Philippe posted Feb 5 '16, 19:23:

Hello Joseph thank you very much for sharing this project.

I have this problem/question:
If i play a note and i take of too fast, it appears thas the note off is not recognised.
Same thing when i play twice the same note, the second is not recognised.

Do i ask too much for the raspberry? maybe my keyboard is too old (when i plug it with another machine, it works well)? do i have to change something in the config/soft (yes i'm a noob)?

Thanks in advance.

NB: The latency is very good, i look forward to add the samplerbox in my machines set!

Mirco posted Feb 7 '16, 23:05:

What do you sue for midi connection ? Serial port or MIDI Usb cable ? With midi serial i have a same problem bit it is fixed with USB MIDI cable


Philippe posted Feb 8 '16, 14:05:

I use midi from DIN connector, is it that you call serial port?
I'll try to get a midi/usb adaptator.


Philippe posted Feb 12 '16, 18:44:

Yes, i bought the M-Audio midisport UNO and my problem is fixed, thank you very much Mirco!

So, I must use a raspberry pi 2, because of the use of 3 Usb connectors (1 for soundcard, 1 for sound storage and 1 for the adaptator) and it seems that the MIDI serial doesn't work well (in the plug'n play way at least).
Now, the box! ;)

My next project is a trigger/midi box to play with the sampler box.
I'm looking at the Microdrum from Massimo Bernava.

Pavel posted Feb 18 '16, 18:43:

It does seem that serial midi fails when transmitter uses feature called "running status" to compress midi messages by 30%. Can be solved by software update.

Using USB instead of serial midi would work faster in either case.

Florian Ostertag posted Apr 25 '16, 18:09:

Had the same problem. In my case it was a thing called "Midi Running Status" some keyboards "compress" Midi if more than one key is pressed at the same time. The Midi Keyboard only sends one Note On Command and then just keys and velocity. You can read about it here:

The decoding of this kind of compression doesn't seem to be implemented in Samplerbox.

jjjjj posted May 4 '16, 10:56:

Is there plans to update the script to support running status?
I believe I have the same problem with my keyboard.
Sometimes notes get stuck and start to repeat in a weird fashion.
I use a midi/USB cable and still I get this problem.

rolf meurer posted Jan 22 '19, 01:18:

about running status:
// RsByte = Running Status Byte
// RsByte is cleared at power up.
// RsByte stores the status when a ChannelMessage Status (0x80 to 0xEF) is received.
// RsByte is cleared when a SystemMessage(0xF0 to 0xF7) is received.
// Nothing is done to RsByte when a RealTimeMessage is received.
// Any data bytes (not the staus bytes) are ignored when RsByte is cleared.

implementing this into the source code will solve the problem.


  (not published)
  I want to post as guest