David B. Benson wrote:I'd like you to start a new thread to analyze the horizontal movements of the antenna mast dish. The idea is to look for jerks or jolts corresponding (I assume) to the various large sections of exterior wall which ripped away early on.
The goal is to establish the time of these occurances with the eventual idea of determining how much PE was released by the vertical motion and hence available to rip off those large sections.
I happen to have a little data, so here goes.
Starting with data from Sep 2007 (just in case any of you thought you were breaking new ground!), there is this post which has manually obtained data for several locations. The post shows locations, data, and graphs for everything, I'm only going to reproduce a graph for horizontal motion here, where distance is in pixels and only differential motion is shown:

Then there is data for the top center antenna dish, obtained via programmatic means. Horizontal is labeled cX, again in pixels.
- Code: Select all
frame,cX,cY
800,471.2903427,51.03925234
801,471.3431203,51.02417962
802,471.3658679,50.98480421
803,471.293125,51.02125
804,471.2013845,51.10383889
805,471.199071,51.06569343
806,471.2635706,51.08745476
807,471.2738324,51.04926424
808,471.2097471,50.96853794
809,471.2041298,50.98879056
810,471.2406802,51.07128842
811,471.2765252,51.05172414
812,471.292633,51.09822647
813,471.2321539,51.01055245
814,471.2145871,51.14044605
815,471.2580645,51.07273877
816,471.2121023,51.08172177
817,471.2122977,51.08608414
818,471.180791,51.10985562
819,471.142004,51.12408759
820,471.1652787,51.0429212
821,471.1114911,51.09233926
822,471.1437372,50.98562628
823,471.1225052,51.15072264
824,471.0266571,51.17651297
825,471.0629229,51.11028417
826,471.0475852,51.18394886
827,471.1228719,51.23019985
828,470.9548239,51.2611026
829,470.9219805,51.18004501
830,471.0588633,51.16170501
831,471.1466571,51.18044572
832,471.0783046,51.08692529
833,471.0364912,50.76982456
834,471.1154372,50.9692623
835,471.0625,51.17674731
836,471.1123512,51.0141369
837,471.1369863,51.07010475
838,471.162623,51.13180328
839,471.1207495,51.22553782
840,471.1136364,51.20974026
841,471.0244063,51.24472296
842,470.9424874,51.33213515
843,471.038358,51.48654105
844,470.9842965,51.3548995
845,470.9851695,51.22387006
846,470.9255464,51.36680328
847,470.9385965,51.29082321
848,471.0512658,51.12468354
849,470.9539386,51.28638184
850,470.9541724,51.27154583
851,470.9569409,51.32455013
852,471.0269231,51.35153846
853,471.0047554,51.26970109
854,470.9926764,51.31091877
855,470.9706304,51.4004298
856,470.9594395,51.21755162
857,470.9647887,51.4029734
858,471.0546407,51.23428144
859,470.9979564,51.42983651
860,471.0138522,51.25923483
861,471.0304799,51.26523995
862,470.9641854,51.37008427
863,471.1612694,51.19106218
864,471.1346298,51.25504862
865,471.0505391,51.32816712
866,471.0510899,51.30722071
867,470.9772584,51.37776374
868,470.899661,51.51389831
869,470.9639364,51.38569682
870,471.083125,51.308125
871,471.0169173,51.45676692
872,470.980755,51.55218357
873,471.040522,51.46085165
874,471.0006906,51.38328729
875,471.0874613,51.51547988
876,471.0791223,51.58510638
877,470.9882716,51.52407407
878,470.9730837,51.46108836
879,470.9242424,51.50505051
880,470.8491379,51.51231527
881,470.8770227,51.3961165
882,470.9603012,51.52019165
883,470.7916145,51.52565707
884,470.8632316,51.60050891
885,470.9125249,51.45858184
886,470.9000684,51.45174538
887,470.8587838,51.45945946
888,470.8215278,51.66805556
889,470.8713561,51.55006337
890,470.8175549,51.74733542
891,470.8717608,51.72292359
892,470.7887574,51.68047337
893,470.8215819,51.82158185
894,470.8212121,51.62787879
895,470.7982063,51.75080077
896,470.7303088,51.75488343
897,470.697619,51.81071429
898,470.6973913,51.77565217
899,470.6718213,51.73825888
900,470.5660848,51.84912718
901,470.437203,51.95790903
902,470.5391484,51.9739011
903,470.5141509,51.97843666
904,470.4898922,52.04043127
905,470.4993734,52.03258145
906,470.5171982,52.1163227
907,470.4777303,52.14521049
908,470.4531526,52.19210371
909,470.3826331,52.21848739
910,470.3691015,52.31674208
911,470.3728814,52.3874092
912,470.3337364,52.50906892
913,470.3613596,52.7149672
914,470.3503299,52.9640072
915,470.3678297,53.04139562
916,470.3569536,53.15364238
917,470.3590376,53.52560148
918,470.3176321,53.79503501
919,470.3787781,54.01286174
920,470.3541667,54.19730392
921,470.309434,54.39874214
922,470.2458308,54.79617048
923,470.2934017,54.78859705
924,470.209901,55.38613861
925,470.2417127,55.83425414
926,470.1894374,56.11595867
927,470.172453,56.41718839
928,470.0412587,56.99090909
929,470.1494322,57.50149432
930,470.1500335,57.94373744
931,469.9909237,58.38441004
932,469.8076923,58.83767927
933,469.8512545,59.59617682
934,469.8236776,60.15743073
935,469.7249566,60.55935148
936,469.7009512,61.20273484
937,469.6165803,62.06957809
938,469.5511811,62.44544432
939,469.6545117,63.60877627
940,469.6117455,64.0685155
941,469.6295353,64.86505411
942,469.5927222,65.99020294
943,469.4819413,66.63769752
944,469.5115995,67.79242979
945,469.4853556,68.35146444
946,469.5079365,69.59218559
947,469.4701493,70.28731343
948,469.5395349,71.66627907
949,469.5231116,72.3416009
950,469.5232975,73.59737157
951,469.5299728,74.66348774
952,469.4724409,75.95396729
953,469.506993,77.08391608
954,469.5300896,78.18053777
955,469.5793413,79.55389222
956,469.5595305,80.65343767
957,469.5134228,82.10178971
958,469.4646226,83.63443396
959,469.504483,84.70591751
960,469.4820717,86.22576361
961,469.4793673,87.99037139
962,469.4733879,89.21494371
963,469.4725957,90.4622544
964,469.4336232,92.17623188
965,469.4522613,93.86557789
966,469.5096207,95.55140187
967,469.4946063,97.11434736
968,469.5729627,98.59949463
969,469.5910931,100.659919
970,469.6708595,102.5807128
971,469.754717,104.1674528
972,469.7673378,106.0760626
973,469.800578,108.0289017
974,469.9124088,110
975,469.9712526,111.9507187
976,470.0104348,113.9373913
977,470.037931,115.8896552
978,470.0788177,118.0164204
979,470.1131285,119.9650838
980,470.2109375,121.921875
The horizontal portion; don't know if the last jog is real:

A little noisy, but pretty accurate. How accurate? This validation animation lets you be the judge:

The dot is located according to the position output from the routine. The discontinuities are from keyframe jumps; the important thing is to note the position of the dot relative to the dish blob and satisfy yourself that it looks reasonable, if not more honest than hand-eye data would be.
This should be sufficient fuel to get things going.
PS as with many of my (few) threads, it will be somewhat unfriendly to those with slow and/or expensive bandwidth. (if I understand correctly, that will be the case for most of us in the future)
