Re: stepping midi sequence. one chord per trigger.

Pete wrote:

I've been following this discussion with some interest, but didn't
contribute because any suggestions I have will be of linited generality.
However...

In article <j91Uk.103680$b76.61442@xxxxxxxxxxxxx>, cat <here@xxxxxxxxx> wrote:
Michael R. Kesti wrote:
cat wrote:

I'm totally lost trying to create an effect.
<snip wonderful clevernes>
Please don't hesitate to ask any questions.

I should have said I'm on OS X for logic.
Have you looked at 'Max'? It's expensive, but is very flexible for setting
up that sort of scheme. On Linux there's the free 'PureData' which is
rather similar. Don't know if it exists on OS-X as well.

I've done virtually all my MIDI work in BeOS, which is why I say my
comments will lack generality (:-/), but I have my own software that
can be set up to do the sort of thing you want. (Haven't yet, but
like you this is something that I think would be very useful to have
-- just one of the things I've never got around to!)

I would approach it rather differently than Michael, because my facilities
differ. My software system has small processing modules that are connected
in flow paths graphically (which is why I suggested Max). One such module
is a triggerable MIDI player, another delays events a desired amount, and
a third will convert a note-on into a note-off. Using this set, my thoughts
would be to create a MIDI sequence of just the desired note-ons, so these
would be triggered as required by the drum-stroke events. As well as being
sent to the synth, the note-ons would go through a delay and then to the
on/off converter. This way, each note is sure to be turned off (and notes
can overlap -- not possible with the clock-based method).

No, notes may overlap but the intended effect of their overlap may not be
achieved.

One reason I'm stressing this is that in the discussion so far, the
fact that each MIDI note is an on/off event pair seems to have been
glossed over. In Michael's scheme, you'd have to ensure that no note
in the sequence would exceed 24 clocks (though they could be shorter
-- unlike their fixed length in my approach).

This is true and for this reason it works out better to use sounds that
have no sustain phase when using this mode of operation. Without sustain
the relative timing of NOTE OFF messages is irrelevant. If sustained
sounds are used and notes in the sequence are less than one quarter note
in apparent duration then their actual duration is a function of the
seuqnecer's tempo setting. If notes exceed one quarter note in apparent
duration then their actual duration is a function of both the sequencer
tempo and when triggers occur.

It's a strange mode of operation and includes some often unexpected
consequences.

And of course you'd be
in trouble if your input triggers came in faster than 24 clocks apart!

Yes, but it's not as bad as one might think. My converter has no notion
of tempo and sends those 24 clocks "back-to-back", as it were. The clocks
therefore require about 8 milliseconds to transmit and other overheads
come to about another millisecond or so, so tital porcessing is on the
order of 10 milliseconds per trigger. This means that trigger would have
to occur at an effective tempo of 6000 BPM to cause trouble. Even then,
the trouble would be overrunning the transmit buffer because the receiver
is buffered, too.

I think one reason I haven't played with this yet myself [maybe I will,
now!] is that I don't trust myself to be able to provide exactly the
trigger sequence needed (I'm not a drummer -- keyboard...). Not sure
how to handle error recovery.

Anyhow, good luck in your quest.

I just remembered a NAMM show demo of a system that determined the
tempo of a mic'd drum kit and, in real time, generated MIDI clocks to
sync a sequencer to that tempo. This allowed one to simply start the
sequence at the appropriate beats and it would follow the drummer's
timing. Unfortunately, I don't recall who had developed this or where
one might obtain it, but something such as that might be a far better
solution!

--
========================================================================
Michael Kesti | "And like, one and one don't make
| two, one and one make one."
mrkesti at hotmail dot com | - The Who, Bargain
.

Relevant Pages

  • Re: Software for algorithmic MIDI rhythmic loop composing?
    ... >Are there any MIDI software tools out there that permit some MIDI note ... could a top hat sequence automatically trigger to play at ...
    (comp.music.midi)
  • Re: Sequencing
    ... >>> about all these pay sites that have midi files of all these songs? ... >>> full band and looking to just sequence stuff outside of drums, guitar, ... > you need a click track that can be sent to the drummer - either a MIDI ...
    (rec.music.makers.percussion)
  • RE: Access Append Query to update Sequence Generator in Oracle
    ... There are two Triggers apparently for TASSIGNMENT and there doesn't appear ... to be a Trigger for TASSIGNINTERVAL ... SELECT COUNTINTO assignmentVariable10 ... Also committing should not make any difference to the sequence or trigger. ...
    (microsoft.public.access.queries)
  • RE: Best Way to Auto Increment with Oracle
    ... What if some code in a trigger on the table did a NEXTVAL on the sequence? ... you are threading (say in Java. ... You then execute the insert statement. ...
    (perl.dbi.users)
  • Re: stepping midi sequence. one chord per trigger.
    ... sounds are used and notes in the sequence are less than one quarter note ... in apparent duration then their actual duration is a function of the ... tempo and when triggers occur. ...
    (comp.music.midi)