>>516715198
>That might be true, but in OPs screenshot it talks about 7 mic locations, my post is only to explain why doing any kind of triangulation on that basis is a dead end because you don't have reliable sync.
You don't need a sync between cameras to do this analysis. The analysis uses the time difference between the sound of the gunshot and the sound of the impact to calculate the distance the shooter is from that camera. Then that's what gets compared -- the distance from the camera location gets drawn as a circle on the graph, and the location of intersection is the location of the shooter. No footage is being stitched together, and time is never compared between cameras, only distance.
I checked the guy's math. He did it right. He's spot on here. I didn't check all of the pins, but I checked a couple and it does work. And on top of that, the fact that there even is an intersecting location is itself a form of proof of correctness -- if the math was done incorrectly the circles would be some variety of random, and if you draw that many random circles you will not get one location where all the circles overlap. So either:
>1. He did the math right and found the location of the shooter
>2. He lied and didn't do any math and drew circles to fabricate a result (which I don't believe because spot checking his math does show its correctness)
or
>3. He did the math wrong and he needs to go buy a lottery ticket right fucking now because the probability of drawing 7 random circles and having them all approximately intersect is astronomically low