off with loss of massive amount engine oil. 
I think we are looking at a simplified dm not specific divided between radial engines and water cooled engines. 
It is no doubt that the sensitivity of engine is bullshit. I think in some way it need to be made more reliable as it was known for. 
I fly P 51 and it goes down just as fast but in one piece not in small bits like p 47. 
So things are just opposite of what I expected. 
And I find it very unbelievable and problematic. 
it is to me not the fact that it cant take 30 mm. It is more like you flying a plane known to take the pilot back but we fly a glass   I accept being shot down. But the way it happened is just way off. Hope it get fixed when they make razorback

EABB0DDB-4610-4AC9-A065-96540AF9B9BA.jpeg

Edited by No.322_LuseKofte
  • Upvote 2

Share this post


Link to post
Share on other sites
2 hours ago, No.322_LuseKofte said:

Looking at real damage on photos and what we got here. 
If you overheat the oil it stop within 30 sec. 

same happen after a bullet in one cylinder. 
there are pictures and written statements that it could fly with one cylinder/ piston shot off with loss of massive amount engine oil. 
I think we are looking at a simplified dm not specific divided between radial engines and water cooled engines. 
It is no doubt that the sensitivity of engine is bullshit. I think in some way it need to be made more reliable as it was known for. 
I fly P 51 and it goes down just as fast but in one piece not in small bits like p 47. 
So things are just opposite of what I expected. 
And I find it very unbelievable and problematic. 
it is to me not the fact that it cant take 30 mm. It is more like you flying a plane known to take the pilot back but we fly a glass   I accept being shot down. But the way it happened is just way off. Hope it get fixed when they make razorback

EABB0DDB-4610-4AC9-A065-96540AF9B9BA.jpeg

Agreed, the engine DM imo is the worst part of the P-47 DM.

 

P-47 got hit and lost a few cylinders and still made it back to base safely.

w-397-hendricks-damage-2-jpg.410145

w-397-hendricks-damage-3-jpg.410146

 

I think this is 37mm (report says 40mm but could be less) damage to both wings.

w-395-rife-rt-wing-close.jpg

w-395-rife-left-wing.jpg

Edited by Legioneod
  • Upvote 2

Share this post


Link to post
Share on other sites

 

That's the Jug I know and love...and want.

 

 

 

 

  

5 hours ago, Voyager said:

Tunnel vision; locked onto a single thing and neglected to check six or my periphery. 

 

 

Target Fixation

  • Upvote 3

Share this post


Link to post
Share on other sites
18 hours ago, Legioneod said:

Agreed, the engine DM imo is the worst part of the P-47 DM.

 

P-47 got hit and lost a few cylinders and still made it back to base safely.

 

 

 

I think this is 37mm (report says 40mm but could be less) damage to both wings.

 

 

It's possible it is a 40mm Bofors. Germans captured hundreds of them during the war (Poland, BEF, Norway) and when the Germans captured the Kongsberg factory in Norway, the factory kept producing guns and repairing 40mm guns and ammunition for the Germans.


file.php?id=4300&sid=dbac281ce147ff04b74c4fc82b36b725
Image result for 4 cm Flak 28

  • Thanks 1
  • Upvote 1

Share this post


Link to post
Share on other sites

I started watching the video and figured I’d never make it to the end and give up watching, but not so! Thanks so much for posting it irregardless of whether you think the P-47 is modeled correctly for damage or not. I think the guy who produced it went to a lot of work researching the ‘47 and damage and is pretty well on the mark. Interesting side notes as well. Gonna see if I can check out the other videos he made as well.

 

I must confess though, I haven’t flown the P-47 much at all and have always thought of the big fat Jug as one of my WW2 faves, if not THE favorite of that era. For some bizarre reason I’ve probably logged many more hours in the P-40 I guess perhaps cuz of all the babysitting it takes to keep it running smoothly thus keeping me a bit busier? Things in my private life have overruled me getting on the PC at all, ever since final release of Bodenplatte even...cringe...I know...smack me with a wet noodle. I just need to put real life on hold and jump back in that VR cockpit.
 

Anyways, interesting note on the tail warning alarm. First time I’ve heard of it. 
 

I guess the only thing I do know is that after watching that video I don’t envy the devs at all on damage modeling. With all the mathematical hypothesis available and thousands of different scenarios based on projectile type, angle of penetration, velocities, deflection capabilities, aircraft material types, etc., etc. let alone the constant groans coming from the forum if I were a dev sorting through it all I’d probably be divorced and an alcoholic by now. No. Seriously.

  • Haha 1

Share this post


Link to post
Share on other sites

I agree it is not easy but the overall results still have to be believable.  I have posted elsewhere about the P-47 DM vs the US Ballistics Laboratory results, but on a slight different tack it is possible to compare DMs in a systematic manner by limiting variables: something that anecdotes from playing experience cannot do.

 

Using my old Airfield Attack test, in which a number of planes make low passes over an airfield defended by LAA, it is possible to change one variable at a time and count the results. The only thing changed here is the plane model: the waypoints and LAA are identical.   Since the LAA only fires HE, we can get a measure of vulnerability to just one kind of damage type. With this test it is difficult to count and locate individual hits, or their results in detail: that would take the enormously time consuming single aircraft tests, which I am not prepared to do again. What you get is the number of aircraft hit, and of those, the overall results. 

 

I have done 5 runs each with the P-47 and P-51. I am 90% sure that the difference in results is a fair reflection of the DMs, although obviously the bigger the sample the more sure you will be, so I aim to do another five runs each. Here are the results so far, hope you can read them.  At Rally means the aircraft made it away from the target area, in control. Lt Damage = superficial damage decals and/or missing ailerons etc. Hv Damage = anything streaming fuel or smoke, most of which would lead to aircraft loss before long in the game.

 

On the basis of runs to date, P-47s are somewhat more likely to be hit, and if they are hit, much more likely to be downed than P-51s.  The first variable is just a function of their size - the courses and speeds are identical. It is the second variable that looks odd.

 

Capture.thumb.JPG.47d16d77428c198a43bc5dac837fe9c1.JPG

 

Edited: table now shows results of 10 runs each, confirming the pattern. While there is considerable variation in outcomes from one run to another, which is why these tests have to be run with a good sample size, P-47s are over four times as likely to fail to make it to the rally point. If you count all heavily damaged aircraft as lost, P-47s are twice as likely to be lost.  In terms of surviving a BoX ground attack mission, the P-51 is a better bet.

 

This is partly explained by the fact that P-47s, being bigger, are easier to hit. What you cannot see from this test is how often an aircraft hit, is hit more than once by the same burst.   It does look, however, as though vital systems are being hit more easily in the P-47, or when hit, failing more easily.  Disentangling these two possibilities requires a single plane test.

 

It would be easy enough to run these Airfield missions for a German radial and in-line. 

 

Edited by unreasonable
  • Like 1

Share this post


Link to post
Share on other sites

Someone on the dev team needs to do a run through of all the DMs and FMs, and see if they all still make comparative sense, given their variations in age of creation...

 

I don’t think anyone here is surprised with the results. And the devs know already. 

Thanks for the work. 

Share this post


Link to post
Share on other sites
On 12/7/2019 at 10:54 AM, unreasonable said:

   It does look, however, as though vital systems are being hit more easily in the P-47, or when hit, failing more easily.  Disentangling these two possibilities requires a single plane test.

 

It would be easy enough to run these Airfield missions for a German radial and in-line. 

 

 

Part of the difficulty with testing the P-47 systems lies in determining exactly what systems have been hit. 

If the engine is dying quickly from oil system or engine/cylinder hits vs. a turbocharger hit, the expected result could vary wildly. 

Ideally, if the engine has lost too many cylinders or the "axle mount" the whole assembly spins on is destroyed the engine should fail rather quickly. 

If the oil system is hit or only 1 or 2 cylinders are destroyed, there would still be time. 

If the turbocharger's internal blades are stopped and the exhaust bypass outlets aren't opened in time, the engine exhaust has nowhere to go and should result in a quick failure.  It's the same as putting a potato in a car's tail pipe. 

If the turbo is damaged but the mechanism still 'windmills' letting some exhaust out, there should be a significant drop in power until the exhaust bypass outlets are opened - provided the damaged turbo unit still lets exhaust through at a reasonable rate.  It would kind of be akin to a 90% clogged sink drain. Getting the outlets opened should (obviously) subtract the power of the turbo but, if the actual engine and oil system are undamaged, it should still be able to run without failure at non-turbocharge-boosted peak power.

 

TL;DR  If the engine is in fact modelled accurately, P-47 engine damage testing is going to be nearly impossible if it cannot be determined by the tester what point(s) of the engine system(s) have been damaged.  

Edited by Mobile_BBQ
  • Thanks 1
  • Upvote 1

Share this post


Link to post
Share on other sites

@unreasonable, if you have time maybe add 10 runs of a Yak, or Lagg or something. Just for the purposes of a larger data set, and just in case the Mustang is unusually tough, and to isolate the Jug a bit more.

Share this post


Link to post
Share on other sites

If I remember, in IL-2, radial engines are a degree tougher than water-cooled inlines as far as damage modelling, this was stated by Dev's. 

 

R2800 is an 18 cylinder compared with all other radials in game which are 14 cylinder. 

 

I would guess that currently it shares the same damage formula as all other radials without the benefits of higher cylinder count. 

 

Going on past experience I would think a future update will give more granular damage model to cover the wider engine variety that we now have, of course sooner would be better than later. 

But a big overhaul of engine damage with detonation modelled would be another of my guesses now that other things like pilot physiology have been introduced successfully. 

 

Cheers, Dakpilot 

Share this post


Link to post
Share on other sites

Well to me, P-47 results do not seem that bad from Unreasonable's testing. As Gambit21 suggested some other aircraft should also be tested to prove if P-51 isn't just too tough. I would suggest testing Il-2, which should be a standard for a though combat attack plane. If the results from P-47 are similar to Il-2 then there might be a problem with P-51 and not necessarily with P-47.

Share this post


Link to post
Share on other sites
On 12/5/2019 at 4:51 AM, unreasonable said:

divorce maintenance, mortgages, substances  and mistresses to pay for

 

(sigh) what an exciting, modern life those software people live! Well, for me it's back to my coal-shovelling wife, mine-working kids and the good, boring old railroad building job. 

 

At least we get to sing a lot. "Gonna see miss Eliza/gonna go to Mississipi"

Edited by danielprates

Share this post


Link to post
Share on other sites
2 hours ago, Rudolph said:

Well to me, P-47 results do not seem that bad from Unreasonable's testing.

28 downed aircraft against 6 for the P-51? Heavy damage is comparable (18:17), but instant kills differ alarmingly.

 

2 hours ago, Rudolph said:

As Gambit21 suggested some other aircraft should also be tested to prove if P-51 isn't just too tough. I would suggest testing Il-2, which should be a standard for a though combat attack plane. If the results from P-47 are similar to Il-2 then there might be a problem with P-51 and not necessarily with P-47.

Fully agree, but such tests are time-consuming, so they'd be better done by the community. I'd gladly contribute. The more plane types, the better, the more test runs, the better.

Share this post


Link to post
Share on other sites
2 hours ago, Rudolph said:

Well to me, P-47 results do not seem that bad from Unreasonable's testing. As Gambit21 suggested some other aircraft should also be tested to prove if P-51 isn't just too tough. I would suggest testing Il-2, which should be a standard for a though combat attack plane. If the results from P-47 are similar to Il-2 then there might be a problem with P-51 and not necessarily with P-47.

I'd recommend reading the test report of the P-47 ballistics testing vs various types of rounds, it doesn't match up with what we see in game, though I understand it's difficult to translate into a DM.

Imo if they can get the P-47 correct then they can use it to judge other aircraft DM, whether they be too resistant or too fragile since the P-47 had actual testing done to it to determine probability of survival.

Edited by Legioneod

Share this post


Link to post
Share on other sites
6 hours ago, Gambit21 said:

@unreasonable, if you have time maybe add 10 runs of a Yak, or Lagg or something. Just for the purposes of a larger data set, and just in case the Mustang is unusually tough, and to isolate the Jug a bit more.

 

I had a start to do exactly that by creating a Fw190 A8 and Bf109K4 version of the mission, but when I ran them I found that several of the planes missed waypoints at some point, circling instead, thus invalidating the comparison.  No idea why this happens - 109s worked fine in the old tests - unless it is something to do with pilot physiology, but I am not asking the AI to make hard turns. Posted about that in the ME section. 

 

I will make a version with a Spitfire and a Yak as well, see if it is only the US planes that can now fly this mission.

 

7 hours ago, Mobile_BBQ said:

 

TL;DR  If the engine is in fact modelled accurately, P-47 engine damage testing is going to be nearly impossible if it cannot be determined by the tester what point(s) of the engine system(s) have been damaged.  

 

What we can do is see whether the overall results match our sources - in the case of the P-47 they do not, see my earlier thread for details.  If the in lines are turning out to be much safer ground attackers than radials, then something is wrong. It is up to the developers to tweak the details. 

  • Upvote 1

Share this post


Link to post
Share on other sites

The DM modelling in computer game is hard for one simple reason. 

 

You have component hit boxes and HP pool on them, the more granular you go, the more accurate you get, but there comes a point where you have to do compromises. Hence, when you have an object that is big, even with lots of durability, it still can seem weaker because the way you model the other things. 

 

When HE ammo has X amount of secondary effects (shrapnel), a larger object, even with nominally better durability, can be weaker in absolute because it will absorb more damage by the virtue of its size. 

The P-47 might be sturdier on paper, but by the virtue of it size, it catches more damage in total, and seems to be weaker when actually being shot at. 

 

If you have 100 hp on wing spar, but it is so small, that you only catch small % of shrapnel, vs 200 hp wing spar that hoovers up more % of the shrapnel, the end result might feel like the sturdier spar is weaker. 

 

This is issue in many games that include granular hitboxes in its hull. Mechwarrior is poster-boy of those games, where huge HP pool parts that are bullet magnets can seem like weak tissue paper, and small hitbox parts with no hp to write home about might seem indestructible. 

 

I am 80% sure that the granularity of component hitboxes play a big role of how the DM of p-47 might feel a bit off, even if there is effort made on the devs part to make the individual components stronger than average. 

 

Edited by Cpt_Siddy

Share this post


Link to post
Share on other sites
3 hours ago, Cpt_Siddy said:

 

 

I am 80% sure that the granularity of component hitboxes play a big role of how the DM of p-47 might feel a bit off, even if there is effort made on the devs part to make the individual components stronger than average. 

 

 

That is very likely true, but does not explain why single hits on an engine, for example, have a much higher probability of stopping it than they should.  If you have an estimate that one 20mm HE shell hit on an engine has a 10% probability of stopping it, then your "if hit then engine takes RNG*xHPs" distribution should have 10% of outcomes over the "stop engine" HP threshold.  It also does not explain why splinter hits from a wing hit can cause a leak in a fuel tank designed to defeat 50 cal direct hits.

 

Anyway, just sticking to observable results, I have managed to make Spitfires IXes and, with a bit more waypoint tweaking, 109Ks behave themselves, so I can do comparative runs on those two in line types. Should be able to post results Tuesday, after which I will try to tame the Fw190s, which really did not like my original mission. 

Share this post


Link to post
Share on other sites
On 12/5/2019 at 4:33 PM, BlitzPig_EL said:

Sooner I hope.  None of our guys will fly the thing anymore.  We are all about ground pounding, and it's sad that what should be the best single engine attack machine in the ETO is less survivable than the Mustang.  Hell, I'd rather take the P40 to be honest.

 

The P-47 is my favorite WWII plane, but I rarely bother with it in game. It's a real pity. The 3D model and the sounds are great in IL2.

Share this post


Link to post
Share on other sites
55 minutes ago, Jade_Monkey said:

 

The P-47 is my favorite WWII plane, but I rarely bother with it in game. It's a real pity. The 3D model and the sounds are great in IL2.

 

Well, according to the test, it is weaker than a Mustang, which is not right, however out of 460 planes flying straight over 20 mm FLAK only 28 were downed out of 91 hit. This does not make it a glass plane, maybe something is off in the Mustang and will be tweaked, maybe the DM of P-47 will be tweaked, if testing shows further discrepancy. I am quite curious about the other planes and their results (good job Reasonable). Devs are definitely willing to change things, P-47 DM has already been tweaked in 3.008 and we have a responding tester here:

 

Edited by Rudolph
  • Thanks 1

Share this post


Link to post
Share on other sites
3 hours ago, Rudolph said:

 

Well, according to the test, it is weaker than a Mustang, which is not right, however out of 460 planes flying straight over 20 mm FLAK only 28 were downed out of 91 hit. This does not make it a glass plane, maybe something is off in the Mustang and will be tweaked, maybe the DM of P-47 will be tweaked, if testing shows further discrepancy. I am quite curious about the other planes and their results (good job Reasonable). 

 

 

Thanks - positive feedback is always motivating! 

 

As to whether the results make the P-47 a glass plane depends on your reference point. US Ballistics Lab tests estimated quick kill probability of 6% and 7.6% by a single hit for the two 20mm types they tested. Going up to 12% and 14% including all aircraft that would not RTB (my "losses", roughly).  This was only from one angle, but it is still the best data we have. 28/91 is 31%.  Some proportion of the P-47s hit in my test were hit by more than one shell, especially the ones downed, so let's say the actual number of hits was double the number of planes hit: 28/182 =  15%

 

So on that basis the P-47 is about half as robust as the US Ballistics Lab thought it was.  Interestingly that is close to the ratio of BoX/Report lethality I discovered in my in depth tests of the damage from single hits. 

 

Edit:

 

Capture.thumb.JPG.b60243b1c87b160e9af885a542d48254.JPG

 

Here is the summary of test results to date including the 109K and Spitfire IXe.

Edited by unreasonable
  • Like 2
  • Thanks 1

Share this post


Link to post
Share on other sites
4 hours ago, Jade_Monkey said:

 

The P-47 is my favorite WWII plane, but I rarely bother with it in game. It's a real pity. The 3D model and the sounds are great in IL2.

 

Concur.  I love the aircraft and how it's been beautifully modelled in this sim, but only use it for the occasional non- combat pleasure flight offline.  It's just not something that's capable of filling the role it really played in real life.

 

 

Share this post


Link to post
Share on other sites

Added P-47, P-51, Spitfire IXe and Bf109 K4 results summary to post above. Will add Tempest and Fw190 A8 soon.

 

Since this is a general comparison I will start a new thread when I have done the full set of BoBP single seaters.

 

Edited by unreasonable

Share this post


Link to post
Share on other sites
17 hours ago, unreasonable said:

 

That is very likely true, but does not explain why single hits on an engine, for example, have a much higher probability of stopping it than they should. 

Bigger parts = more % that something important will be hit + bigger % more of shrapnel will hit it from that single hit. 

 

Bigger hit boxes on important parts explain perfectly why they blow up in computer games. This is literally the feature that i just mentioned that is issue on any game that does not make parts granular enough. Did you read my post? 

Share this post


Link to post
Share on other sites
Just now, Cpt_Siddy said:

Bigger parts = more % that something important will be hit + bigger % more of shrapnel will hit it from that single hit. 

 

Bigger hit boxes on important parts explain perfectly why they blow up in computer games. This is literally the feature that i just mentioned that is issue on any game that does not make parts granular enough. Did you read my post? 

 

Yes I read it: did you read my post before making your snarky reply?  A larger object will absorb more of the damage in real life, as well as in a computer game. It will also be, other things being equal, more robust.  It is up to the developers to set the HPs of each part, whatever their granularity, so that hits from a given weapon type produce believable results. This they demonstrably have not done.  It does not matter how granular the pieces are if the distribution of damage points is too high relative to the HPs.

 

 

Share this post


Link to post
Share on other sites
6 minutes ago, unreasonable said:

 

Yes I read it: did you read my post before making your snarky reply?  A larger object will absorb more of the damage in real life, as well as in a computer game. It will also be, other things being equal, more robust.  It is up to the developers to set the HPs of each part, whatever their granularity, so that hits from a given weapon type produce believable results. This they demonstrably have not done.  It does not matter how granular the pieces are if the distribution of damage points is too high relative to the HPs.

 

 

 

 

Yes, but in real life, the larger objects don't share common HP pool and are granular. R-2800 has many parts like cylinders, air intake manifolds, exhaust manifolds, cooling air guides, etc...

Any of those parts can absorb hit and not lead to catastrophic engine failure. In game, it is hard to differentiate the hit on such parts, more you add, the more computationally heavy your sim becomes. This is why large parts, even with substantially larger HP pool, are still prone to getting smoked compared to small parts with small hp pool that are hiding behind some other parts. 

 

P-51 might have an engine with 1 hp, but as long it is small, and covered by other, non essential, parts that might absorb the damage, it will survive more encounters with enemy fire. This is issue in every game that has granular damage model but compromise in parts granularity. 

 

Edited by Cpt_Siddy

Share this post


Link to post
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
Sign in to follow this  

×
×
  • Create New...