I am isolating a midi problem to try and figure what is going on in an old file.
 
For testing I load the same patch prog in Dimension Pro (1.5.5.13) and Rapture Pro
They are playing the same midi (each has their own version of the MIDI)
I used the Grandmother's Song patch in Leads for testing, but any will do it.
 
The Dimension Pro will play once until it get to a certain point and the synth goes dead.
It will never play again.
Only by loading a new entry of DPro; deleting the old and restarting can I get it to play again.
No problem with Rapture Pro. (I would rather use DPro it is throughout the project)
 
I have looked for odd controllers, but see none.
 
Any suggestions on what might be happening or where to look?
 
It is from a very complex project started at least ten years ago which I am trying to resurrect and clean up.
 
Have replaced all the DXIs that Cakewalk has dropped with the 64bit VSTs.
Working totally 64bit.
 
I reduced the problem area down to four tracks (two instruments two MIDI) maybe 50 measures.
 
We can't seem to upload attachments so I am not sure how to go forward.
 
Any suggestions?
 
 
 
 
 
 

When you say there are no odd controllers, are you including all MIDI events in your review. I would suspect a patch change message telling the synth to change to a silent or absent patch which would remain in effect until another patch change was received, but that is certainly not the only possible issue. If the effect is always at the same point, look at everything in the event list near that point. 

Attachments are not allowed on this forum. Upload the project or midi file to a file hosting service such as dropbox, a Google drive or MS One Drive and add the link to a post.
 
Often times it is RPN or NRPN data which can be spotted and removed from the event list. Pretty sure a patch change would do it because CW sfz engine synths do not support them. It is interesting RPro does not shut down too. Maybe they added some (much needed) filtering to the plug-in.

The controllers are such arcana to me. In the effected area that the problem occurs:
 
10-pan 64
1-Modulation 5
91-external effects depth 33
93-chorus depth 22
10-pan 64
notes on either side.
 
No patch changes.

Have you tried removing all the data other than notes from the problem area then adding the data back until the problem reappears?

Well that stopped the problem. Slowing put them back in to see what was bad.

The problem was with
RPN Pitch Bend Sensitivity 1535
But if I remove it no glitch, but notes afterward sound funny.
 
I looked up RPN...gobbledy gook.
1535 represents +3 or -3 semitones?
tried 0, but glitch back.
 
Oh well, this is enough time wasted, I guess I am stuck using Rapture Pro.
 
Thanks for you help, at least I learned where the problem lies.
 

Pitch bend sensitivity is set by element in the synth. The elements in Grandmother's Song are set to +-2 semitones.

"Grandmother's song" was just for the test.
I went here
http://odasan.s48.xrea.com/dtm/wheel.html
to see what the numbers meant.
1024 does not produce the glitch, but the notes still sound funny.
 
I have hacked and split up this MIDI so many times I deserve trouble I fear.
 
I use many different patches with this midi.
 
I wonder why it doesn't glitch Rapture Pro and only Dpro?
 
 
There is a setting in VST plug-in properties:
Do not intercept NPRNs I wonder if this would do anything.
It says somewhere you have to replace the synth after changing these settings so I will try it after dinner.
 

Yes, that setting usually fixes the problem too.
 
The only reason I can think of why RPro is immune to the problem in this case is the plug-in has some additional filtering to ignore the instruction.