Jump to content
Bill

Pgf Site Sun Angle Investigation

Recommended Posts

parnassus

Paranassus:

See opening post.

Bill

Bill, I see the chart but can't see how you got the 27 degrees elevation. How did you get it?

Share this post


Link to post
Share on other sites
Bill

Paranassus:

You build a model in a 3D software, and test various sun angles and elevations.

I tested 180 degrees, 182, 184, 186, 188, 190, 192, 194, 196, 198, 200, 202, 204, 206 for rotational position, (0 being camera line of sight center)

and I tested 24, 27, 30, and 33 degrees elevation.

This one was the closest match so far to shadows in the PGF.

So the process is one of trial and error, testing a range of options and seeing what yields the closest result. Thus far, the above was the closest match.

Bill

Share this post


Link to post
Share on other sites
Guest

Shadows do not appear to be correct on several of the Moon shots. Take this picture for example. The shadow on the LEM is due East and yet the shadows on the rocks in the foreground are South East? If two light sources are at work on the Moon's surface, they would combine together and the shadows would fall accordingly, not at random points. Unless the sceptics are saying that Sunlight is falling in the middle of the picture and there is Earth light at the forefront of the picture? And that would go against known physics.

Enter the

. They built a miniature moon scape replicating this scene and proved that a single light source actually could produce the effect seen.

Just thought this was interesting and possibly applicable.

Share this post


Link to post
Share on other sites
xspider1
BFF Donor
:wacko: great post DDA!

Share this post


Link to post
Share on other sites
Bill

DDA:

Thanks.

Nice to know models have been used in other circumstances to test issues of shadow angle. They help resolve issues of what variables may be in play, irregular ground, things leaning instead of straight up, etc.

The Bluff Creek sunlight model is still a work in progress, and I'm still open to suggestions of any variables I should test further.

Bill

Share this post


Link to post
Share on other sites
parnassus

Paranassus:

You build a model in a 3D software, and test various sun angles and elevations.

I tested 180 degrees, 182, 184, 186, 188, 190, 192, 194, 196, 198, 200, 202, 204, 206 for rotational position, (0 being camera line of sight center)

and I tested 24, 27, 30, and 33 degrees elevation.

This one was the closest match so far to shadows in the PGF.

So the process is one of trial and error, testing a range of options and seeing what yields the closest result. Thus far, the above was the closest match.

Bill

Thanks, I was hoping for a sundial.

Share this post


Link to post
Share on other sites
Bill

Paranassus:

You may get what you hoped for. The first and second tree Patty walks behind,both have shadows on the ground visible in the film (as Roger moves forward into his final position for the "walk away" ending). Once an accurate site model is made, and oriented true north, those tree shadows will become a sundial.

We're not there yet, but moving toward that goal, day by day.

Bill

Share this post


Link to post
Share on other sites
parnassus

Bill, it's interesting to play around with the angles; using my own version of sundials, and in 20 minutes, it looks like, if the Green/McLarin film date was say, June 15?, 1968, that it was shot midmorning. It looks to me as if Green (and hence Patterson) was shooting almost north, maybe a little west, at the point where PattyBob does the famous head turn. Is that what you get? or do I have to back to trig tutorial.

Edited by parnassus

Share this post


Link to post
Share on other sites
Bill

Paranassus:

Yes, the camera for the lookback is pointed near due north. This needs more factual corroboration, but with further analysis, this should be verified.

Bill

Share this post


Link to post
Share on other sites
parnassus

Paranassus:

Yes, the camera for the lookback is pointed near due north. This needs more factual corroboration, but with further analysis, this should be verified.

Bill

Bill,

Is this done in Bryce? seems like it must be pretty fun stuff though pretty tedious at times?

How do you place the subject in the site, if the subject is not seen in the same place on multiple views?

Edited by parnassus

Share this post


Link to post
Share on other sites
Bill

Paranassus:

Yes, it's done in Bryce because that is my strongest 3D application, so I can get things done quickly and easily. Yes, testing a whole list of alternative settings can be tedious, but a lot of good research always is that way, from what I've found.

The figure positions, relative to each other, must conform to a path estimates by subject diminishing size, which translates into further distance, plus estimations of the body angle to camera, which also indicates a path direction. But again, alternatives are tested as well.

At this point, I'm simply evaluating the preliminary results and studying what variables might still be in need of testing.

Bill

Share this post


Link to post
Share on other sites
parnassus

Bill,

So do you try to match the posture and shadows on Patty's body with the Sun Control angles in Bryce? is that what I'm getting from your chart....there aren't many other good shadows because of the time of day and direction....how close does it seem like you can get to the solar angles?

Edited by parnassus

Share this post


Link to post
Share on other sites
Bill

Paranassus:

Yes, I think you are getting it correct.

As far as how close we can get, I can't say yet, until I get all the variables accounts for.

Bill

Share this post


Link to post
Share on other sites
parnassus

Paranassus:

Yes, it's done in Bryce because that is my strongest 3D application, so I can get things done quickly and easily. Yes, testing a whole list of alternative settings can be tedious, but a lot of good research always is that way, from what I've found.

The figure positions, relative to each other, must conform to a path estimates by subject diminishing size, which translates into further distance, plus estimations of the body angle to camera, which also indicates a path direction. But again, alternatives are tested as well.

At this point, I'm simply evaluating the preliminary results and studying what variables might still be in need of testing.

Bill

so you go through frame by frame and use a surrogate figure in Bryce to match up, to have the program estimate size and angle? that sounds to me like an impressive amount of work!!

Share this post


Link to post
Share on other sites
Guest

We need to solve the sun angles with some geometry. Note that true North from Roger's POV is past Tree 3, otherwise, we wouldn't see the shadow on Patty's back on the right side of Tree 3 (based on 12:20PM PST after a time zone correction for Bluff Creek). If the shadow couldn't be seen then Tree 3 (at center-frame) would be pointing due North and Roger would have the sun directly behind him. If Patty had walked past 1 more tree and the shadow showed up on the left side of the tree -- and the assumed time of day was accurate -- then we could interpolate a decent estimate for true North.

ShadowsLMS.gif

I'm not sure how valuable it is to determine the sun angle on Patty for frame 352. As Roger pans the camera to the right to keep Patty center-frame, the sun angle is constantly changing relative to Roger's POV and he faces due North when the position of the sun is directly behind him.

But without another shadow for reference, we need to know how close Patty got to Tree 3 to extrapolate true North. For this we need to determine the sun angles by solving the "Tree1->Tree3->Camera" triangle, then adding Patty's trackway and the shadows into the mix. I think we have enough info to pull this off, providing we can assume a few things, such as the FOV, Patty's approx step length and her body orientation relative to the camera. Then it is a matter of making the trackway fit over the Tree triangle and the shadows until we get what's seen in the film.

The camera position might also come out in the wash. Lots of work tho..and in the end who's gonna buy it?

Share this post


Link to post
Share on other sites

×