Aces High Bulletin Board

General Forums => Wishlist => Topic started by: pervert on April 08, 2012, 12:36:32 PM

Title: A more accurate damage model.
Post by: pervert on April 08, 2012, 12:36:32 PM
I ran some tests on the 190 d9 and its ability to seemingly take rad damage at will, what I found out is that the nose spinner is set to trigger a radiator hit, you are actually less likely to get a rad hit by shooting at were the radiator actually is, with the top half of the cowling surrounding the radiator being less susceptible than the lower half, the front part being very susceptible and the spinner being a dead cert.

Problem is that the game doesn't take into account the angle the shot is fired from only that it has registered as a 'hit' the propeller is not modelled to take damage from weapons fire, given the size of the hub spinner it makes a very large target and even shot from behind and the side can register a hit anywhere even shooting the very tip of the hub.

50 cal it did not matter what distance it was fired, if it hit the spinner with just 1 round you had a radiator hit. Just for a laugh I bailed out and tried again with my .45 it takes 3 rounds of this ammo to register a 'hit' on the spinner.

Although it makes sense for me with frontal shots I am presuming this is were I am getting these 1k spraying radiator hits from, its like pressing a button, even from the front would a 50 cal round be able to penetrate the hub and spinner or even make it through a spinning prop every time to the radiator?

I heard WW1 was a testbed for a new damage system, is this purely cosmetic? Ie holes showing etc or does it take in to account were it has been hit from? That was over 2 years ago, is there any update or plans for a reworking of the damage model in the pipeline?

Put into the context of the game, this simply means because of the simplistic modelling holding down your fire button and hoing a D9 is a viable tactic, it only takes 1 of those thousands of rounds to ping the spinner and 2 minutes later its game over.
Title: Re: A more accurate damage model.
Post by: LilMak on April 08, 2012, 01:44:59 PM
.50 cal round will penetrate 1/4 inch steel @ about 2000 yards. Standard 6gun American plane will throw around 80 rounds per second. So a 190 at convergence of a p-51 should get hit with about 20 rounds of .50 cal in 1/2 a second or so. Chances of one of those missing the prop and hitting the rad might be better than you would think. Still a revamp of the damage model might be in order in the future. Control cables, prop damage, hydraulic failures, turbo/supercharger failure, control flutter, ect ect ect.
Title: Re: A more accurate damage model.
Post by: titanic3 on April 08, 2012, 02:53:11 PM
Instrument panel damage
Pilot doesn't blackout when wounded, instead, more sluggish controls or none at all (no rudder or being able to swivel your head around, etc etc)
Engine fires
Being able to open your cockpit (under a certain speed) to see after an oil hit.
Diving to put out a fire
Radio damage (no vox, can still type)
Partial damage (instead of the 100% or 0% we have now)
Bomb bay door jams (stuck close or opened)
Structural failure after damage is taken and the plane continues to pull large amount of Gs
Title: Re: A more accurate damage model.
Post by: Raphael on April 08, 2012, 02:57:28 PM
Instrument panel damage
Pilot doesn't blackout when wounded, instead, more sluggish controls or none at all (no rudder or being able to swivel your head around, etc etc)
Engine fires
Being able to open your cockpit (under a certain speed) to see after an oil hit.
Diving to put out a fire
Radio damage (no vox, can still type)
Partial damage (instead of the 100% or 0% we have now)
Bomb bay door jams (stuck close or opened)
Structural failure after damage is taken and the plane continues to pull large amount of Gs

Amazing! +1 to that
Title: Re: A more accurate damage model.
Post by: pervert on April 08, 2012, 02:57:58 PM
.50 cal round will penetrate 1/4 inch steel @ about 2000 yards. Standard 6gun American plane will throw around 80 rounds per second. So a 190 at convergence of a p-51 should get hit with about 20 rounds of .50 cal in 1/2 a second or so. Chances of one of those missing the prop and hitting the rad might be better than you would think. Still a revamp of the damage model might be in order in the future. Control cables, prop damage, hydraulic failures, turbo/supercharger failure, control flutter, ect ect ect.

I'm quite sure thats true, the bit I don't get is were the bullet changes direction and goes backwards into the radiator  :rofl
Title: Re: A more accurate damage model.
Post by: titanic3 on April 08, 2012, 03:06:51 PM
I'm quite sure thats true, the bit I don't get is were the bullet changes direction and goes backwards into the radiator  :rofl

It's almost impossible to fly directly, 90 degrees, behind a target. There is going to almost always be an angle at which the bullets are impacting. Especially with wing mounted guns, which fire at an angle to the point of convergence.

So the rounds might miss the rudder and tail, but might hit the cockpit or engine a millisecond later.
Title: Re: A more accurate damage model.
Post by: pervert on April 08, 2012, 03:27:18 PM
It's almost impossible to fly directly, 90 degrees, behind a target. There is going to almost always be an angle at which the bullets are impacting. Especially with wing mounted guns, which fire at an angle to the point of convergence.

So the rounds might miss the rudder and tail, but might hit the cockpit or engine a millisecond later.

I'm talking about the angle the bullet hits the spinner at not going through the plane, I know for a fact the radiator is not in the spinner/hub  :)
Title: Re: A more accurate damage model.
Post by: icepac on April 08, 2012, 05:42:14 PM
Isn't the radiator at the leading edge of the cowling?
Title: Re: A more accurate damage model.
Post by: pervert on April 08, 2012, 07:06:37 PM
Isn't the radiator at the leading edge of the cowling?

(http://www.clubhyper.com/reference/images/fw190engine090701bg_5.JPG)


Title: Re: A more accurate damage model.
Post by: MK-84 on April 08, 2012, 07:38:29 PM
Also to point out, is that hitting an aircraft from behind does not and should not mean that a component in the front of the aircraft doesnt get damaged.  A shot into the rear of an airplane doesnt have to mean that the bullet was stopped instantly. 
Title: Re: A more accurate damage model.
Post by: pervert on April 09, 2012, 05:54:50 AM
Also to point out, is that hitting an aircraft from behind does not and should not mean that a component in the front of the aircraft doesnt get damaged.  A shot into the rear of an airplane doesnt have to mean that the bullet was stopped instantly. 

Quite certain it doesnt work that way even if it did i doubt a 50 cal round could pass through an entire jumo engine then penetrate the rad.
Title: Re: A more accurate damage model.
Post by: PanosGR on April 09, 2012, 08:12:38 AM
109 or La7 slat damage. Never saw a hit resulting in a loss of slat control
Title: Re: A more accurate damage model.
Post by: Paladin3 on April 09, 2012, 12:33:34 PM
Instrument panel damage
Pilot doesn't blackout when wounded, instead, more sluggish controls or none at all (no rudder or being able to swivel your head around, etc etc)
Engine fires
Being able to open your cockpit (under a certain speed) to see after an oil hit.
Diving to put out a fire
Radio damage (no vox, can still type)
Partial damage (instead of the 100% or 0% we have now)
Bomb bay door jams (stuck close or opened)
Structural failure after damage is taken and the plane continues to pull large amount of Gs


+1!

Also... I would rather have to evaluate more if the enemy is no longer combat effective. If I get a PK many times I see a puff of an explosion and know to pull off and go hunting, saving ammo. I would like it to be harder to figure if the bad guy is dead or not.
Title: Re: A more accurate damage model.
Post by: Brownien on April 09, 2012, 02:36:06 PM
+1 for revampt damage model! especially if theres variable engine damage involved.
Title: Re: A more accurate damage model.
Post by: Paladin3 on April 11, 2012, 02:42:49 PM
That would be cool.

The other night I had a FM2 over an enemy field and was in a couple of tough fights with 109s and a P39. I went across the field (others were attempting a capture) and as I pulled up the ack tagged me... I immediatly knew something was wrong. The rear end floated some, but oddly. So I glanced around to see what was going on and saw i still had a tail, and all of my control surfaces, so I wrestled with the controls. Finally, realizing I was going in I wanted to know what was up and it said my vertical stab was gone. I looked back, and nope still there.

End of story - I like flying without the automated list open and trying to feel my way out of it while visually picking up the damage. Variable levels of damage would ROCK!
Title: Re: A more accurate damage model.
Post by: Skull_001 on April 13, 2012, 07:25:54 PM
+1
Title: Re: A more accurate damage model.
Post by: Noir on April 16, 2012, 10:09:34 AM
 :aok
Title: Re: A more accurate damage model.
Post by: TWC_Angel on April 16, 2012, 10:43:43 AM
+1!
Title: Re: A more accurate damage model.
Post by: Wmaker on April 18, 2012, 02:02:38 PM
Wow, that's an interesting find regarding the Dora! :uhoh

WTG for testing it pervert! :aok

I just always thought that damage was so frequent because of the placement of the radiator in general.

Hopefully the radiator's hit area gets looked at. Dora's spinner is huge compared to the overall area we are talking about.

A pic showing the radiators in different Dora subtypes. D-9 has Jumo 213A which can be seen on the left:
(http://i46.photobucket.com/albums/f147/Wmaker/Dora_radiator.jpg)

Title: Re: A more accurate damage model.
Post by: pervert on April 18, 2012, 02:37:42 PM
I actually filmed it but everytime I play the film to get a screenshot the camera is zooming all over the place wmaker but I filmed it in a custom arena so I do not know if that has anything to do with it  :mad: but if count the side profile hits as surface area then you really do have a huge hit zone for the radiator! No wonder it takes radiator hits all the time!

Its debatable even when it takes rounds head on if it hit the spinner if it would deflect of the spinner, or penetrate it? In the former a rad hit would have to be perhaps expected but the latter it would miss it completely. I still do not understand the rad hits from behind which drove me to carry out this test and could not replicate it in the tests with any great frequency, unless it is the ballistics of the plunging round? But shooting the top half of were the radiator is housed results more frequently in an engine oil hit? Shoot the bottom half and it is a pretty consistant rad hit as you would expect.

Just out of curiosity Wmaker is that radiator in 2 halves or a single rad? I would think it mad to have such an achilles heel on a warplane.
Title: Re: A more accurate damage model.
Post by: bustr on April 18, 2012, 06:19:59 PM
Is the spinner's back plate armored? If the backplate is armored is it thick enough to resist a .50 API or ball?

If not, does the 50cal then deflect up and into the radiator and along the reduction gear housing into the engine compartment? Or does the 50cal fragment and it's parts spray around and out into the radiator structures? What happens when a .50API, 20mm AP or 20mm HE or 20mm Minengeschoß hits the spinner or prop roots from the front?

No one has asked why it seems normal to be abel to hit the spinner/radiator with such repeated success in this game opposed to what kind of regional damage mapping is in place. Thats a very small target at the speeds and distances we fight in this game. I suspect the gemans determined that in real life also before putting the radiator in that spot.

------------------------------------------------------------------------------------------------------------------------------------------

How is the acceptance of damage programed for the frontal crossection at the coordinates describing the spinner, prop, radiator, and engine back to the wing leading edge region of the D9?

if round type xx hits coordinates x, y, z for aircraft 123
 then damage zzz is recorded and
 functions a-z are triggered?

Or is the aircraft mapped in granular damage regions described as a relationship to:

armor factor of region
type of round & destructive nature
angle of entry
velocity at contact
list of damage results for contacted region

Or is the aircraft mapped in All or Nothing damage regions and you are S-O-L getting hit with a BB from spinner tip to wing root?

-------------------------------------------------------------------------------------------------------------------------------------------

Offline when I was performing shooting tests against D9's in the drone circuit. I accomplished a single radiator/oil cooler hit by shooting too far forward from underneath hitting the engine compartment rather than the fuel tank region under the pilot. Hitting the engine region on top or from the side seemed to do no damage and I was inside of 100 on full zoom.

Is damage mapped differently from air to air sources versus from ground to air sources?

-------------------------------------------------------------------------------------------------------------------------------------------
Go here for pretty drawings: http://translate.google.com/translate?hl=en&rurl=translate.google.com&sl=ru&u=http://www.airwar.ru/other/draw.html

(http://www.albentley-drawings.com/images/FW190D9F.jpg)
Title: Re: A more accurate damage model.
Post by: Wmaker on April 18, 2012, 07:16:57 PM
Just out of curiosity Wmaker is that radiator in 2 halves or a single rad? I would think it mad to have such an achilles heel on a warplane.

Yes, the radiator was split to left and right halves.
Title: Re: A more accurate damage model.
Post by: Ruah on April 19, 2012, 05:39:50 AM
radiator damage is the most frustrating and poorly modeled parts of the game - which is why I have been using liquid cooled engines less and less. There is no way that any radiator damage should be registered if the opponent is shooting from your da or near 6 - the bullet simply has too much to travel through (and would do a lot more damage <i.e. kill the pilot, take out the engine, etc) to reach the radiator in the front.  

On 109s - both radiators (housed in the inner wing) are modeled for damage, but a hit to either one kills the whole system.  This means that a single ping to your inner wing is an instant dive home or ditch attempt.

I flew the typhoon a fair bit last tour for kicks - and I suffered maybe 3 radiator hits the whole time and never from a stray rear shot.  In a dora, 1 in 5 sorties ends with a radiator hit. . . so far this scenario I have not taken a radiator hit, but I consider myself lucky for that.

Yah. . .back to the sans-radiator plane.

and yes, the radiator was in 2 halves just like in the 109 for redundancy.  In the case of the 190, they were just placed a lot closer.
Title: Re: A more accurate damage model.
Post by: pervert on April 19, 2012, 10:47:50 AM
Thanks guys, I thought it was 2 but only from some user comments I had read elsewhere difficult to find that sort of info  :salute
Title: Re: A more accurate damage model.
Post by: Delirium on April 19, 2012, 11:46:09 AM
How did you test it, Pervert?
Title: Re: A more accurate damage model.
Post by: pervert on April 19, 2012, 12:23:24 PM
How did you test it, Pervert?

Custom arena KOTH map a jeep with a 50 cal shooting one round at a time at zack1234 and cobrajet, they spawned in a FW190 D9 and opened up damage and told me over range vox when they took damage and what it was, when they took a rad hit they respawned. The shots were mainly close range 50 - 100 yds so I could zoom in and be accurate, some of the longer range 1k shots were hampered by low res graphics and it became very hard to put a single 50 cal round accurately on the plane or see where it landed.

Once I seen the spinner was a dead cert hit zone for the rad I drove around taking shots at the spinner from different angles, it doesn't seem to change power on longer distances but the calibre of the round makes a difference, it took 3 rounds of a pilots .45 to get a rad hit were a .50 cal took just one.

(http://jpeghoster.com/images/54583612060600425491.jpg)

(http://jpeghoster.com/images/33325378422584546167.jpg)

(http://jpeghoster.com/images/20222905850404334736.jpg)
Title: Re: A more accurate damage model.
Post by: Delirium on April 19, 2012, 12:28:34 PM
Interesting... would you mind if I did something similar in testing the P38's propensity to suffer pilot wounds?
Title: Re: A more accurate damage model.
Post by: pervert on April 19, 2012, 12:52:40 PM
(http://jpeghoster.com/images/77028631394903926675.jpg)

Red zone is 100% radiator hit
Yellow zone is about 90% for a radiator hit
Blue zone is a toss up between radiator and oil with oil more being more likely than the radiator

(http://jpeghoster.com/images/74146605058457242572.jpg)

Red zone here represents the actually radiator position in the dora.

Visual aids!! while I have photoshop open and the jpeg hoster is working!  :)

When you paint the hit zones on the dora for the radiator compared to where the radiator actually is you can understand in game why it takes radiator hits so much
Title: Re: A more accurate damage model.
Post by: Delirium on April 19, 2012, 12:56:57 PM
I can understand why shots on the prop hub could spiral into the radiator, particularly when fired at from the front.
Title: Re: A more accurate damage model.
Post by: pervert on April 19, 2012, 01:04:48 PM
Interesting... would you mind if I did something similar in testing the P38's propensity to suffer pilot wounds?

At a guess I would say the head area is modelled as there are shots that result in an instant death and is very hard to hit, failing that from doing tests on the dora anywhere you hit round the cockpit or from behind will result in a pilot wound. I will certainely go in and play test dummy for ya Del, just give me a shout in game, a mouse would probably be more of an advantage for accurate shooting from a 50 cal.

I can understand why shots on the prop hub could spiral into the radiator.

I could live with that as well so long as the angle is correct shooting the tip of the spinner from behind it would be a very magic bullet that would deflect backwards and into the radiator. Also the fact that the radiator is spilt into 2 halves would suggest that it is some sort of margin of fail safe in the event that one of the radiators is punctured similar to the 109's.

How this translates in game flying a Dora constantly is something very irritating, within 2 minutes of getting this radiator hit your engine is dead. When I hear the magic rad hit sound I go all out knowing I have only 2 minutes to vent my fury at the modelling on someone! Take one up yourself and fly it for a few days you'll notice the pattern of rad hits starting to form with depressing frequency  :)
Title: Re: A more accurate damage model.
Post by: pervert on April 19, 2012, 01:16:39 PM
Bustr I think even though it would be nice to have that degree of modelling considering the amount of shells whizzing around it would be impossible to factor all that in mate. I'm no ballistics expert I don't even know if you could predict what would happen to a hub spinning at several 1000 rpm when a bullet hits it  :headscratch:

What I would like is the hit zones tightened up and the direction of an impacting round to be considered when determining what damage is taken.
Title: Re: A more accurate damage model.
Post by: pervert on April 19, 2012, 01:27:51 PM
Also if I think it was Krusty? could furnish us with were exactly the extra armour is supposed to be on a FW190 A8 (I remember seeing some drawings but have been unable to find them in search  :headscratch:) we could do a comparison test with an A5 to see if the extra weight that people complaining about the A8 is modelled.
Title: Re: A more accurate damage model.
Post by: bustr on April 19, 2012, 02:29:41 PM
Since you have shown that the expectantcy of a radiator or oil hit response for yourself is statisticly rather high in game. Along with the eye opening prop spinner visual demo. Did you test single shots along the engine compartment below the exauhst stack line back to the wing leading edge while the D9 was static?

I'm not sure how damage is mapped. It could be mapped to the 2D skin framework as coordinants or regions with hit values before the 3D wrapping. Your experiences would indicate to resolve your (motion/instantaious location) in the environment to the probability of being hit forward of the wing, that the whole area from the tip of the spinner back below the exauhst stacks responds as radiator or oil damage. At least it's no longer a hit bubble.

Can you test many different types of aircraft for the single hit to the spinner to see if it's included in the radiator or oil damage response? I understand that location is the issue with the D9 radiator. But, the spinner part if it holds true across many aircraft may answer indirectly if you have any chance to get the spinner removed from the damage equation with the D9.

Some nights my squadmates cannot go within 3k of an active airfield with a D9 for fear of a radiator hit. Other nights they dive through the ack with a D9 all night vulching the runway with only holes in noncritical areas. I've never had my radiator ack holed in a D9 but very often by other aircraft and wirbel. I'm not sure if some of this kind of damage might be for other reasons though I cannot see how latency or quality of connection would increase radiator/oil hit precentages.

Otherwise the 8th Airforce produced an analysis report of small arms AA fire statistics which directly stated fighters like the D9 and it's allied contemporaries were never designed with the expectaion of flying low through small arms AA fire or taking many hits to the engine compartment and survive. Those activities were better served by radial engines.
Title: Re: A more accurate damage model.
Post by: Noir on July 02, 2012, 10:51:35 AM
HTC please don't forget this for the upcoming version! I quick fix would be to double the hit points on the 190D9 radiator, to emulate the dual radiator  :aok
Title: Re: A more accurate damage model.
Post by: 10thmd on July 03, 2012, 12:54:48 PM
I'm curious as to HTC view on this matter. I always felt the d9 suffered from a glass jaw.
Title: Re: A more accurate damage model.
Post by: Hazard69 on July 04, 2012, 12:04:57 PM
Interesting... would you mind if I did something similar in testing the P38's propensity to suffer pilot wounds?

Please, pretty please do! :x :x

Most likely area for a pilot wound in my limited experience in the 38, seems to be the rear, lower rear fuselage and bottom of the cockpit. :noid

 :salute

Take one up yourself and fly it for a few days you'll notice the pattern of rad hits starting to form with depressing frequency  :)

Not just the Dora, Yak seems to have the same issue as well. How so many manage to aim so accurately on so tiny an airplane, I'll never know. :lol
Title: Re: A more accurate damage model.
Post by: Bruv119 on July 18, 2013, 04:08:02 AM
wish granted!      :aok
Title: Re: A more accurate damage model.
Post by: Zacherof on July 18, 2013, 04:10:47 AM
wish granted!      :aok
:banana:
Title: Re: A more accurate damage model.
Post by: Randy1 on July 18, 2013, 06:31:28 AM
Just what we need, more 190 vulchers.
Title: Re: A more accurate damage model.
Post by: olds442 on July 18, 2013, 07:51:17 AM
Instrument panel damage
Pilot doesn't blackout when wounded, instead, more sluggish controls or none at all (no rudder or being able to swivel your head around, etc etc)
Engine fires
Being able to open your cockpit (under a certain speed) to see after an oil hit.
Diving to put out a fire
Radio damage (no vox, can still type)
Partial damage (instead of the 100% or 0% we have now)
Bomb bay door jams (stuck close or opened)
Structural failure after damage is taken and the plane continues to pull large amount of Gs

Also super charger and turbo charger damage. Also control surfaces flapping and losing pistons