Thread by @nafomono

This 🧵 is about the terror attack on Kostyantynivka on 06.09.2023 and why the narrative
from ru sources and other "pundits" regarding the direction of the attack is wrong and complete bullshit. Its a long thread, so strap in.🙄

Facts:
The attack occured on Kostyantynivka, on Tsiolkovskoho Street.
-the cam is looking in SE direction downwards, approx 5-6m above ground
-the cam is on the left side and likely on a lantern/post
- the distance from the cam to the street crossing of the epicenter is about 65m

Facts:
From the debris and lack of impact crater we know:
- not a storm shadow, 1.3t warhead
- not a kalibr, 200-500kg warhead
- proximity fuse and shrapnel -> S-300 like, ballistic missile
- subsonic, otherwise we would hear the missile only after detonation

Facts:
- the cam must be some kind of outdoor webcam with atleast IP-54 certification to handle rain and dust
- It's therefore safe to assume the mic is within the webcam enclosure and that the enclosure has only a few openings.

Part 1: Come to your senses
As humans we prioritize our visual perception in analysing things.
That also explains why everyone is looking at the data wrong, quite literally so.
Let's take a look at the audio. Yes, you heard right, THE FUCKING AUDIO.

We will prove the mic has unidirectional-like polar pattern and can't hear anything from behind and the missile could only come from a southern direction towards the mic. This is how the patterns and a cardioid mic look like.

These patterns visualize mic sensitivity based on the angle to the mic.
For cardioid pattern, obviously, the mic must have inlets in the back to hear anything.
So far I have found no outdoor-waterprotected webcam, which has such a mic.

For the analysis I took the video from @JulianRoepke and cut it from 00:00:00 -> 00:00:05.333

For starters, we see that both channels are the same, when substracted from another the resulting track is 0. So its a webcam with a cheap mono mic.

https://twitter.com/JulianRoepcke/status/1699453405415055825?s=20

So NO, you have no spatial hearing, you can't hear the rocket from behind. You think you do, because most people have asymetric-hearing. Lets see audio and video side by side. We can see the boom-wave at the beginning.

The superposition (interference) of the boom wave and transition to the missile sound gives us a starting point, from when the mic first picked up the missile.
BoomWave 00:01.531 - 00:02.865
Missile 00:02.865 - 00:03.866 explosion
Missile Car 1: 00:03.733
Missile Car 2: 00:03.766

They argue that the mic can hear the missile from behind and that it flies above the street, which reflections supposedly show. Lets do the math. The cardioid pattern has an angle of 30° in both directions in the back, limiting the deadzone, where the mic cant hear anything.

This image shows the variables and equations used to calculate x2 and with it the angle alpha. So, ballistic missiles can have a true parabolic trajectory, or a straightened out one, with a steep angle at impact.

With the speeds involved and the short distance traveled, we can safely assume the missile was in its linear descend phase. From the time-tags in the audio, we know the mic hears the missile for about 1 sec.
Sonic speed in 20°C is 343m/s.

I chose the camera angle to the street as g=20°, the height of the camera h=6m, x1=65m.
With d=343m, we can determine the impact angle a missile has to stay under, to be subsonic.
We can also only hear the missile after it passes the deadzone of the mic., of course.

With these values we get: x2=194,61m, alpha=40,96°.
s=38m, timestamp=03.713
Such an angle leads to a height of the missile in distance to the epicenter 8,68m(10m).
For half-sonic speed we get: alpha=32,79° and a height h=6,44m at (10m) from epicenter.

Which is about the height of the buildings. The shortest distance to the mic s=30,16m and the corresponding time-stamp 03.528s. I think we can safely assume, that everyone! on this street would notice a ballistic missile flying above their heads.

Also, we would see the actual missile in the footage for any speed at/below half-sonic. If we take the 3D case into account, the XY plane looks like this.
The distance from the middle of the street to the camera is approx 5.8m.

According to their bullshit analysis, the missile comes in a 5° angle from the right side of the cam.
We can see that it crosses the pattern from the side and doesn't even directly approach the mic or fly over it. The distance to the mic would be even greater.

The observed volume increase wouldn't be that profound, even if the vertical approach angle changes by a few degrees. So this great narrative went to shit really fast, almost at sonic speed.

For unidirectional pattern, you get the inverse problem, with very steep angles for sonic speeds, so that it passes the mic at greater x distance and always in front of it.

Lets further analyze the audio to find more ways why it is bullshit, so we can skip the needless discussion of a full 3D calculation.
This image adds 2 markers for the previous calculation half-sonic/sonic speeds and the timepoints the missile would have had the shortest distance

The spectrogram shows the frequencies on the left. Brighter color is higher amplitude. We can see a linear-like volume increase throughout the frequency spectrum and past the flyby-timestamps.
The audio is missing the reduction in volume once the missile passed the camera.

If the mic had cardoid pattern, we would also have simultaneous amplification of the volume as the missile passes the pattern. Higher-freq. noise doesn't travel very far. So if the missile is producing noise in higher freq., we will only hear them when the missile is near impact.

When the missile is entering the urban area, you will also hear echos of it, especially as the cam and mic are looking down the ground. This is the step and sudden pickup of high freq. noise, you can see in the spectrogram.

To further demonstrate, I made a little experiment, indoors and therefore with echo. I used an cardioid mic and white noise-track from the phone and made recordings, one coming from an offset angle downwards ~45° from behind and one from the front also coming in with ~45°.

I will keep it short about the reflections. Here is the wiki link. If this simple physics is too much for you to grasp, noone can help you.
en.wikipedia.org/wiki/Reflectio…

Part2: lets reflect:
"But, but, but the people in the video saw it". No they didn't.
The guy in the white shirt started turning right after the first boom wave was heard. We usually turn into the direction we hear something louder.

He is starting to turn counter clockwise, because he most likely heard the boom from the east and had the impression that something passed him from east to west. After him the other people on the left start turning, because if one guy "sees" something, you are curious what it is.

By the time they hear the missile approaching, they were already looking in the wrong direction. At the time of the second car reflection, the missile had supposedly already passed them. So they are still looking at something, which is already behind them.

Take notice of the guy on the right. He apparently heard the boom wave coming from his front east direction. That is also the direction he is looking up, when the missile is audible. As we have now abundantly discussed, he is the only one who most likely actually saw the missile.

Conclusion:
ruZZians and tabloid reporters have no basic understanding of physics and are posting and spreading bullshit to thousands of followers.
What a shocker and new revelation.
I hope it's possible to state that fact, without being called names. 😅

Usually I wait if some news and clarification comes up days after an event, so that I don't have to waste time debunking something, which had already refuted.
As people still debate about it, I decided to give it a look. It was a ruzzian missile, coming from southeast direction.

I've deactivated comments to keep that thread clean, if you want to discuss it, you can do so here:

https://twitter.com/nafomono/status/1702690991117901891?t=T-XCTt2x5n3B0N69sjmdkA&s=19

We can now see on the bottom tracks what the sound from passing by looks like. You have a minimal peak at closest distance. As the distance increases it quickly subsides and the audio stays static without a noticeable volume ramp. The last track is heavily amplified.

@threadreaderapp unroll