Filter bugs -- (FW 1.11.9 -- Editor 1.11.7)

First, my setup is as follows:

Inputs -
A - Analog Rytm
B - Analog Four
C - Iridium
D - nothing

Outputs -
A - Analog Rytm
B - Analog Four
C - Iridium
D - Analog Heat/ 2 Akai MB76’s

The purposes of my Midihub patch are to:

  1. Provide discrete input and output ports for each device to communicate with Max MSP.
  2. Route MIDI clock from the Rytm to the rest of the devices.
  3. Forward program change messages from Rytm to A4.
  4. Remap select performance macro CC’s from both Rytm and A4 to Iridium’s CC mod sources.

My issue:
I’m currently getting notes flowing to Iridium (MIDI output C), despite filtering/ bypassing every potential note source, as well as filtering preceding each MIDI output C module.

Here’s a video:

My patch:
04_ghost_notes.mhp (619 Bytes)

Many thanks.

I’d also like to add that I think it would be swell – for moments such as these – to have something like Max MSP’s “print” object for monitoring events at any stage of a pipe/ pipeline.

Going ahead and updating with a simpler patch that still allows errant note data to flow to Iridium:

The purposes of this patch are to:

  1. Provide discrete input and output ports for each device to communicate with Max MSP.
  2. Route MIDI clock from Rytm to both A4 and Iridium.
  3. Forward program change messages to only A4.

Here’s the patch:
06_AR_clock_only.mhp (315 Bytes)
If I bypass the final “Output MIDI C” module, ghost note data stops.

Quick question - is it constantly sending some notes to MIDI OUT C, or is it a single stuck note which got there some time earlier?

Bypassing the output makes it send note off messages for the notes it’s aware of as being on. Does unbypassing the output continue to produce the notes?

Un-bypassing allows errant note data to flow and create new note-on events on the Iridium.

My fix has been to remove all pipes/ logic – aside from MIDI to USB (and vice versa) – leaving Max to do the rest.

Ok, we’ll look into it. Do you know which of the inputs is actually inputting the notes? Do they come via USB or DIN-5 MIDI?

Bypassing the final Output MIDI C cuts off the data flow from the Input MIDI A (DIN-5) -> Virtual Pipe A.

Also, Input USB C (the USB port running to Output MIDI C) was bypassed in the video, and notes were passed.

Just checking in to see if there’s any updates on this.

1 Like

Hi, thank you for bumping this - we didn’t manage to reproduce the issue at the time and somehow missed marking down this thread in my todo, so forgot to update, sorry about that!

I think we could reproduce it if you could send us recorded MIDI clips of data coming in from MIDI A port. I see in the patch it’s already forwarded to USB A, so recording the first USB MIDI port should be enough, but recording the rest of the USB ports could be useful, in case there’s some interplay going on.

You could either post it here zipped or .mid file, or PM it to me directly, or send to hello@blokas.io

This method previously helped us find and fix some corner case issues in Dispatcher & Micro Scale pipes.