Thoughts on Damage Model

[Saxman]

Yes, I know, but its the best I could come up with. .25 increase for every 25% damage isn't enough, as you will only hit 1.5% loss of effectivness per % damage cause, and that is the max you can do with each damage %. I know what I've stated isn't perfect, but its better than the above.

What do you guys think of the idea as a whole?

That's why 1% damage = 1% effect is the best option. Damage a wing at 25%, it loses 25% of its lift. Damage it to 50% it loses 50%, etc. The damage directly correlates to the effect.

I'll certainly admit that it IS simplified as compared to what the real-life effects would be, but under the circumstances I think this is the most flexible and easily implemented approach. It's purely mathematical, meaning there's no real basis for complaint about someone saying the game randomly tore their wing off, and it's good enough to simulate the effects of having your wing riddled with bullets.

[Nemisis]

That's why 1% damage = 1% effect is the best option. Damage a wing at 25%, it loses 25% of its lift. Damage it to 50% it loses 50%, etc. The damage directly correlates to the effect.

I'll certainly admit that it IS simplified as compared to what the real-life effects would be, but under the circumstances I think this is the most flexible and easily implemented approach. It's purely mathematical, meaning there's no real basis for complaint about someone saying the game randomly tore their wing off, and it's good enough to simulate the effects of having your wing riddled with bullets.

yes, but mine wouldn't be random either, and I do admit it would me harder to implament. I'm just suggesting a more realistic option. What happens when you shoot up an alreay shot up wing? each bullet causes more problems then one hitting an undamaged wing. Your tearing off MORE of the already lacking flight surfaces which means that it hurts you more then a bullet would if your in a fresh plane.

[EskimoJoe]

I have a feeling HiTech is sipping coffee from his I heart HTC mug, slowly reading over this thread and raising his eyebrows at some of the interesting points brought up in this thread...

Hey, nobody said I couldn't wish here in the wishlist! 

Give ye olde damage model an update, eh? 

[Nemisis]

I bet everytime someone starts an "IN" thing, the moderators start laughing there tulips of cause we just told them we expect them to close it.

[Saxman]

yes, but mine wouldn't be random either, and I do admit it would me harder to implament. I'm just suggesting a more realistic option. What happens when you shoot up an alreay shot up wing? each bullet causes more problems then one hitting an undamaged wing. Your tearing off MORE of the already lacking flight surfaces which means that it hurts you more then a bullet would if your in a fresh plane.

The key here is to balance realism with playability and keeping things simple enough for to be done without enormous expenditures of processing power. Yeah, it would be great if airfoil surface damage responded EXACTLY as it would in real life, but that would be unfeasible. I also don't expect it to ever be quite to the level that Il-2's is, because while Il-2 has the benefit of only rendering for only one to a handful of players at a time, Aces High needs to track this information for hundreds of players at once over a network connection. The simpler the equation is the better.

The other problem with % effect increasing exponentially as % damage does, rather than effect increasing with damage linearly, is that eventually you'll reach a point where % effect will exceed 100%. For example, if you start at 1% damage = .25% effect, and double the effect with each % point of damage, You'll reach 128% effect at only 10% damage.

A MUCH better way to do it would be this:

Rather than having effect increase exponentially with the amount of damage inflicted, have DAMAGE increase exponentially with the number of hits. Say, a .50cal round causes 1pt of damage baseline. You fire 100 .50cal rounds at a target, and all 100 hit. Rather than each round causing 1pt of damage for a successful hit, as the part takes fire the amount of points of damage inflicted by the round increases to reflect the weakening of the part. This will also account for other factors such as range, hit pattern, etc. (so if you hit the target in a five foot zone on the wing at long range it may only cause 4% damage if the wing is undamaged. However if you make the same shot while the wing is damaged above a given level, a hit under those same conditions may cause 25% damage instead).

That way, % effect can never exceed % damage, while at the same time, weakened structures will be more susceptible to continued punishment.

[Nemisis]

The key here is to balance realism with playability and keeping things simple enough for to be done without enormous expenditures of processing power. Yeah, it would be great if airfoil surface damage responded EXACTLY as it would in real life, but that would be unfeasible. I also don't expect it to ever be quite to the level that Il-2's is, because while Il-2 has the benefit of only rendering for only one to a handful of players at a time, Aces High needs to track this information for hundreds of players at once over a network connection. The simpler the equation is the better.

The other problem with % effect increasing exponentially as % damage does, rather than effect increasing with damage linearly, is that eventually you'll reach a point where % effect will exceed 100%. For example, if you start at 1% damage = .25% effect, and double the effect with each % point of damage, You'll reach 128% effect at only 10% damage.

A MUCH better way to do it would be this:

Rather than having effect increase exponentially with the amount of damage inflicted, have DAMAGE increase exponentially with the number of hits. Say, a .50cal round causes 1pt of damage baseline. You fire 100 .50cal rounds at a target, and all 100 hit. Rather than each round causing 1pt of damage for a successful hit, as the part takes fire the amount of points of damage inflicted by the round increases to reflect the weakening of the part. This will also account for other factors such as range, hit pattern, etc. (so if you hit the target in a five foot zone on the wing at long range it may only cause 4% damage if the wing is undamaged. However if you make the same shot while the wing is damaged above a given level, a hit under those same conditions may cause 25% damage instead).

That way, % effect can never exceed % damage, while at the same time, weakened structures will be more susceptible to continued punishment.

That looks to be a good compromise, I like it. And BTW, I wasn't suggesting exponential increase in damage done by rounds. I was saying that at 10% the structure would weaken enough, combined with previous damage weakening cruicial parts, that it would likely be .75% efectivnes lost for every % damage done. And likewise it would hit 1% loss of efectivenss for every % of damage done untill 25% is reached at that point, the damage done would likely be great enough to cause weakening of each sequential part.

[AirFlyer]

That's why 1% damage = 1% effect is the best option. Damage a wing at 25%, it loses 25% of its lift. Damage it to 50% it loses 50%, etc. The damage directly correlates to the effect.

I like this idea, but it does have one problem that would end up being odd. When damage to a surface begins to reach the larger end of the percentage range (80 - 100) it's going to seem a bit unrealistic. If a wing has 99% damage done to it, that means it would lose 99% of effective lift, which in reality would make it entirely useless. On the other hand, even if it did have 99% damage, with this system the wing would still technically be there and likely be creating more then 1% lift of its original potential in realism. I can't think of how to put it in an equation but I think the only way to solve this would be once the percentage of damage begins to climb into its higher end, the Damage:Effect ratio would need to lessen to make things a bit more realistic.

[Nemisis]

so a graph of effectivness should look like this in your oppinion? would look like this:

[Boozeman]

Booze,

Y'know I could have sworn my original post did this already....

Scenario 1:

An F4U-1C and F4U-1A both put a half-second burst into the inner half of a target's left wing--say, a P-47--under full convergence range. Using my settings, that's at 200yds. The 1C will likely take almost any opponent's wing right off at that range. Depending the target, however, the 1A may not have caused enough damage to finish off the airfoil. Under the current model, the enemy escapes and continues to fight without problems and possibly gets a shot of its own at the Corsair before it can be finished off.

Now, let's change the damage model so that the amount of damage caused directly correlates to the amount of lift lost:

The 1A puts in his half-second burst and fails to destroy the wing, however the hit was sufficient to do 60% damage to the airfoil. The target has suddenly lost 60% of the lift produced by that part of the wing, leading to a stall of the left wing. The pilot is unable to recover, perhaps because the aircraft was already at a high angle of attack while maneuvering against the 1A, and the stalled wings snaps the aircraft into an unrecoverable spin.

Scenario 2, with Wing Spar damage added:

The 1C and 1A both put their half-second bursts into the wing, striking the main spar. The 1C's cannon destroy the spar and the wing shears off. The 1A, meanwhile, not only shreds part of the wing's surface (damage to the wing spar shouldn't take the place of damage inflicted to the airfoil itself) but seriously damages the main spar, perhaps within 60% destroyed. Although the P-47 is flyable, he can no longer press the fight as the damaged spar wouldn't be able to withstand the G-loading of hard maneuvering and threatens to snap under the stress. Compounding the situation is the shredded airfoil reducing the amount of lift the wing can generate, causing the aircraft to roll to one side. While the P-47 may still be flyable, in all probability he's out of the fight (perhaps the creak that plays under high G-loads can occur at increasingly lower airspeeds and G-loads as an audible queue of damage to the spar?)

This is why I say the effectiveness of cannon have to some degrees been exaggerated by AH's damage model. I'm not arguing that cannon wouldn't inflict more damage, my argument is that the all-or-nothing, your wing is either there or it isn't damage modeling makes machine gun fire less effective than it SHOULD be. Cannon would remain more potent, but machine guns gain an additional benefit by not needing to destroy a component entirely to have an effect.

Nemisis,

The problem there is what is the rate of increase? An equal, tit-for-tat correlation between the amount of damage inflicted and amount of lift/effectiveness lost is cleaner, easier to understand, and less arbitrary.

Saxman, I get your point, but still, no bias for cannons in your scenario. In any of your cases, the cannon armed bird will always do more damage - with the same ratio as it does now.

What your proposal essentially does is lowering the effective damage threshold - for both weapon types. It will not alter the effective power difference between them. So even if you get noticeable damage for machine gun fire earlier, the cannons will still disable/rip-off that part much earlier - just as it is now. Or from a more pessimistic POV, if you don't get enough machine rounds on target, that part may not be crippled at all. But the same with a cannon, it may very well cripple that part. You see the ratio does not change.

Apart from that, I'd welcome a higher fidelity damage modeling like your suggestion. But in that case I'd also lower weapon effectiveness across the range. It's is too easy to kill in AH, it should not be any easier than now. 

[OOZ662]

You're looking at it from a top-to-bottom view. "No matter what, a cannon will always deal more damage." Look at it from bottom-to-top: "A machine gun will always deal less damage." Until you get the kill, there is absolutely no benefit at all to landing MG rounds on target. It's essentially the same as waiting for the Golden BB, because you can (theoretically) pour hundreds of rounds all over a plane and have it not bother him at all. Not only is that entirely unrealistic, but it gives a slight disadvantage and "loss of reward" for firing MGs (reward for cannons being structural failure...why not reward the MGs with chipping away at performance?).

I agree with this wish, and add that the Ctrl-D list should display dark yellow (to make it more easily visible on the light blue sky) text for damaged but not failed components.

[Anaxogoras]

I like this idea, but it does have one problem that would end up being odd. When damage to a surface begins to reach the larger end of the percentage range (80 - 100) it's going to seem a bit unrealistic. If a wing has 99% damage done to it, that means it would lose 99% of effective lift, which in reality would make it entirely useless. On the other hand, even if it did have 99% damage, with this system the wing would still technically be there and likely be creating more then 1% lift of its original potential in realism. I can't think of how to put it in an equation but I think the only way to solve this would be once the percentage of damage begins to climb into its higher end, the Damage:Effect ratio would need to lessen to make things a bit more realistic.

I had the same thought.  Make it damage %/2.

[Boozeman]

Not only is that entirely unrealistic, but it gives a slight disadvantage and "loss of reward" for firing MGs (reward for cannons being structural failure...why not reward the MGs with chipping away at performance?).

This is the error of your thinking - in the "new model" a cannon round (unless of course the big calibers, but those are essentially grenades anyway) will also start with chipping away performance and not causing structural failure at once - just as a mg round, but at a 3-4 times higer rate. The ratio remains the same as it is now, it's just the planes that "cripple" sooner.   

If anything, if done correctly, cannon round will do much more damage than than mg rounds as we have now. A 20 mm HE blast for instance will tear more wing area off than 3-4 mg bullet holes.   

Just look what a 30 mm cannon round does to a wing:

http://www.youtube.com/watch?v=ZoLLDi-M3fk

For reference, a German 20 mm mine carries 1/4 th of the HE material as this 30 mm mine, so overall results will be much less severe, but still magintudes above than what a single MG round will ever do. 

[Saxman]

Booze, you're still missing the point. Under the current model you're more likely to see instant results from firing cannon than with MGs. A part is either there or it isn't. The threshold for seeing results from MG fire is SIGNIFICANTLY greater than with cannon, where only a couple hits is often enough to break something.

I'm fully aware that landing a cannon hit is going to degrade performance under the new model as well. However since it's also much more common to knock a part off with only 2-3 cannon hits outright, the cannon aren't going to take as much advantage of this as MGs, which may take a good 100rpg to deal damage under the current model. I mean seriously, how often do you hit a fighter in the wing with a 30mm in the game and that plane remains flyable, anyway (barring incidents of rubber bullets from net lag)?

[Boozeman]

Booze, you're still missing the point. Under the current model you're more likely to see instant results from firing cannon than with MGs. A part is either there or it isn't. The threshold for seeing results from MG fire is SIGNIFICANTLY greater than with cannon, where only a couple hits is often enough to break something.

Yes because thats the difference in destructive power. It's significant, like 3-4 times in favor for 20 mm cannons, and like 10 times for 30 mm. 

Take a threshold of X lbs of damage to knock a wing off. Therefore it will take Y machine gun round to knock it off, or 1/4 Y of cannon rounds.

Please note that if you fail to reach the threshold with cannon, you will have no damage on the wing either.

I'm fully aware that landing a cannon hit is going to degrade performance under the new model as well. However since it's also much more common to knock a part off with only 2-3 cannon hits outright, the cannon aren't going to take as much advantage of this as MGs, which may take a good 100rpg to deal damage under the current model. I mean seriously, how often do you hit a fighter in the wing with a 30mm in the game and that plane remains flyable, anyway (barring incidents of rubber bullets from net lag)?

I think your numbers are a bit off, I highly doubt that (small) cannon shells do 33-50 times the damage of mg round. I'm flying both .50 cal and cannon planes often, and I cannot confirm your observation. In my experience, 2 hispanos are worth 8 .50, and all other 20 mm a good 6 .50. That is for a sustained burst to knock a part off.

30 mm and up are overkill for fighters.   

[Saxman]

I fly MG-armed rides extensively, and there's times where yes, I end up putting 400-600rds into a target's wing before it breaks off, and this is within convergence range. Probably netlag, but it DOES happen.