This long trialogue is laughable pseudo science
There's really not much "science" involved, pseudo or otherwise. Simply accurate measurement, and some mathematics.
It's not rocket science.
Why dont you perform a trace yourself ? If you have any doubt about my data, generate your own.
You "accept" both Chandler's
and NIST's, even though they are
different
As you have seen, my data is
similar, but you reject it because... ?
Just grab a video (cam#3 or Dan Rather), track the motion and plot it yourself.
and will be rejected everywhere but here
Nope. It took an awful long time for folk here to accept the increase in data accuracy. Others still reject it, or consider the differences irrelevant.
Have you contacted Chandler yet ? (You should request his data, y'know)
Do you know what software NIST used yet ?
Do you know what software Chandler used yet ?
Do you understand why even slight differences on a velocity plot become much larger when derived to acceleration yet ?
Do you understand that, by definition, my velocity and acceleration graphs "match up" ?
You admit you don’t know much about structure yet you believe everything these anonymous double talkers say.
Why do you "believe" NIST, or Chandler ? Do you have either set of data ? Do you understand how it was gathered ? Have you checked it by tracing the motion yiourself ?
Find someone willing to put his credentials on the line publicly.
lol. Who are you ? Who cares ?
Your issue is interpretation and ill-advised repitition of absolutes which are simply not true.
I'm talking about your "AT FFA" meme.
Yet again...
Do you understand that the black line is the NIST linear fit, their estimation of approximate and average acceleration during that period.
Remember...
NIST say their
estimation of
approximate and
average acceleration is
equivalent to the acceleration of gravity during that period.
Do you understand that even if noise error reduced the magnitude of the data that the trend, or shape, would not change that much ?
The whole >g thing is an attempt to deny that WTC 7 fell at FFA for about 100' and you know it.
You really don't seem to "get it".
Even on my data, the period you could call roughly freefall (which includes the period over-g) accounts for about 80ft of the descent.
Who is denying what, and where ?
The problem, and it's your problem, is that you have latched onto a "catchphrase" which is inaccurate and misleading. You need to change that.
FEMR's velocity curve on the NIST graph is first <g
Yep, though it would be better to be looking at the acceleration graph, wouldn't it.
then parallel to the FFA line at ~1.75 s [= g]
Nope. Not parallel. Now, tell me, what do you know about reading that velocity graph if a section of a curve has an average gradient greater than the one NIST plonked on there ?
Nope. About 1s. Again, use the acceleration graph. Much clearer. Same base data.
then it stays right on the FFA line for ~1.5 s. [= g from 2.25 to 3.75].
Nope. Pretty close, but also wobbling about a bit.
Then it goes to <g again.
Yaaay.
[qimg]http://img221.imageshack.us/img221/2116/femr3.jpg[/qimg]
Please don't upload copies of my graphs to other locations, as the quality is degraded. You're welcome to link directly to them.
This graph greatly exaggerates the discrepancy
What discrepancy ?
Bit more than that, and APPROXIMATE FFA, ffs.
Again, please link directly to my images rather than upload them to other locations.
You thought the interval between 2.75 s and 3.0 s was >g but that's not where FEMR has it. He has WTC 7 falling at g there. Those two data points are NOT WTC 7 falling at >g.
I don't have
any sustained period AT "g".
The data dots are simply imperfect as even FEMR has said.
All data extracted from video contains some noise, absolutely.
That
includes the data gathered by NIST and Chandler.
Everything derived from that data also contains some noise. It's not EXACT.
Sticking a straight line over a wobbly line does not magically make that straight line more accurate than the wobbly line. It just makes it straight, averaged, approximated, simpler.
Bored.
Your belief that WTC 7 was going from >g to <g to >g to <g again is shown to be wrong.
Not by you. You're simply showing continued inability to read and interpret simple graphs.