So It's Back To First Principles (Part 2)

[phiphi-the-frenchie]

What is your own reaction time to hit the spacebar after hearing a sound?

[roger-knight]

In this video Corey’s crack from shot 2 occurs at 00:01:19.663, and Trump is holding his ear with his hand.

Do you finally give up the claim that shot 2 hit Trump’s ear, or shall we continue wasting time on this?

[kincses-zsolt]

@roger-knight

Looking frame by frame, we can see the crack of the first shot at 00:00:38.071.

What is the source of the audio?
Did they use local mic or got electric connection to podium mic?

Great question — you’re digging into some real-world AV sync analysis, and you’re absolutely right to consider both sound propagation delay and processing/transmission delays depending on the setup.

Step 1: Sound Propagation Delay
So the ambient audio recorded from 300 ft away will lag behind the video by about 263 milliseconds due to sound propagation alone.

Step 2: Direct Audio Feed (Podium Mic)

If Analog:

• Analog audio cables (like XLR) introduce negligible latency — typically <1 ms.
• Audio and video can stay nearly perfectly in sync if the video is also transmitted analog.

If Digital:

• Digital processing introduces latency:
  • Analog-to-digital conversion (A/D): 1–5 ms
  • Digital processing or mixing: 5–20 ms
  • Network transmission (e.g., Dante, AVB): 1–10 ms
  • Digital-to-analog conversion (D/A): 1–5 ms
• So total delay can be in the range of 5–30 ms, possibly more in complex setups.

So, if audio came via a digital feed, you could expect 5–30 ms of delay — but this is far less than the 263 ms from ambient mic pickup at a distance.

Check whether mouth movements and spoken words are misaligned.

Yes, I was about to ask that.

You’re spot on — George H. W. Bush famously said, “Read my lips.” This was during his 1988 presidential campaign, and it became one of his signature lines.

@flamecensor

I think the North TV elevation gave Trump some protection from the Water Tower.

The tree line too.

@sonjax6

It’s well worth the read; there are a lot of details.

Long story short: “I’m not responsible, since I was NOT actually doing my job/assignment.”

This is not a valid defense in any functional organization. It essentially admits to dereliction of duty, while simultaneously attempting to dodge responsibility.

In well-run organizations, authority and accountability go hand in hand. If someone leads a team, they are accountable for its outcomes — even if they delegated work. Leadership isn’t just a title; it’s responsibility.

Even if someone wasn’t directly doing the task, being in a leadership role means owning the outcome. That’s basic accountability.

(And this is not unusual. - Parkinson’s law?)

You’re very close — but what you’re describing is not Parkinson’s Law, but rather the Peter Principle.

Parkinson’s Law

“Work expands to fill the time available for its completion.”

Peter Principle (the one you’re thinking of)

“In a hierarchy, every employee tends to rise to their level of incompetence.”

• Coined by Dr. Laurence J. Peter.
• The idea is that people are promoted based on success in their current role, not based on the skills required for the new one.
• Eventually, they reach a role they’re not competent at, and stay there — resulting in leadership positions being filled with people who aren’t effective.

@sonjax6

James Copenhaver says he was shot by the first bullet and we see his arm react at the time of the first shot.

More than a year after, I still don’t know.

• From podium video we might think the first bullet pinched Trump’s ear. But I’m not sure.
• From south bleacher video we might think the first bullet hit the corner of railing. Perhaps due to some magic audio lag/delay.
  Both cannot be the truth - at the same time.
  So I try to consider that the 2nd bullet grazed Trump’s ear. And maybe the “ouhh” was said since he heard the sonic boom. (Remember, on Corey’s footage a guy plugged his ears immediately.)

@roger-knight

Many participants in this forum struggle to accept that the first shot struck the corner of the bleachers. Some claim that Copenhaver’s arm was hit by the first shot, while others suggest the presence of a suppressed weapon—even though no such sound is detectable in any of the available audio recordings.

I still don’t know.

@roger-knight

either shot 1 grazed Trump’s ear at approximately 1:18:662, or it was shot 2 that did. Which of the two are you claiming?

I lean to say it was shot 2.

@roger-knight

We have multiple videos from different angles all confirming that the first shot hit the corner of the bleachers.

“Sounds” like that. But again, the sound propagation has some delay. So I think we heard the 1st shot sound while saw the impact of the 2nd one. It is possible.

@phiphi-the-frenchie

you will see that I think it’s shot 2 that hit Trump’s ear.

I started thinking that.

@flamecensor

designed to discourage reading.

I think so too.

@roger-knight

Looking frame by frame, we can see the crack of the first shot at 00:00:38.071.

But maybe he just heard the sonic boom of the 1st bullet very close.

@roger-knight

At this point, Trump is already covering his ear with his hand, which makes it impossible for this shot to have struck his ear.

However, the recorded audio might be off by hundred of milliseconds. We should clarify the audio source.

@sonjax6

(two of which are circled in blue). Then there’s another 4 feet from the post to Mr. Copenhaver.

But why his left arm injured? Why the left one?

@sonjax6

but it sure as hell looks like Trump’s hand is almost at his ear if not right there when the 2nd shot happens.

Based on the soundtrack? That might be off by hundred milliseconds.

@sonjax6

“who you gonna believe, me or your lying eyes?”

ears instead?

@sonjax6

I keep a pretty extensive list of bookmarks

Unfortunatelly the playlists gets renumbered.
I posted several links, then I saw they point to different list items. So I should have to remove the list reference from the links.

@roger-knight

Do you finally give up the claim that shot 2 hit Trump’s ear

In case of Corey’s footage the sound source is the phone’s mic. We can consider the sound propagation delay. (But his phone was in custody, and a half of second is magically missing between shot 3 and 4.)

[phiphi-the-frenchie]

Sorry for you but I have again verified and I maintain the following:

@phiphi-the-frenchie

My reasoning is different:
The bullet takes 146 ms to travel from Crooks to Corey, then another 31 ms to reach Trump.
For shot 1, Corey’s crack is at 1:18.657, so 1:18.688 for Trump.
For shot 2, Corey’s crack is at 1:19.513, so 1:19.544 for Trump.
Trump begins raising his hand at 1:19 + 20 frames, or between 1:19.833 and 1:19.875, or about 1.2 seconds after shot 1 and 0.3 seconds after shot 2.

[flamecensor]

My

@phiphi-the-frenchie

own reaction time to hit the spacebar

is unknown, imprecise, but the imprecision is close enough to equal for each of my 2 spacebar tap events.
So the time difference of
0.668 second is a good estimate of the time Trump needed to make an animal reaction to his injury.
After re-watching many times, I also think that

1. Trump heard the rail impact behind his head
2. Corey heard shots from Northeast behind him, because his camera work became agitated before shots you heard.
3. your 146 millisecond judgement has no basis in reality…you simply guessed a ~800 meter/second bullet speed, saw Corey-to-Crooks distance ~118 meters, and made a calculation based on information you didn’t have, and never will have.

[roger-knight]

@kincses-zsolt

I lean to say it was shot 2.

I have a very simple and straightforward way to prove that Trump is covering his ear during Shot 2:

Download this YouTube video:

https://www.youtube.com/watch?v=1ejfAkzjEhk

Add timestamps with this command:

.\ffmpeg -i "Shots fired at Trump rally.mp4" -vf "drawtext=text='%{pts\:hms}':x=10:y=140:fontsize=24:fontcolor=yellow:box=1:boxcolor=black@0.5" -c:a copy "Shots fired at Trump rally with timestamps.mp4"

Add a waveform panel with this command:

.\ffmpeg -i "Shots fired at Trump rally with timestamps.mp4" -filter_complex "[0:v]scale=1280:-2, pad=1280:ih+500:0:0:black[base]; [0:a]showwaves=s=1280x500:mode=line:colors=white[wf]; [base][wf]overlay=0:H-h[v]" -map "[v]" -map 0:a -c:v libx264 -pix_fmt yuv420p -c:a copy "Shots fired at Trump rally with timestamps and waveform.mp4"

Extract the key frames:

.\ffmpeg.exe -ss 00:00:36 -to 00:00:40 -i "Shots fired at Trump rally with timestamps and waveform.mp4" frame_10%04d.png

Unlike @phiphi-the-frenchie, you know how to download a YouTube video and use ffmpeg.

If you run the three commands above, you only need to check one frame—Shot 2 at 00:00:38.906 (frame_100088.png).

The yellow arrow marks the crack from Shot 2. This example is clear because there are no delays: the bullet passes right by Trump’s microphone, so the crack sound is heard almost instantaneous.

ChatGPT confirms:

The crack is not truly instantaneous; it always arrives at the speed of sound. But if the bullet passes extremely close to the microphone, the delay is so tiny (fractions of a millisecond) that it can be considered practically immediate.

Do you agree with the above?

[flamecensor]

@kincses-zsolt

a half of second is magically missing between shot 3 and 4.)

I thought it was more than 1/2 second missing but, however much was missing, it’s because Corey’s phone heard the 3 tonal percussive shots(1 @ B Major, 2 @ E Major) aimed at Hercules2.
Other editing: the pixels in the margins left and right are blurred…Copenhaver’s stomach shot reaction ends in blurred pixels East.

[phiphi-the-frenchie]

@flamecensor

is unknown, imprecise, but the imprecision is close enough to equal for each of my 2 spacebar tap events.

What a nice scientific and accurate method, I am impressed!

@flamecensor

your 146 millisecond judgement has no basis in reality…you simply guessed a ~800 meter/second bullet speed, saw Corey-to-Crooks distance ~118 meters, and made a calculation based on information you didn’t have, and never will have.

Where do you see I use these 146 ms?
I use 31 ms for the time for the bullet to go from Corey to Trump and even if the bullet’s speed is different this time will vary only of 10 ms for a 30% error and my reasoning won’t change.

[flamecensor]

@phiphi-the-frenchie

Where do you see I use these 146 ms?

You typed that yesterday:

@phiphi-the-frenchie

The bullet takes 146 ms to travel from Crooks to Corey,

[phiphi-the-frenchie]

Are you doing this on purpose or are you stupid?

[phiphi-the-frenchie]

Are the data I give for Comperatore’s video correct, yes or no?

[flamecensor]

You asked me a question.
I answered.
Or, how did you calculate the time of 146 (milliseconds?) for the time of a bullet(never fired) from Crooks, to Corey?

[roger-knight]

I allready have proven to you that you are completly off with your numbers.

@phiphi-the-frenchie

For shot 2, Corey’s crack is at 1:19.513,

No, Corey’s crack for shot 2 is at 1:19,663 and Trump is already holding his ear. We don’t even need to calculate the time that is needed from Corey to Trump, because your claim has allready been disproven with this frame. Please look at the yellow arrow.

@phiphi-the-frenchie

Trump begins raising his hand at 1:19 + 20 frames, or between 1:19.833 and 1:19.875

Wrong again—Trump begins raising his hand at 1:19.413, before Shot 2.

Do you finally give up, or do you want to keep wasting everyone’s time?

[kincses-zsolt]

@roger-knight

I have a very simple and straightforward way to prove that Trump is covering his ear during Shot 2:

First of all I had to check his lips:

** https://youtu.be/O696aD_ssIk?t=712
Hmm. “Probably 20 million people” - here the echo ceased.
“And you know that’s a little bit old” - without background echo.
Interesting. They suddenly changed from ambient to podium.

But those studio people know to deal with round trip time. Same issue when your computer reads the clock from time.windows.com

C:\>ping time.windows.com
Pinging twc.trafficmanager.net [104.40.149.189] with 32 b
Reply from 104.40.149.189: bytes=32 time=25ms TTL=110
Reply from 104.40.149.189: bytes=32 time=25ms TTL=110
Reply from 104.40.149.189: bytes=32 time=26ms TTL=110
Reply from 104.40.149.189: bytes=32 time=25ms TTL=110
Ping statistics for 104.40.149.189:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 25ms, Maximum = 26ms, Average = 25ms

From other hand, the loudspeaker itself should be delayed to avoid the amplifier excited. (I experienced that many times in elementary school - bad amplifier settings. (Oh, by the way, a professional singer was unable to sing the anthem due to his echoed voice.))

RSBN:
** https://youtu.be/Rr63RgGN-Yo?t=23792
** https://youtu.be/Rr63RgGN-Yo?t=24173

Fox NEWS:
** https://youtu.be/zBzysZQqatM?t=340

BTW:

That is phone’s mic.
** https://youtu.be/O696aD_ssIk?t=1138
2nd report comes before he touched his ear.
So still I don’t know.

@sonjax6 Walktrhough
** https://youtu.be/88qoHWVEPIg?t=184

Now I slow it down (0.25):
“Happened”
1st crack
“Ouhh”
1st report
moves his shoulder
starts raising his hand
2nd crack
touches his ear
2nd report
lowers his hand
3rd crack
3rd report

So the first crack comes before he touched his ear, but the report comes after.

[phiphi-the-frenchie]

I have explained to SonjaX6 how I proceed to get my results.
Do the same and we’ll talk about it again.

What tool do you use to get your numbers?

[kincses-zsolt]

@roger-knight

Corey’s crack for shot 2 is at 1:19,663 and Trump is already holding his ear.

I slowed it down. I can hear the crack before his hand reached his ear.

[roger-knight]

@phiphi-the-frenchie

What tool do you use to get your numbers?

I use FFmpeg, one of the most reliable and widely used multimedia frameworks in existence. It is trusted by both open-source and commercial projects.

For our case, FFmpeg can automatically set timestamps with millisecond precision using the following command:

.\ffmpeg -i "Shots fired at Trump rally.mp4" -vf "drawtext=text='%{pts\:hms}':x=10:y=140:fontsize=24:fontcolor=yellow:box=1:boxcolor=black@0.5" -c:a copy "Shots fired at Trump rally with timestamps.mp4"

If your video has a different name, simply change the input and output filenames. This is a game-changer, because now we can refer to every frame with milliseconds precision.

---

Adding Audio Visualization

We can also add an audio waveform to each frame. This way, every frame shows both the precise timestamp and the corresponding audio graph. This makes frame-by-frame analysis much clearer and far more professional.

Use this command to overlay the waveform panel:

.\ffmpeg -i "Shots fired at Trump rally with timestamps.mp4" -filter_complex "[0:v]scale=1280:-2, pad=1280:ih+500:0:0:black[base]; [0:a]showwaves=s=1280x500:mode=line:colors=white[wf]; [base][wf]overlay=0:H-h[v]" -map "[v]" -map 0:a -c:v libx264 -pix_fmt yuv420p -c:a copy "Shots fired at Trump rally with timestamps and waveform.mp4"

---

Extracting Screenshots

If you want to capture screenshots from a specific time sequence, you can extract the key frames with this command:

.\ffmpeg.exe -ss 00:00:36 -to 00:00:40 -i "Shots fired at Trump rally with timestamps and waveform.mp4" frame_%03d.png

Just change the filename and time range as needed. In this example, you’ll get a screenshot of each frame between 00:00:36 and 00:00:40. Be careful not to specify a long duration, or your hard drive will quickly fill up with images.

---

Important:

This workflow provides a reliable and professional method for frame-by-frame video analysis. Instead of relying on subjective impressions (e.g., “I hear a crack here” vs. “I hear a boom there”), everyone can reference the same precise timestamps and synchronized audio waveform. It saves time and ensures consistency across analyses.

[sonjax6]

@flamecensor

1st time I put a ‘timer’ into a video,

Thank you, that’s very helpful.

[roger-knight]

@kincses-zsolt

I slowed it down. I can hear the crack before his hand reached his ear.

Before we continue this debate about shot 2 hitting Trumps ear, let’s agree on one point:

Trump starts raising his hand to his ear. This means the shot must have already struck his ear — otherwise, he would have no reason to protect the injury. Therefore, if we want to determine when the sound of the second shot should appear in Comparator’s video, it must be before he begins raising his hand.

Here’s why:

• A human typically needs 0.2–0.3 seconds to react.
• The sound of the gunshot takes an additional 0.022 seconds to reach Trumps ear from Corey’s camera microphone.

In total, this adds up to roughly 0.3 seconds of delay. So we should subtract about 0.3 seconds from the moment Trump starts raising his hand.

If you can indeed hear the crack of the shot before that point, then we can begin analyzing whether this aligns with the conclusion that the second shot hit Trump’s ear.

[flamecensor]

@roger-knight

let’s agree on one point:

Yes, let’s agree that Shot 1 hit Trump’s ear, Copenhaver’s arm, the TVbox’s metal outer cladding, ToughOldBird’s elbow, FlowerLady’s left shoulder/back, and probably some upper body parts of David Dutch.
Copenhaver’s arm is reacting for itself, but the other 3 are reacting to bullet fragments, from behind them, after hitting the TVbox’s outer metal cladding(with or without black vinyl):