Hi, I have tried to fix TFF and I have identified a bug. It has to do with the scaling of the search window and when that is fized the tracking code does not crash (I also enlarged the track candidate array to 5000, but I don't think that matters very much). The problem is that I have actually rewritten/modified a bit of code to make it more clear what goes on, so other can touch it. It looks reasonable when you look at the local tracks, but you need a better test to be sure. If somebody has an analysis where they could test and see if it reasonable I would appreciate it. The modified files are BrLocalTrack, BrTrackFollowFinder, and BrTrackFinderBase, and I can mail the files. Short description of problem : The projection length (next_row_z - trackcandidate_z) was also used when scaling with number of rows, but trackcandidate_z is always 0 so the scale factor goes up and down where it should depend on (last row where hit was added - next row) (That is at least what I beleive. Cheers Peter -- :-) --------------------------------- )-: |Peter H L Christiansen aka PAN @ NBI | |EMAIL : pchristi@nbi.dk | |OFFICE : Tb1 @ NBI | |PHONE : 353 25269 | |SNAIL : Hans Tavsensgade 35, 4th | |PHONE : 35 349336 | :-D --------------------------------- \-:
This archive was generated by hypermail 2b29 : Fri Mar 30 2001 - 08:18:41 EST