Jump to content

When Was This Riding Footage Shot?

Recommended Posts


A longer shadow would indicate either a lower sun angle (later in the day or later in the year) or a more sideways sun position relative to the camera view and riders. But overall, I think we can say that filming this footage in late October is still a possibility.


Here's an example I mentioned about footage flipped by TV producers.


Why they flipped it, who knows. But film editors do, and there's nothing suspicious about it.

I think the PGF shown on YouTube, the Discovery Channel one, has the rider/tentpole and PGF frame 1 flipped, and I've seen a lot of past forum debate about why that was flipped, so it happens.

But in calculating the shadow on Gimlin, the left-right orientation doesn't affect the angles calculated, or time of year concern.


Link to post
Share on other sites

I did a bit more testing on the sun angle issue. I had mentioned the camera distance might be a factor, so I decided to test it with renders at different camera distances.

What was interesting is the farther the camera is from subject, the flatter the shadow looks, creating the illusion of a lesser rotational angle.

My first chart had the camera at a scaled 70' from subject, and I re-rendered that, and then did a camera at 120' away (with subject image scaled to match 70' render, for ease of comparison). The further camera produced a flatter shadow, (the red dotted lines at base of hoof, and the 70' shadow head, show the head in the 120' render is more flattened toward the hoof baseline) making the sun appear more to the side at a lesser rotation.

The diagram is as follows:


A 25mm lens on the camera would scale to about 120' away, for the size the man and horse are in the real footage. A 20mm lens would scale to about 96' away, and a 15mm lens would scale to about 72'. I say "about" because these are rough calculations.

For this test, I set the sun elevation at 38 degrees, and the actual rotational angle at 150 degrees ( 0 degrees being straight ahead, and 90 degrees being exactly to the right on a perpendicular to the line of sight)

But the main point of this particular post was to show the camera position, distance from subject, was a variable in the evaluation of the shadow and sun angle.


footnote: The camera distance is calculated from an estimation of the horse height from rear hoof to top of the hind quarters, as being 5', and the 70' camera was 14 times further, while the 12' camera was 24 times further than that scaled measure. If the horse is scaled differently, the camera distances would be adjusted accordingly.

Link to post
Share on other sites

Nice job Bill! My initial thought was that the sun angle (altitude) looked too steep for Oct 20th, but now that I've had a chance to look at the 3D geometry a bit more (and your graphics), I think I was too hasty ruling out Oct 20th. ONLY if the shadows aligned 90 degrees from the camera was the date no later than Sep 10th. But I stated this was not the case and more likely the shadows were angled greater than 90 degrees and the bigger the angle, the shorter the shadow and the later the date. But how much later? It's tricky to estimate the sun angle from the camera position relative to the horse and rider. The apparent length of the shadows are very deceiving. Too much of the backsides of the horses & rider are lit up by the sun to be casting shadows at 90 degrees relative to the camera. But what angle must the sun be to shorten the shadows according to what we see in the photo? That is the angle we must determine and its feasibility relative to what we see in that frame.

The following graphic shows an apparent sun angle of 54 degrees (Sep 10th) along with the sun angle of 38 degrees (Oct 20th 12:30pm DST).


Here is another estimate.


But these are based on the shadows aligning at right angles to the camera, which was not the case. So what configuration would it take for an Oct 20th date? Bill looks like he's pinned this angle down to 140-150 degrees (50-60 degrees past 90). The following animated GIF is my stab at estimating this angle. You might have to click to animate.


This shows the length of shadow (radius of the protractor) for Oct 20th if the sun was at right angles to the camera and the "apparent" length of the shadows on the film. I have also superimposed an overhead view of the shadows from ground level for Oct 20th. Think of it as a piece of protractor standing on end with the center going thru Gimlin's head to the ground. The next frame shows the protractor pushed over and laying flat on the ground simulating a distance from the camera of ~80 feet.

Note that the alignment of shadows matches the protractor pretty well. The angle of the shadows on the ground relative to the camera must have been approx 149 degrees (59 degrees past 90). All this shows is what the shadow angles must have been for an Oct 20th film date. In particular look at the shadow on the right box on the packhorse. If the sun was at right angles to the camera, this box would be in shadow, which it clearly isn't. As a matter of fact all the shadows in this frame are angled lower left to upper right. Is a 149 degree shadow angle reasonable? We need to measure more frames to pin it down, but IMO we can no longer rule out a filming date of Oct 20th. And round and round we go.

Link to post
Share on other sites


I was frankly surprized by this test, because I didn't think an angle of around 38 degrees elevation would compress as much when the rotation was adjusted, to resemble the photo.

But it does show that with proper testing of digital models, we can discover new things to factor into our analysis.


Link to post
Share on other sites

I was just piddling around the web and found this PDF of the Kodak K-100 user manual. I thought I would make it available to anyone who doesn't have a copy of their own. Sorry for the OT post. BTW, it's a big file.


Link to post
Share on other sites

Other factors that might affect the shadow examinations - just throwing out some thoughts here.

The angle of incline or decline of the road. Is there an appreciable difference in the length of the shadow if the road is actually a 10 degree decline compared to a 10 incline?

The angle of pitch of the road? Same idea but the angle of the road from side to side, not from foreground to background.

Link to post
Share on other sites
This topic is now closed to further replies.
  • Create New...