hi jakob,
I play a little bit with the transport.
I get always the same problem.
Here a way to reproduce.
1. Start Hydrogen
2. Enable jack transport.
3. Start in songmode with a song.
4. Hit the taptempo key "backslash" some times.
On this point all works right. Tempochange do nothing.
5. Stop sequenzer and hit the "Backspace" key to reposition to start.
6. Start play.
now the position is wrong.
also in state playing taptempo makes jumping in timeline.
I can fix the problem removing this lines of code in jack_output.cpp. see patch
after removing this, you can use tap tempo in state playing without jumping.
and backslash key go to the song beginning.
Not working is use taptompo in the middle of a song in state STOPPED and jack
transportmode is on. Then after tapping and start haydrogen jumps to an other
position.
But in state PLAYING it works well.
--- libs/hydrogen/src/IO/jack_output.cpp (Revision 325)
+++ libs/hydrogen/src/IO/jack_output.cpp (Arbeitskopie)
@@ -213,8 +213,8 @@
void JackOutput::locateInNCycles( unsigned long frame, int cycles_to_wait )
{
- locate_countdown = cycles_to_wait;
- locate_frame = frame;
+// locate_countdown = cycles_to_wait;
+// locate_frame = frame;
}
/// Take beat-bar-tick info from the Jack system, and translate it to a new
internal frame position and ticksize.
@@ -276,11 +276,12 @@
void JackOutput::updateTransportInfo()
{
+/*
if ( locate_countdown == 1 )
locate( locate_frame );
if ( locate_countdown > 0 )
locate_countdown--;
-
+*/
if ( Preferences::getInstance()->m_bJackTransportMode ==
Preferences::USE_JACK_TRANSPORT ) {
m_JackTransportState = jack_transport_query( client,
&m_JackTransportPos );
@@ -345,17 +346,17 @@
WARNINGLOG( "Frame offset mismatch; triggering
resync in 2 cycles" );
must_relocate = 2;
} else {
- if (
Preferences::getInstance()->m_bJackMasterMode ==
Preferences::NO_JACK_TIME_MASTER ) {
+/* if (
Preferences::getInstance()->m_bJackMasterMode ==
Preferences::NO_JACK_TIME_MASTER ) {
// If There's no timebase_master, and
audioEngine_process_checkBPMChanged handled a tempo change during last cycle,
the offset doesn't match, but hopefully it was calculated correctly:
m_transport.m_nFrames =
m_JackTransportPos.frame - bbt_frame_offset;
// In jack 'slave' mode, if there's no
master, the following line is needed to be able to relocate by clicking the
song ruler (wierd corner case, but still...)
if ( m_transport.m_status ==
TransportInfo::ROLLING )
H->triggerRelocateDuringPlay();
- } else {
+ } else {*/
///this is experimantal... but it works
for the moment... fix me fix :-) wolke
// ... will this actually happen?
keeping it for now ( jakob lund )
m_transport.m_nFrames =
H->getHumantimeFrames() - getBufferSize();
- }
+ //}
}
}
@@ -369,9 +370,9 @@
if ( must_relocate == 1 ) {
//WARNINGLOG( "Resyncing!" );
relocateBBT();
- if ( m_transport.m_status == TransportInfo::ROLLING ) {
- H->triggerRelocateDuringPlay();
- }
+// if ( m_transport.m_status == TransportInfo::ROLLING ) {
+// H->triggerRelocateDuringPlay();
+// }
}
if ( must_relocate > 0 ) must_relocate--;
> Hi List
>
> I have been trying to fix bugs in the transport / jack sync part of Hydrogen
> for the last few days.
>
> I thought I was on the right track for a while, but after playing with it for
> a while, it appears that the opposite is true.
>
> Hydrogen appears to be far from robust, when for instance, interacting with
> buggy counterparts like seq24. Playback will jump back and forth when
> adjusting the tempo, sudden segfaults happen etc.
>
> The reason for this is that the basic design of Hydrogen's audio engine is
> too
> primitive to handle things like tempo changes etc, in a sensible manner (see
> my mail from yesterday). The hacks required to build our features on top of
> it, become too complicated and buggy. I honestly doubt that we can meet the
> featureset that we currently plan for, in a reasonably bug-free way, without
> rewriting major parts of the core engine.
>
> Anyway I'm really fed up with i right now, and I can't keep debugging it for
> fear that my head will explode :-)
>
> What's worse, I have no Idea whether my commits to the engine (hydrogen.cpp
> and jack_output.cpp) have made matters better or worse (though I think,
> better
> - I have been trying to test it, but I can't find any real pattern in what
> triggers the bugs, sorry).
>
> Hrmpf.
> - Jakob Lund.
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Hydrogen-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/hydrogen-devel
>
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Hydrogen-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/hydrogen-devel
