Conference napalm::commusic_v1

    Thanks Karl,  VERY interesting stuff.
    
    I have a few questions I'd appreciate answered at some point when you
    have the time:
    
    >The samples are set up to only mildly respond to velocity - fairly
    > constant volume,	with only a bit of boost based on keyboard
    > velocity.
    
    What is the difference in samples across the velocity domain?  Are
    they actually just "E string - light pluck" and "E string - heavy
    pluck"?  I need to try and synthesize this (I don't have these samples
    for my S-550).

>	I find a folk-oriented finger-picking style quite easy to emulate.
>	Rarely do I depart from straight tonic/3/5/flat7 notes, but I
>	don't ever play 1/3/5 with one hand.  I use widely-spaced intervals,
>	lots of 5ths and 6ths, and often hit non-tonic 3rd or 5th low notes.  
>	A 3rd bass note is good as a passing tone for transposing up a 4th.
>	The circle picking pattern isn't hard, I just start with a chordal
>	note somewhere in the middle of the keyboard, and start the rhythm.
>	Often I find my thumbs overlapping with this style.
    
    Could you give a few examples of chord voicings?
    
    Do I take it correctly that you generally stick to 3 or 4 note
    voicings?

>	Strumming... about 18 months ago there was a series of notes about 
>	truly emulating rhythmic up and down strumming - someone from one
>	of the music comix posted some freeware which would generate strums.
    
    I remember someone posting some trick that involved using two
    keyboards, but I don't remember the specifics.  Or it might have
    been in some article I've read and not in a  notesfile.
    
>	I don't have a pointer.. perhaps there's a keyword, or a helpful noter
>	with a better memory than me will post a pointer.  Well, my strums
>	are just for punctuation, and is/are generally one widely-spaced 
>	(intervallicaly speaking) arpeggiated chord, low notes to high.

    Do you use the freeaware that you mentioned or do you just do it
    manually?  I take it that the arpeggiation is extremely fast?
    
    I've tried to actually play a strum on the keyboard as an arpeggio but
    it just doesn't seem to sound right.  I'm planning to examine the
    strums in the VFX demo with the step editor to see if that reveals
    anything.  They are so convincing, I wonder if they weren't played
    on a guitar-to-MIDI converter (which I suppose IS a good way to
    "emulate" guitar strums huh?).
    
    
>	I found effects to be critical in getting a good sound.  As a sampler
>	or good imitative synth patch obviously lacks the guitar's body
>	resonance, I used a small-room reverb patch with a lot of midrange.
>	But then, in order to give some sustain and space, I also ran it 
>	through a crisp stereogen reverb hall patch. 
    
    Another thing I've learned is that a lot of times you just have to hear
    the sound IN CONTEXT.   I've done stuff that I didn't think sounded
    very guitar-like and then when I put it into context with everything
    else, it sounded pretty good.
    
    It's one thing I learned from listening to sound-patch demos - often
    a sound seems useless when played by itself but in context sounds
    great.  There are classes of sounds that often fall into those
    categories and guitar patches and bass guitar patches are among them.
    
    (Also somethings sound great solo and lousy in context - that's often
    the case with string patches.)
    
    On lead guitar, my input is to use as much modulation/expression as
    you can.  Add vibrato with the mod wheel on almost EVERY sustained
    note - consciously force yourself to use the pitch wheel.  With the
    possible exception of Al DiMeola, no electric guitarist plays two
    phrases in a row without a bend of some kind.
    
    Also you need to force yourself to go beyond the things that guitarists
    do that are easy on keyboards.
    
    For example, it's fairly easy to play a note and then bend up or down
    to another.   However, guitarists start with a bent note and then
    release.  (I've generally heard this referred to as a "ghost bend").
    
    This isn't the same as playing E and then bending down (with the pitch
    wheel) to D.   To my ears, it sounds more real if you move the pitch
    wheel up before you play the note - partly because you you want to get
    a little bit of the upward part of the bend into the note (I.E. you
    want to "slur" into the note) but I also think it's because the the
    tension corresponds closer to what the guitarists tension is like when
    you do it that way.
    
    The point of "rest" of the guitar is "unbent" and on the keyboard it's
    centered wheel.  The bend on the guitar has the tension of the bend,
    and on the keyboard it has the tension of holding the wheel.
    
    Anyway, I've just found that doing it that way gives you more natural
    sounding bends (for whatever the reason).
    

>    What is the difference in samples across the velocity domain?  Are
>    they actually just "E string - light pluck" and "E string - heavy
>    pluck"?  
    
    Yes, using my nylon string EMAX presets, there's no velocity switch,
    but I have done a lot of experimenting in inversely mapping velocity 
    to filter cutoff - a louder note is sharper, a softer note kicks in
    the filter, which softens the attack.  AKAI samplers allow you to 
    choose where to start the attack in the waveform (which is also avail.
    in 'mono' mode on EMAX, but it's binary - play the attack/don't play
    it).  The lead guitar samples I discussed definitely have a second
    layer, but they're not velocity-switched, they're delayed and thus
    time-dependent; if I hold the note then 3 seconds later the harmonic
    kicks in; hold the note and the LFO kicks in.
    
>    Could you give a few examples of chord voicings?
    
    Yes, a D chord would be F#2,A2,D3,A3,D4 or E4 for a 2nd.  I mostly
    just wing this stuff, it's not scientific, but I've played closely
    with guitarists since my 'teens and just have a good feel for the
    voicings.
    
>    Do I take it correctly that you generally stick to 3 or 4 note
>    voicings?
    
    Mostly, but see above - I never strike all notes simultaneously,
    but in a finger-picked style, do staggered arpeggiation.
    
>    (STRUMS) Do you use the freeaware that you mentioned or do you just do it
>    manually?  I take it that the arpeggiation is extremely fast?
    
    I don't have the freeware I mentioned; handplayed arpeggiated bottom-
    to-top 'roll' across the keyboards, using a preselected chord voicing.
    
>    Another thing I've learned is that a lot of times you just have to hear
>    the sound IN CONTEXT.   I've done stuff that I didn't think sounded
>    very guitar-like and then when I put it into context with everything
>    else, it sounded pretty good.
    
    Other than the lead playing in my Mahavishnu cover, the only keytar
    pieces I've done are solo, NAKED, improvs.  But it is sometimes true
    that a so-so sound WILL work fine in an arrangement.  But for me if the
    sound works well as a solo sound, it'll work well (maybe with a bit of
    EQ) in a multitimbral arrangement.. I'd rather start with a good sound.
    
    karl

    Does anyone else have any keytar tricks they'd like to share ?  This
    starts to look like Dave and I talking to each other...
    
    anything coming up on CVIII ?
    
    karl

>Does anyone else have any keytar tricks they'd like to share ?

Commusic is amazing. It adds to my vocabulary (keytar!) and inspires
me to try different things, like, for example, emulating a guitar
with a keyboard. %-)

Last night I fooled around with the apeggi-8 (sic) pipe tool on
the Bars and Pipes sequencer. I got a pretty decent emulation
of folk guitar picking. When I thought about the way you alternate
an "appeggio" on a guitar going low, high, middle, high, middle etc.
(1 5th 3rd 5th 3rd 5th for example) I got some nice sounds. So
I just play the root note and those dang computers play the rest.
The nice touch is when I hit the root note stacatto or legato the rest
of the phrase followed. I just set it up to play every note as an appeggio
this time, but for a real song I could go in and set up each
phrase as I want it. Since Bars & Pipes can do it in real time it is
like having one of those appeggio producers like on a Roland D5.

For really fast (but impossible) picking I added a trill tool too!
21 notes for each one played. Yahoo! ;-)

Dave

	An accompanying guitar should be a so-so voice.   Otherwise it will
	distract from the solo.  A solo guitar should
	use delay vibrato, a little portamento at big leaps up, buzz on loud
	notes (tiny distortion triggered on a high level would be fine), 
	and have fret noise.  Squeaks will do, very low level.
	
	When I have actually written for guitar, or orchestra or its 
	instruments, my goal was to push the envelope, and develope new
	sounds.  but if you do on midi imitations, it won't sound like the
	instruments.  If you do it with the actual instrument, it can be
	the most effective part of the piece.  Imitating standard styles
	is of no interest.

	Study Bach's cello arpeggii for guitar voicing. Also The Berlioz/
	STrauss book on orchestration has a guitar section.  Berlioz
	played guitar somewhat.;

	Accompanying voices should not be so rich they come forward.
	Solo voices should have all the enhanvcements, e.g. vibrato,
	dynamic change, accents, etc.
Tom

    ...good note this ["Keytar" - I like it]...
    
    A couple of observations:
    
    1)	The pitch bend NOTE (.-?) from DB is interesting. I reckon Dave is
    right; not so much for the resultant sound, but for the inevitable MIDI
    Controller message which reverts the pitch back to 'normal'. I have
    previously used the 'slide UP' technique, and had the problem of
    positioning the 'pitch-normal' message AFTER the bent note has
    terminated, before the next note sounds [I'm using Creator]. 
    I think Dave's idea makes this easier.
    
    2)	In my experience, good guitarists rarely hit the full six strings.
    It make look like that, but they mainly hit the top 3 or 4.
    Also, you rarely see good guitarists sweeping up & down the fretboard
    inbetween chord changes. They find an inversion of the 'next' chord
    near to the 'current' chord. This makes a difference to the note
    spacings of the Keytar chord.
    
    3)	As Karl mentions, adding a ninth or sixth (or other tension)
    definitely adds colour - even if nothing else in the multi-timbrality
    of the recording uses that ninth/sixth [but beware if the tension is
    only a semi-tone from the melody].
    Guitarists do this all the time, and are often unaware of it, IMHO.
    
    Jim Burke, in response to the earlier cry for more participation.

    ... and here I am wanting a MIDI controller for my guitar so I can make
    it sound like a keyboard! I guess I had the silly thing in reverse!
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    	Seriously, as a guitar player, I'm enjoying this note as well. That
    D chord voicing is basically an open position chord using all (or most)
    of the strings. One of the keys (no pun intended) to the voicings you
    would use is that the strings on the guitar are tuned in intervals of
    fourths, with one third thrown in for good measure (damn these
    unintentional puns!).
    
    	To save you buying a book, from low to high pitches, here are the
    fundamental chord voicings they teach us at the beginning:
    
    I'm going to use the convention "En" for indicating pitches, since I
    don't know where they are relative to 'real' music theory.
    
    Shape	Voicing (in open position, i.e. the first five frets)
    -----	-------
    
    Open	En	An	Dn+1	Gn+1	Bn+1	En+2	
    
    (this is how the open strings are tuned; I believe the low E is E2 (?))
    
    
    E chord	En	Bn	En+1	G#n+1	Bn+1	En+2
    
    A chord	En	An	En+1	An+1	C#n+2	En+2
    
    C chord	En/Gn	Cn+1	En+1	Gn+1	Cn+2	En+2
    
    	Notice the number of notes in the voicings that correspond to the
    open strings from the open string tuning. By keeping these shapes and
    using the open strings as a reference, you can move these shapes around
    on the 'fretboard' to play in different key signatures. The guitarist
    accomplishes this by using a capo (that mechanical device that clamps
    onto the fretboard) or by 'Barring' with his first finger (the one
    directly beside the thumb) and using the other fingers to play the
    other notes. By deciding which notes of the shapes we play, we can also
    play all those neat inversions that you keyboard types are so involved
    with. It goes wihout saying that you can also derive other chords such
    as minors, diminished, augmented, sevenths, ninths, etc., by adding the
    proper notes to the existing chord and subtracting duplicate or
    unnecessary notes (notice all the tonics in the 'open' position
    voicings of E & C ). For authentic guitar voicings, however, remember
    that usually we guitarists play voicings that do not span more than
    five frets (one fret = one semitone), unless we have really big hands
    or are masochists/geniuses/virtuosos.
    
    Legitimate guitar C9 chord -> Gn	Cn+1	En+1	Bbn+1	Dn+2	Gn+2 
    
    (don't jump on me, now; we guitar players call these ninth chords, even
    though it has the flatted seventh).
    
    Unlikely guitar C9 chord -> Gn	Cn+1	En+1	Gn+1	Bbn+1	Dn+2
    
    (Gn+1 is played as an open string, assuming that we are using my pitch
    conventions from above; the reason that this is unlikely as a 'guitar'
    voicing is that Bbn+1 would have to be played on the same string as
    Gn+1, which is not possible; moving the voicing to eliminate the open
    string won't help either, because Bbn+1 and Dn+2 will run interference
    with each other; this all makes perfect sense when you realize that the
    unlikely guitar voicing is a credible piano voicing).
    
    Any questions?
    
    Cal Connolly
    
    HPO, Burlington, Canada