Do Subpixel Video Measurements Reveal Errors in the NIST account of 9/11?

Major_Tom

Active Member
So how can these tracking tools be applied to debunking various claims about WTC1, 2, and 7, official or otherwise?




First, it is important to know how the collapse initiation sequences are described in the most accepted version of collapse events available to the public: the NIST reports.

The case of WTC1 provides a wonderful example of how to test the NIST description of collapse initiation movement, so let's start with WTC1. How did the NIST describe the early movement and behavior of WTC1?


All descriptions of WTC1 early movement found within the NIST reports and all the photographs they used to support their claims are reproduced on a single page for any reader to review:


3.2: NIST WTC1 Misrepresentations



To remove any ambiguity of what each description is stating, descriptions of WTC2 collapse initiation motion are also given next to the WTC1 descriptions, so the two cases can easily be compared.

There is no doubt, after reading and comparing WTC1 and WTC2 descriptions, that the NIST understanding was that WTC1 tilted as much to the south and WTC2 tilted to the east during the collapse initiation movement.

By comparing the NIST descriptions of WTC1 to WTC2, it is unmistakable that the NIST drew a direct analogy between the 'tilt and descend' movement of WTC2, which anyone can see by using the visual record, and the same type of movement for WTC1. One can easily see over what angle the NIST understood WTC1 to 'tilt and descend' and over what angle the NIST understood WTC2 to 'tilt and descend'.

One can easily see that the NIST understood WTC1 to tilt a bit more to the south than WTC2 tilted to the east during the collapse initiation sequence.

>>>>>>>>>>>>>




Using the tools introduced in this thread and applying a frame-by frame viewing tool like virtualdub, freely available to anyone, it is not hard to fact-check the NIST description of WTC1 early movement and see that it is grossly inaccurate. It is absurdly inaccurate.

The entire concept of the WTC1 collapse initiation movement and the NIST hypothesis of the initiation mechanism that resulted in such movement can be fact-checked and found to to be grossly inaccurate simply by applying the tools examined in this thread in a systematic way.
 
Last edited by a moderator:
If the top of 1WTC did not tilt as much as NIST claims it did.... aside from them being wrong and that's pretty bizarre... how does the fact that the tilt was less change the mechanism they propose for the collapse?

You can't track the inside "stuff" because you can't see it. We know that some of had come apart because the antenna dropped inside the tower... likely destroying part of the hat truss or its drop was a consequence of a partially destroyed hat truss. The first explanation makes little sense and so can assume that the hat truss lost its ability to support the 360 ton antenna and it did so with some amount of asymmetry in the loss of support because the antenna tilts AND drops.

We can only theorize as to how the hat truss lost capacity and "buckled" under the antenna. My hunch is that the bottom chords of the truss failed when they lost support of the core columns beneath it. Or the bottom chord failed when columns below it with their attached loads turned into "hangers". And this seems to suggest that the column lines that became hangers lost axial alignment at some place and so instead of their loads being directed down, the were hangers and pulling down on the hat truss. Imagine standing on a chair with your arms above your head and holding a 2x4 which is resting on some walls on either side. Someone removes the chair and instead of holding the 2x4 up... you are now hanging on to it like a chin up bar. If your weight is substantial the 2x4 will deflect. And if you weigh too MUCH the 2x4 will FAIL in tension.

This seems to be a decent theory to explain why the antenna plunged into the top of the tower. How many hangers could the hat truss carry before the bottom chords parted from tension? I suggest that once one chord failed it began a cascade of failures which each involved a momentary increase in load on the hat truss from the hanger. Once the truss loads have exceeded its capacity it folds rapidly like cheap card table...
 
If the top of 1WTC did not tilt as much as NIST claims it did.... aside from them being wrong and that's pretty bizarre... how does the fact that the tilt was less change the mechanism they propose for the collapse?


This is how FEMA described what they considered the failure mode which matched observables in their 2002 study (chapter 2), before the NIST took over the investigation:

External source:
"Review of videotape recordings of the collapse taken from various angles indicates that the transmission tower on top of the structure began to move downward and laterally slightly before movement was evident at the exterior wall. This suggests that collapse began with one or more failures in the central core area of the building. This is consistent with the observations of debris patterns from the 91st floor, previously discussed. This is also supported by preliminary evaluation of the load carrying capacity of these columns, discussed in more detail in Section 2.2.2.2. The core columns were not designed to resist wind loads and, therefore, had less reserve capacity than perimeter columns. As some exterior and core columns were damaged by the aircraft impact, the outrigger trusses at the top of the building shifted additional loads to the remaining core columns, further eroding the available factor of safety. This would have been particularly significant in the upper portion of the damaged building. In this region, the original design load for the core columns was less than at lower floors, and the column sections were relatively light. The increased stresses caused by the aircraft impact could easily have brought several of these columns close to their ultimate capacity, so that relatively little additional effects due to fire would have been required to initiate the collapse. Once movement began, the entire portion of the building above the area of impact fell in a unit, pushing a cushion of air below it. As this cushion of air pushed through the impact area, the fires were fed by new oxygen and pushed outward, creating the illusion of a secondary explosion."




So this was the FEMA theory when the NIST took over the investigation in 2002. They saw visual evidence that elements in the central core gave way first.

Their initiation causes differ because their descriptions and understanding of early motion are incredibly different. It isn't a question of 'tilting less'. FEMA saw an entirely different initiation sequence than the NIST did. It is as if they were not even looking at the same building. The NIST saw WTC1 as tilting to the south much the same way WTC2 tilted to the east, but a little bit more.

>>>>>>>>>>>>


Is isn't difficult to use the methods described in this thread, along with the free frame-by-frame video viewer VirtualDub, to see for yourselves how badly the NIST botched up their description of the early movement of WTC1 when compared to that of FEMA given above.
 
If the top of 1WTC did not tilt as much as NIST claims it did.... aside from them being wrong and that's pretty bizarre... how does the fact that the tilt was less change the mechanism they propose for the collapse?
It depends on what you define as the "collapse mechanism" - or more accurately how you distinguish the higher order more significant features of the collapse from the details which MAY be significant if anyone has a defined purpose or objective which relies on those details. There may be such a purpose but I've never seen one published and justified as legitimate.

Explaining that briefly and working down from the bigger picture issues towards details:
Both "Twin towers" collapse mechanisms can legitimately explained as two or three significant stages. The "two stage" version is:
1) "Initiation" stage running from aircraft impact through to a defined line such as "release" or "top section descending bodily" (precise definition not needed at this stage); AND
2) "Progression" stage from end of initiation to completed global collapse"

(For those who prefer a three stage explanation insert "1.5) Transition stage" - doesn't change what follows.)

Significant facts of explanation working down the hierarchy of taxonomic significance could be:
A) The "initiation" stage was a "cascade failure" of columns;
B) There is evidence for both "core led" and "perimeter led";
C) Whichever it was doesn't change the next stage;
D) All columns failed in two sub-stages (at least)
D1) Columns failed in cascading axial overload mode until there was not sufficient remaining strength to support the "Top Block"; THEN
D2) Top Block [released|descended|fell (choose your preferred word)] and failed all the columns which had survived to that point.

Now I'm asserting that "A>B>C>D1>D2" are true independent of which columns failed in what sequence. That is the "big picture" reality >> "initiation" was a cascade failure which "dropped" the "top block". Whichever columns or grouping of columns went first they all failed.

And the visual evidence being referenced here does not require sub-pixel accuracy. The dropping of the antenna is obvious to Mk 1 Eyeballs.

Ditto "tilt" - which is a feature of the cascade failure which has zero effect on the explanation provided we understand how the cascading worked.

I have never seen anyone explain why the difference - perimeter v core led matters OR to whom it matters OR for what purpose it matters. Ditto for "tilt" which is of little significance in the progressing of the cascade failures.

Clearly core first would be convenient for persons proposing CD by core cutting. BUT there is no viable hypothesis for that scenario - a few nonsense attempts by T Szamboti et al notwithstanding.

"tilt" has been subject of a lot of wasted discussion where both "sides" accepted the Bazant/Szamboti endorsed starting point assumptions for the B&Z 2002 "limit case" paper. That didn't happen in the real event. Despite both Szamboti and Bazant heading down a false trail in pursuit of those inapplicable assumptions.

You can't track the inside "stuff" because you can't see it...
True if you mean "visually track". Doesn’t necessarily mean that you cannot legitimately work out what happened. I first postulated how ROOSD got started back in 2008 - before the term ROOSD was coined but more importantly it was working blind - wasn't till ~2010 before femr2 confirmed from visual analysis what I has postulated from engineering reality.
 
Last edited:
It depends on what you define as the "collapse mechanism" - or more accurately how you distinguish the higher order more significant features of the collapse from the details which MAY be significant if anyone has a defined purpose or objective which relies on those details. There may be such a purpose but I've never seen one published and justified as legitimate.


One would simply need to read the stated objectives within the NIST report and review their logic used to arrive at their conclusions to see such a purpose clearly defined.

To save readers time and effort, I've extracted the NIST stated objectives and the logical sequence of arguments used to derive their results from the report.

The method in which the NIST went about determining a collapse sequence is reproduced from the Executive summary. The condensed form of the NIST conclusions is given in the final reports, in the opening pages of NCSTAR 1-6. That is where their executive summary is located.

An outline of their entire argument given in the executive report is reproduced in this link:

Purpose of the NIST reports


This purpose is stated in the introductions of each report. It is also clearly stated on the opening page, in the opening paragraph of their website featuring their collective work on the collapses. It is also clearly stated multiple times within the Executive Summary of findings (in NCSTAR 1-6) as can be seen in the above link.


The executive summary contains a flow chart of the steps taken by the NIST to arrive at the determination of probable collapse sequences for each building. The chart serves to show the general methodology used throughout the reports, and is available at the above link.

Note that section E.2.1, in the executive summary specifies 11 steps of the NIST approach in its most explicit form, listed as A. through K, available for viewing at the above link.

>>>>



In much more detail than the Executive Summary, NCSTAR 1-6 section 9.2 clearly describes how the NIST was reliant upon the visual record to establish a collapse hypothesis for each building. The direct quotes showing in what ways the NIST was reliant upon the visual record are reproduced at the above link.




I have never seen anyone explain why the difference - perimeter v core led matters OR to whom it matters OR for what purpose it matters. Ditto for "tilt" which is of little significance in the progressing of the cascade failures.

I agree that you most probably have never seen it, but that does not mean the information wasn't freely available since at least 2002. It was also reproduced in detail on my website for years.
 
From NIST NCSTAR 1-6 section 9.3.1 titled 'Probable Collapse Sequence of WTC 1':





on pg 298.

They use this description of tilt to justify their claims on collapse sequence.



In fact, a simple application of the tools used within this thread shows the FEMA description of early movement was way better than that of the NIST. It stood in the written record until it was overruled by...the NIST.

Likewise, One can verify that overpressurizations leading into the collapse initiation sequence did not begin on the 98th floor as the NIST claims. There are 3 separate indications of overpressurization from as far down as the 88th floor before the 998th floor ejections first become visible.
 
@Major_Tom

The question Jeffrey Orling asked was in two parts:
If the top of 1WTC did not tilt as much as NIST claims it did.... aside from them being wrong and that's pretty bizarre... how does the fact that the tilt was less change the mechanism they propose for the collapse?
FIRST he accepts the moot conditional that NIST was wrong: "If the top of 1WTC did not tilt as much as NIST claims it did.... aside from them being wrong..." THEN
SECOND he asks this question "... how does the fact that the tilt was less change the mechanism they propose for the collapse?"

So he accepts that NIST could be wrong - that is not the question - he asks if the "tilt" factor would change the mechanism. Which it wouldn't because tilt was consequential not causal.

But first recall the simplest answer which is the obvious one - the collapses occurred on 9/11 2001 and the features of those collapses were written into history on that date. Nothing that NIST wrote years later - whether right or wrong - can change the reality of what actually happened. That is the trivial but undeniably true response to the NIST involvement.

So - with NIST out of the discussion - I responded to the actual question - does "tilt" change the collapse.

And the simplest response to that is to note that "tilt" is a consequence of the collapse mechanism - not a cause. It occurred because there was a cascading failure of columns and - for each Twin Tower - more columns failed earlier on one side causing tilt. Whether it was x degrees or y degrees tilt does not change any of those main features viz:
1) Collapse in two (or three) stages;
2) Initiation stage was a cascading failure of columns;
3) Which continued until not enough columns surviving to support the "Top Block";
4) Which "fell";
5) Thus failing all those surviving columns;
6) At that stage all columns had failed; AND
7) From there global collapse was inevitable.

"tilt" occurred during that sequence and the tilt that happened is what happened as a consequence of that sequenced failure - whether or not NIST explained it correctly. Whatever NIST claimed cannot change what happened.

You correctly identify the difference between the early FEMA report and the later NIST report. One difference is that FEMA identifies core failure as the leading process whilst NIST identified perimeter failure. So what? What difference does it make? To whom does the difference matter? Why? As with "tilt" the core led v perimeter led detail does not change those seven main features. There may be some persons who have a legitimate interest in the details - If you have identified such persons and their legitimate reason(s) on your site please link to it. But it does not change the reality that the details are not significant in the broad explanation of the mechanisms.

Whichever it was does not change the validity of those steps 1>>7 that I listed. And I identified the possibility that there may be persons with a legitimate interest in those differences of details. BUT they do not change the broader explanation - the "big picture" as described by those seven steps.

Since the question was NOT about NIST accuracy most of your comments are not relevant to the question I responded to.

And I remain unconvinced that sub-pixel accuracy of analysis is needed to identify the evidence of the descending WTC1 antenna which suggests "core led".
 
Last edited:
The detailed understanding of the mechanisms which led to the collapse are only significant if they reveal some sort of "defect" in the design or construction. That is.. was there something about THESE particular designs which MAY have made them collapse at all, sooner, completely and so on... would more traditional steel frames performed "better"? And what would better be.... last longer? only partially collapse or maybe not collapse at all with the fires "burning out" leaving charred wrecked buildings.

These buildings stood apart from almost any other high rise in their design... and approach to egress and fire suppression (maybe). That is the question that perhaps needs to be answered and this means not treating these buildings as garden variety high rise steel structures found in every major city in the world. My sense is what "freaked out" to public was the idea that every high rise could/might collapse if the fires were not knocked down in a hour or two.... or would definitely collapse from a 7 hr fire or "office contents". I think this is the issue that NIST did not properly address and the left many dissatisfied with their findings.

By any name/acronym describing the collapse... can any high rise collapse from "the top down"?
Can any high rise collapse from a single column failure?
 
@Major_Tom



You correctly identify the difference between the early FEMA report and the later NIST report. One difference is that FEMA identifies core failure as the leading process whilst NIST identified perimeter failure. So what? What difference does it make? To whom does the difference matter? Why? As with "tilt" the core led v perimeter led detail does not change those seven main features. There may be some persons who have a legitimate interest in the details - If you have identified such persons and their legitimate reason(s) on your site please link to it.


So from this point of view there really was no reason for FEMA to make a report at all. Or the NIST. And if the reports are wrong, so what?

Also, any interest in the thread title is a waste of time.


FEMA would have no interest in whether the conclusions of their own report is correct or not. Neither would the NIST. The NIST was wrong about their proposed initiation scenario, but such a person who would have any legitimate interest in that has yet to be identified.



I think this is the issue that NIST did not properly address and the left many dissatisfied with their findings.

According to Ozeco, a person who may have a legitimate interest in that has yet to be identified.


>>>>>>>>>>>>>>>


Would it be OK to show how tracing methods can be applied to WTC1, 2 and 7 to fact-check various claims floating about, including official ones? Isn't that the purpose of the thread?
 
I have no wish to chase these discussions further off topic.

My recent posts were IMO at the limit of relevance to the topic viz my expressed opinion that:

A) Sub pixel accuracy is not needed to identify some features of the cascade failure of initiation including:
(i) Tilt at either WTC1 or WTC2;
(ii) Antenna downward movement at WTC1 and the link to the debate about "core led v perimeter led"; AND
B) Unaided "Mk #1 Eyeball" is sufficient for those analyses.

And I outlined the reasons why the degree of tilt does not affect explanation of collapse mechanisms and reasoning showing where the core v perimeter distinction fits into the taxonomy of the explanation.

If either Jeffrey Orling or Major_Tom wish to follow their latest themes of discussion I'm prepared to respond in an appropriate thread. At this stage I don't think this thread is appropriate BUT I am open to guidance from Mick.
 
You correctly identify the difference between the early FEMA report and the later NIST report. One difference is that FEMA identifies core failure as the leading process whilst NIST identified perimeter failure. So what? What difference does it make? To whom does the difference matter? Why? As with "tilt" the core led v perimeter led detail does not change those seven main features. There may be some persons who have a legitimate interest in the details - If you have identified such persons and their legitimate reason(s) on your site please link to it.

These are interesting questions from both a scientific and philosophical point of view. I don't want to brush them aside as they have to do with:

a) NIST and FEMA responsibility as government agencies.
b) The scientific method and the inability of the NIST to provide meaningful channels to receive feedback.
c) A massive breakdown in communication within the engineering community.

one of my favorite topics:

d) Limits of human perception and limits within scientific feedback

among other weighty topics.


If largely anonymous posters on a small forum can share simple techniques exposing how WTC1 is misrepresented in the NIST report, then there is some serious, large scale miscommunication going on. Those are interesting questions, but this thread is about tracing methods.

I hope the Ozeco questions don't disappear as off topic. Maybe they can be moved or addressed at The 9/11 Forum if they are too off topic here.



And I remain unconvinced that sub-pixel accuracy of analysis is needed to identify the evidence of the descending WTC1 antenna which suggests "core led".

Let me make the arguments first. You may be right. You may not be.
 
A brief outline of tracing previously applied to WTC1...


This image shows some points that were traced and some key points on the building that provide information from two different viewpoints:





The following graph gives a nice concept of what can be measured from the Sauret viewpoint from about 9.5 seconds before the visible collapse begins:




Larger version here: http://femr2.ucoz.com/photo/6-0-855-3?1431639481


This is antenna and NW corner movement after the raw movement is subtracted from movement of a static point.


Since the west face visibly fails from south to north, and the north face fails from east to west, watching and measuring the failure of the NW corner along the 98th floor helps one identify the moment when all columns have failed. Synchronizing video from multiple perspectives allows one to determine the position of the upper portion at the time of NW corner failure. This is the moment:






The motion over which all columns fail as seen from the east:




The motion over which all columns fail as seen from the west:







That is the basic simple argument. A much more complex argument and mapping of early WTC1 behavior can be seen at this link
 
Also, tracing of the downward movement of the Upper NW corner from the Sauret viewpoint allows one to determine the precise moment that NW corner dropping (column severance) begins:




Larger version here


>>>>>>>>>>>>>


The NIST descriptions linked earlier have nothing to do with what is witnessed here and seem to come from a different world. Remember, they were under the impression that WTC1 tilted as much to the south as WTC2 tilted to the east, but a little bit more.

For comparison, here is a gif of WTC2 tilting as columns failed from east to west. Final column severance can be determined in this case also using tracing techniques.





>>>>>>>>>>>>>

Any of the claims made here can be fact-checked or refuted using the tracing techniques tested within this thread. Feel free to refute anything or make alternative mappings.

If they are as good, or better, or refute a claim, I'll use yours to construct the maps.
 
The collapse was a dynamic event... and it progressed through time. Using tracking (precise movements of building parts, debris, smoke, ejecta etc.) of what we can see of the collapse are the results of what caused those "movements". We may not (likely not) be able to know with complete certainly what caused each observation/movement. But certainly we can make well educated guesses based on physics, engineering, the design of the towers, and what we (think) know about the planes hitting the buildings (mech damage and fuel "delivery").

When you observe an energetic burst of smoke (ejecta) you are safe in attributing this to an over pressure... of gas (air?). When you observe the top section moving down... you are safe to assume that the top no longer had adequate axial support. But was this from weakened columns? mis-aligned columns? Loss of columns? or a combination of these causes? The general statement - "failed" columns hardly explains how the tops got moving downward.

The tracking of multiple points that we can see showing different movements seems to indicate that the rigid 3D form was under stress, distorted, warped and likely lost its integrity. How would, what would cause the WTC tops' frames to distort / warp in a measurable amount? To the casual observer it looks like the tops simply move down as (integrated)_ "blocks". But the pre movement "distortions" of the facade tell us that something was going on to/with the frame.

Educated guess informed by the structure, and the effect of heat and assumed column damage is all that one can do to envision how the static structures, were distorted and warped and collapsed. And the guess have to make "sense" as a progression through time and "match" the observed movements.

The antenna movement is a tell tale that the structure which supported it failed before the top moved at all. That would be inside the core... likely involve the hat truss and multiple columns the hat truss was supported by.

Is IB of part of the facade indicative of floor truss pull in from sagging... Would this have pulled the perimeter core columns or the belt girder out as well as the facade in? Or perhaps the IB indicates buckling and the facade's inability to redistribute"new" axial forces as we the membrane like facade would be expected to do. What can we "read" about the interior ( or the forces in play) from the IB of part of the facade?

The only reason to figure this out... is to know if THESE designs performed better or worse than the standard column format. Would a standard gridded design not have collapsed as we saw? Partially collapsed? Or collapsed completely as the twin towers did? Could more robust and redundant fire suppression systems saved the towers?

So from my perspective... this "exercise" is about the engineering of the towers... was it good or not? Was it more or less sustainable given the assaults it endured than other approaches might have been?

Aren't questions that NIST did not address? Wasn't that what they were supposed to determine?
 
NEW THREAD

The preceding 14 posts have been split from the thread "Subpixel Motion Tracking: Methods, Accuracy, and Application to Video of Collapsing Buildings" to allow exploration of issues in three related but distinct themes raised by Jeffrey Orling, Major_Tom and econ41.

HOW THE DISCUSSIONS STARTED

This is how these discussions started:

There is no doubt, after reading and comparing WTC1 and WTC2 descriptions, that the NIST understanding was that WTC1 tilted as much to the south and WTC2 tilted to the east during the collapse initiation movement.
.......................
Using the tools introduced in this thread and applying a frame-by frame viewing tool like virtualdub, freely available to anyone, it is not hard to fact-check the NIST description of WTC1 early movement and see that it is grossly inaccurate. It is absurdly inaccurate.
..................
The entire concept of the WTC1 collapse initiation movement and the NIST hypothesis of the initiation mechanism that resulted in such movement can be fact-checked and found to to be grossly inaccurate simply by applying the tools examined in this thread in a systematic way.
So Major_Tom's initial focus was on tilt, on NIST errors in assessing tilt and he concludes that the NIST explanation is "...grossly inaccurate. It is absurdly inaccurate."

Jeffrey Orling responded:
If the top of 1WTC did not tilt as much as NIST claims it did.... aside from them being wrong and that's pretty bizarre... how does the fact that the tilt was less change the mechanism they propose for the collapse?
...which I read as saying - in effect - 'how does that detail of tilt change the mechanism and make the NIST explanation "....grossly inaccurate. It is absurdly inaccurate."?'

So I responded - saying in effect - "the details don’t change the overall broad picture and are only important to those persons who have a reason to make the details important."

Major_Tom was focussing on one aspect and claiming that an error in one detail falsified the whole explanation. So I parsed the issues in a "top down hierarchy" and showed that none of the main features of the actual collapse mechanism OR the NIST explanation are changed by the error of detail.
It depends on what you define as the "collapse mechanism" - or more accurately how you distinguish the higher order more significant features of the collapse from the details which MAY be significant if anyone has a defined purpose or objective which relies on those details.
.............
Significant facts of explanation working down the hierarchy of taxonomic significance could be:
A) The "initiation" stage was a "cascade failure" of columns;
B) There is evidence for both "core led" and "perimeter led";
C) Whichever it was doesn't change the next stage;
D) All columns failed in two sub-stages (at least)
D1) Columns failed in cascading axial overload mode until there was not sufficient remaining strength to support the "Top Block"; THEN
D2) Top Block [released|descended|fell (choose your preferred word)] and failed all the columns which had survived to that point.

I have never seen anyone explain why the difference - perimeter v core led matters OR to whom it matters OR for what purpose it matters. Ditto for "tilt" which is of little significance in the progressing of the cascade failures.
That is how it started - multiple sub-themes have been raised.

One sub-theme which I raised was this:
- he asks if the "tilt" factor would change the mechanism. Which it wouldn't because tilt was consequential not causal....
...which actually goes to a common source of error issue - lack of comprehension of the cascade failure of the initiation stage for Twin Towers collapses. Consequences of that include all the wasted discussion of (a) the Bazant "inspired" Szamboti claimed "Missing Jolt"; AND (b) similarly wasted discussions of whether or not "tilt" could cause or prevent axial impacts of descending columns on their lower parts. Classic examples of missing the "bleeding obvious" - arse about logic caused by starting from false premises AND the wrong end of the argument.
 
Last edited:
I always thought the core gaves way first because it simply make sense for toppling and initiation. Initiation by perimeter columns is not very logical because you clearly see they are already destroyed and the bending of them is not the cause of initiation but the result of the fact that the core gives way first. The perimeter theory is created in favour of the idea that the collapse is predicted beforehand. Harley Guy will love it. No subpixel information is needed for this...
 
I always thought the core gaves way first because it simply make sense for toppling and initiation. Initiation by perimeter columns is not very logical because you clearly see they are already destroyed and the bending of them is not the cause of initiation but the result of the fact that the core gives way first. The perimeter theory is created in favour of the idea that the collapse is predicted beforehand. Harley Guy will love it. No subpixel information is needed for this...
Agreed that sub-pixel information is not needed.

The relationship between core and perimeter contributions to the failure is more complex.

As (or if?) discussion progresses we may need to examine the cascade failure of the initiation stage. What should be without dispute is that:
A) It reached a stage where all columns had failed;
B) It passed through the "threshold" stage where the sum total support from the surviving columns became less than the weight of the Top "Block" - so the Top "Block" [descended|fell|got lower] (there are reasons to be pedantic over word meanings);
C) Whichever columns or groups of columns went in what order does not change those two statements.

Also I advise taking care to not lock up on one plausible explanation - "...bending of them is not the cause of initiation but the result of the fact that the core gives way first..." - when:
P) there are other possible explanations; AND
Q) You are arguing without a defined context or fixed points of known truth.

The "context" is an overall explanation of the cascade mechanism and some of the known points of truth include that ALL column failures - core AND perimeter - up to stage "B" above - "Top Block" falling - were contributory to the failure - not mere consequences. Whether core failing pulled the perimeter inwards OR the more popular explanation that sagging joists started the IB matters little. The resulting failure of the perimeter was as much part of the cascade as core column failure. I can explain that more fully at an appropriate time and if discussion requires the explanation.

Meanwhile let's see if Jeffrey Orling and/or Major_Tom want to follow up the discussion of the points they have raised.

The OP "Do Subpixel Video Measurements Reveal Errors in the NIST account of 9/11?" raises at least three points for discussion:
1) What errors are alleged to exist in the NIST account;
2) Whether they are legitimately described as "errors"; AND
3) Whether or not sub-pixel accuracy is required to reveal them.

The latter point IMO remains valid whether or not there is agreement that NIST made errors.

In fact if we don’t limit the scope to "alleged errors by NIST" the more open question could be of more interest:
"What aspects of WTC collapse need sub-pixel accurate measurement to explain what happened?"
 
Last edited:
Looks like interest has died.

However - I must blame myself for the OP question - "Do Subpixel Video Measurements Reveal Errors in the NIST account of 9/11?"

Which conflates two questions - so let's separate them.
1) Are there errors in the NIST account of 9/11? (Presumably the NIST explanations of the WTC collapses?)

My own position is that there are inaccuracies in the NIST reports but I am not aware of any that have been demonstrated to be:
a) Significant;
b) To a defined demographic sector of interested parties;
c) Which have a legitimate purpose which requires them to pursue that issue; AND
d) Has standing to support NIST being required to address the issue of concern.

AND
2) Do we need sub-pixel accuracy to reveal them (OR conversely to rebut the claims that there are errors.)

My emphasis is on "need". e.g. the antenna movement at WTC1 is obvious to macro inspection of the video record - Mk 1 eyeball - sufficient IMO to raise valid questions about "core led". So that much does not "need" sub-pixel accuracy.

In contrast the early motion of WTC7 was detected by high accuracy femr2 analyses. And was not visible to the unaided eye. Such early motion raises a barrier to claims of CD at WTC. So in that example sub-pixel accuracy supports rebuttal of CD claims.
 
How do we know that subpixel tracking is reliable in the first place? Won't there be errors from gaps in coverage between sensors?
 
How do we know that subpixel tracking is reliable in the first place? Won't there be errors from gaps in coverage between sensors?

Maths! Emphasis mine:
In the experiments reported in the Siggraph paper, the researchers also measured the mechanical properties of the objects they were filming and determined that the motions they were measuring were about a tenth of micrometer. That corresponds to five thousandths of a pixel in a close-up image, but from the change of a single pixel’s color value over time, it’s possible to infer motions smaller than a pixel.
Content from External Source
-- https://news.mit.edu/2014/algorithm-recovers-speech-from-vibrations-0804

That's a slightly different example, as they are making assumptions that the motion is going back and forth and extract the periodic components, but in building collapse you can assume the motion isn't back and forth, and use that to help you extract the DC component. Of course, the MIT setup was way fancier, but it was achieving way more impressive measurements. Simplifying assumptions (such as things obeying Newton's laws) let you home in on most-likely ranges of values for parameters like position and speed - you end up with probability distributions for them. Less information in the inputs, such as worse sensor coverage due to larger gaps between the cells, and you just end up with wider bell curves.
 
How do we know that subpixel tracking is reliable in the first place? Won't there be errors from gaps in coverage between sensors?
Sub-pixel tracking requires that the feature being tracked affects multiple pixels. Gaps really have nothing to do with it.
 
What about what @FatPhil said?
@Mick West 's right. An object has to be seen part in one cell, part in the adjacent cell, and the change proportions of it that that are in those two cells (i.e. the influence it has over the cell's total reading) over time will tell you about the movement of the object in the direction connecting those two cells. Gaps between the actual sensors just mean that you'll get bigger jumps in the data.

Are you familiar with anti-aliasing of lines in computer graphics? A shallow sloped line (a couple of pixels wide) will cause pixel values to change very little as you examine it from left to right. However, slowly it will have less influence in the higher pixels, and more influence over the lower pixels - effectively the slice through the line is moving down with a sub-pixel-per-step speed (and its position can be defined to sub-pixel accuracy). Big gaps between sensor cells have the same kind of effect as simply deleting alternate rows. Some of the data's missing, but you can still extrapolate from the data that remains that you had a downward sloping line with the same gradient as before.

[Edit:]Alternatively, just think of looking at a moving object through a fence. You can still work out what's walking/riding/driving behind the fence, even through there're bits of it missing.
 
@Mick West 's right. An object has to be seen part in one cell, part in the adjacent cell, and the change proportions of it that that are in those two cells (i.e. the influence it has over the cell's total reading) over time will tell you about the movement of the object in the direction connecting those two cells. Gaps between the actual sensors just mean that you'll get bigger jumps in the data.

Are you familiar with anti-aliasing of lines in computer graphics? A shallow sloped line (a couple of pixels wide) will cause pixel values to change very little as you examine it from left to right. However, slowly it will have less influence in the higher pixels, and more influence over the lower pixels - effectively the slice through the line is moving down with a sub-pixel-per-step speed (and its position can be defined to sub-pixel accuracy). Big gaps between sensor cells have the same kind of effect as simply deleting alternate rows. Some of the data's missing, but you can still extrapolate from the data that remains that you had a downward sloping line with the same gradient as before.

[Edit:]Alternatively, just think of looking at a moving object through a fence. You can still work out what's walking/riding/driving behind the fence, even through there're bits of it missing.
But in the case of a fence or missing line, the dimensions of the missing part are known, and therefore it is easy to compensate. In a camera, the dimensions and shape of the gaps is unknown.
 
Cameras are designed and manufactured from well-known components, why would that be unknown? Apart from the fact that you could easily measure it, e.g. by looking at a slanted edge?

That, and when I'm looking through those annoying speckled designs on bus windows I haven't got a clue, nor do I care, what the dimensions of the dots in its design are - my brain, just like imaging software, can reconstruct the world outside..
 
I think I would like to echo and expand on this that others have said: Subpixel level tracking is not needed because if the column weakening pattern was substantially different from what NIST calculated, it would be obvious because:

1. If the damage was more concentrated in the core, the hat truss would fail, and the core and it's antenna would fall by a number of feet before the walls began moving, due to the lack of vertical rigidity in the floors. Like what happened with WTC7.

2. If it was only the exterior walls that were damaged, then vice versa.

To expand on the hat truss failure, the outriggers are supported on just 30 exterior columns, and the sum of their capacities at fl106 is around 12 million pounds. This is around a sixth of the building weight at fl98.
 
Why not use subpixel on high Res photos to get post impact displacements to get the extent of impact damage to the core?
 
Thus website does not load.
Yes. Major_Tom's site has been "off-line" for some time. A comprehensive collection of data no longer available. And I have had no contact with him for at least 2-3 years. He was an outstanding researcher of the video record who, together with three colleagues, approached WTC Towers collapse research from a "truther side" starting perspective. Then honestly rebutted a number of truth-movement memes. The example I recall from Major_Tom - the photo-shopped graphic used by (I think) S Jones showing alleged molten steel in the debris pile. It was a collage of four separate photos and Major_Tom identified where the added bits of imagery were sourced in the debris pile relative to the central feature of heated steel. He was the first person to put the "ROOSD" explanation into mainstream 9/11 debate and was met by denials, insults and personal attacks on another forum which was renowned for hostility to truthers. (It had in fact seen a "split" of members over those issues. )

His colleague "femr2" was the person who specialised in accurate interpretation of video measurements including the related issues of "sub-pixel" accuracy.
 
Last edited:
One thing that really unsettles me is the similarity between the NW corner peak and the camera shake trace
484379556.jpg
It makes me wonder if the NW corner trace is corrupted by camera shake.
Perhaps @Major_Tom missed an optical aberration in the camera?
 
Last edited:
How do we know that subpixel tracking is reliable in the first place? Won't there be errors from gaps in coverage between sensors?
That question has to be considered in the context of some other issues. And the serious debate took place on another forum. Given this forums "rules" about no-click and referencing sources I have chosen throughout the debate on this forum to avoid attempting "proof" of sub-pixel accuracy. Because for the topics debated here it is not needed. In this thread the OP question can be answered, as I have answered it - in the negative. Sub-pixel accuracy is NOT NEEDED to reveal any potential errors in the NIST account. And that does not presume that there are any such errors. So read my arguments earlier in this thread for my reasoning.

That said "sub-pixel" tracking was only one out of a dozen factors involved in correcting/ensuring the accuracy of video measurements. It was not a stand-alone single factor. Most of the discussion on that other forum (JREF >> ISF) arose out of the discussion of the NIST<>Chandler "contention" over free fall of a zone on the perimeter face of WTC7. Both NIST and Chandler employed crude measurement techniques which MAY have been good enough for the original purpose. But there were a total of (from memory) a dozen other factors as potential sources of error. For example factors of 3D geometry - the point of view did not align with the frames of motion. One big issue was that neither NIST nor Chandler accurately identified the start and finish of their measurements.
One thing that really unsettles me is the similarity between the NW corner peak and the camera shake trace
484379556.jpg
It makes me wonder if the NW corner trace is corrupted by camera shake.
Perhaps @Major_Tom missed an optical aberration in the camera?
For reasons similar to my previous comment - it is probably of no consequence. However the rigorous work on fine motion analysis was done by femr2 - not Major_Tom. There are cross-links to femr2's "ucoz" site in the Wayback archive of Major_Tom's site but I have not tested either those links or whether they are also Wayback archived.
 
Last edited:
Wait ..it should be possible to get a 3D model of building deformations using sychronized and located videos from different sngles. Does anyone know of a software that does this?
 
As one who has been through what remains on the internet, this "work" was a bunch of amateurs playing with very noisy data. No one on this forum who participated in those experiments can repeat them. They do not have the required skills.
 
Back
Top