Turns out what I was describing before is “Object Tracking using Time Difference of Arrival.”
This analysis was done for shots 1 through 10 by YouTuber using 4 microphones: @greg_nichols
" Trump Rally Audio Analysis: Counter-Sniper Location Revealed!"
He concludes the first 8 all came from Crooks’ position and that shots 9 and 10 were from just north of the audience very near a red farm equipment. However the first 8 shot locations drift quite a bit, I think due to the movement of the phones.
The first part of his video is analyzing snick-boom; but then he gets into the arrival time which is what I was hoping for.
This could still be confirmed by other investigators; by finding other recordings; perhaps fine tuned by factoring phone locations that changed between shots.
I don’t think this rules out other locations that are very near Crooks position.
In a pinned comment, he said he found that the red farm vehicle shown at the end of the video is a Case 8230 harvester. The top of the cab is 29 feet off the ground and would have made an excellent vantage point.
I just posted this in a different thread regarding the ESU officer who most likely fired shot 9 and from where. He was definitely paying attention to building 6.
Supersonic cracks are very useful for determining direction of shooter if you have three or more microphones. Further, the direction is even better if you have them in a plane perpendicular to the bullet trajectory. They are not good for determining range other than to provide a time difference between crack and boom of gun report. Shooter location is best determined from direct line of sight gun reports. The number of data points for gun reports depends upon whether you want 2-D or 3-D solution for location. And the microphones need to be distributed to cover all dimensions. The problem here is that microphones are mostly distributed in a 2-D geometric plane and won’t be able to help with elevation. And worse, the microphones in which we have data are mostly along the same line without much of a second dimensional separation.
VT,
Gun flash is too fast sometimes to be captured in video. Moreover, the flash would be larger and directional with the flash suppressor instead of a dot. The white dot is most likely the reflection off the lens of his red dot scope that happened to glint some light. Any relation to a shot may be the cringing factor of the shooter reflexing to the loud reports nearby.
Thanks for the pointer. I am new here. Now I need to figure out how to get the audio data I need stripped from these mp3 files and must figure out locations from the video. I don’t know if someone has done this before. Normally, engineers on my staff take care of these details but I am not at work for this.
One thing you have missed in your speed of sound calculation is the elevation or altitude at Butler PA at 1043 feet. I don’t know exactly how much of a difference it makes, the air does get thinner the higher it gets.
For you detail oriented guys out there calculating CRACK-THUMP from the audio files to determine distances and all, do keep in mind the velocity of the bullet changes over the distances. Here I have 6 ballistic tables (one for each Zero Range: 25-36-50-100-150-200yards), considering a 5.56x45 caliber with 55 grains, which is the most common. I used Hornady’s ballistic calculator and also gundata (on the right, considering 100yard zero). Altering the zero range itself won’t affect the speed, just the other factors important for the guys working with trajectories on the other threads. For the audio thread, do ignore the circles in blue, they are just demonstraing the amplitude of corrections for scenarios given the different zero ranges configured on the sights, meaning the lesser the amplitude, the less need for on-the-fly compensations for that particular zero range. Just ignore those for audio work.
Sanity check me, please… On Bodycam file 2 at 3m32s you can actually hear something very interesting when the cop in blue uniform directs ESU operators to a ladder and they all seem a bit confused about which ladder can be used. When the camera is already on the small steps just before the black tactical ladder is reach you can hear:
I don’t understand what was said in that clip, but here is a video that walks through the ladder issue very nicely using video evidence. Aside from the media’s initial confusion over the red ladder, there is nothing conspiratorial about the ladders.
Looks like a great start into doing an audio analysis. Interesting to see how the first few shots line up with someone hiding in the roof space. How some of the microphone locations do move over time and working in a 3D rather than a 2D space are a couple of improvements that can be made.
Yes. A coincidence that the 0.2 second time difference is so close to the timing of the two ends of a typical primary “N-wave” front propagating as the “cone” from the course of a supersonic projectile as the front passes any fixed point adjacent to the trajectory.
Check out
Greg Nichols
Trump Rally Audio Analysis: Counter-Sniper Location Revealed!
This analysis shows BOTH shot #9 and #10 came from the location of the RED Combine. Possible Sniper on top of the machine.
Also! the press photo of the bullet trace suggest the bullet was in an upward trajectory, or is this an optical illusion from the photographers shooting angle?
Seems the only way to get an upward angle for Trumps ear is shooting from the side Vents @ 9 ft ceiling level. However this would miss the stand railing hit BUT might account for the Hydraulic line of the speaker lift.
Lastly the first 3 rounds differ from #4~8 in sound by a METALLIC clang which is common for an AR style rifles buffer and spring recoil, Shots #1~3 dont have that one possibility is shots #1~3 are from a bolt action 5.56nn.
Leo, do you have a source for the explanation of a primary “N-Wave” front timing. I would be interested in learning more about that. Thanks in advance if you can provide a link.
Use ffmpeg. It is a very versatile audio/video editing/conversion tool. The command I use is:
ffmpeg -i inputfile.mp4 outputfile.wav
the -i is used the specify the input file, which can be any format. The last parameter is the output file and without further options ffmpeg creates a file based on the file name extension (in the above case “wav” which is an uncompressed audio only file).
This was some time ago, in either a written article or a video. It gives a real world explanation to this.
It was about use of a Mk12 SPR with an Ops Inc 12th model suppressor in Afghanistan. Because the muzzle blast was reduced by the suppressor it was harder to know where the shooter was. If the bullet passed a tree or rock, the resulting crack at that spot would draw the target’s attention to that spot in the flight of the bullet.
What if the 1st shooter came down a bit rough and was hurt. Then he could say that he looked on the roof and TMC pointed his gun at him and then he fell.
