[PD] Getting Sysex output to work on Linux

Miller Puckette mpuckett at imusic1.ucsd.edu
Sat May 31 23:02:26 CEST 2008


I'm still curious why the "raw" MIDI doesn't work for you.  I need to 
find someone with a similar USB MIDI device to try to figure this out.
I did recently find out that the raw interface has a bug that prevents it from
opening /dev/midi for reading only and for writing only; the only operation
that works is for both reading and writing.  I've got a fix for that
(s_midi_oss.c; change two occurrences of "devno == 1" to "devno == 0").

But this wouldn't help if, as I read your letter, you simply don't see any
input or output devices at all, for instance in the midi settings dialog.
That sounds like a different problem from the one I was having.

cheers
Miller

On Mon, May 26, 2008 at 10:11:53PM +0900, PSPunch wrote:
> 
> Ken,
> 
> > OSS has, IIRC, been deprecated for many years now.
> > 
> > ALSA should be the correct one to use on Linux.
> 
> Deprecated in terms of people abandoning it, or is it literally not 
> functioning any more?
> 
> 
> To be specific, what I am trying to get working is a Korg padKontrol 
> which is basically a USB-MIDI device. As soon at it is connected to my 
> Ubuntu box, it is recognized as an Alsa-MIDI device. When routing this 
> to Pd, only Sysex messages are dropped. (No response when routing 
> directly to kmidimon either)
> 
> 
> Since Miller had suggested its usage, I was hoping OSS to work but not 
> quite yet.
> I understand that this classic API takes a different approach from ALSA 
> and expects applications to route directly to each other. The list of 
> MIDI devices that Pd offers to connect to is simply blank.
> Is there such thing as a wrapper that lets Alsa-MIDI devices be 
> recognized as OSS devices?
> 
> 
> --
> David Shimamoto
> 
> 
> > 
> > And, also, for the past year or so there's been a new JACK MIDI API which is a lot cleaner (and strongly-timed) than ALSA, although it of course uses ALSA for transmission to actual MIDI devices.
> > 
> > -ken
> > 
> 
> 
> _______________________________________________
> PD-list at iem.at mailing list
> UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list




More information about the Pd-list mailing list