[LAD] Re: MIDI 2

[Will Godfrey]

Thanks for this info.
Some bedtime reading in order I think :/


On Sun, 17 Mar 2024 13:38:01 +0100
Florian Hülsmann  wrote:

>Sorry, quick correction
>
>> you'll see duplicate ports in your DAW  
>
>was wrong, as the kernel doc explains:
>
>> When a device supports MIDI 2.0, the USB-audio driver probes and uses the 
>> MIDI 2.0 interface [...] as default instead of the MIDI 1.0 interface  
>
>Am So., 17. März 2024 um 13:21 Uhr schrieb Florian Hülsmann :
>>
>> Hey Will,
>>
>> tl;dr: it should work without issues, just don't expect DAW/plugin
>> support for MIDI 2.0-only features yet :)
>>
>> ALSA has pretty good support for USB MIDI 2.0 from Linux 6.6 on
>> (rawmidi + seq) [1][2], translating transparently between MIDI 1.0 and
>> 2.0/UMP. Obviously you'll loose precision and maybe a few features
>> translating from MIDI 2.0. So far I only tested with a RasPi Zero 2
>> acting as a mock MIDI 2.0-only device [3], I'd expect "real" MIDI 2.0
>> controllers will present as both legacy USB MIDI and USB MIDI 2.0, so
>> you'll see duplicate ports in your DAW on recent kernels and legacy
>> only on < 6.5 kernels.
>>
>> Regarding everything above ALSA:
>> - JACK doesn't have a special port type for UMP and using the existing
>> MIDI ports for that feels weird (byte buffer vs. packets etc.), so
>> let's wait for/make that happen
>> - no free DAW yet with MIDI 2.0 support
>> - VST3 and CLAP already allow high resolution parameters and note
>> expression by their own event abstractions, however this is useless
>> without DAW support
>> - CLAP plugins can also interface MIDI 2.0 directly
>> - LV2 currently limited to MIDI 1.0
>> - WINE doesn't implement the new Windows MIDI Services yet
>>
>> [1] ALSA details: https://docs.kernel.org/sound/designs/midi-2.0.html
>> [2] General overview (ADC 23): https://adc23.sched.com/event/1PueB
>> (video not up yet)
>> [3] https://gist.github.com/cbix/97a341c2857fd4f55d0cd19ccf6c354b
>>
>> Flo / cbix
>>
>> Am So., 17. März 2024 um 12:25 Uhr schrieb Will Godfrey
>> :  
>> >
>> > I've seen quite a few adverts for MIDI2 keyboard controllers now, so has 
>> > anyone
>> > tried interfacing with this yet?
>> >
>> > If so, what sort of problems have you had to resolve.
>> >
>> > --
>> > Will J Godfrey {apparently now an 'elderly'}
>> > ___
>> > Linux-audio-dev mailing list -- linux-audio-dev@lists.linuxaudio.org
>> > To unsubscribe send an email to linux-audio-dev-le...@lists.linuxaudio.org 
>> >  
>>
>>
>>
>> --
>> Florian Hülsmann
>> 
>> http://cbix.de  
>
>
>


-- 
Will J Godfrey {apparently now an 'elderly'}
https://willgodfrey.bandcamp.com/
http://yoshimi.github.io
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
___
Linux-audio-dev mailing list -- linux-audio-dev@lists.linuxaudio.org
To unsubscribe send an email to linux-audio-dev-le...@lists.linuxaudio.org

[LAD] Update of zita-jacktools

[Fons Adriaensen]

Hello all,

zita-jacktools 1.7.1 is now available at the usual place:



zita-jacktools is a collection of Jack audio processors
implemented as Python classes. These can be combined to
create complex graphs that are completely controlled by
a Python script and can be interfaced to anything that
Python can handle. Main applications are automated
measurements, sound installations, listening tests, etc.


New is this release:

class JackBw8filt

  Up to 100 8th order Butterworth bandpass filters.
  Each one can be configured separately and be lowpass,
  highpass or both in series. Mainly meant for measurements
  that require this sort of filtering.


classes JackFwplay amd JackFwcapt

  These two classes allow to process an audio file via any
  combination of Jack apps that function correctly in Jack's
  freewheeling mode, and record the result.
  

All examples having a GUI now use PyQt6

-- 
FA

___
Linux-audio-dev mailing list -- linux-audio-dev@lists.linuxaudio.org
To unsubscribe send an email to linux-audio-dev-le...@lists.linuxaudio.org

[LAD] Re: ome Faust questions

[Stéphane Letz]

> 
> Hello all,
> 
> Triggered by a recent post, I've been dabbling a bit with Faust,
> and have some questions :-)
> 
> (For the jack-gtk architecture)
> 
> - When started the app prints some useless info (a nuisance) and
>  it also autoconnects (a definite no-go).
>  Is there any way to avoid this ?
> 
> - The faust2jack script says it adds 'preset' functionality
>  to the GUI. But there seems to be no way to access this.
> 
> - The way the PRESETDIR is handled is weird. It is done by 
>  prepending a single line temporary file to the merged cpp file
>  using cat.
> 
>  What would be wrong with
> 
># add preset management
>if [ $PRESETDIR = "auto" ]; then
>   PRESETDEFS=-DPRESETDIR=\"/var/tmp/\"
>else   
>   PRESETDEFS=-DPRESETDIR=\"$PRESETDIR\"
>fi
> 
> and then adding PRESETDEFS to the compiler options ?

Sure, I guess this specific thing came by coping an already written « way of 
doing », lack of Unix culture, years of history…  

> 
> (General)
> 
> - The various faust2*** programs work by merging the compiled dsp
>  file with the architecture file. While doing this, also all
>  include files from the /usr/include/faust directory are recursively
>  expanded in-line. Now I wonder why this is done. Surely the C++
>  preprocessor would do this anyway ?
> 
> 

This -i option is there to produce a kind of « self-contained » Faust 
architecture (so only inlining the « faust/xxx » files). And possibly not 
really needed in this specific example : history, copy/paste…etc...

Stéphane, 
___
Linux-audio-dev mailing list -- linux-audio-dev@lists.linuxaudio.org
To unsubscribe send an email to linux-audio-dev-le...@lists.linuxaudio.org