Skip to main content
Topic: Strange attractors (Ties running wild) (Read 3562 times) previous topic - next topic

Strange attractors (Ties running wild)

Seen this before?
Receive 'Petite Valse' from fellow in France (scanned in color at 72 DPI sent as a .gif). Photoshop and IPhoto can neither one handle making a good 300 dpi lineart copy for midiscan to use. Print it out on the inkjet, scan the print copy, midiscan eats it up!! Does better than it does on music store mint copy scores. Go figure. Edit in midiscan, convert to midi, import to Noteworthy, nice job NWC. Good sounds till the long slow piano part that goes for 6 dotted quarter notes in 3/8 time at d=66. Sounds of silence revisited in spite of much hardwork with AWE contol panel.
However--this piece will do just fine if those ties are replaced with slurs--in fact probably the composer meant slurs and midiscan did ties.
So untie those notes and set them free!
Highlight-good-tie button is depressed-good-release the tie button-bad-the tie just flipped LEFT to the closest equal pitch note-??
Untie that and it flips left again or it does not have a depressed tie button and you can't untie it. Delete the note and the tie jumps left..
After the "strange" notes have no depressed tie buttons I play the tune just to see what happens sonically. The ties don't play. After I stop play, the "strange" tie symbols are gone. However the next time I play I get the Dreaded Stuck Notes. Stuck notes go away after a save and reopen.
Other files imported from .mids will sometimes do this, some don't.
Clueless in Carolina.

Re: Strange attractors (Ties running wild)

Reply #1
I am having trouble following your question and how it relates to NWC. If you have a MIDI file that you think is being improperly imported by NWC, feel free to send it to us for review at support@noteworthysoftware.com.

Re: Strange attractors (Ties running wild)

Reply #2
The behavior was:
editing an imported .mid file, desiring to remove a tie
Highlight the lead note
click the depressed tie key
tie key pops up
tie symbol redraws connected to an equal pitch note to the left of the original, the right end of this new tie symbol would variously remain in place, move left to the original start tie note, move right to some other equal pitch note.
Or delete the tie note and the same sort of left jump would occur.
Tries at removing the bogus ties would cause them to do more left jumps until a tie spanning several equal pitch notes was drawn. Eventually, when highlighted, the lead tied note would have a raised tie button. Deleting a lead tie note with an inactive raised tie button would cause another left jump.
Playing the piece would clear the bogus ties.
Save/reload cycles cleared the stuck notes.
Not reported as a bug because I cleared the problem by play/save/reload and was unable to duplicate.
My question was "Have you seen" and the apparent answer is no. The other time I saw similar behavior was on imported files with the 32 bit beta a long time back. A review of one of the files from that era found one case of a tie/slur marking with neither tie nor slur button depressed, deleting the lead tie note cleared that right up.

The behavior will not repeat on a reconstructed version of the file I was editing at the time.
Conditions are not the same, I am now 1.50a and have neither ability nor will to duplicate the editing sequence or the start up / shutdown sequence of NWC and other programs I did that day. I think I remember WIN 95 being flaky shortly after the editing was over. Does untying ties involves windows system calls?

 

Re: Strange attractors (Ties running wild)

Reply #3
If you get flaky lines or junk on your screen while using NWC, the first thing to try is a Refresh, on the F9 key. If the unusual display is corrected after pressing F9, then you can chalk it up to one of two things:

1 - NWC screwed up somewhere and corrupted the screen display (temporarily)

2 - Your display driver screwed up and did not paint the screen properly

Either way, if an F9 clears it up, you can be sure that you do not have a problem.

By the way, the 1.50a patch does fix some things in this area, but not specifically in the tie display.