mån 2005-01-24 klockan 16.13 skrev webhead:
> Mathias Lundgren wrote:
> > lör 2005-01-22 klockan 03.05 skrev webhead:
> > > Hi all,
> > >
> > > Concerning that MC505 idf:
> > > Apparently, the only problem I have is getting the drums channel to
> > > work. It would seem that the drums are on channel 0 (?). Whenever I
> > > record some events it plays back correctly but editting them only gives
> > > none-drum sounds. I'll probably look some more at the midi data to
> > > figure this out.
> > >
> > >
> > > Other problems are:
> > > 1) recording/playing midi data gives double sound, obviously I need to
> > > switch off the midi thru on muse, so far I'm avoiding this by pulling
> > > cables :(
> > >
> > Hi!
> > Have you tried using the midi input filter? There's a thru filter there
> > to the right. Now I'm not completely sure that it's working 100% as of
> > today, because of MusE's new internal controller handling, and I'm not a
> > midi expert so I'm actually not sure exactly how it is supposed to work.
> > Would be happy if someone could give me some hints. I know that people
> > have successfully turned off midi thru sometime in the past, but things
> > might have changed since then...
> I'm not quite sure how that thing works... Maybe I need to twiddle it
> a bit to figure it out...
Me neither. But it looks like you should be able filter out midi through
for specific messages IIUC. BTW, I hope you saw Werner's suggestion
about recording input for specific midi channels. I never noticed the
fields in trackinfo, but it definitely seems like _the_ solution (if
it's still working, don't have the gear to verify).
> > >
> > Midi clock sync is an issue that there's been some work on lately.
> > Actually, in the cvs for the 0.7-branch it works somewhat to use MusE as
> > slave, but it's not stable (the sync might run fine for several minutes,
> > but in some situations the algorithm unbalances itself. changing
> > position while running is not the best thing to do either ...). The new
> > clock-sync stuff has as of today worked, in it's current state, for at
> > least 3 persons, all of them with different setups. When I get the time
> > (as soon as possible) I'll try to improve the algorithm for the sync,
> > which is quite crude right now, and is probably the main source for the
> > problems.
> For some reason, pressing the jump buttons (to switch to a different
> measure) on my MC505 does indeed move the cursor (well, the red
> line/arrow thingy) around. It's not quite as it should be though,
> instead of moving forward it moves three parts forward and then jumps
I guess it's MMC controlling this, and I haven't begun looking into this
yet (my old keyboard doesn't support it it seems, only midi clock). I'm
planning to look into MMC and MTC after getting midi clock more stable.
> Anyhow, the main problem here is, and I think this IS the main reason
> you want muse to be a slave, is to start recording as soon as I press
> play on an instrument. That way, the midi data appears at the start of
> a track and should be nicely laid out...
Hmmm, I'd say this sounds more like a question for the master midi
device. Since MusE is slaved, it shouldn't start playback itself. On my
old keyboard, there's a button called synchro start, which starts to
send out midi clock messages, so MusE can begin to calculate tempo
values. If I press a key on the keyboard, it sends "start" and
everything begins to roll. If I want to achieve what you're talking
about, I record enable MusE, press synchro start, and playback +
recording begins, with MusE as slave. Is this what you're thinking of?
> > I don't think they're trivial at all, you're right on the weak spots. If
> > you try any of the above, I'd be very happy if you could give us some
> > more reports on your proceedings.
> Well, I've never liked computers for actually composing the music, the
> tunes and paterns all get sequenced on the instruments themselves,
> after all, they're short repetitive parts, no hassle and I prefer a
> 'hands on' method here.
> Computers come in real handy when working on the song structure which
> is much less visual on the instruments themselves. Also, as a song
> consists of large structures, keeping an overview and the ability to
> move big pieces around is just impossible on some of the instruments.
> A great feature would be to be able to tune in volumes and panning,
> visually over the track (think Cool Edit). Often when I record some
> intruments (guitar for instance) it's natural to mess up a part here
> and there, for now I can splice and remove certain parts which works
> but isn't exactly user friendly. I do believe that was being worked
> on, right?
Are you talking about automation for volume and panning of audio tracks?
In the upcoming 0.8 automation for a lot of audio-related things has
been added and is already working (LADSPA effect parameters, volume +
panning for audio tracks, etc). There are a lot of things that aren't
working in 0.8 yet though.
> I've never found a neat sequencer for paterns btw, or something to
> compose and use an arpegiator...
I've been thinking of adding the possibility of having a part-library
where you can store certain parts, name and categorize them, sort them
etc etc.. This would be great for drum parts, since they're usually
similar between projects. Controller sequences would also be great to
save between projects, to just be able to apply them on a certain part.
Then add a nice way to import and export them to/from the library, to
share with others. But time... :-) (I'm suspecting things like these
might already exist, in one way or another, in the .idf files)
> Don't get me wrong here, I'm happy something such as muse even exists,
> at least I'm not being forced into a windows enviroment this way :)
Thanks for sharing your opinion!