Yet another example of ridiculous puffy ack

[Lusche]

Question- how does AI work? Way I hear is the bogie is in the middle of a box that the AI fires into randomly, and size of the box is determined by a combination of the bogie's speed, G's, and distance from ack. Is that right?

Right

And does the AI only target the highest bogie in range?

I think it's rather based on net distance, i.e. closer bogie get's shot with ack somewhat "sticking to it" so that it's not constantly jumping around... but don't quote me on that part

[mthrockmor]

Last week I was turning G's just above the dreaded 3k, maybe less then 3,500 alt. I'm in a Fw-190A5, speed above 300 knts pulling near blackout G's. I am guessing somewhere around 5-6G's. I have a Spit 16 firmly on my tail and not changing things. The enemy CV was close enough to put some puffy ack in the area. Suddenly, in the middle of a near blackout turn I have the wiz-bang-pop-engine dead experience. CV puffy killed my engine while in a tight turn, at beyond icon range of the CV.

In real life this was clearly one of those Vegas shots. (ie right up there with winning the jackpot.) Frustrating but there it is...

Boo

[715]

I would expect the bomber to get shot down...oh wait...bombers are immune to puffy 

When this topic comes up many, including myself, say the CV puffy flak never seems to kill bombers before they drop.  But that's anecdotal evidence and so I decided to get a bit of data.  I did offline tests so they wouldn't be affected by enemy fighters or manned 5"ers (and so I knew where the enemy CV was located).  I confirmed that the default offline settings for gun lethality matched the MA settings.  I took a formation of B17s to 6800 ft and attacked the enemy task force ten times (I should have done it 100 times to get more accurate data, but I'm lazy).  I calibrated the bombsight, but I didn't drop (because I didn't want to have to restore all the destroyed ack with the object command).  The results are interesting:

Probability that all three bombers survive to drop on CV:  90%
Probability that one out of the three is destroyed before dropping on CV: 10%
Probability that two out of three are destroyed: 0%
Probability that all three are destroyed before dropping bombs on the CV: 0%

The margin of error is large, +-10%, because I only did ten tests.  Still, the chances that the task force puffy flak will protect the CV are very near zero.

So why does it seem the puffy flak is biased against fighters?  I suspect it isn't.  I was hit several times without any damage or non fatal damage.  A B17 can take a lot of damage, a small fighter- not so much.  So a fighter probably would have been killed by the first hit.

note- the drones flying around the task force drew it's ack fire but not the puffy flak as they are too low to be targeted.

[gyrene81]

When this topic comes up many, including myself, say the CV puffy flak never seems to kill bombers before they drop.  But that's anecdotal evidence and so I decided to get a bit of data.  I did offline tests so they wouldn't be affected by enemy fighters or manned 5"ers (and so I knew where the enemy CV was located).  I confirmed that the default offline settings for gun lethality matched the MA settings.  I took a formation of B17s to 6800 ft and attacked the enemy task force ten times (I should have done it 100 times to get more accurate data, but I'm lazy).  I calibrated the bombsight, but I didn't drop (because I didn't want to have to restore all the destroyed ack with the object command).  The results are interesting:

Probability that all three bombers survive to drop on CV:  90%
Probability that one out of the three is destroyed before dropping on CV: 10%
Probability that two out of three are destroyed: 0%
Probability that all three are destroyed before dropping bombs on the CV: 0%

The margin of error is large, +-10%, because I only did ten tests.  Still, the chances that the task force puffy flak will protect the CV are very near zero.

So why does it seem the puffy flak is biased against fighters?  I suspect it isn't.  I was hit several times without any damage or non fatal damage.  A B17 can take a lot of damage, a small fighter- not so much.  So a fighter probably would have been killed by the first hit.

note- the drones flying around the task force drew it's ack fire but not the puffy flak as they are too low to be targeted.

if you pushed your bombers to 300+ mph the stats would have been a little different...

[bongboy1]

Another puffy whine...

[W7LPNRICK]

There was a reason for that chicken wire over the trenches in basic training obstacle course, so dummies wouldn't stand up even though you can see the tracers shooting overhead & hear the grenades(fake charges) exploding next to you in the 55 gal drums. Even though they tell ya over & over don't stand up, occasionally some dip stick panics, tries to stand & gets tangle in the chicken wire. So? The game wants to be realistic...don't fly through the puffy! 

[kvuo75]

..don't fly through the puffy!  

and that can magically be done by staying under 3000msl (why 3000?)

which kinda argues against the whole point of the automatic random number generated puffy ack.

just get rid of it, it doesn't deter cv killers. FFS i used to just take sets of 234's at 1500 msl and scream in and kill cv's. it's so easy i got bored with it.

[Karnak]

The reason it seems to have more effect on fighters is that when a fighter gets hit it often is destroyed or disabled.  Bombers get hit a lot more, but they just soak it up.

If a fighters wing has, say, 25 hit points and a bomber's has 150 hit points and the flak hit does 40 hit points, well, the fighter dies in one hit and the bomber has to take four hits to the same spot or it just flies on unaffected.

[kvuo75]

The reason it seems to have more effect on fighters is that when a fighter gets hit it often is destroyed or disabled.  Bombers get hit a lot more, but they just soak it up.

If a fighters wing has, say, 25 hit points and a bomber's has 150 hit points and the flak hit does 40 hit points, well, the fighter dies in one hit and the bomber has to take four hits to the same spot or it just flies on unaffected.

indeed, but the conclusion is, nobody should ever fly a fighter near puffy ack. they risk getting shot down by a random number generator.  bombers have 3 planes who can survive getting tagged a few times, fighter has 1 plane that cannot.

btw, why is 3000MSL the magic altitude?

[Chalenge]

Even if you attack a bomber properly (from the front) you risk getting hit by your own puffy. Bombers are not immune though. That was a stretch.

Attacking from the front at strats where the puffy is probably at its most accurate I have been hit three times out of more than one-hundred attacks and only one of those was fatal.

Try flying 1.5-2k off a bombers wing as it passes over a puffy protected target. You can literally see the box the puffy ack stays within. The box extends further behind the bomber than in front and so that is the most likely place you will get hit. If you come up behind the bomber you have a decreased closure rate and therefore a greater risk of getting hit from both puffy and the bomber guns because you are spending more time in the riskiest zone.

[Crash Orange]

The reason it seems to have more effect on fighters is that when a fighter gets hit it often is destroyed or disabled.  Bombers get hit a lot more, but they just soak it up.

If a fighters wing has, say, 25 hit points and a bomber's has 150 hit points and the flak hit does 40 hit points, well, the fighter dies in one hit and the bomber has to take four hits to the same spot or it just flies on unaffected.

That is how it is in the game, and it's also how it's extremely unrealistic. Puffy ack shouldn't do the same damage to smaller target as to a larger one, because it is not one shell hitting or missing, it's a proximity-fuzed shell going off near the target and creating a random dispersion pattern of hundreds of fragments. Try shooting birdshot at a target with to-scale plane silhouettes on it and see what happens to those targets inside that pattern. A 4-engine bomber that is 4 times the size of a fighter will on average get hit by 4 times as many pellets. That's how it should work in the game but doesn't.

Also, the effects of maneuvering vs. flying straight and level are IMO undermodeled in the game. A RL gunner has to be able to guess where the plane will be several seconds after firing in order to have any chance of hitting it. That's orders of magnitude harder with a target that is maneuvering effectively randomly from the gunner's POV. The AI gunner doesn't have that problem, it's calculating hits as of the moment of the hit, it "knows" exactly where the plane is and only take a moderate penalty to hit if the target is maneuvering. Buffs flying straight and level less than 10k up in full daylight should be easy pickings for the AI gunners for this reason - that's a big reason why daylight bombing raids on the Third Reich were flown at 20k+ IRL, the flak would have massacred them at 8-10k.

[kvuo75]

i still think just making all puffy ack mannable, and if it isn't manned, it doesn't fire is a good solution.

[Pand]

i still think just making all puffy ack mannable, and if it isn't manned, it doesn't fire is a good solution.

This

[whiteman]

taken multiple puffy hits and flown away and have had the golden BB, i think it works just fine. Without the auto puffy CV's are even bigger sitting ducks.

[Noir]

puffy ack is good at killing defenders and nmy fighters cruising at 25K, not defending the CV.

It's awesome to ruin the fun too