Please find below a recent John E Hoover clip, which purportedly, includes many “suppressed” gun shots…Please see parentheticaI notation below to see exactly where they are in the clip. Can these additional shots be factored into your audio analysis ? P.S. I’m new to this site so please bear with me (2:05, 3:07, 3:12, 3:18, 4:14) https://www.youtube.com/watch?v=oKQNR0b3KtU&t=20s
I’VE INCLUDED A JOHN E HOOVER CLIP WHICH, PURPORTEDLY, SHOWS MANY “SUPPRESSED” GUN SHOTS…I KNOW U’RE BUSY SO HERE’S EXACTLY WHERE THEY ARE
2:05, 3:07, 3:12, 3:18, 4:14
@cmartenson I have watched all the videos in this series plus a couple others and offered some suggestions early on on YouTube about how live audio equipment was likely affecting Trumps mic. I am quite familiar with field mics and audio systems. The recent analysis to put a second shooter back on the table seems to rest on two things: a) the difference in delay of the echos recorded on Source #4 for shots 1-3 vs 4-8.
b) the differences in the spectral visualization between the reports of those groups.
I am mildly unconvinced that mic directionality did not cause this or is a factor.
- Mics have directionality (i.e. some are omni-direction and others directional) and a phone mic could have feedback prevention circuitry that affects everything
- Mics have different performance characteristics (polar pattern) in terms of accurate capture based on the distance and angle of the sound wave wrt the mic element
Assertion #1: The same sound will sound different in the same mic based on the distance and angle that the wave passes by the mic element. Said another way, the accuracy of a captured frequency will change based on the directional performance (polar pattern) of the mic. This could contribute to the difference in the spectral visualization. You can demonstrate this phenomenon using a directional mic (because it accentuates it the most) by speaking into it at different angles and distance. At times it can sound full and robust while other times thin or tinny (due to missing frequencies).
Assertion #2: The timing of the echo will change based on the orientation of the mic. The two shot groups sound the same on the Stewart film which IIRC does not change its aspect to the sounds. Arguably, if the Source #4 mic was pointing in a different direction for the two groups of shots, the sound waves of the echo it captures may have traveled a different path ergo taken longer for one group versus the other because the mic was capturing from a different direction.
A muffled sound implies the lack of higher frequencies wrt mids and lows. The most common cause of this on a phone is face/finger-on-mic. The spectral analysis of Source #4 confirms what our ears hear. Lastly, low frequencies travel farther than highs. So the echo sound can change in tone based on distance traveled. These two characteristics may be things to factor in/rule out in the analysis. My two cents.
As an electrical engineer who has developed shooter location systems and viewed shot spectragrams, the spectragrams show different propagation reflection geometry in near field between the two shot groups 1-3 and 4-8. Spectragrams 1-3 indicate the lower frequencies were attenuated which is consistent with shooting through an aperture such as a window or small hole in a wall. The second group of shots has two bands of frequency cut-off which indicate there is a near field reflector to create a multipath cancellation with the direct path. I would consider the roof vertical flanges the source of these reflections. The near field gun report path of these 4-8 shots probably straddled two of the flanges where the forward path of the shots and impinged on the two roof flanges to the microphone of the phone at location (6) from Chris’s video.
Using timing information of the rifle reports from the videos in the forward clear area from AGR buildings toward the stage I can do audio analyses to synchronize the multiple videos to a common global clock reference. Then from microphones in the open area (away from reflectors or non-line of sight, I can narrow down the potential shooter locations. All I need for every microphone of data collected is the following:
- Estimated location in a common coordinate reference frame to the area
- Estimated time of peak gun report amplitude (not supersonic shock cracks) of all eight shots in each telephone’s video stream relative to its own individual video file time.
If someone can provide this information is a place on this website (its too damn filled with so much information and junk) where I can find this, I will solve the problem. I have spent days trying to interact with this sight to be find a place to respond where I maybe found.
7 Likes
Mic would have to be extremely directional to capture the first echo and delete the second, then delete the first echo and capture only the second. Seems quite unlikely to me, especially if the phone is recording out the back. As for the rest, I can’t comment.
You’re also missing the third point - the difference in the report time between the sets when comparing Trump’s mic vs source #4 indicates at least a 12.75’ jump in origin toward source 4. I’ve long been saying that triangulation of the origin from the reports using multiple recordings is the way to crack this case.
1 Like
We mentioned the need for a canonical location for the best versions of all pieces of evidence a few hours ago. Hopefully it will come together soon.
It seems a bit overloaded at the moment. I think there was an email mentioned in some videos. If you include the term “Expert: Shooter location analysis based on audio” to the email address, it may get found. I would state the email address mentioned in the videos, but it might be wrong. I’ll send you a message in a bit if no-one with authority finds this.
Meanwhile, can we get as many likes as possible on his post? There’s a bigger chance of it being found if it has 10+ likes in a sea of babble.
I really only want data from the microphones in locations 1,2,5 and 6. These are in direct line of sight of the shots without reflectors. GThere will be no echos or direct path will be the first arrival. This data will yield a solution with bad gdop (geometric dilution of precision) but at least they will point to a narrow region of possibility. It will be best to get information in three significant figures for shooter location (milliseconds and feet) but narrowing down the general location will give someone a better region to narrow down the search.
Source 4 is not good to use as a source of shot timing since it is in an area that would not receive direct line of sight acoustic path. Only microphones 1,2,5,6 can help. We need another microphone downstream of shooters but in the westerly area away from the podium to get good gdop.
1 Like
There are a lot things that could be the cause of that audio. For example, on the first time at 2:05 the camera also vibrates, it could be the person tripping on something as walking, or hitting the cellphone on his own body as walking.
Comparing sounds must involve analyzing the spectrum of frequency’s and have a known example of suppressed gun fire frequency’s spectrum to compare with.
-
Does anyone know how we have established the shots fired were .223 rounds?
-
My AR-15 S&W is chambered for .556 and will safely shoot .223 also. But a gun chambered for .223 will not shoot .556 safely. Also, shooting .223 from a gun chambered for .556 will result in loss of velocity and accuracy. Could mixed rounds in a single gun account for the difference in audio signatures and accuracy? (shots 4 through 8 were all over the place.) Did the “first shooter” fire the accurate rounds 1-3, and then a confused Crooks rapid fire the 4-8 rounds that went all over the place?
-
One of the videos Chris did has bodycam of one of the cops(?) on the roof counting cartridge casings out loud (1, 2, 3, 4, 5), while another video has someone saying there are 8 cartridges. Doesn’t anyone know why?
It’s possible, but I don’t like this theory because I think it’d be odd for someone to load a magazine with different rounds and then just happen to shoot three rounds of a certain kind, and then pause, and then shoot five rounds of another kind.
It’s actually the cop with the bodycam who counts five shell casings early in the video and then later in the same video counts eight. I personally counted five pretty clear casings on the north side of the ridge, and then some other stuff on the other side of the ridge that might have been casings, but it was hard to tell.
2 Likes
Chris Martenson has setup a drop box with all the better quality recordings that day.
1 Like
Well, if you are sticking to the principle of forming hypothesis based only on hard data, then you have to rule out a single shooter based on your excellent and dogged audio analysis. Whether one can or cannot yet imagine where a second shooter would be is not a “hard data point” but a matter of belief or imagination, so by this principle should have no bearing on whether to accept or reject a second shooter.
2 Likes
That’s a great question. My interpretation atm is that shots 4-8 were pre-planned and did not get influenced by 1-3 missing, Trump ducking, etc. Specifically, I suspect that Crooks thought he was part of a plan with another sniper to take out Trump and save democracy. His pre-defined role was to wait for the other sniper to shoot first, then to fire off ~5 shots of his own to finish or secure the job. So even though pro sniper 1 missed (because Trump moved his head) and Trump was ducking, shots 4-8 were Crooks simply executing his role.
2 Likes
Chris,
I believe it would be very interesting for you to convert echo time for each microphone location to feet based upon the speed of sound at the Butler location that day. For each microphone location draw a circle with radius of that distance on a plot map of the rally. Each circle should show all of the possible buildings of the sound reflection. Intersected circles should define the reflection site and could help determine if multiple shooter locations were involved.
As an electrical engineer, the “Squelch effect” on sound tracks, is the result of a circuit in the preamplifier called Automatic Gain Control {AGC). The AGC circuit maintains a good signal to noise ratio and prevents overdriving the amplifier and speakers.
1 Like
Forget the moving video, focus just on the Trump mic. Differential audio delay of .007 seconds implies shots 4-8 were ~15 feet closer to Trump than shots 1-3. Since Crooks was seen in D Stewart video with his head sticking up right at the roof ridge just before the first shot, if it was Crooks who shot 1-3, then for Crooks to shoot 4-8 15 ft closer he’d have moved way over the ridge onto the visible side of the roof - nobody noted this, Stewart video does not show this, would have been absolutely stupid of Crooks to become so visible. Conclusion: Crooks cannot have taken all shots 1-8.
This argument does not depend on making any assumption about the speed of a moving videographer.
The challenge with this idea is that the delay of the echo, to be converted into feet, would have to make a lot of assumptions about the distance between the shot and the reflective surface generating the echo before its arrival at the microphone. The echo travels from gun to surface to mic. Without knowing the distance between gun and surface and between gun and mic, you have one data point (echo delay) and two unknowns.
I ruled it out based on audio and ballistic evidence. The distance to Trump’s stage was about 730 feet from that tall building. Given the snick-boom delay of .22 seconds (or .213 for that matter), from that distance the implied average speed of the bullet is only about 1880 ft/s. In contrast, ballistics data on multiple 223 rounds have the slowest 223 round averaging about 2300 ft/s over that distance, 30% faster than the data implies on the assumption of a shot from 730 feet. This huge mismatch rules out a shot from anywhere modestly close to 730 feet.
The .007 second difference in the chart is the difference between shot 1 and shot 3, not the difference between shot 3 and shot 4.