Is there automatic duplicate message filtering in RK-002?

Redfox

Hi everyone,

I'm working on a project using the RK002_onControlChange callback to handle MIDI CC messages from a Arturia Beatstep controller. The controller's encoders can operate in either absolute or relative mode, and I need to map the relative mode to specific SysEx commands.

When the encoders are set to relative mode, they send a CC message with a value of 0x41 for every clockwise rotation and a value of 0x3F for every counter-clockwise rotation. Naturally, as you turn the knob, multiple identical CC messages are sent (e.g., B4 1E 41 for clockwise and B4 1E 3F for counter-clockwise).

The issue I'm facing is that only the first CC message is processed, while subsequent identical messages are ignored. I suspect that the RK-002 library might have some form of duplicate message filtering.

To investigate further, I tried sending a Note On message before rotating the knob again, and in that case, the CC message was processed correctly. This suggests that there is an internal mechanism filtering repeated CC messages.

I also attempted to use the generic RK002_onChannelMessage callback, hoping it would bypass the filtering, but unfortunately, that didn't resolve the issue either. It seems the filtering is applied regardless of the callback I use.

Is there a way to disable or manage this duplicate message filtering behavior? I would like to ensure that all CC messages are processed, even if they are identical or occur in quick succession.

Thanks for your help!

albert

There is no filtering like that in the RK002. If it is really behaving as you describe, it would be a bug inside the library. Can you share the code which you're using to test ?

Redfox

Here the subroutine,
boolean RK002_onControlChange(byte channel, byte nr, byte value) { boolean ret = true; if (channel == ch_cmd) { if (nr == cc_cmd) { // SysEx Dial if (value == 0x3f) { // Down ret = false; const byte sysexMsg[] = {0xF0, 0x26, 0x79, 0x03, 0x1f, 0x7F, 0xF7}; RK002_sendSystemExclusive(sizeof(sysexMsg), sysexMsg); // Invia il messaggio SysEx const byte sysexMsg_off[] = {0xF0, 0x26, 0x79, 0x03, 0x1f, 0x7F, 0xF7}; RK002_sendSystemExclusive(sizeof(sysexMsg_off), sysexMsg); // Invia il messaggio SysEx } else if (value == 0x41) { // Up ret = false; const byte sysexMsg[] = {0xF0, 0x26, 0x79, 0x03, 0x20, 0x7F, 0xF7}; RK002_sendSystemExclusive(sizeof(sysexMsg), sysexMsg); // Invia il messaggio SysEx const byte sysexMsg_off[] = {0xF0, 0x26, 0x79, 0x03, 0x20, 0x00, 0xF7}; RK002_sendSystemExclusive(sizeof(sysexMsg_off), sysexMsg); // Invia il messaggio SysEx } } } // RK002_printf("received cc message (chn=%d, ccn=%d, val=%d)",channel,nr,value); return ret; }
Regards,
Claudio

Redfox

Hi Albert,

I’ve read your response, and since the RK-002 doesn’t have routines for filtering out duplicate events, I started thinking about what else could be causing this behavior. I modified my setup a bit and eventually discovered that the issue was caused by my Numa Compact 2 keyboard. Since its MIDI output is modified to power the RK-002, I used it as the output. The operating system has a MIDI MERGE function, so I connected the MIDI OUT of the Arturia BeatStep to the MIDI IN of the Numa Compact with a regular cable. Then I connected the RK-002 to the MIDI OUT of the Numa Compact, which then goes to my Ketron EVM Expander.

Everything was fine, except that the Numa Compact filters out duplicate MIDI messages (who would have thought, right?).

Thanks for your help, and sorry for wasting your time.

The solution would be to connect the RK-002 directly to the Arturia BeatStep, but its MIDI output only has two pins, so there's no ground. I need a clever solution—let’s see what I can come up with.

albert

oh wow.. very unexpected behaviour of the Numa indeed: good to know. You're really diving deep into it, nice!