Well now, it isn't very often I post about a problem I've been having. Maybe someone will see this and have a clue.
In the piece I currently have up in The Listening Room, my rendition of "Honey Pie" by The Beatles, I used 5 Saxes from JABB.
About 1/2 way into the project, whenever I'd start playing the project back in Sonar from anywhere but the very top, the Saxes would all start way out of tune. I tracked the cause down to pitch bend - I would have to activate each track, wiggle my keyboard's pitch wheel so it would re-set things to Zero - I'd do that on each Sax track, then I was good to go again--until I had to stop. Then when I'd hit Play again--same thing.
I scrutinized the PVR for each track, and did the same with the Event List view for each - NO stray pitch bend data was there. Nada - Nothing. No stray MIDI data of Any variety.
It made editing the project cumbersome, and added much time to the project - highly annoying.
Anyone else have that happen before? If so - Did you find a solution?
I've seen something similar in GPO - caused by setting VAR1 way high (most obvious on brass instruments). I don't have JABB so I have no idea if JABB also has VAR1/VAR2 but if it does that's where I'd start looking.
Hi, Rob - Thanks for replying! You're a new member, but apparently are looking in the Tech Forums - it's great you're wanting to help out.
Good idea, what you brought up about Var 1 or 2 - but that's not it. In the Sonar Event View, absolutely everything which is in a MIDI track is displayed. After looking for stray pitch bend events, I scoured the list for anything else that might have crept in, but there are no Var1 or 2 events. There's just nothing there which would cause the sudden pitch shift.
And that takes it back to the pitch bend itself and not other controllers, because when I wiggled the wheel back to zero on each track, the problem would be solved. See what I mean?
Actually Randy, I'm not quite a new member - I joined back in Feb 2007 but somehow or other the board stopped recognising my login and password so I created a new one. You had some kind words to say about my first GPO piece (first movement of a symphony) back in May of 2007.
Unfortunately I have no experience with JABB - but I did have a problem that sounds rather like yours in GPO and it turned out that I'd wound the VAR1 up to 127 on some of the brass instruments; my host (overture) saved the settings so they persisted across reloads even without any midi data in the controller lanes.
Hi Randy, I have an old Kurzweil K2000 that I use for all my playing/ programming, and both potentiometers for the pitchbend and modwheel are intermittent and produce crazy results.
There have been times where I inadvertently had a mid track armed for record, and even though during playback the modwheel data was playing like it should, sometimes it would become all jagged sounding because the keyboard was still turned on and sending data. I don't think this is the case with your situation because I believe you said more than one of your tracks starts acting weird and out of tune.
Can you make an SMF and import it into a brand new Sonar project and see if it acts the same way? I wonder if it is a clock source issue.
I assume you are running your sample libraries on the same computer as Sonar? hmmm.... ya got me stumped.
I was just going to suggest something along the lines of DPDan's suggestion. That is my M-Audio keyboard developed a problem with the pitch wheel to the point I had to take the cover off and disconnect it. It would spew out pitchwheel data and render some......interesting results.
I troubleshot the problem by arming a midi track in Sonar, hitting record and let a few seconds pass. I then looked at the event list and could see dozens of pitchwheel messages.
Maybe a longshot, but the only thing I can think of outside of your file. As Dan suggested you could export/import to a new Sonar project or maybe if there's someone with Sonar and JABB out there they could open your project and see what happens. That would certainly rule out the file and Sonar and point to possible hardware problems.
Dan, Steve and Rob - Thanks Much you guys for the input. I'll run a test by importing the tracks into a fresh project, as suggested. The theory that random data is being generated and creeping in is dificult to believe because you Gotta believe me, I have thoroughly examined all the tracks in every way possible, and there's just no extra data there.
This is the only time this has happened - I'll run the test, maybe I'll find something, and if I don't--well, I'll just hope to not go through it again since it easily added a full day of extra work on that project.
I use JABB and Sonar ... I'm curious, after you have applied pitchbend, are you ending the pitchbend at 0 (no pitch bend)?
IOW, whenever I do apply pitchbend (I.E., trombone shakes, falloffs), I make sure I end the data with the little black dot back on the "0" line (sometimes after the actual note so a falloff dosen't scoop back up to pitch!), ensuring the wheel data is ending at "0" (none) (I'm usually painting it in the piano roll view).
I'm in the habit of always zeroing-out pitch bend after using it because I've had this flooey thing happen in the past.
Hi there, Frank - So good of you to offer some help.
Yes, I have the same habit as you, to always make sure I have a nice big dot showing Zero in the PRV after using pitch bend. So, afraid it's not that. You see, I've done hundreds of projects with this keyboard and Sonar, and this particular project, "Honey Pie," is the first time this has ever happened. I followed my usual routine, and once the out of wack bending started happening, I thoroughly scrutinized all tracks--as said in earlier posts on this thread. All zeros are in place, all "T"s are crossed and "I's" dotted.
I'll run the test of bringing everything into a fresh project page and see if something clears up. Just too busy with other projects to have done that yet.