Conference napalm::commusic_v1

    Oops, forgot a delay modulation frequency control.
    
    len.
    

    
    hmm. I'd much more like a box that had fewer controls, that just sensed
    whether or not what was coming in was FSK or not, and if it was
    FSK, decoded it into MIDI clock/MIDI timecode, and if the signal
    wasn't, passed it through an audio output. Maybe needs a 'record
    FSK to tape' switch (or MIDI sysex command), a 'FSK sense' indicator,
    not much else. Couple of jacks, MIDI IN, MIDI OUT, (would it need
    MIDI THRU?) TO DECK, and FROM DECK...
    

    Interesting thought, though. 
    
    /pjh

    Well, I use a Yword YMC-10 FSK-MIDI box permanently wired into/out
    of tape channel 8.  As the YMC-10 has sufficient gain to record
    well on tape, it doesn't require being run thru a board.  Also I
    don't understand your need to delay the sync track.  If I need to
    'delay' a MIDI track, I just enter the required offset in Performer.
    
    When I said 'permanently wired' I was lying.. it's wired thru the
    patchbay, so that with only moving one patchcord I can record on
    and play back from tape ch. 8 normally.
    
    karl

    
    
    Why would you want to delay the sync signal?  Isn't that just
    going to throw your sequencer out of sync from everything else
    on tape?
    
    Ralph
    
    

    Hmm, I see people haven't discovered delayed sync as a clever trick.
    
    First off, if you delay every track by the same amount, they remain
    in sync.  If you are playing along live to some sequenced tracks
    (something I don't do a whole lot) then you must delay the click;
    no problem if you drive your click from the sequencer, as I usually
    do.
    
    Second, if you delay every track by, say, half the maximum delay,
    then you can make some tracks a little early (by setting the delay
    to a little less than half) and some tracks a little late (by setting
    the delay to a little more than half) relative to some fictitious
    time reference.
    
    Third, yes, you can do this with MIDI time offsets or with a "MIDI
    delay" (i.e., not a MIDI-controlled delay, but a unit that delays
    MIDI messages).  But the delay has to be channel specific, and may
    represent some nontrivial edting unless you have a sequencer that
    supports this capability (specifically, my MC-500 does not, but
    the MC-500 Mark II with the "Turbo" MRC-500 software will).  Also,
    it is a nontrivial proposition to randomly modulate a MIDI time offset
    so as to provide a quasi-"human" feel.
       
    .1's suggestion for automatic mode detection and direct conversion
    to MIDI clock is interesting, but since my MC-500 already provides
    the conversion function why duplicate it?  The automatic mode detection
    might be both useful and practical, but I have to think a little
    bit about whether or not it would always do what I really want.
    Manual control is at least guaranteed to do the latter, assuming
    I don't forget what I'm doing.
    
    len.
    

How come you say "If you are playing along live to some sequenced
tracks then you must delay the click."??

    Because if you don't you won't be playing along.  If the tracks on
    tape have been delayed relative to the FSK sync, and you're taking
    your click from that sync, then the click had better be delayed
    appropriately.  Of course, you could just listen to the tracks already
    on tape and play to the click implied by them, which probably makes
    more sense anyway.
    
    len.
     

    re the last few.. len, if you monitor the sync (FSK in your case?)
    track off the simul-sync head, or whatever the 38 uses, just like
    you monitor the already-recorded audio tracks, then your newly-recorded
    track, whether audio OR in the MIDI sequencer, will be in the pocket.
    
    karl

    Sorry, but I just can't comprehend why this would be useful.  Could you
    perhaps elucidate? 

-brad the ignorant

    re .8 - bang, bang, I shot you down.  Yes, I "monitor" the sync
    off the simulsync head.  That's *not* the point of delaying the sync.
    
    I suppose I have to reiterate the example I gave a long time ago
    in the "Delayed Sync" note I posted in this conference.
    
    I recorded a piece with 21 drum parts (3 per 7 tracks, with one
    track of sync).  First I laid down the sync track at rock steady
    tempo.  Then I recorded each audio track, with the sync delayed
    by some nominal amount and slightly (*very* slightly; perhaps some
    of the confusion is coming from the amount of modulation applied)
    modulated the delay, by a different amount and frequency, for each
    track.
    
    Net effect: each track was more or less in sync, but not perfectly.
    Of course, perfect sync would have resulted in perfectly stacked
    sounds, with almost no variability of interest, and it would not
    (indeed did not, I tried it) sound like 21 distinct drummers. It
    sounded like one drummer playing a wierd drum that sounded like
    21 drums stacked up.  But with the slightly varying delays (around
    a shared "center point"), it sounded like 21 drummers playing
    together.  The effect was a dramatic increase in the apparent "size"
    of the sounds (especially due to stereo localization effects of the
    *relative* delays between voices).
    
    A delayed sync is also useful for "advancing" in time tracks with
    slow attacks (e.g., strings) that otherwise sound "late" relative
    to other voices with fast attacks.  The alternative is to adjust
    the MIDI times of the note ons, making the track difficult to edit
    and "read" (all the times are offset by some constant), as well. 
                              
    len.
    
    

    
    
    hmm....sounds interesting.  Do you ever run into any problems
    getting things to sync to tape after the sync signal has been
    run through a signal processor?  Seems the only way I can
    consistently sync to tape is to run my sync track directly out
    of the 38 into the sequencer, ie. no eq, processing, or other
    added noise.
    
    ralph
    
    

    Well, I think I understand that application in .10 although I still
    don't understand why you once said "you better delay your sync...".
    It sounded like I might be doing something wrong.
    
    Anyway, I don't seem to have any of these problems.  I think I do
    pretty much what Karl does.  This is my recording process.
    
    1) Record a sync track (using FSK from the ESQ-1).
    
    2) Rewind and record a click track.  The click track is driven
       from the sync track.  Now the click track is rarely just a
       metronome.  It's usually just a rough mono mix of all the sequenced
       stuff.  It's the click track I use for adding "live tracks".
    
    Rationale:
    
    Putting the sequenced stuff onto the click track has several
    advantages:g
    
    1) My unit is FSK which doesn't have song position pointers.  Having
       a real track for the click means I don't have
       to rewind the tape to the beginning for each live track.  This
       makes it MUCH easier to do punches.
    
       In fact, I would do things this way even if I had SMPTE.
    
    2) Having the MUSIC (as opposed to just a metronome) on the click
       track allows me to find out where I am in the song, as well as
       the ability to play along to the rest of the "band".  
    
    3) I don't need to use the MIDI stuff to add live tracks
    
    Thus the only other time I use the SYNC is when I do the final mixdown.
    
    I have found it vital to record the sync track and the click track
    in separate passes.  This is because if I do it on the same pass,
    I find that occasionally the sequenced stuff driven from the sync
    track is "behind" the sequenced stuff on the click track.
    
    Perhaps THIS is the delay problem that Len is talking about.  However
    the solution for this would be the opposite of delay (whatever you
    might call reading the tape a few micro seconds ahead).
    
    Anyway, by doing things the way I do it, I have circumvented the
    problem (apparently) and I derive other benefits from it as well.
    
    I thought this is "how everybody else did it".  I'll never think that
    kind of thought again.
    
    	db

RE: len

    Now I understand.  That's kind of a slick idea (especially for live
    drums).  Initially I was trying to figure out why you would fool with
    the sync track when you could use a sequencer to do the same thing
    using clock offsets (e.g., copy trk 1 -> trk 2, sel trk 2, mung clock
    offset by -2 to +2 ticks, repeat). 

    MTP can do this *real* easily - but not for LIVE drums.  8-) 

-brad

    Re: the delayed/modulated sync "trick"
    
    For those interested in reading the original note about this, it's
    #250.0 in this conference.
    
    I've saved that note and have used this effect.  It's really worth
    trying to see how it sounds.  Something I never would have thought
    of had I not been reading this conference.
    
    - John -

    After having thought about it a bit, I also like db's trick of putting
    a "rough mix" track down to play along with.  I have *never* thought of
    that, but you can bet I'll start doing it. 

    And like John said, it's amazing the things that one can learn in this
    conference. 

-b

    I been away vacating and then I had a coupla all day meetings, but
    I'm back.
    
    I too generally lay down a "rough mix" for a "click" track, it's
    a whole lot more interesting than two tones (one for the downbeat,
    another for the rest) and it's a lot easier to tell where you are
    verse and chorus wise.
    
    No, the delay I'm talking about is not the delay associated with
    interpreting the sync signal inside a sequencer.  Dave's right,
    you should lay down the sync track first before recording *any*
    audio.
    
    My remarks about delaying the click were based on the assumption that
    the click came from a sequencer driven by the sync, and that it
    had *not* been recorded as audio on a tape track.  Obviously, once
    the click's on tape it need not (indeed, *should* not) be delayed.
    But putting it on tape, you must delay the sync driving the click
    just as you'd delay the sync driving any other audio signal, otherwise
    it will have a different relative time position from all the other
    audio tracks.  One virtue of not putting the click on tape (if you're
    willing to put up with the two tone click instead of a rough mix
    click) is that you can delay the sync to the click to control the
    live track's relative time position on tape, assuming the players
    can lock to a click.  It also saves a track, if that's an issue
    for you.
    
    len.