File talk:Ukraine Sunset orange Helmet Comp.png

I've been trying to find correlation between the three videos (activist, RT, BBC) of the sunset sniper incident. From the longer BBC video, it seems the crew (3 men) film the hotel after runnning, crouching, from behind the pillars to right next to the little stairs at the far north end. This should be visible in the RT 7-hour video with that field of view. But so far, I haven't been able to spot them running across - and I even have a rough time to look now. At 6:39:42 in the RT video, is that them standing in between the columns during the fast zoom in for "white balance"? I think so, by color match - green coat, orange helmet. I think so, but can't match the actions yet, nor the audio (to follow on that). --Caustic Logic (talk) 12:29, 11 April 2014 (UTC)

RT Video Timeline
Another update is that, in my video Ukraine: Dusk Sniper(s), I was wrong on the RT video timeline and the blank spot. That's irrelevant. The 5 PM shooting is present - at time-stamp 6:23:59 - and it's not the single pop that odd-sounding activist video portrays. Considering the last signs of sunlight fade around 6:42 in the video, putting sunset just after that, and another spot where an azimuth is reached, it seems the RT times are one hour, 19 minutes, or so, ahead of real local time. Sunset was 5:24 - about 6:43 on video. Shooting at 6:24 on video = app. 5:05 real time (the sun says no later really than 5, so maybe it's 5:03?). So, these guys are seen just minutes before full sunset, 5:20 or so, a good 15 minutes after the shots are fired. That's why you only hear some kind of banging he treats as shooting (or ??), which I should be able to match with the RT audio... but no luck yet. --Caustic Logic (talk) 12:29, 11 April 2014 (UTC)

I'm trying to get an overall grasp on the timeline so I can say what the (approximate) offset is between time-stamps and real time, and using solar azimuth to do it. To turn video time into Kiev local PM time, you will subtract an hour and some minutes. It's the number of minutes that needs set - once seeming 19-20, looking wider, it looks more like 35 minutes. The reliability of these readings and best time-stamp will vary.


 * 5:07:41 (or a bit earlier?) - Overall hotel shadow lines up w/south end of palace. On map = 231 degrees = 3:31 PM. Offset from 5:08 = 1:37
 * 5:12:46 - crown edge on edge of south door section = 233 degrees = 3:36 PM. Offset = 1:37
 * 5:38:10 - hotel crown shadow covers entire palace roof. On map = app. 236 degrees = 3:54 PM to 5:38 = 1:44
 * 6:03? (est between 5:54 and 6:06) 6:06 = crown shadow lined up with south edge of palace (extended line) = 248 = 4:53 PM. Offset = 1:10
 * 6:20:26 - I measured something and got 249 deg = 4:58. Offset = 1:23.
 * 6:38 hotel overall shadow passes the other way, letting sun shine briefly on the palace south face - to about 1/3 of the span between the pillars and the south door section.  On map = 249 deg = 4:58.  offset 1:40
 * 6:40:45 last glow on OP roof - full sunset will be (7?) minutes later. 6:48 (est) to 5:24 agreed sunset = offset 1:24.

If 1:35 were the exact offset, a 6:24 shooting (per audio) would mean 4:49 PM. Hm! The best reading says that was about 4:51. --Caustic Logic (talk) 00:23, 13 April 2014 (UTC)


 * I said "video stopped at 1:21." Is it not some minutes after the shooting? I do not see the shot on the video. (did not really look.) could it be that the video starts 30 seconds after the shot? -- Petri Krohn (talk) 07:46, 13 April 2014 (UTC)
 * What? You said that where? The blank patch is at 1:18 in my last-part-only RT video, but you don't have that. If it's related to a shot, we can't say, but I found it's not related to the shot, which is actually about 12 shots. The black patch is at 4:59 in the video, which should be about 3:26 PM. Mt bad. Shots, as I've said, can be heard at 6:24, equating to app. 4:49 PM. !! Some agitation, movement of medics, can be seen over the following minutes, one last burst of 2-3 shots comes about 3 minutes later, and I hear no others anywhere. !! --Caustic Logic (talk) 11:35, 13 April 2014 (UTC)