3D study of CE 133B
ShutterBun
This is a 3D model I created yesterday in response to a debate about the
shadows in CE 133B over at a.c.jfk yesterdat, and I thought there might be
some interest here as well.
The debate had to do with the fact that Oswald's rifle appears to be held
up in the "11:00" position, while the shadow of the rifle on the ground is
closer to the "9:00" position. As this model shows, the shadows are
perfectly consistent with the sun's angle that day.
The 3D program I use allows the sun to be calibrated to both the date &
time, as well as the geographic location. Obviously, this kind of
modeling could prove invaluable in photographic research & reconstruction,
as it allows us to view the scene from any angle (as with Dale Meyers'
Zapruder model) and, if we were really nail it down accurately, to use the
shadows to pinpoint the exact time the photos were taken. I've settled on
11:15a.m. as the time for this photo, as it seemed to match closely. A
more painstakingly accurate model could provide a more precise estimate.
Next up, I'll be recreating the CE 133A pose. If anyone has any
suggestions for improving the models, or requests for alternate angles,
feel free to ask, and I'll be happy to oblige.
(and yeah, I realize my "Oswald" model doesn't particularly resemble LHO,
but he's just a stand in. Facial modeling is tres difficile.)
http://img297.imageshack.us/img297/9454/oswald133bfinaldate.jpg
http://img24.imageshack.us/img24/2265/oswald133b34anglesk.jpg
David Von Pein
Excellent.
Anthony Marsh
> This is a 3D model I created yesterday in response to a debate about the
> shadows in CE 133B over at a.c.jfk yesterdat, and I thought there might be
> some interest here as well.
>
Well, it's cute, but what is it supposed to prove? Are you claiming that
you perfectly duplicated the shadows in the backyard photo? Do you not
notice that in the backyard photos it appears that the shadow on the
ground from Oswald's body falls at a different angle than the other
shadows?
> The debate had to do with the fact that Oswald's rifle appears to be held
> up in the "11:00" position, while the shadow of the rifle on the ground is
> closer to the "9:00" position. As this model shows, the shadows are
> perfectly consistent with the sun's angle that day.
>
What was the sun't angle that day? Can you prove the exact day and exact
time the photo was taken? If so then you should be able to tell which
photos were taken when relative to each other. And maybe when the missing
photo was taken.
> The 3D program I use allows the sun to be calibrated to both the date &
> time, as well as the geographic location. Obviously, this kind of
> modeling could prove invaluable in photographic research & reconstruction,
> as it allows us to view the scene from any angle (as with Dale Meyers'
> Zapruder model) and, if we were really nail it down accurately, to use the
> shadows to pinpoint the exact time the photos were taken. I've settled on
> 11:15a.m. as the time for this photo, as it seemed to match closely. A
> more painstakingly accurate model could provide a more precise estimate.
>
You may not be up for this, but certain people here question when some
famous photos were taken in Dealey Plaza. By studying the shadows you
should be able to narrow it down.
> Next up, I'll be recreating the CE 133A pose. If anyone has any
> suggestions for improving the models, or requests for alternate angles,
> feel free to ask, and I'll be happy to oblige.
>
What is the length of your rifle?
> (and yeah, I realize my "Oswald" model doesn't particularly resemble LHO,
> but he's just a stand in. Facial modeling is tres difficile.)
>
> http://img297.imageshack.us/img297/9454/oswald133bfinaldate.jpg
>
> http://img24.imageshack.us/img24/2265/oswald133b34anglesk.jpg
>
Please be very careful. This is getting very close to actual research.
ShutterBun
Appreciate the cautious, yet somewhat hopeful skepticism, Tony.
Granted, you've been at this a lot longer than I have, but I seem to
recall that you were an early proponent of computerized
reconstructions, even in the earliest days of rasterizing and
raytracing.
Now then, point by point:
1. Do the shadows appear to fall at different angles? Well, from
"Oswald's Camera" I suppose they might, but this is the whole reason I
included alternate angles. Keep in mind, I did NOT set out to
reconstruct shadows. I merely built a model, set the "sun" element to
the correct date, time, and location, and let the shadows fall where
they may. In point of fact, the "golfclub" shaped shadow on the
support beam was quite accidental, but once I noticed it, I knew I was
on the right track. Again, these images are not merely "photoshopped
recreations, " they are true 3D models, and behave in highly accurate
ways. Short of building a time machine, this sort of reconstruction
is arguably our best bet to reconstructing the true events of when &
how the backyard photos were taken.
2. I used March 31 as the date, which seemed to be the consensus
amongst the written record. I didn't know the exact time, but I
seemed to recall that Marina mentioned "late morning" or possible
"around midday" as the time of the photos. At that point I had to
guess, but as there were certain known features (buildings, etc.) I
feel that I got a pretty accurate timeframe. I can't tell you right
now what the sun's actual angle was, but as you can see from the
overhead shots, it was quite high, but also somewhat easterly.
3. I'm working on a full-on model of Dealey plaza right now. And in
fact, I've already tackled one such myth via 3D animation, namely the
"3 shadows" in the Altgens photo. Using 3D modeling, I was able to
prove that the shadows were produced by Moorman, Hill, and Brehm (with
a slight contribution from Brehm, Jr.) That video can be seen here:
http://www.youtube.com/watch?v=rGfgUpD71kA&fmt=18 (it runs kind of
fast, so some frame scrubbing might be in order) The only problem
with doing analyses like these is that no matter how careful and
honest one is, people still come out of the woodwork to poke holes &
criticize (which is fine ) but then attack the whole methodology as
some kind of "voodoo", which I resent. The principles are sound. If
I've misplaced something, or made something the wrong scale, I can
accept that. But so many people insist that the whole idea ought to
be ash-canned, which aint cool.
4. The rifle in my "backyard photo" is 40 inches (the "ruler" is
divided into 1 foot lengths, with the inches from 36 to 48 marked by
red stripes). Oswald's height is approximately 5'10" (difficult to
pinpoint due to his somewhat slouchy posture) But, as you're so fond
of pointing out, within 1 inch is close enough for a WC defender ;-)
Just for fun, here's my preliminary version of CE 133A
http://img44.imageshack.us/img44/6921/oswald133abi.jpg
timstter
This is very interesting work, Shutter Bun. Well done!
Of course, the idea that the backyard photos are fakes is absurd, LOL!
Regards,
Tim Brennan
Sydney, Australia
*Newsgroup(s) Commentator*
Anthony Marsh
Well, I appreciate your effort at doing something. I can't vouch for it
because I know nothing about your qualifications or the program you used.
You log on with an alias instead of your real name, and the alias appears
to be a typo as if you meant to write Shutterbug, as in a photographer. So
you may be a professional photographer who does not want his professional
name associated with a controversial subject.
> Granted, you've been at this a lot longer than I have, but I seem to
> recall that you were an early proponent of computerized
> reconstructions, even in the earliest days of rasterizing and
> raytracing.
Of course, which is why your attempt is interesting.
>
> Now then, point by point:
>
> 1. Do the shadows appear to fall at different angles? Well, from
> "Oswald's Camera" I suppose they might, but this is the whole reason I
> included alternate angles. Keep in mind, I did NOT set out to
Yes, I am not talking about the HSCA analysis of the nose shadows or
anything like that. What I mean is that when I look at the shadows they
appear to point back to the light source being in different positions for
different objects. There was an expert of digital forgery who pointed out
that conflicting light sources are often the first clue that a photograph
has been retouched or altered.
When I look at Oswald's shadow I can visualize the sun high and to the
right. But when I look at the steps I visualize the line going back to the
left in order to create the shadow on the post. I don't know if your
software automatically recreated the objects from the photograph or
whether you had to create 3D objects in space from scratch. But for an
example the shadow of the tip of the rifle falls on the post exactly as
seen in the backyard photo.
> reconstruct shadows. I merely built a model, set the "sun" element to
> the correct date, time, and location, and let the shadows fall where
> they may. In point of fact, the "golfclub" shaped shadow on the
Yes, did the software automatically calculate the position of the sun or
did you have to look it up and plug in the numbers? What were the angles
for the sun at that time? How do you prove the exact day and exact time
that each photograph was taken? What margin of error is there? Could you
work it backwards by duplicating the shadows in the photo and therefore
knowing the exact date and time?
> support beam was quite accidental, but once I noticed it, I knew I was
> on the right track. Again, these images are not merely "photoshopped
> recreations, " they are true 3D models, and behave in highly accurate
> ways. Short of building a time machine, this sort of reconstruction
> is arguably our best bet to reconstructing the true events of when &
> how the backyard photos were taken.
>
I guess your model is about halfway between Paul Burke and Dale Myers, but
I wonder how you determined the exact dimensions of each element or if you
just used a nominal or average value. For example, in real life were all
the steps exactly the same width, thickness, length? Just as an aside
someone could attempt to model the TSBD and use values for the bricks
which they found listed in a construction booklet, but pick the wrong kind
of brick and get the dimensions wrong. Or one could, and one did, actually
measure the dimensions of the bricks, and use that value.
Fine with me. Just wanted to tip you off that various kooks have
theories which negate those lengths. For instance that the weapon shown
was actually the 36" carbine that Oswald ordered. Or that Oswald was not
really 5'10" tall.
ShutterBun
Thanks again for the feedback. Glad you didn't "jump all over it" as some
might, as my efforts were, if anything, sincere. As far as my name,
ShutterBun isn't a typo, it's the name I use as a photographer in the
Hollywood goth/fetish club scene. Google it and there may be some NSFW
samples available. Suffice to say, tend to focus on a certain part of the
female anatomy, hence the moniker.
As far as the software I use, it is called Cinema 4D, which is a fairly
mainstream/high end software package, approximately equivalent to what
Dale Myers used. When building a model, the user can specify the sun to
act as a light source, and the software will automatically calculate the
position (according to some complex math) and behavior (according to some
other mega-complex math) for that time of day, in that location.
Fortunately these days, one need only plug in the location, and the dat &
time, and the software does the rest. Keep in mind, I made the model
first, then hit the "render" button, at which point all of the shadows
fall where the software tells them to. Once again, just to emphasize: the
shadows are NOT created by me, they are generated by the software. It
takes approximately 30 minutes to go from a "wire frame & stick figures"
to a fully realistic scene. Therefore, I was most gratified to see that
this model I had built resulted in a very close approximation of the
actual photo. The "golf club" shaped shadow on the support beam for the
stairs was a real treat. I hadn't even noticed it at first, but there it
was, in my final render.
Perhaps it would be helpful to post some "wireframe" shots, so you can see
how things look before they are rendered, and how the objects appear
within the software itself. I will post some as soon as I can.
As far as the dimensions, well, the one thing I knew I had to get right
was Oswald & the rifle. The "Oswald" character was calibrated to be
approximately 5'10" tall, but since he was leaning on his hips slightly
and somewhat "relaxed" you can see that his actual height appears somewhat
shorter. I figured this was reasonable. The Carcano was modeled somewhat
quickly, but steps were taken to ensure that it had the same overall
dimensions, with regard to lenght, width, and height. Other
considerations were merely cosmetic.
Other items, like the walls, the trees, and the staircase were more or
less "eyeballed." The trees were made from pre-existing models I had,
scaled to approximate what I saw. The walls were placed where I figured
they were. The staircase was something of a challenge, but really I
wasn't overly concerned about it, since the main issue was not *where* the
pictures were taken, but rather with how Oswald's shadows behaved.
My Oswald model is a bit beefier than the real thing, but that was due to
expedience. His main role was to stand where & how I positioned him, and
hold up his props.
I think one of the drawbacks is that the "alternate angles" I posted were
a bit dark. If you'd like, I can post a "Sun's eye view" of the scene, so
that you can see exactly how the rays hit the scene. Trust me, when you
see it from that angle, all the shadows make sense. As a bonus, I think I
can even model a "Marina the photographer" figure, so that we can see
about where she was positioned. Gimmee a day or two for that.
Again, thanks for the feedback. If nothing else, I hope I've at least
demonstrated that this type of analysis can indeed be beneficial, even for
the layperson. It's certainly easier to see the results of a geometric
comparison presented this way, as opposed to the hopeless prose of the
HSCA, etc.
John McAdams
wrote:
Impressive.
The HSCA proved the shadows are consistent, using vanishing point
analysis.
But this is a powerful tool, much more technically advanced.
.John
--
Kennedy Assassination Home Page
http://mcadams.posc.mu.edu/home.htm
Greg Jaynes
That's pretty cool. You did that in one day? I like it.
Respectfully,
Greg Jaynes
Anthony Marsh
But don't you have to specify the location of the camera? That has to be
a guess.
> takes approximately 30 minutes to go from a "wire frame & stick figures"
> to a fully realistic scene. Therefore, I was most gratified to see that
> this model I had built resulted in a very close approximation of the
> actual photo. The "golf club" shaped shadow on the support beam for the
> stairs was a real treat. I hadn't even noticed it at first, but there it
> was, in my final render.
>
> Perhaps it would be helpful to post some "wireframe" shots, so you can see
> how things look before they are rendered, and how the objects appear
> within the software itself. I will post some as soon as I can.
>
> As far as the dimensions, well, the one thing I knew I had to get right
> was Oswald & the rifle. The "Oswald" character was calibrated to be
> approximately 5'10" tall, but since he was leaning on his hips slightly
> and somewhat "relaxed" you can see that his actual height appears somewhat
> shorter. I figured this was reasonable. The Carcano was modeled somewhat
> quickly, but steps were taken to ensure that it had the same overall
> dimensions, with regard to lenght, width, and height. Other
> considerations were merely cosmetic.
>
> Other items, like the walls, the trees, and the staircase were more or
> less "eyeballed." The trees were made from pre-existing models I had,
> scaled to approximate what I saw. The walls were placed where I figured
> they were. The staircase was something of a challenge, but really I
> wasn't overly concerned about it, since the main issue was not *where* the
> pictures were taken, but rather with how Oswald's shadows behaved.
>
I am not concerned about Oswald's shadows, but if his shadows are
consistent with the shadows from other objects.
ShutterBun
> But don't you have to specify the location of the camera? That has to be
> a guess.
That's right. And to be honest, it would probably take several hours of
fiddling to really get it zeroed in. In this case, I at least knew what
the nominal focal length and field of view were for Oswald's camera, and
more or less lined it up as closely as practical with the model I made.
Some of the backyard buildings required some tweaking and adjusting, since
again I didn't have any blueprints, and could really only rely on what I
interpretted from the photos. The building next door to the house has
since been turned down, so I wasn't able to find any decent shots of the
surrounding features.
Assuming I can further refine the camera angle & position, I ought to be
able to put a model of Marina in the scene, approximating her position &
posture (as suggested by the camera's position)
avon
wow that is an excellent bit of 3D modelling. I have been looking for
an excuse to post this... its some art guys work on the CE133 pic. and
its just a bit of fun.
http://modernartobsession.blogs.com/modern_art_obsession/2008/03/artist-intervie.html