Does Sonar mess with synths like this on your system? Solved

It's usually better to leave "Zero Controllers when Play Stops" OFF.  Instead check the next box, "Patch/Controller Searchback Before Play Starts".  This is will ensure, 1) the default controller values will be reverted to regardless of where playback begins, and 2) any controller values in the sequence will be where Sonar sets those values if playback begins at a later measure in the sequence .  Not all synths respond well to these MIDI commands.  There are sometimes anomolies that are often worked around by adding your controller data close to (a beat or so before) the event you want affected by that controller.

JG
www.jerrygerber.com

I have had a similar horrible experience trying to figure out why the pitch would get lowered by a whole step after some, but not all, playback of a midi track that had a particular Reaktor 5 sound loaded up.

It ended up being that I had to remove the check from the zero controllers box, and as soon as I did that, the problem went away.

I never did figure out what got changed in the settings, although I did look, when this issue was occurring, I just left it alone after removing that check box and it started working again.

I have seen other posts where people have commented that they too have removed that check box to get things working properly again, for various synths.

I hope that helps, 

Bob Bone

czyky


Thanks for your comments robert_e_bones. Besides my own issues dealing with this, I have, like you, seen other mentions of this anomaly. That's a big part of the reason I'm trying to encourage some forum users to try to replicate my (Drop Dead Simple to do) test case scenario (see above "DDS" instructions), so that CW support can see/acknowledge/face/whatever the problem and put it on their to-do list--so that this bug/design flaw/feature/whatever can be eradicated.

I'm hoping somebody will find a minute--literally--to try this scenario, to help isolate this long-standing Sonar issue.

PS. Best wishes on you DC gig. I feel a rather tenuous kindred spirit with your adventure. A few years ago I had email discussions with a tech person stationed in the basement of the White House regarding some software that I had written that ended up being used by WH personnel.

We got to know the Secret Service folks REAL good one day.  I was the senior systems programmer for the DC Y2K effort, and one day one of the network guys did a 'ping' test of all connected devices to see which were still active, and apparently there had once been a terminal connected in the bunker I was in to the FBI system at the headquarters building across the street and above ground.


I will never forget him wandering in to tell me he wasn't sure the physical location of one of the terminals, when all of a sudden the power to the entire bunker was CUT, and a FLOOD of dark-suited FBI's and Secret Service folks came in - they thought that someone was attempting to hack into the FBI systems.  DOH! ")


They were there in like 10 minutes from when the ping went out.  Amazing!


Bob Bone

I just tried it. The problem with your recreation steps is that I wasn't aware that I was supposed to be taking note of the settings in CUT1 before I hit the space bar. So I don't know if any of them changed afterwards. Also, you say that Sonar immediately stops playback after starting. It doesn't on my system. The NOW time just keeps going forward in silence until I stop it. I was also confused by the fact that you'd instructed me to create an INSTRUMENT track for Rapture, and then you talk about whether or not I had a MIDI track. If I have an instrument track, then a MIDI track is going to be part of that isn't it? 

But anyway, I did everything I could to order, and I didn't notice any change in the sound of the patch after I'd started playback and then stopped. Apart from that, I was confused by the instructions 

czyky

I'm trying to help support pinpoint a long-standing bug that has flummoxed a bunch of users over the years. 

You imply that the problem may be evident on earlier versions of Sonar, so I tried your test with version 7.  Sorry, nothing to report, i.e. no Rapture settings were changed.

I've got Rapture v1.22 with Sonar v7.0.2 32-bit running on Windows 7 SP1 64-bit.

Right-click on that Cut1 Depth 'widget', go to Change MIDI Config, and select Reset.
 
My guess is that the parameter was accidently 'learned' to Rapture's internal MIDI Learn feature.  When you 'zero controllers' with the transport, the MIDI message (like CC1, CC7, CC 11 or CC64) resets the 'bound' parameter to default values.  If you ever right-click (sometimes, shift+right-click) on most Cakewalk synth parameters, it will 'listen' for the next incoming MIDI message until it 'learns' something.
 
All Cakewalk synths with an internal Learn feature have a distinct Midiconfig00.bin file (or Midicfg00.bin).  The number refers to multiple MIDI learn configurations (00-09 in Rapture / LE).  So if you ever get anything flaky happening in Dimension Pro, DropZone, z3ta+, etc., try deleting that associated file first.  It will be regenerated the next (or first) time you use it.  "Change MIDI Config -> Reset" does the same thing.

If you only knew how many times I got "caught" by this in the beginning ... I also reported this as a a bug to Rene Ceballos; long after I should've known better.  Project5 used to blast out 256 MIDI messages [4 iterations of 4 CC / PB resets over 16 MIDI channels] by simply clicking on the interface or transport, with no global setting to defeat it.
 
The advantage (and sometimes the disadvantage) of the internal MIDI Learn system is that it acts across all instances of the synth, and persists through all projects.  Couple that with less-than-full documentation, and it's easy to have something slip in unnoticed.
 
If a user happens to start experimenting with MIDI Learn, or an errant right and left click occurs, it's not immediately apparent that the MIDI routing is going to"stick".  The important thing is that you're now [solved], and maybe that will prevent someone else from tearing their hair out.  Despite the potential for erratic results, I've never been a fan of dumbing-down a synth to prevent that from happening.  I'd rather see more information dispersed where the end users can find it.
 
Here's one hint:  If a parameter is using internal MIDI Learn, you can see it physically move or update when you vary the "bound" MIDI message.  The same result won't happen when you're using the MIDI / Modulation Matrix or external control from your DAW.  The exception might be ACT (which I don't use).