Conference napalm::commusic_v1

    I'll probably get beat up bad for this, but what the hell...
    
    Unless you wanna invest BIG BUCKS in sequencers and computers with
    giga-acres of memory, you're probably gonna be disappointed. Making
    the crowd wait while you load up the next sequence, patch change
    and drum pattern(s) is unprofessional. If you can't do it fast,
    don't do it.
    
    Edd
    

	RE:.1

	Are we really that naive?  We aren't planning to have sequencers run the
show, just add parts and texture.  I've seen bands do it that are worse off than
us, so it can be done on a reasonable budget.  We may not do it on every song,
but would probably like the option to.

	What specifically are the limitations?  How long does it take to dump a
new song into the XYZ sequencer (seconds? minutes?)?  Some people have been
noting about various systems able to sequence only several minutes of music
before exceeding capacity.  Will we find most low-budget sequencers overly
restrictive?

    > "Are we really that naive?"
    
    I doubt it. At least I hope not.... ;^)
    
    Seriously, your "every song" option is gonna run you cash. Dumping
    via tape is slow and unreliable, so that means you're looking at
    discs, floppy ($) or hard ($$$$). What's a Fat-Mac go for??
    Even if you can set everything up in 1 minute, that's an eternity
    for those on the dance floor.
    
    To just use it for "parts" is tricky also. You have to be sure it's
    at the right tempo and start it at *exactly* the right time.
    
    I'm not trying to discourage you, we both know it can be done. MIDI
    costs. I just want you to make sure you what ballpark you're playing
    in.
    
    Edd
    

Re: .0
    
    I can't give U $$$ off the top of my head (don't have time to look 'em
    up right now, either), but I think that the best bet for a sequencer is
    either the new Korg unit or perhaps the MC500 Roland unit.  Former
    holds around 15K notes, latter around 40K notes.  Price is somewhere
    from 450 to 1000.  They both load REAL fast (Korg about 3 seconds when
    I saw it!).
    
    Drum box?  I know the TR707 with expander cartridge would have enough
    patterns/songs/tracks to just about kill a whole set.  Not sure
    what the TR505 or other drum boxes have to offer.
    
    Going this route will save you from having to sink $$$ into a blasted
    computer to do MIDI loads during the show.  Maybe I'll ramble more
    when I get time.  Good luck.
    
8^)

    Just for reference...
    
    One generic "pop" song, with sequenced bass, piano, and horns, (but
    no drums) will suck up 5-10K notes. Velocity info will add another
    33%.
    
    My entry on the Tape ran around 6K with no velocity.
    
    Edd 
    

    I talked to Andy Mendleson, who taught the 24-trk class I went to
    and plays in "Body English" about this.  He said that using sequencers
    and drum machines live is a big pain in the ass.  What happens if
    you can't load a song or a hardware failure or bad cord/connection
    rears it's ugly head?  The show must go on.
    
    Most small time bands can't afford tech roadies and duplicate hard-
    ware, like Tears for Fears or the Cars, for gigs.  What Andy did was
    to record the songs on a small portable 4 track and just set that
    up to run.  Alan has a Tascam 244, doesn't he?  The quality will
    be good and you just leave the 244 onstage.  Plug it in and send
    a line out to the snake.  I think this will be a lot more reliable.
    
    DD
    

    You guys are all way off my base.  My old band was three pieces
    and we added synths to almost all of our stuff, driven off a sequencer
    (a tired old MSQ-100 and MSQ-700).  We weren't a synth band, and
    we only used synths for strings and brass sorts of things where
    the song called for it in addition to the basic guitar/bass/drums
    mix.  We loaded up the sequencers between sets.  We had no problems
    with fitting things into the sequencers.  A typical set would have
    five or six songs with synths, five or six without.  It took about
    2 minutes to load the sequencers from cassete tape, plenty of time
    to fit into a break.  We had almost no reliability problems.  When
    things didn't work, it almost always turned out to be cockpit problem
    (i.e., pilot error, not equipment failure).  When things broke hard,
    we did without them - we faked things on the guitars, and just didn't
    sound as "produced" as with the synths.  The sequencers were no
    less reliable than anything else (including my acoustic drum kit -
    heads and sticks break, you know).
    
    OK, real advice:
    
    	1) get a TR707.  Best sounds for the buck.  Do you really need
    		a drum machine, though, given you've got a live drummer?
    		We never used a drum machine (well, not quite true,
    		I used a TR606 Drumatix as the source of my click track
    		so I could play in sync with the sequencers even when
    		the synths were tacit (that's musical jargon for "silent");
    		I played on every number.  We never tried adding 727-style
    		percussion, but that would have been trivial.
    
    		A 707 with an M64C (bad news - they're about $80 each)
    		will hold 12 songs based on three distinct sets of 64
    		1 bar drum patterns.  More than enough for a set.  Note
    		that you can't reuse the M64C from set to set - and
    		you still need to tape load the 707 itself to get 4
   		songs into it - the M64C only holds 8 itself.  In fact
    		the way you load the M64C is by dumping the 707 to one
    		bank of the M64C, reloading the 707 and dumping that
    		to the M64C's other bank, then reloading the 707 with
    		its "permanent" stuff.
    
    	2) You can pick up an MSQ-700 for a song these days.  The -700
    		is easier to use in performance situations than the
   		-100; you put one song on each track, up to 8.  The
    		-100 is more useful for editing - it can do bar by bar
    		inserts and deletes - the -700 behaves exactly like
    		an 8 track recorder - i.e., no "in place" inserts and
    		deletes, only at the end.  The -100 will store multiple
    		songs, but back to back with a lot of empty bars between
    		them, so you'll need a index of bar numbers for where
    		each song begins, rather than just a list of what song's
    		in what track.  Getting to the next song requires "fast
    		forwarding" the bar number, which can be a pain in
    		live performance situation.  To get around this we used
    		to segue a couple of songs together.
    
    	3) Cheap sequencers will not remember tempo.  This means you have
    		to set the tempo for each number.  Better have a poop
    		sheet.  We used to make this more manageable by grouping
    		songs by tempo, sometimes even just playing a succession
    		of songs that were at almost the same tempo at exactly
   		the same tempo.  Not as bad as it sounds.
    
    	4) If the synths EVER go tacit, somebody in the band who will
    		be playing at the time needs a click track.  We solved
    		this by having me play all the time and feeding me a
    		click.  Songs that didn't start with drums or synths
    		were still a problem - the guitarists couldn't hear
    		my click, and my playing metronome for them was deemed
    		bad stage presence (to which I agree).  The guitarist
    		who ran the sequencers (he wanted to, otherwise I would
    		have done it) simply got good at watching the countdown
    		in the sequencer's display window.  Note that this will
    		require enough light to see an LCD, if that's what the
    		display is.  As a side effect, the guitarist and I got
    		rhythmically VERY tight - we were both playing against
    		click tracks (this does wonders for your time after
   		a few months), but the bass player never really did
    		get locked in.
    
    	5) Very few sequencers can handle tempo changes (accelerando,
    		ritardando); pick your material accordingly.
    
    	6) Make MULTIPLE copies of your data tapes.  Store them separately
    		but bring them ALL to a gig.  Record in stereo, play
   		back (for loads) in mono.  If one track turns out to
   		bad, you've got another one on the same tape.  Not all
    		tape problems run across the full width of the tape.
    		Use short data tapes (a la Radio Shack) if you can get
    		them.  They *do* work better than audio tapes.  Do NOT
    		use noise reduction.  Record at the highest levels
    		practical (about +3 VU).  Order your material on the
    		tape in the order it will be used.  Make a new tape
    		if you have to reorder things.  Preceed each data segment
    		with a verbal/audio description of its contents.  If
    		you reuse a data tape, erase it completely (i.e., make
    		an erase pass over the existing data, without recording)
    		before you record over it.
    
    This is all based on real experience, not theorizing.  I have done
    this and it works.  We used to always blow people away with our
    arrangements.  "You guys have those horns on tape?".
    
    len.
    

	Re:.7 

	A real good checklist.  Thanks, Len.

	I've seen a couple of bands use the tape method as well.  Still, to
generate the tapes it'd be useful to be able to sequence what goes down so we
can easily change patches, parts without going nuts. 

	We aren't a synth band by any means, just want to add the extra parts we
usually do when recording that are a problem doing live.

	Monitoring.  I assume you just fed the keys and stuff through the PA
system and mixed the click track into the monitor for the drummer?  We typically
use the house PA system and may not have the luxury of selectively panning
click tracks.  Do all sequencers generate a click track that can be used
with, say a set of headphones?

    The story with respect to sequencer supplied click tracks is not
    real good.  The MSQ-100 and MSQ-700 supply an audible click, but
    no audio out for it.  You'd have to attach a contact mic to the
    box and feed that to something.  That's why I took my click from
    a TR-606 synced to the sequencers - I programmed a trivial drum
    pattern with a ride beat consistent with the song I was playing
    (i.e., straight or shuffle time) and selected the pattern as each
    song came up - another opportunity for cockpit error (and I did
    screw up now and again - it's interesting trying to play straight
    time with a shuffle click track in your ear - the tempo's ok, but
    you have to resist the shuffle ride REAL hard).
    
    I took my click through headphones.  I would worry about a click
    through a stage monitor being audible to the front rows of the
    audience.  The monitor mix is usually inadible to the audience
    because it's so much like the hall mix that it's effectively masked.
    A click track might stick out like a sore thumb.  Since I sang as
    well, my 'phones also served to mount my mic and feed me my own
    vocal mix.  What I actually did was set up a small mixer with the
    following inputs:
    
    	1) the PA mix
    	2) my own mic
    	3) the TR606
    
    and the output of this mixer drove my phones.  I effectively had
    my own monitor mix.  It was actually better than this, 'cause we
    had the board set up so the voices and keyboards went to different
    amps and drivers, so I had separate feeds for voices and keyboards.
    My Simmons bass drum, and my whole Simmons set when I went with
    all electronic drums, went to the keyboard mix.  I had a volume
    control on my own mic (send to the PA) so I could kill the mic entirely
    when I wasn't singing, set a low level for backup vocals, and a
    high level for lead vocals.  Another opportunity for cockpit error.
    I kept a set list with detailed instructions for each song (somewhat
    encoded of course) so I knew what to do when.  I still screwed up
    occasionally.  I also needed a wire list to get this mess set up
    in a timely fashion.  I ended up with 18 signal wires and 7 AC power
    sources!  But it *did* work.  For me at least.
    
    More modern sequencers I think do supply an audio out for the click.
    Dave Dreher - how about the MC500?  Yamaha QX-nn owners?

    len.

    NO click track except when recording. No audio out. Clicks on
    quarter notes only.
    
    You could use it for a click track though; record all your data
    and move it to track 2. Now record nothing on track 1 while listening
    to the clicks via contact mike.
    
    Edd

    Side note:  Hey Len, did you guys ever get out of sync?  What did
    you do during a gig when this happened?
    
    Todd.
    

    
    	I'm surprised that your band would 'allow' yourselves to be
    Phase-Locked to some click trak or electronic meter. Music has
    rymithical nuances ya know - I've always percieved meter thats
    *too* precise quite dry...
    
    	What would impress Joe Jas? How bout the wizards in Rolandland
    or Korgland coming up with a scheme to lock the sequencer or MIDI
    (et al) tempos to WHAT THE BAND IS DOING rather than vice versa.
    [ a possible place to start is with an opto interrupter on the bass
    drum pedal ]
    
    	Be careful as your attention get focused on these *discrete*
    matters associated with computers - you can easily get lost in the
    "acres of memory" information needed therein - and forget about
    some significant property of the whole. 
    
    	Do you really want to play tonights set at a tempo within a
    gnats ass of whats on the album, or of last nights?
    
    Joe Jas
    
    

    re .12 - yeah, those damn soulless machines are taking over; obviously
    any music played at a constant tempo has no feeling.  There *is* a tempo
    dial on the sequencer and you can set it to anything you want,
    whenever you want.  Our audiences never complained, they in fact
    seemed to be having a pretty good time, but what do they know? 
    Anyway, even playing with a click track my time's not *perfect*,
    and there's a lot more to expression than just tempo.  Don't forget,
    the only things that were LOCKED to the sequencers were the synths,
    and their tracks were real time sequenced without quantization.
    Methinks thou doth protest too much.  We were tight but far from
    mechanical.  And we *did* do ritards at the end of some songs; if
    the synths were not playing I just killed the click track or ignored
    it and the band followed me.

    Regarding the sequencers taking tempo cues from the band - I myself
    don't understand why such a device hasn't been built.  All it has
    to do is generate MIDI clock messages at a rate dictated by some
    kind of human input.  The closest I have seen is on the Polaris,
    whose internal sequencer accepts a foot pedal input that sets tempo
    based on two sequential pedal pushes.  The problem with using the
    bass drum or the ride cymbal is they're not constant; e.g., an omitted
    1/8 note on the bass drum or cymbal might be interpreted as calling
    for halving the tempo.  I'd find having to play a constant ride
    or bass drum pulse far more objectionable than playing at a constant
    tempo.  Putting in enough logic to deal with this sort of variability
    might not be worth the trouble.
        
    re .11 - yes, we did on occasion get out of sync.  I was going to
    add a few things to my comments, this is as good a place as any
    to do so.  I had a kill switch on my click track feed, so I could
    shut it off if I wanted or needed to (we didn't use click tracks
    except when we were using the sequenced synths; no synths, then
    I'm the clock); we also used the RESET button on the sequencers
    when things got too out of whack to recover.  As I said earlier,
    when things got too screwed up we just did without the synths.
    Usually I could hear us drifting off and correct for it before it
    was obvious to anyone else (if you can *hear* the click track, you're
    not playing *with* it - when you're in the groove, it's masked by
    your own playing).  Sometimes the guitarists would miscue; the only
    thing that differs in this situation from any other miscue within
    a band is that the sequencers can't listen and adapt, and somebody has
    to shut them off.
    
    A few other things about data tapes - ALWAYS verify them when they're
    made.  Also, play them back on the same machine they were recorded
    on.  Finally, don't make copies by actually copying the tapes from
    one deck to another.  Make 1st generation copies directly from the
    sequencer.  It doesn't take that much longer.
    
    len.
    
    

    Thank you for the info, Len.  I thought there had to be a kill switch 
    in there somewhere, just in case.
    
    Regarding the constant tempo characteristic, I feel that this is
    a non-problem.  Lack of dynamics will make music seem mechanized, 
    constant tempo won't.
    
    Todd.
    

	Thanks all for the responses.  We're going to persue this with caution
and common sense.  The temptation to add this kind of stuff can be very hard to
resist, but we want to do it in a musical (not gimmicky) way.

	RE: constant tempo

	Maintaining a steady tempo throughout a song is very important. That's
what makes it groove.  Changing tempos within a song (except when called for, of
course) is a sign of not paying attention.  

	Well, we hashed it over last night.  Between us we have two 4-tracks,
two synths (one non-Midi) and a non-Midi drum machine.  For the next month we're
going to experiment using our existing equipment, generating tapes and seeing
how well we can manage playing with click tracks, synth parts and other stuff
and how much musically it *really* adds. 

	Now, how about some suggestions on how we should lay down tracks.  It
seems that the easiest way to do it is lay down the drum machine as the click
track using Len's method, then overdub one to three synth parts.  What'd be
nice is to have the drummer just use the headphone output on the tape deck
for the click, but then...how does he hear the vocal monitors when he sings?
Ok, so we need to figure out how to get the PA mix into the headphones, but
alas, our puny rehearsal PA has no headphone out. 
	
	This should be interesting.  I remain to be convinced that this is
going to give us a much needed bang, but it'll be fun trying things out.
	

    If your going to use the 244, try this.  Record click track on trk-1.
    Record synths on trk-2 and trk-3.  Send the monitor signal from
    the live mix into ch-4.  The drummer will listen to the 244 headphone
    jack or set him up a monitor to listen to the line-out signal of
    all 4 channels.  Then send just the synth tracks to the PA snake
    via an auxillary send from the 244. The audience hears the band
    and just the synth tracks and the drummer hears click track, monitor
    mix, and the synths.  Any variation along this idea should work.
    
    DD