On April 21, 2015 10:21:07 PM Dennis Schulmeister wrote: > Hi Tim, > > On Tue, 21 Apr 2015 01:02:30 -0400 > > "Tim E. Real" <[email protected]> wrote: > > > Looking around, the issue is in Audio::start_rolling(). There's an > > > if-statement for the PRECOUNT state. At that point some values are > > > calculated but nothing else. Therefor the state newer changes to PLAY. > > > Seems like something's has been lost there at some point in time. > > > Commenting the code out (just the "if", not the "else") makes recording > > > start again. Albeit with no preroll, of course. > > > > Robert I think all you did at one point was disable the GUI? > > Because it either wasn't working or interfered with something. > > I could swear long long ago it used to work. Do you remember? > > The problem for me is not the missing precount. The real bummer is that > no matter the settings in the preferences dialog you can't record > anything since MusE won't start rolling. So simply commenting out the > mentioned if { ... } would be perfectly fine for me. > > > Instead of precount, another option is to reserve a 'setup' bar (or bars). > > You learn to do that rather quickly, I guess. :-) > > > But we could allow users to adjust the first bar number from 1 to say, > > > > 0, -1, -2 and so on to help users mark that precious 'bar one'. > > > > A simple trick of numeric display on the timeline - like a sliding rule? > > That would be a real nice feature if the sliding rule would apply to > all displays, like say also the event time in the list editor. The > point is you already got guiding lines for every fourth measure and as > a musician you do think in four, eight or twelve bars so often and then > all your visual clues are one measure off ... > > BTW, thinking of pre-count. For me it is not so much the beginning of a > song where it comes handy. It is on recording and overdubbing.
Ah yes, a very good point, the precount would be needed here for sure. Guess it's time to try and fix it, eh? > I do > use the hardware sequencer inside one of my keyboards quite often and > having the metronome run for a measure or two before recording starts > is a huge helper. I think it is fair to assume MusE being the time > master if precount is to be used. Not sure however how it can be done > in a sensible way using the Jack API. Might be worth a question on LAD > to get Paul and the over dev's opinion. The slow-sync feature for sure. It allows a client to delay the start of the transport. We already use it for other things. Possibly IIRC that is what our precount feature did use, I need to check again. Just need to reexamine the situation to find the (apparent) trouble and a fix. > > > Coda? DC: Al Coda? > > Slightly OT: Might be interested to know I have an idea to use MusE > > Markers > > > > as 'Jump' commands (and therefore loops too) > > I think some arranger workstations allow to do a similar thing. I can > see where something like that could be useful. > > > Yes the sustain is treated specially in MusE, it needs to be muted > > > > when the transport is stopped, its state remembered and then > > recalled when the transport rolls again, although the sustained > > notes themselves are not reactivated mid-note. > > > > Without this, sustain might remain stuck on, in stop mode. > > But it still allows you to play your keyboard, with sustain, in stop mode. > > Makes perfectly sense. On a side note, there are actually four > switch-type cc events. Are they treated the same or is it just sustain? Mm, no I think only sustain is treated specially, currently no others. If you can think of any others which might cause problems if they were stuck 'on' in transport stop mode, let me know. Casting a wider net, one could argue a few other CCs might qualify for this. And what about CCs that MusE doesn't know about, for example ones that are part of some synthesizer plugin. I imagine some of them might cause problems. Maybe a user programmable 'reset/restore on transport stop/start' feature for any desired CC ? Meh... maybe too much. Sustain is extra-extra-special, so good enough? T. > Will do more testing and report back. > > Dennis ------------------------------------------------------------------------------ BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT Develop your own process in accordance with the BPMN 2 standard Learn Process modeling best practices with Bonita BPM through live exercises http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_ source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF _______________________________________________ Lmuse-developer mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/lmuse-developer
