Ableton Forum

an unexplicable midi problem just popped up for which I need some help:

my footpedal is sending midiCC4 to miditrack1 via midiyoke, this track outputs to miditrack2 containing BFD2-plugin, to control the variable hihat position and to trigger the auto-pedal event in BFD2.
This way I can record notes and the CC4 envelope in the same clip on track2.
Track1 is set to Monitor In, Track2 is set to Monitor Auto.
In preferences, the midiyoke-in port is track- and remote-enabled. The midiyoke-out port is disabled.

I arm Track2 for recording and I can hear the notes played on my Alesis Controlpad and the hihat-chicks when the footpedal is fully pressed.
As soon as I start recording a clip however, extra hihat-chicks are sounding in addition to the ones I play by fully pressing the footpedal (extra chicks either at the moment of the pedalpress and/or upon release of the pedalpress).
These extra chicks are not recorded in the clip and also not shown in BFD2-midimonitor!
When I stop pressing the pedal no hihatchicks are heard.
Now: when I select Monitor In on miditrack2, no extra chicks are produced!

I thought that with Monitor set to Auto, record-enabeling the track and firing a clip-recording, effectively turns monitoring to IN, but it does not seem to be the same as Monitor IN.

I cannot get my head around this: there is no feedback, only two tracks, the footpedal produces a nice smooth CC-curve, I reloaded the BFD-plugin and built a new kit, checked the routings, nothing works....I had this setup for a long time and never had this problem.

any insight is highly appreciated!

in addition:
the outputs of miditrack1 are set to: 2BFD, Track In
if I change this to: 2BFD, 1-BFD (or any other BFD-channel from 1-16)the problem is gone.

PLEASE DISREGARD THE ABOVE POSTS.
my analysis was not proper.

The problem is:

a sustainpedal connected to a joystick sends continous CC4 messages via Rejoice (or Fergo MidiJoystick) through midiyoke to miditrack1.
The output of track1 is set to miditrack2 containing BFD2 plugin.
The input of track2 is set to Alesis Controlpad.
The purpose is to control the hihat-height and to trigger the hihatpedal-sound and to record both notes and cc4-data in the same midiclip on track2.
When track2 is armed for recording the cc-data displayed in Rejoice correspond excactly with those received in BFD2, however when I start recording a clip in track2, now and then an extra cc4 message is received by BFD2 which was NOT SENT by Rejoice.
The result is that extra hihat-pedal triggers are sounding which are not played with the sustainpedal.
Furthermore, when changing the monitor of track2 to IN (instead of auto) NO EXTRA CC4 messages are generated in BFD2!
With both monitor-settings,the recorded clip does NOT contain the unwanted midiCC data. The CC4 datacurve looks smooth.
Somehow additional midiCC data are generated in the receiving miditrack when recording is active and the monitor is set to Auto.

Does anyone have an explanation for this behaviour?

Try playing with the global Overdub button. Good luck!

yur2die4 wrote:Try playing with the global Overdub button. Good luck!

to achieve what?

Is it possible that your clip is looping? Picking up extra signals on it's second time around? I guess I should have asked if you were in session or arrangement view and how you play/record the actual clips. Another possibility is seeing if it needs to be filtered by midi channel, or midi device.

I'm not sure how many other devices you have on hand. I'd suspect you are only using the drum kit, game controller, and possibly another device??

For each of the midi channels, have you specifically selected the appropriate device, or are they accepting all inputs?

the clip is not looping, it is in the process of being recorded in sessionview, and that is the moment when the rogue midiCC-data are somehow generated. These rogue-data are not recorded in the clip.
The input of the recording track is set to Controlpad midichannel 1, and the input of track1 is set to midiyoke1 channel2, so the filtering for midichannel and mididevice seems to be OK.
Your remark about another device reminds me that I could try sending, for example, a modulationwheel-signal through midiyoke into the recording track to see if produces the same results. (although the peeps at Fxpansion just tried this and could not repro....)
Thanks for thinking!

amazing: your suggestion for playing with the overdub button worked: turning on OVR stops the rogue midi-data when recording, same as switching monitor from Auto to In!
Could you explain the rationale behind this?? why did you suggest this?

When the monitor button is activated Live is listening for data. When you have overdub turned on, it listens while a clip is playing back so that you can add more data on top of it. When Overdub is off(Auto on), Live just plays the clip back, automatically blocking out any additional incoming data. Overdub is a powerful and fun tool to learn to abuse

I understand, but why would the OVR have influence while recording a clip? afterall during recording the clip is not being played back.
The same goes for monitor Auto or In: during recording of a clip the data sent to the VST should be the same regardless of the monitor-status.
Maybe I'm still not getting the complete picture....
Thanks again for the input.

I managed a workaround: midimapping the tracklaunchbutton to the same controller as the monitorselectbutton (which was troublesome at start since switching from monitorAuto to Monitor In needs CC values 0 and 64, and the tracklaunchbutton needs CC 127-value).
This switches monitoring to In when I record-launch the track and switches it back to In when play-launching.

Still wondering what is going on with the extra CC's being generated and what the Monitor status and OVR have to do with this.