File talk:Ukraine Sunset Sniper Shadow.png

From A Closer Look On Syria
Jump to navigation Jump to search

Azimuth

I've been looking into the various sources for the sunset sniping incident and trying to get one coherent reading. It's been a bit of a headache, but now it's falling into place, and I may have a very interesting bit of fakery to expose soon. Double-checking first. First, I'm setting times on all ends, as possible. BBC's footage I can't time, except by correlation, in time... RT's live feed, had that read wrong before, right now, with only the finer points to settle. But this video, the activist one filed right by V. Melnychuk (or so I reason) as he's shot, should be easy enough. Petri - you're usually better at solar elevation than me. How tall is that post, and how long is its shadow? Bonus point for azimuth readings. For a first reading, I just said 2-3 feet, 32-38 feet. That said 2.5-maybe 5.5 degrees elevation, weighted to the lower end, so about 4:45-5:00, weighted to later. Other clues suggest it's actually a ghhair after 5, say 5:03. Anyway, a good tight range should be reachable here, thought I'd give you a chance to have a say. Also, taking related thoughts, if anyone has any. --Caustic Logic (talk) 13:43, 10 April 2014 (UTC)

How wide would the angle of view of a wide-angle lens be? This does not look like a camera phone. The lines are too fish-eye distorted for regular use.
Can you give a link to the video source? -- Petri Krohn (talk) 08:41, 11 April 2014 (UTC)
Do we know the exact spot where the shot is taken? It would be easy to take bearings to the Maidan pillar and a corner of the Ukraine hotel. -- Petri Krohn (talk) 08:44, 11 April 2014 (UTC)
It's a full-on fisheye view, don't know how that effects things. Not too much I suppose. Video source updated, below. The one I had on the page was pulled since. The spot he fell is about the very middle of the north plaza of the OP, right on that dividing line. The shadow stops (x distance) from the base of the stairs. The time is important - it'll be our best benchmark, ad be used for setting other videos (RT, BBC) --Caustic Logic (talk) 09:27, 11 April 2014 (UTC)
The sun is above a very high building in the distance. Could be this one:
I get something like 67.5 / 247.5 degrees. Would that be correct? -- Petri Krohn (talk) 12:38, 11 April 2014 (UTC)
More exact numbers: south edge of south tower: 243.16, north edge of north tower: 74.20. From the video, starting at 1 minute one would have to measure the relative position of the sun to the two towers. Using a ruler on by 24 inch screen with the video stopped at 1:21 I get 16 mm on the left (south) and 23 mm on the right. Doing the math i get 247.68 degrees. -- Petri Krohn (talk) 12:56, 11 April 2014 (UTC)
NOAA calculator has that azimuth (247.66) passed at 4:51 PM. So that's a general match, but a bit earlier than I thought. I guess a good idea would be to compare to the altitude or azimuth at video's end, 2.5 minutes later. I want an extra good reading on this one! --Caustic Logic (talk) 14:22, 11 April 2014 (UTC)
Why is it I get azimuth 247.64 at 17:21 on February 20th? -- Petri Krohn (talk) 14:15, 11 April 2014 (UTC)
Don't know. That's 4:51 at GMT +2 and DST off, general Kiev coordinates, over here. --Caustic Logic (talk) 14:22, 11 April 2014 (UTC)

I guess it is because I am still at the start location in Nebraska!

Location: Lat 39.833 Lng -98.583 Time Zone -6 DST? (no)

Horrible ergonomics! Why can't they have the start location read 0.0 - 0.0 - 0:0 or something?

Here is what I get after I move the "large pin" between the pillars at the October Palace:

Location: Lat 50.449787 Lng 30.5274208 Time Zone +2 DST? (no)
Date: Day 20 Mon Feb Yr 2014 Local Time: 16 :41 :00 PM

Equation
of Time
(minutes):

Solar

Declination
(in °):

Apparent

Sunrise:

Solar Noon: Apparent

Sunset:

Az/El (in °) at
Local Time:
-13.73 -10.8 07:00 12:11:38 17:24 245.61 5.68

You seem to have a typo on the time 16:41: I get an azimuth of 247.68 degrees at 16:51:28 -- Petri Krohn (talk) 21:18, 11 April 2014 (UTC)

No, i stopped .66, close enough - it had a default 'and 20 seconds' I didn't change. Both .66 and .68 pass within that minute. People forget what a tiny fraction that second digit after the . really is. :) So this is saying about 4:50, not 5:00-5:05. That I'll have to plug back in and re-consider. I've been having the same problem with Nebraska, and even pins - I had to dig up the coordinates and enter them to get it to stay put. I didn't think you'd have the same problem. --Caustic Logic (talk) 23:14, 11 April 2014 (UTC)
Wait, 16:41? Wherever I said that, oops. --Caustic Logic (talk) 09:11, 13 April 2014 (UTC)

I think being exact with a seconds place helps here, although the right answer I think will be a range. 4:51:28 is the center - at least for what the azimuth says, and it's more precise than any altitude reading we can get here. My best reading so far (see here) says the RT 7-hour video has a time-stamp offset from real local PM time most commonly coming up as an hour 35 or 37 ahead (all with rough readings and rounding off on both on protractor degrees and video minutes). But with all outliers but one (1:10) settling around 1:20 and 1:40, the 1:35 offset seems best. That makes athe shots heard at 6:24 in the video events of just about 4:49 PM. That's just 2.5 minutes off from this reading, which in turn is from 1:29 in the video - about 50 seconds after the shot. So roughly 1.5 minutes off, really. Both ends could use some more tightening, but this getting about solid enough to call audio fakery in one of these videos. And I don't think it's RT's. --Caustic Logic (talk) 09:11, 13 April 2014 (UTC)

Ergo, exact shooting time for Mr. Melnychuk can be best put as 4:50:39 PM +/- ... what, 2 minutes? --Caustic Logic (talk) 12:21, 13 April 2014 (UTC)

Petri, I checked your azimuth finally, and improved it I think.. You identified the right building, but put the sun over it, rather than a bit north. I measured to a line between (as named on Wikimapia) "Volodymyrska vulytsia, 49А" and ""Kyivproject" (Kyiv)" - a bit closer to the latter but not right over it.

Ukraine Sunset Azimuth.png

Between 247.5 and 249.5 I would select 248 or 248.5. The second gives a time of 4:55:37 - minus :49 seconds to actual shooting time = 4:54:48. Or, if it's just 248, just over what you got, 4:53:03 - :49 = 4:52:14. This adjustment makes the RT time match just a bit more of a stretch, but it's all still quite close and, again, there's no other shooting there to match it with, so my estimated offset must be a bit less accurate than I thought, and it's only about 1:30, not 1:35 ahead. I think it's best to let this video set the time and the other left to match it - and now I say the time is 4:53 (4:52 oops) +/- 2 minutes. --Caustic Logic (talk) 13:14, 14 April 2014 (UTC)

That's Azimuth, and I started out mainly concerned with altitude. This might be settled already, but like I said, this one deserves extra effort. I'll be back on it soon. --Caustic Logic (talk) 23:14, 11 April 2014 (UTC)

Or not. It's pretty well set I think. --Caustic Logic (talk) 09:11, 13 April 2014 (UTC)
I did not look at the center of either building, but the northern and southern edge. If we want to improve on this, we should get a better photo that shows approximately the same view. Then we could place it on top of the video image and align them. It would then be easier to match the view to the exact features of the buildings. -- Petri Krohn (talk) 15:45, 14 April 2014 (UTC)
It's not right over the edge either, though. I don't think a photo would help. Those are the right buildings, the right one with a very distinct shape. What matters is accurate lines to the same place - as I picked, apparent centerline (based on footprint, est.) As you can see, my guides only worked approximately, but mainly that's to find the right buildings in the first place. The blue lines and maybe camera placement could be refined. But I don't think anything will change much from what I mapped out there. I'm for calling it good with imagery, and minimal debate to set the best reading of it. As of now, I estimate bullet impact at 4:52:14, +/- a generous 1.5 minutes. --Caustic Logic (talk) 11:59, 18 April 2014 (UTC)

The Video

Fullest version I've seen is the standard - 3:03.

Sources:

The Audio: The audio in this video (I checked and it's the same in all) is rather odd. IIt's got some wrongly-plugged-in microphone effect, perhaps, so every squeak on the camera's body is loud as hell, but there's virtually no background sound from the Maidan presenter and crowd next door, only sounds very nearby, and that single sharp pop gun shot. I'll have more to say on this, but thought I'd open it up to any first thoughts first. --Caustic Logic (talk) 09:27, 11 April 2014 (UTC)

The thing about that is it leaves out most sounds, making it easier to cut to zero and insert new sounds, without complex mixing. I didn't even think about it until I noticed how it mismatches with the audio from 6:23:59 and forward in the RT 7.5-hour video - a burts of 12 shots, followed by two more 2.5 minutes after. That seems like the only possible hotel-based gunfire at the right time, or ever, and the offset range puts this moment at about 4:49, +/- a loose 7 minutes or so. Nearest single shots, 30 minutes distant either side, and too quiet/distant to be from the hotel anyway. The RT audio or the fisheye video both claim to show a shooting at that time, one clearly believed to be from the hotel, one audibly close to that. But one has to be altered. One has goofy sound that initially seems fine (especially at the start where you can see the guy on the phone). But the other one is also the alleged sound of that time, and when it's plunked into the video, the actions make even more sense. Or so I think. Inviting feedback on this. --Caustic Logic (talk) 12:23, 18 April 2014 (UTC)