CC Messages Crash CopperLan

[initself]

I have the following CopperLan configuration set up:

Atari Sequencer Out >> MIDI In 7
Atari Sequencer In >> MIDI Out 7
Nord Lead In >> MIDI Out 1
Nord Lead Out >> MIDI In 1
Controller Out >> MIDI In 8

When sending CC messages either from the Nord to the sequencer, or from the sequencer to the Nord, my AL-88 unit will disappear in the CopperLan software completely, causing the synth to hang on whatever notes were last transmitted. The MIDI Out 1 light stops flashing on the unit altogether. It comes back eventually, but then inevitably gets overwhelmed by the CC messages and crashes again.

Is the CopperLan software simply unable to handle that kind of traffic? Crashing the entire unit seems like a drastic action to take for such data.

I am about to give up on the AL-88/CopperLan solution, too many headaches, not enough music making.

Please advise.

[initself]

Instead of CC messages, I just slammed on the keyboard and sent a ton of notes to the unit. CRASH.

[CopperPhil]

Hi,

Please note that we are not the manufacturer of Alyseum products.

We did performance test with the AL-88, sending full speed MIDI simultaneously on each port without crashing the device.

CopperLan is able to handle far more messages/second than MIDI, and contains a security feature to avoid crashing hardware equipment running at MIDI speed if it receives more messages/sec than it is physically possible to get out. If the AL-88 comes back when you stop sending, this may be a symptom that this flow control has been activated.

Finally, your issue can have two causes:
- you are sending more than 1000 note on/off or CC per second... (can be due to a MIDI loop re-injecting and multiplying the messages)
- the AL-88 has a problem, so please contact Alyseum (info@alyseum.com) to get more support.

Hope this helps, and let us know when your issue is done.

Best regards,

/Phil

[initself]

I have three of them, all upgraded to the latest firmware, all showing the same symptom.

I understand you aren't the manufacture. I also have nothing but respect for all the effort you put into supporting this forum, and others.

I don't think slamming my controller with two hands 4 times would generate > 1000 notes, but it does look like I am suffering from the problem you describe.

Performing the same tests directly recording into an Atari ST running Cubase, no sweat. I could look in MIDIOX to monitor the stream, but I do not believe there is a loop, because my test setup is so simple.

The way it handles note on/off resets in Cubase VST (reported in a previous thread) shows that there is something off about the way these units handle "too much data".

[CopperPhil]

Just for clarification: your setup is using IN & OUT 7, IN & OUT 1, IN 8. Can you tell me the connection done between these ports (virtual MIDI cables and/or single connections) and eventually some computer's VMIDI? Are you using MIDI clock in your setup? an arpeggiator?

I guess that you're sending MIDI stream to the OUT 1, right? If so, it would be great if you add a connection to the port currently sending to OUT1 in order to also send data to a computer's VMIDI. Then you can spy what is sent to the AL-88 using MIDI-Ox. If you have some USB-MIDI dongle, you can also listen what's issued from AL-88's OUT1 in MIDI-Ox. Comparing both AL-88 in and out streams could be very helpful. You can send me the files for analysis (pca@copperlan.org). It will also allow me to try to reproduce.

[initself]

Ok, installed a firmware upgrade provided by Alyseum to one of my units. This upgrade did appear to fix the rebooting of the CopperLan Manager with an overflow of notes! Thanks!