Re: [Faudiostream-users] Faust dropping short MIDI notes?

2019-01-30 Thread Stéphane Letz
How can I simulate the problem here ? (cannot reproduce it with my virtual MIDI 
keyboard…)

Stéphane 


> Le 30 janv. 2019 à 15:04, Mykle Hansen  a écrit :
> 
> 
>> On Jan 30, 2019, at 2:47 PM, Stéphane Letz  wrote:
>> 
>> This standard MIDI handling code may drop some MIDI events. You can more a 
>> more robust path (using a queue of incoming events) that is used when 
>> dealing with MIDI synchronization messages  (start, stop, clock…)
>> 
>> You can « force »  that MIDI sync path by explicitly adding an appropriate 
>> metata like in the following line: 
>> 
>> on = button("[midi:key 60][midi:clock]on »);
>> 
>> Can you try that and report? If this solve your problem, then we may have to 
>> switch this more robust mode on by default…
> 
> Ah ok … so this is an issue of the time-resolution of UI events.  
> It makes sense that your suggestion would work, but alas,
> I tried it & it didn't seem to change anything. 
> Faust is still dropping more than half the events.
> 
> My modified test program is below — can you double-check my syntax?
> 
> -m-
> 
> ——
> 
> 
> declare author "Mykle Hansen";
> declare name "short Midi note example";
> declare options "[midi:on]";
> 
> import ("stdfaust.lib");   
> 
> //on = button("[midi:key 112]on");
> on = button("[midi:key 112][midi:clock]on"); // sletz suggestion:
> 
> count = +( on > on' ) ~ _ : hbargraph("# of button presses", 0, 10);
> 
> process = os.oscsin(440 + count ) * on;
> 



___
Faudiostream-users mailing list
Faudiostream-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/faudiostream-users


Re: [Faudiostream-users] Faust dropping short MIDI notes?

2019-01-30 Thread Mykle Hansen

> On Jan 30, 2019, at 2:47 PM, Stéphane Letz  wrote:
> 
> This standard MIDI handling code may drop some MIDI events. You can more a 
> more robust path (using a queue of incoming events) that is used when dealing 
> with MIDI synchronization messages  (start, stop, clock…)
> 
> You can « force »  that MIDI sync path by explicitly adding an appropriate 
> metata like in the following line: 
> 
> on = button("[midi:key 60][midi:clock]on »);
> 
> Can you try that and report? If this solve your problem, then we may have to 
> switch this more robust mode on by default…

Ah ok … so this is an issue of the time-resolution of UI events.  
It makes sense that your suggestion would work, but alas,
I tried it & it didn't seem to change anything. 
Faust is still dropping more than half the events.

My modified test program is below — can you double-check my syntax?

-m-

——


declare author "Mykle Hansen";
declare name "short Midi note example";
declare options "[midi:on]";

import ("stdfaust.lib");   

//on = button("[midi:key 112]on");
on = button("[midi:key 112][midi:clock]on"); // sletz suggestion:

count = +( on > on' ) ~ _ : hbargraph("# of button presses", 0, 10);

process = os.oscsin(440 + count ) * on;



___
Faudiostream-users mailing list
Faudiostream-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/faudiostream-users


Re: [Faudiostream-users] Faust dropping short MIDI notes?

2019-01-30 Thread Stéphane Letz
This standard MIDI handling code may drop some MIDI events. You can more a more 
robust path (using a queue of incoming events) that is used when dealing with 
MIDI synchronization messages  (start, stop, clock…)

You can « force »  that MIDI sync path by explicitly adding an appropriate 
metata like in the following line: 

on = button("[midi:key 60][midi:clock]on »);

Can you try that and report? If this solve your problem, then we may have to 
switch this more robust mode on by default...

Stéphane

> Le 30 janv. 2019 à 14:05, Mykle James Hansen  a écrit :
> 
> Hi,
> 
> I’m seeing a problem with Faust’s MIDI support, on OS X 10.11.6,
> in an app built with faust2caqt .  It appears that if a NOTEOFF
> message follows a NOTEON message too closely, Faust may not
> receive either message.  In my case, the timing between NOTEON
> and NOTEOFF is about 6ms, which is very short.  But I have 
> confirmed that the OS and various other applications are all 
> able to receive and process these messages.
> 
> Below is a simple Faust program to demonstrate the problem.  When I
> send the NOTEON/NOTEOFF pairs I describe, Faust sees them
> only intermittently, about 20-30% of the time.
> 
> I haven’t been able to this in any other environments, because I haven’t
> gotten the [midi:key] metadata label to work anywhere else.
> (I’m hoping it’ll be supported in the next release of FaustLive. :)
> 
> Any advice would be much appreciated & I’m happy to give more info as needed.
> 
> Thanks,
> -mykle-
> 
> 
> ——
> 
> declare author "Mykle Hansen";
> declare name "short Midi note example";
> declare options "[midi:on]";
> 
> import ("stdfaust.lib");
> 
> on = button("[midi:key 112]on”); // replace 112 with whatever MIDI note is 
> handy for you ...
> 
> count = +( on > on' ) ~ _ : hbargraph("# of button presses", 0, 10);
> 
> process = os.oscsin(440 + count ) * on;
> 
> 
> 
> 
> ___
> Faudiostream-users mailing list
> Faudiostream-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/faudiostream-users



___
Faudiostream-users mailing list
Faudiostream-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/faudiostream-users