8 Jun
2011
8 Jun
'11
1:01 a.m.
On Tue, 2011-06-07 at 23:25 +0200, Ralf Mardorf
wrote:
> On Tue, 2011-06-07 at 22:50 +0200, Robin Gareus wrote:
>> BTW. IIRC you should not use the -X... options for JACK-midi but rather
>> use a2j instead. AFAIK there were/are [timing?] issues with the built-in
>> JACK midi backends. (see Jeremy's workshop slides on how to launch a2j
>> using qjackctl's "Execute Script" option.)
Well, actually this comment of mine is somewhat outdated. Sorry.
jackd2's -Xalsarawmidi is indeed a very good. I'm not up-to-date with
current jack1 and tschak implementations of JACK-MIDI backends. There
were issues in the past. Those might be resolved now.
Nevertheless, a2jmidid is a good solution for all versions of JACK.
>> HTH,
>> robin
>
> Yesno! Perhaps it's better to avoid -Xseq, but I'm using JACK2 from svn
> with -Xalsarawmidi + a2jmidi_bridge to get a jitter free hw MIDI output
> of Qtractor's ALSA MIDI. I don't believe latency tests,
There's no need to _believe_ proper measurements. They're accurate
within systematic and random error components.
> but I believe
> what I hear and what I see for the waveforms. No jitter! Before Jack2
> from svn I experienced 1 - 3 ms jitter for hw MIDI (ALSA), depending to
> the used MIDI interface, I can't make music even with 1ms jitter, I'm
> able to hear this, see the following from jack devel list:
> [..]
> On Sat, 2011-06-04 at 03:54 +0200, Ralf Mardorf wrote:
> Similar good results as before :)
Now that's good news.
Cheers!
robin