John Costella's theory: No pincushion distortion present in the Stemmons sign

[Chris Bristow]

I have never found an explanation for John Costella's observation about the right pole changing its angle of lean from frames 193 to 228. The poles are fine until you correct for the pincushion distortion and then they no longer match. In the uncorrected version (I think it is the Grodon version) they do not change at all. It is not till you run a pincushion correction program that they deviate. It is as if they had no pincushion and so were over corrected and that caused fr193 to lean differently than frame 228. The fact that they do not deviate in the Groden copy is as much of a problem as the corrected version in which they deviate.
 I have seen an explanation floated that  a leaning pole will change it's angle as you change your position relative to the pole. So the explanation is that Zapruder panned to the right between 193 and 228 thereby changing his position and causing the pole to appear to change.
 But this explanation is backwards and in no way can it explain the issue. That is because when a pole is leaning away from the observer (As the Stemmons pole was leaning away from Z) it will move in the same direction as the observer moves. When a pole is leaning towards the observer it will move against their motion. The rule is:A pole leaning away moves WITH the observer and a pole leaning towards the observer will move AGAINST the observers motion.
 The pole in the Z film moves Left from 193 to 228 as Z pans right. It moved against Z's motion. So the explanation is backwards and cannot explain this anomaly.
 also I did a full scale test of this and found the 3 or 4 inches Z panned between 193 and 228 would only cause about 1/8th of a degree variation. To get the 1 1/3 degrees seen in the Z film I had to walk 11 feet to the right.
 I heard  some skeptics point to the side of the sign and say it proves that there was pincushion distortion but that still does not explain the pole in Grodon that does not deviate(It should from pincushion) nor does it explain the right pole that deviate from 193 to 228.
 I have also tested the small amount of difference between fr193 when the pole is lower than 228. But that small amount of keystoning is only a 3 degree difference and gives an infinitesimal result and does not explain the issue of the leaning pole.
 Can anyone provide a theory as to why the poles don't match in 228 and 193.
PS it is not rotation of the frames or distortion near the sprocket areas as noting else in the Z film distorts that way.

[Dan O'meara]

I have never found an explanation for John Costella's observation about the right pole changing its angle of lean from frames 193 to 228. The poles are fine until you correct for the pincushion distortion and then they no longer match. In the uncorrected version (I think it is the Grodon version) they do not change at all. It is not till you run a pincushion correction program that they deviate. It is as if they had no pincushion and so were over corrected and that caused fr193 to lean differently than frame 228. The fact that they do not deviate in the Groden copy is as much of a problem as the corrected version in which they deviate.
 I have seen an explanation floated that  a leaning pole will change it's angle as you change your position relative to the pole. So the explanation is that Zapruder panned to the right between 193 and 228 thereby changing his position and causing the pole to appear to change.
 But this explanation is backwards and in no way can it explain the issue. That is because when a pole is leaning away from the observer (As the Stemmons pole was leaning away from Z) it will move in the same direction as the observer moves. When a pole is leaning towards the observer it will move against their motion. The rule is:A pole leaning away moves WITH the observer and a pole leaning towards the observer will move AGAINST the observers motion.
 The pole in the Z film moves Left from 193 to 228 as Z pans right. It moved against Z's motion. So the explanation is backwards and cannot explain this anomaly.
 also I did a full scale test of this and found the 3 or 4 inches Z panned between 193 and 228 would only cause about 1/8th of a degree variation. To get the 1 1/3 degrees seen in the Z film I had to walk 11 feet to the right.
 I heard  some skeptics point to the side of the sign and say it proves that there was pincushion distortion but that still does not explain the pole in Grodon that does not deviate(It should from pincushion) nor does it explain the right pole that deviate from 193 to 228.
 I have also tested the small amount of difference between fr193 when the pole is lower than 228. But that small amount of keystoning is only a 3 degree difference and gives an infinitesimal result and does not explain the issue of the leaning pole.
 Can anyone provide a theory as to why the poles don't match in 228 and 193.
PS it is not rotation of the frames or distortion near the sprocket areas as noting else in the Z film distorts that way.

Costella insists that the Stemmons sign is pasted on in each frame. This means:
a) Whoever was pasting it on was doing a great job until z228, then they cocked it up and put on a wonky pole. But when the Z-film runs there is no sudden transition from one to the other. Everything is smooth.
b) Whoever was pasting it on changed the angle of the pole ever so slightly in each frame so it looked smooth but by the time it had moved from one side to the other the poles didn't match. Why would anyone do this?
c) Costella distorted z228 himself then measured his own distortion!

[Dan O'meara]

Is it my imagination or is the pole slightly bent as it reaches the sprocket hole demonstrating the very pin-cushion effect Costella argues is not there?

[Chris Bristow]

Costella insists that the Stemmons sign is pasted on in each frame. This means:
a) Whoever was pasting it on was doing a great job until z228, then they cocked it up and put on a wonky pole. But when the Z-film runs there is no sudden transition from one to the other. Everything is smooth.
b) Whoever was pasting it on changed the angle of the pole ever so slightly in each frame so it looked smooth but by the time it had moved from one side to the other the poles didn't match. Why would anyone do this?
c) Costella distorted z228 himself then measured his own distortion!

Thanks for the reply.
  First, if Costella distorted 228 himself it would no longer be an exact match for the official frame 228 you can download from the National Archives. It would be obviously different from all other copies of 228 online but it is consistent with every copy of 228 (And 193 for that matter) I have ever come across. The National Archives version proves he did not alter 228 or any other frame.
 
"Whoever was pasting it on was doing a great job until z228, then they cocked it up and put on a wonky pole. But when the Z-film runs there is no sudden transition from one to the other. Everything is smooth."
    On this point there is some confusion. The pole changing smoothly(Or changing at all) in the pin cushion corrected Z film is a result of the pin cushion correction program.
Prior to correcting the pincushion distortion the pole never changes at all. In the uncorrected Groden version you can see the pole does not change. So instead of the correction program correcting the opposite and inward lean of the poles it takes the straight poles we see in Groden and causes them to lean outward away from each other. If the Groden version had shown the effects of pincushion the pole would lean left in 193 and right in 228. If that had been the case the correction program would have removed the lean in 193 and 228 and then the pole angle would match. But because the pole in fr193 of the groden film does not lean, the correction program corrected for a left lean that was not present, so it caused the pole to go from straight to a rightward lean instead of taking a left leaning pole and correcting it to straight up.
A simpler way to address the pincushion issue would be to leave the Costella pin corrected version out of the subject entirely. We would being addressing the same fundamental issue if we only look at the apparent LACK of pincusion distortion in the Groden version. The fact that the pole in Groden does not lean as it should from pincushion is the fundamental question because it is that lack of pincushion in the uncorrected version that creates the anomaly in the corrected version.
 It is interesting to note that in the Groden version the pincushion distortion is obvious at the corner of the wall that sits above and behind the Stemmons. As you go from 193 to 228 the corner of the wall leans one way in 193, has no lean around 208 when the pincushion is pushing and stretching objects downwards rather than left or right. Then by 228 it is leaning the opposite direction as 193. So the wall shows distortion in Groden while the pole does not. Then in the corrected version the wall no longer leans and changes but the poles start leaning and changing.

For anyone not familiar with the effects of pincushion it works like this. Pincushion distortion has to rules. 1.) It displaces pixels outward from the center of the image. So if you want to find the direction the pixel moves just draw a line from the center thru the pixel and beyond it. That line represents the direction the pixel will move. 2.) The farther from the center the pixel is the farther the distortion will move it.
So in the pole image in 193 the bottom of the pole is farther from the center so it gets displaced more than the top. The bottom gets displaced more to the right than the top and  that creates the inward lean. When the pole in screen center is gets displaced downwards directly away from the center. The bottom is farther away so it gets pushed further down not left or right. When the pole is on the right side of the frame it will lean inward to the right.
 

[Chris Bristow]

The pole has a natural lean of about 2 degrees. The issue is it does not changes from 193 to 228. When I say the uncorrected version is straight or does not lean I mean relative to the other frame. Also take into account frames 193 and 228are  not rotated to the exact same place. There is about 1/3 of a degree difference. But the amount of pincushion that is missing is over one degree.
To see how much the pincushion distortion should change a pole you can look at the anomaly in the comparision of 193 to 228. The amount of difference you see between 193 and 228 represents the amount of pincushion correction applied. So pincushion would lean the poles inward to the same degree you see them leaning outward when you look at 193 next to 228.

[Jerry Organ]

[Chris Bristow]

When I use the pin corrected frames I find that frame 228 only needs to be rotated one degree not two degrees to make it align with the wall in 193.
 

NOTE: I DELETED MOST OF THIS POST ABOUT KEYSTONING. WHEN I TOOK A FEW TEST PHOTOS I HAD THE OBJECTS TOO CLOSE AND IT MAGNIFIED THE RESULT.

[Dan O'meara]

Please bear with me chaps.
I've had a slightly wrong-headed idea about pincushion distortion and that it was a very localised effect as the images passed from the 'central display' to the area with the sprocket holes. I'm now getting the impression it's something that affects almost the entire image.
Just to clarify an earlier point Chris - you were quite adamant that Costella had not distorted any images then went on to explain about the 'pincushion correction program' which, I assume, distorts the original film to compensate for the pincushion effect. So, the flopping pole is created by distorting the images using the 'pincushion correction program', it's just that other aspects of the original which were flopping around in the original due to pincushion distortion are no longer flopping around because the pincushion effect has been corrected (I'm sure I could've put that better)
Turning now to Jerry's excellent graphic (once again).
As I understand it, these are 'non-corrected' images from the Z-film, so the pole should be staying straight as the corner of the wall 'flops around'. The first thing that stands out to me is the slice from z228. It is clear, when compared to the other frames that the top of the 'wall corner' is leaning towards the left in the way indicated by the arrow. What is equally clear to me is that the bottom of the pole is doing exactly the same thing. At about the halfway point the pole bends noticeably in the direction of the arrow, this 'bending' gets more pronounced as it reaches the sprocket hole.
Should this be happening?