https://docs.google.com/spreadsheet/ccc?key=0AtjXdIjRQo5NdEkzYjVNcm5nWXFmWlZkTVhsZm5zSlE
The above spreadsheet, for the first 60k frames, takes the negative1 provided numbers and extrapolates/finds variations between the GOUT HK, then US, then UK releases, and recalculates the totals. I used AVIDEMUX 2.5.4 on the mac and AVIDEMUX 2.5.6-1 on a PC. (well only for the HK and they matched) Please recommend other frame accurate programs to use to verify frame counts. or let me know that avidemux suxxx ballz. (for reasons unknown avidemux 2.6 on the PC dropped the frames for timebased values...)
How the spreadsheet works, first is the -1 data (start, count, end and description), then a numbered column and the values of the frame shift between the prior data and the new calculations. Then have a column for the frame range, and brief description of when the new frame appears. (gave up on these last two columns quickly)
Next comes 5 columns, the outside edges are spot verify (SV), where i'd verify that that frame was the first or last shot of that sequence, when it wasn't would drop in the number and track back until the numbers work. The middle three columns are the start / end and the number of frames in the shot. When a variation between the frame count of the shot occurs, that's got a box around it and the formula references the Shifting offset previously mentioned.
Unfortunately started with the R3 GOUT (Hong Kong) which doesn't match the US/UK GOUT. But have the values and can work backwards.
So anyone have a frame accurate program to recommend to verify these numbers. Guessing there's something between my avidemux process and -1's previous method which is returning the differences.
Was able to find some other initial errors, for instance a shot of R2 was 1000 frames too long (just a typo), and found a missing shot of Luke going to eat with Owen and Beru.
*EDIT*
oh -1 would you mind if I recommend this thread to be moved to the Theatrical vs SE forum, seems more in line which what's happening there now.