Gunners ineffective with wind

[SAS_Storebror]

Hi folks,

 

Is it just me or are the gunners really duds as soon as you have wind on the mission?

Test scenario:

Setup a quick mission (n vs n planes), one Pe-2 S.87 "ace" with Blister Turret for player, one Bf 109F-4 "ace" as AI plane.

Map is Stalingrad, I've been using the "n vs n" start point right at Stalingrad City.

4000m altitude, "Escape" situation.

Use any difficulty setting of your choice but for testing purpose, I'd recommend to use "custom" settings with all "normal" boxes checked plus unbreakable and invulnerable.

Start the mission, unpause it and engage the level autopilot to keep your Pe-2 flying straight.

Now hit Ctrl+F2 and watch the scene from the 109's perspective.

 

What happens in my test is this:

The 109 will always approach from dead six. Most deadly thing you can do against a Pe-2. But...

With wind and turbulence set to "0 m/s", the 109 will get hit at about 1.0 kilometer distance. Below that distance it usually gets humilated.

With wind and turbulence set to "1 m/s", the situation already differs much: The gunner misses the 109, always by the same amount and in the same direction, until it gets within ~700m.

With wind set to "3 m/s", the gunner misses until the 109 gets within 400m distance.

With wind set to "6 m/s" or higher, the Pe-2 gunner doesn't hit sh*t anymore.

 

Watching the scene repeatedly, it becomes obvious that the Pe-2 gunner seems to try to compensate the wind.

Which is silly, because for two planes flying within the same medium, there is no relative wind offset. Wind just doesn't matter here. Yet the gunner tries to compensate it and therefore constantly misses in the same direction, by the same amount.

 

This explains why people on WoL complain about deadly gunners - which is their mistake, but still... without wind, the Pe-2 gunner will hit you on it's dead six - yet on our own Server, where we have wind set to 3 m/s and turbulence to 1 m/s, the gunners hardly hit anything.

 

BUT: Now comes the next fun fact: This only happens with Player planes. It's only your gunners suffering from this.

If you try it the other way around, with you sitting in a 109 attacking a pure AI PE-2 (or you're sitting in a russian plane attacking a He-111 for the same matter), the gunners will hit you at 1.5km distance regardless whether you have 0 m/s wind or 12 m/s. They just hit. Every time, everywhere.

 

Anyone else noticed something similar yet?

 

Mike

[unreasonable]

I tried that and I agree that something odd is happening: ramp the wind up to maximum on the ground of 12 m/s - it will be 3 to 4 times as fast at 4000m, say 130 kph - and the gunner is missing the 109 by about three wingspans at close range!

 

However, in my test the gunner misses downwind, and  is actually aiming downwind relative to the 109, judging from the orientation of the gun barrel. Use the city fires as a wind marker.  

 

The only quasi logical explanation I can come up with is that if the gunner's point of aim is based on the target's predicted track over ground but does not take into account the fact that his bullets have an initial total vector equal to the Pe's, he is going to aim too far downwind by the magnitude of the wind.  

 

 

 

 

 

[supersqwack]

my experience suggest that several AI AC gunners, but especially the pe-2, are terminators that rarely miss and more often than not deliver critical damage at a single pass.

It very poorly reflects the historical accounts i've read. Sadly

[1PL-Husar-1Esk]

If the  the speed of medium  is the  same for  both object, it is irrelevant or in other words should not be taken into calculating shooting solution.

[unreasonable]

8 hours ago, 307_Tomcat said:

If the  the speed of medium  is the  same for  both object, it is irrelevant or in other words should not be taken into calculating shooting solution.

 

Yes, but only if both objects and the bullets are moved freely by the medium by the same vector.  Take a stationary firer and a stationary target.  The speed of the medium is the same for both objects, but if you are the firer you certainly have to take into account the speed of the wind, since it changes the ballistic track relative to the objects.

 

So an AA gunner firing at an aircraft has to take into account the wind as does an aircraft firing at a ground target. 

 

It would be interesting to see if AA gunners make an adjustment for wind: I will try to see if I can create a test for that in the ME once I am properly caffeinated.  

 

 

 

 

 

[unreasonable]

Having done some testing, my conclusions so far are that:

 

1) Gunners aim according to the predicted track over ground of the target - while this can be calculated using a wind vector and a plane vector, I doubt that this is what the AI is doing. I suspect it is just taking an X-Y-Z coordinate and extrapolating. 

 

2) The ballistics accurately take into account wind effects.

 

3) The AI  do not take into account the effect of wind on ballistics after the shells have been fired: so in the OP's case they are probably not taking them into account before they fire either. 

 

Additional evidence is in the attached test mission.  This has a Stuka flying (crabbing   ) south over a 61-K  while there is an east wind of 50 m/s, the most you can put into the ME.  You will see the downwind deviation of the shots and the AA's failure to adjust, so at long range hits are impossible.   You start in a tank - just press free camera and walk over to behind the nearby 61-K. 

 

 

Flak Wind Test 4.zip

Edited September 12, 2018 by unreasonable

[1PL-Husar-1Esk]

2 hours ago, unreasonable said:

 

Yes, but only if both objects and the bullets are moved freely by the medium by the same vector.  Take a stationary firer and a stationary target.  The speed of the medium is the same for both objects, but if you are the firer you certainly have to take into account the speed of the wind, since it changes the ballistic track relative to the objects.

 

So an AA gunner firing at an aircraft has to take into account the wind as does an aircraft firing at a ground target. 

 

It would be interesting to see if AA gunners make an adjustment for wind: I will try to see if I can create a test for that in the ME once I am properly caffeinated.  

 

 

 

 

 

You right about ground to ground and ground to air (and other way around) but air to air if the wind speed is the same only speed of the objects matters. That what I think.

[SAS_Storebror]

4 hours ago, unreasonable said:

Gunners aim according to the predicted track over ground of the target

 

That's already the problem I think.

If both objects are moving in the air, ground speed vector is irrelevant.

That'd explain part one of the issue (why player planes' gunner doesn't hit sh*t when there's wind), but I couldn't find any good reason why only player plane's gunners suffer from this - pure AI planes have exceptional gunner precision - both russian and german planes by the way, and both to the very same extent despite the usual Luftwhining we keep hearing about Pe-2 gunners only.

 

Mike

[unreasonable]

I will have to have a think about the issue of the player plane vs AI plane and run your original test again from a player target POV. 

 

In my ground AA test both the target and the firer were AI, the player being confined to a tank as a neutral observer.  My own hypothesis has always been that AI is AI and it makes no difference where it is: that was certainly the case with the (interminable) ground AA tests that I ran last year. It made no difference to the accuracy of the AI firing whether it was firing at a player or at another AI. 

 

But I have never tested AI's AI gunner vs Player's AI gunner: so I have an open mind. But that will have to wait for my morning window of lucidity  

 

Another relevant point that came to mind after HerMuff's thread about waist gunners - if AI gunners fire using target track over ground while ignoring their own, then waist gunners should always miss ground targets (actually any targets) even if there is no wind, since they will not allow for the deflection needed due to their own aircraft's motion.  I have no idea how you can make He111 waist gunners fire at ground targets to check this, unfortunately.

Edited September 12, 2018 by unreasonable

[unreasonable]

Update on Flak wind Test: by changing the Stuka to player plane and then pressing A for autopilot to fly towards it's waypoint, we can see if the the AI gunner on the ground behaves any differently: as I expected, it does not.  The shells all miss downwind again until the range is very low.   

I can do a test that has the player Pe-2 on level auto and a 109 flying to a waypoint behind and over it:  this certainly confirms that the gunner in a player plane is incompetent: is it set on novice AI? It  always seems to allow far too much lead: it is not just a random offset, it happens every time the test is run.  This is with no wind.

 

This would be consistent with the idea that the gunner is extrapolating the track of the target but not taking into account his own plane's vector, or to put it another way he is leading by an amount appropriate to the target speed as though the gun was stationary, rather than by an amount appropriate to the closing speed.  Further testing on that idea to follow.

 

But I cannot get a comparison with AI vs AI since if you put the Pe-2 on AI and it's waypoint on High Priority, the plane will fly straight but gunners will not fire,   while on Medium Priority the gunners will fire but the pilot will manoeuvre once the 109 is in gun range.  This makes it impossible to compare the aim point systematically.  Any ideas?

 

 

Edited September 13, 2018 by unreasonable

[unreasonable]

In addition, He111 gunners told to attack ground targets do not take into account their own plane's vector.  Try this mission, you are player He111 but do not have to take control of aircraft, just leave it to fly straight.   Press RAlt+6 to order fire at ground targets, pause, F11 for free camera and move the camera down to the right of the He111's path where you will see a couple of trucks on the runway.  No wind set.

 

Observe fall of shot: the gunners always miss far in front of the trucks - ie in the direction of travel of the He111.  They are presumably aiming right at the targets.

 

 

 

 

He111 vs Ground.zip