Conference napalm::commusic_v1

    VOICE - A measure of polyphony. A synth that will play only 1 note
            at a time (incapable of playing chords) is a 1 voice synth.
            A voice is the audible result of a 'note on'.
    
            Voices can be layered, that is, 1 'note on' can activate
            more than 1 voice.
    
            Voices can be activated by either or both of 2 methods.
            Method one is the direct result of depressing a key. Method
            two is the result of receiving a 'note on' over the MIDI
            buss. No matter where the catalyst originates from, the
            result will decrease the available voices left.
    
    
    OK - it's a first pass. Debate it. Accept it. Offer your own opinions.
    
    Edd

    
    Voice - A thing dey gots alot 'a in Miami.
    
    	    Well, the confusion is with voicing: pick a synth.
    		Its got how many preset voices? letsee Tuba, brass, lead...
    		What do we call those? 
    
    	    I like "instrument" better. It describes potentially more
    	    than the "voice" and/or preset sound. Look, I have a 6
    	    instrument synth. Sometimes >1 same instruments play at once.
    	    (layer). Sometimes I have DIFFERENT instuments playing at
    	    once (multi-timbral).....usually I have 6 same instruments
    	    playing different notes (chords...)
    
    Whatdya think? Less confusion.
    	    
    
    	    

	Good topic, Edd.  Much needed.

    PATCH - A combination of settings of the sound-producing parameters
		within a synthesizer.  Some synthesizers offer the ability
		to 'program' PATCHES, others have preset PATCHES.  A PATCH
		may sound like an imitation of an acoustic instrument (i.e.
		violin, horn), another electronic instrument (Rhodes, 
		Hammond B-3) or may be entirely unique and non-imitative
		(that Oberheim sound, that DX7 sound).

		I believe the term PATCH comes from the early days of
		electronic synthesis when various modules (i.e. voltage-
		controlled oscillators, voltage-controlled filters, 
		voltage-controlled amplifiers, low-frequency oscillators...)
		had to run 'patch-cords' to and from one another in both 
		signal and control arrangements in order to produce sounds.
		This has since been superceded first by hard-wired circuits
		(sometimes over-rideable) with knobs and sliders.  The more
		cost-efficient alternative of late has been digital access
		(i.e. call up parameters and change values within that 
		parameter via single dial, slider, or increment buttons).

	There are various topics within this conference created for the
		specific purpose of sharing original PATCHES for different
		synthesizers (i.e. CZ series, DX 4-op series, ESQ-1...).

	As with others, please feel free to correct this definition if it
	is found unsatisfactory or misleading.

    Re:  .2
    
    I think the term "instrument" would tend to have people looking
    for the sound of a tuba, zwobar, or EGuitII. That is, the more
    traditional definition of 'instrument'. If I was to say "I have
    a 6 instrument synth..." (in reference to polyphony), I think many
    would easily get confused and wonder "Really? Which 6?"
    
    Re: .3   Sounds good, but I think the hi-end Rolands use the term
    patch as referring to part of the total composition of a sound.
    (partial->patch->tone???). 
    
    Not that we have to use Roland's definitions...
    
    Can we get some input on what we need to attempt to define????
    
    Edd

    re .4 - partial -> patch -> tone is partly backwards.
    
    Roland defines a patch as one or more programmed sounds plus
    "performance" parameters saved as a single recallable entity.
    Patch is always the top of the hierarchy.
    
    The hierarchy for the Roland synths that I am familiar with is as
    follows:
    
    Juno 106			single oscillator -> patch
    
    Super Jupiter (MKS-80),	two oscillators -> tone
    JX-8, JX-10			one or two tones -> patch
    
    D-50			one oscillator or sample -> partial
    				(one or ?) two partials -> tone
    				one or two tones -> patch
        
    When you use two tones to make a patch, you give up half of your
    polyphony.
    
    While it's nice to define our own terminology consistently, the
    fact remains that the industry seems to use "voice" interchangeably
    to mean both timbre and degree of polyphony.
    
    len.
    

    
    
    
    		FORGET IT GUYS. THIS IS BLACK HOLE....
    
    
    politely_refraining_but_arguing_inside_ron

I'd like to revive this glossary note for those of us who are struggling
to master the terminology.  I'll start with one I *think* I know:

MIDI = Musical Instrument Digital Interface.

I think I understand IN (somebody else triggers the patches on this
instrument), OUT (this instrument when operated triggers the patches on
some other instrument--where the cable is connected to the "IN" port). 
Somebody care to articulate "THRU"?  And how about a few of the basic MIDI 
terms.

Other terms that need definitions:

Sequencer
loop
registration function (simple definition: the ability to associate a
    voices or layered sets of voices to buttons.)
layering (simple definition: the ability to couple two or more voices
          together to the same key/note).
polyphony (simple definition: is how many tone circuits (voices) the 
    instrument will activate at once.  However, the definition gets 
    more complicated when you consider MIDI channels, layering, etc.)
pitch bending
sampling (simple definition: using a mike or pickup to "listen" to a sound 
    and storing and representing that sound as a voice within the instrument 
    (e.g., digitally), including the ability to generate a range of
    notes and associate those with the appropriate keys,)
FM (Guess: frequency modulation: an approach to representing and generating
   voices on an instrument.  Used by Yamaha,...?)
SA (an approach to representing and generating voices on an instrument. 
    Used by Casio, ...?)
PCM (simple definition: pulse code modulation--a digital representation of 
     a voice)
...more...

In addition, there are common terms/concepts that probably have other names 
in the world of this notes file:

     Vibrato
     How long the note remains on even after the key is released (even
        with no sustain pedal)
     How long it takes for the note to sound after the key is pushed.
     
     
 

> Somebody care to articulate "THRU"? 
  
  A thru port passes data appearing at the IN port directly back to
    the buss without adding anything to it. Usefull for daisy-chaining
    multiple SGUs...
      

> Sequencer
    
    A MIDI data recorder. It records events such as note-on, note-off,
    velocity, aftertouch, channel data, most anything your synth can
    generate. You can then edit the data. Sort of analogous to a
    tape-recorder...

    > loop
    
    A portion of a SAMPLE that is played over and over again to make
    the sample seem longer.
    
    > Vibrato
    
    A cyclic modulation of pitch
    
    > How long the note remains on even after the key is released (even
    >    with no sustain pedal)
    
    This is the RELEASE portion of the amplitude envelope.
    
    > How long it takes for the note to sound after the key is pushed.
     
     This is ATTACK portion of the amplitude envelope.
 
                                                 
    Edd

RE: .8

Thanks so far.  You just generated some more questions:

>A thru port passes data appearing at the IN port directly back to
>  the buss without adding anything to it. Usefull for daisy-chaining
>  multiple SGUs... 

It is an IN port that also directly passes the IN information to the OUT
port, right?

>amplitude envelope. 

Can you define the common parameters (in addition to attack and release)
commonly applied to amplitude envelopes and what these parameters mean? 
Probably a definition of amplitude envelope itself would be useful here.
(Simple definition: the amplitude envelope is the specification of the
entire duration of a patch note from the time the key is pressed until 
the last noise is heard. Right?)

>aftertouch

>MIDI channel.  

Alex

No, a THRU port is an output.  The only input is the IN port.  What ever
comes in the IN also goes out the THRU.  Anything coming out of the OUT
port originated inside the piece of equipment under duscussion.

The amplitude envelope is the specification of the *amplitude* throughout
the entire duration of a note from the time the key is pressed until 
the last noise is heard.  Usually such envelopes are divided up into sections,
and each section has its own specification.  The specifications are in terms
of a goal level, or how fast the level should change, or some combination.
The simplist envelopes have 4 parts, for the attack, a drop after the attack,
the sustain, and the release.  The release section is used after the key
is released, the sustain region is used while the key is being held down.

Aftertouch is sensitivity to pressure on keys after they have been pressed
all the way down.  Aftertouch in a keyboard refers to the ability to detect
such pressure and send MIDI messages about it.  Aftertouch in an SGU refers
to the ability to act on such MIDI messages.

The MIDI protocol is like a telephone party line, carrying up to 16
conversations at once.  Most messages are tagged as to which conversation
they are part of.  These are the 16 MIDI 'channels'.

RE < Note 923.11 by WEFXEM::COTE "It looks like Fruit Loops out there!" >


>    ... The sustain
>    level is help until the key is released where you enter the decay
>    phase. This is how fast the note dies away.

	Just a quick nit - and I'm sure it was just an oversight.  Instead of
calling this post-sustain period the 'decay phase', it is more commonly called
the 'release' phase.

	Dan
    

RE: Sequencer

    I've always had a problem with the "tape recorder" analogy.  A
    sequencer may *operate* like a tape recorder, but internally is NOTHING
    like a tape recorder. 

    MIDI is nothing more than an ASCII protocol.  For example, there are
    128 different notes available via MIDI (0-127) that may be played at
    128 different levels (0-127).  So a C3 played as hard as possible
    generates: 

    	[MIDI NOTE_ON] [VELOCITY]
	    64            127			(note on value is a WAG)

    Sustain pedal on/off, aftertouch (0-127), pitch bend (0-127) and
    modulation are all transmitted out the MIDI port as ASCII values. And
    these values are recorded as such by a sequencer. 

    Don't make the mistake of believing that a sequencer actually records
    waveform information (or the actual "sound" that's being played from a
    synthesizer).  A sequencer is really nothing more than a computer that
    runs one program - a data recorder that somehow time stamps incoming
    data and stores it in memory. 

    To extend this a bit, this is how quantization is so easy on a
    sequencer.  Time stamp values are adjusted by a certain amount in order
    to adjust a MIDI event to the closest clock value corresponding to the
    quantization value.  For example, assume that a sequencer clock cycles
    240 times per second, and a passage is quantized to 16th notes.  This
    will move a note to the nearest 15th clock tick (16/240 = 1/15), as
    follows: 

	|<--------------------- One Measure ---------------------->|
	|                                                          |
	0....^....1....^....2....^....3....^....4....^....5....^....6
OLD         x        x         x        x            x            x
NEW     x              x              x�             x              x

         � both notes quantized to same clock tick - smart
         sequencers will note this and delete one note on. 

    Do you see how this type of quantization can screw up a triplet, and
    why the quantization value must be divisible by 3 in order to maintain
    proper timing on a triplet? 

    Anyway, all I was trying to point out is that, functionally, there is a
    lot less similarity between a sequencer and a tape deck than one is led
    to believe. 

-b

    re: .12
    
    Yep, typo. Sorry for the confusion.
    
    Edd

MIDI has *nothing* in common with the ASCII coding.  ASCII codes graphic
characters into numbers from 0 to 255.  MIDI codes performance events into
messages of varying lengths (from one byte for some messages up to hundreds
or thousands of bytes for some SYSEX dumps).

> Thanks so far.  You just generated some more questions:
  
    Drat!   ;^)

>It is an IN port that also directly passes the IN information to the OUT
>port, right?
 
    Nope. You've got it almost precisely backwards. (There is an
    exception.) Info appearing at the IN port will cause the SGU
    to react *IF* the SGU is set to the proper MIDI channel (or
    OMNI (receive all)). Whether it is 'used' or not, it appears
    at the THRU port and goes back on the buss. Only data that
    originates at the SGU appears at the OUT port.
    
    The exception is when, thru software, the IN and OUT info is
    merged inside the unit. Then, data originally headed for the
    THRU will appear at the OUT.
    

>Can you define the common parameters (in addition to attack and release)
>commonly applied to amplitude envelopes and what these parameters mean? 
>Probably a definition of amplitude envelope itself would be useful here.
>(Simple definition: the amplitude envelope is the specification of the
>entire duration of a patch note from the time the key is pressed until 
>the last noise is heard. Right?)
 
    The 'classic' amplitude envelope is defined in terms of ADSR.
    Attack, decay, sustain and release.
    
    Three of these terms are rates, the length of time it takes for
    the amplitude to change from one level to another. Attack, decay
    and release are rates. Sustain is a level...
    
                   ^
                  / \_________
                 /             \
                /               \
                A  DS         R
    
    At the beginning (attack) of an envelope, the amplitude rises from
    a point (probably 0) at a rate. Steep attack rates reach their peak
    quickly while slow rates take longer. Assuming the key is held down,
    the amplitude will reach its' highest level and then start its'
    decay. Decay is the rate with which the amplitude changes from its'
    peak to the sustain level. (Confusion sets in; the sustain level
    can be higher than the peak! Go to advanced class...) The sustain
    level is held until the key is released where you enter the release
    phase. Release is how fast the note dies away.
    
    You should note that it is not necessary to complete the entire
    ADSR cycle. Indeed, the only part of the cycle that is gauranteed
    to go full term is the release. (Polyphony constraints aside.)
    Lifting your finger while the envelope is still in the attack
    phase will send you directly to release. Do not pass sustain.
    Do not collect $200.
    
>aftertouch
    
    Unlike an organ, some synths allow you to control various parameters
    after you press the key down by varying the pressure on the held
    key. That's aftertouch...

>MIDI channel.
    
    16 of them. Just like a TV, if you want to receive a certain MIDI
    channel you have to be 'tuned' to it. If you're not, you won't
    respond to any info present on that channel. It is possible to have
    un-used MIDI channels or multiple receivers of the same channel.
    OMNI mode listens and reacts to ALL 16 channels. 
    
    Some synths are not capable of receiving all 16, but it's fairly
    rare. Others can receive on one channel and transmit on another.
    (BTW - Drum machines are usually capable of receiving on all
    channels but there seems to be some informal agreement amongst
    MIDIots to use channel 10. Nothing carved in stone mind you...) 
        
> Alex
    
    Edd (who really must go do some work...)
          

    Now you have me wondering...
    
    Isn't the data coming out of the OUT port equivalent to the data
    going out the THRU port *plus* the data being generated by the SGU?
    
    Putting it another way - if the SGU is not generating any data you
    would not see any difference between THRU and OUT data. Yes???
    
    	Mark_who_seeks_enlightenment
    

RE: .17

    NO!

    MIDI OUT is used to:

	a) xmit keyboard output (if SGU has attached kybd)
	b) xmit SYSEX data (such as patch dumps)
	c) on *some* units, to serve as an output for merged MIDI IN
	   and keyboard generated data
	d) on *some* units, to xmit "overflow" data (eg ESQ, which can
	   be told to send notes 8+n back out via MIDI OUT)

    MIDI THRU is almost always used only to repeat (forward or pass along)
    data that came in the MIDI IN port. 

    Make sense?

    BTW - Jeff, I think if you read the MIDI spec it says that MIDI is an
    ASCII protocol.  Not a standard per-character protocol (since we're
    dealing with packets [eg note on/velocity/aft-tch/mumble), but
    nevertheless ASCII. 

-b up north

If the MIDI spec says that MIDI is an ASCII protocol, then it is wrong.

What we call "ASCII" is a particular coding standard from the American National
Standards Institute.  (an organization that ocassionaly changes its name) MIDI
is not an ANSI standard, but a de-facto industry standard.  It bears no
similarity in any substantive way, technical or organizational, with the
"ASCII" coding. 

    ASCII=American Standard Code for Information Interchange.
    A=65(10), B=102(8), C=43(16), D=1000100(2), etc.
    
    MIDI assigns numbers to notes, with some flexibility for transposition.
    For a keyboard controller, the number corresponds to a key you press
    (given transposition).  It's similar because hitting a key on a
    keyboard controller produces a signal representing a number that
    represents a note name,
    just as hitting a key on a terminal keyboard produces a signal
    representing a number representing a character.

    You can have an terminal that produces ASCII character numbers
    with a current-loop interface, so it bears resemblance to MIDI.
    Tom

Coding members of a set into small numbers for purposes of communication
has been around for a lot longer than ASCII.  And the note numbers are just
part of what is going on in MIDI messages.

    isn't Baudot 5-bit code 60 years old?  and Morse code 150 years
    old?
    Tom

    God said 'let there be light. Dark. light. light. light. dark..."
    
    Richard.

    Re somewhere back there, (over the rainbow?):
    
    About envelopes, Edd said that the sustain level can be higher than
    the peak level.  I don't see how.  The peak level (for a traditional
    ADSR envelope) is defined to be *the peak* level.  The only way
    the envelope could fail to reach this level is if the NOTE OFF to
    arrive before the attack segment has completed.  Should this be
    the case, the envelope will release from the value it had reached,
    and the sustain segment will never be invoked.  The sustain level
    of the envelope is generally defined as a percentage of the peak
    level, and I have yet to see a synth that allowed sustain levels
    greater than 100% (for a traditional ADSR envelope).
    
    Now, it has become fashionable to provide other kinds of envelopes
    besides good old ADSR.  Cf. Casio's CZ series synths, Yamaha's DX
    synths, and Roland's D series synths.  Their envelopes are defined
    by a series of (level, rate) or (level, time) pairs.
    
    Also, to make things more confusing, there are variations on the
    ADSR theme; the Polaris has a 5 stage envelope for the filter, and
    a 3 stage envelope for the VCA.  The Korg Poly-800 had 6 stage
    envelopes.
    
    Your mileage may vary.
    
    len.
    

    For a traditional ADSR, I agree. But with the n stage envelopes
    available on Yams and Casio's I tend to think of the attack phase
    as "the first thing that happens". It stops when something else
    happens....
                                    L2
                                    --------------  R3
                             R2   /                \
                       L1        /                    .
                        ________/                       . 
                       /                                  .
                 R1   /
                     /                     
                         Time ----->                
    *I* think of the attack as ending at L1.
    
    My mileage always varies...
    
    Edd
    

    re.25 re .24 re.?:
    
    Right, Edd, but that's not an ADSR envelope, which is what I thought
    you were talking about way back somewhere.
    
    Actually, Casio defines the "attack" phase as all the envelope stages
    before the one you designate "sustain".  Thus, the attack is not
    limited to being monotonically increasing.
    
    len.
     

    
    	ok, i'm stupid! Velocity please...
    
    			bs

    The term velocity used with respect to MIDI is simply the measure of
    how fast (or how hard) a key is struck.  It is sometimes called "attack
    velocity". 

    Velocity ranges from 0-127; 0 is the equivalent of a MIDI Note-Off
    command, 1 is lowest, 127 is highest. 

    Another like term is "release velocity" which refers to how fast a key
    is released (the speed in which the fingers are lifted from sustained
    keys).  Some synths respond to this (eg Kawai K5, Xpander) but most
    popular units today do not, and not many keyboard controllers transmit
    the info. 

-b

    
    	re .28 - thx, that was my suspicion but i never could validate
    that with any meaningful context. (In otha words, usually accompanyed
    with plenty of other terms of dubious meaning).
    
    				bs

    S'prisingly enuf, the Mirage responds to release velo...
    
    Edd

                                                                   
                
       DCO (digitally controlled oscillator) is generally a sound-source
       which is fed through the amp/filter envelopes etc.          
                
       LFO (low frequency oscillator) ise generally a modulation source,
       used to modulate oscilators and/or filters and/or amplifiers
       with a selection of waveshapes. Typically used to add vibrato by moding
       the DCO with a sine wave.                                   
                
                                                                   
                
       Richard.                                                    
                

Casio, in its literature on their VZ-1, state the following:

    Key Follow Function.  Casio's innovative key follow function lets you 
    vary both DCO and DCA contours in accordance with the pitch of the key 
    played on the keyboard.  The function gives timbres a much more 
    "natural" and realistic sound.

My question is this: is key follow strictly a Casio innovation as they
claim?  

(I have key follow on my HT6000 to adjust the VCF cutoff point and the DCA
decay time.  I find them to be among the most important parameters on my
synth.  VCF key follow is essentially a variable low pass filter in which
the lower the note on the keyboard, the lower the cutoff point.   DCA key
follow adjusts decay so that, for example, on a piano, you can set the
decay on a low note to be much longer than that on a high note.)

I had assumed virtually every good synth had this feaature...

Alex

    Keyboard tracking is something that virtually every synth (except
    the modulars) has.
    
    In my experience- the better synths allow you to turn the tracking
    OFF- and the crappy ones don't...
    
    	-Bill
    

    Yamaha refers to it as rate and level 'scaling'...
    
    All my 4-ops have it.
    
    Edd

>    < Note 923.33 by GUESS::YERAZUNIS "Row K !?!?!?!" >
>    Keyboard tracking is something that virtually every synth (except
>    the modulars) has.

    Well, both the Emax and Kurzweil 1000PX modulars has 'em

Even my FB01 has it.

    I think the issue is what parameters keyboard follow can affect.
    It's one thing to change the amount of envelope modulating some
    parameter as a function of note number, it's another to change the
    shape of that envelope as a function of note number.  Some synths
    do allow envelope rates to be affected by note number.  I doubt,
    however, that this is unique to Casio.
    
    len.
    

    re .35:
    
    	A _modular_ synth is rarely seen these days.  They look, as
    	Eirikur puts it, like a telephone switchboard attempting to
    	mate with a Hammond Organ.  They're big, heavy, you change patches
    	by moving patch cords (the origin of "patch"), and they are
    	usually monophonic (only 1 note at a time).   
    
    	A synth _module_ is different- it's typically a modern synth
    	with electronic patching, in a 19" box without a keyboard.
    	
    
    -Bill

Dynamic voice allocation.

   Keyboard says that the new Casio synth, which they otherwise apparently 
   like, does not have this feature.  Would somebody be willing to give a
   definition?   

Alex

    Given the recent (and ongoing?) brouhaha about the meaning of
    multitimbrality, one touches this one with some reticence, but...
    
    This gets at the issue Chad (I think) was raising about "slots"
    versus polyphony.  Dynamic voice allocation means you can get as
    much polyphony as you ask for (up to the instrument's limit),
    regardless of how much multitimbrality you're using.
    
    Consider Roland's simpleminded implementation of bitimbrality -
    you get two timbres, each with half as much polyphony as the instrument
    is capable of in monotimbral mode.  Suppose it's a 16 voice instrument
    in monotimbral mode - in bitimbral mode, it's two 8 voice instruments.
    If one of those timbres/instruments is monophonic, too bad, the
    unused 7 voices are *not* made available for the other timbre/
    instrument.  This is "static" voice allocation.  If the voice
    allocation were dynamic, voices would be allocated to timbres as their
    polyphony required, rather than by some preassigned scheme.  However,
    you now have to worry about what to do when you run out of voices
    and some timbre's current degree of polyphony demands more voices
    than are available at the time.  Some examples of allocation strategies
    include stealing the lowest note in use, the oldest note in use,
    some inner voice, according to some user specified priority, etc..
    
    Obviously, this is far more interesting (and important) for instruments
    with higher degrees of multitimbrality.
                                   
    len.

Len,

So far your definitions (here and in the multitimbrality note) are
clearest to me.  

Sticking out my neck, would it be fair to say the following?

   Multitimbrality is a measure of the number of fully defined instruments 
   that can be independently activated on a synthesizer regardless of the
   source of activation (splits or MIDI channels).  Because of implementation 
   constraints, most synthesizers impose increasing limitations on the 
   polyphony of each instruments as the timbrality is increased.  Dynamic
   voice allocation ensures that, as long as the full polyphony limit of
   the synthesizer has not been reached, any timbre that demands another
   voice will be allocated that voice regardless of its current polyphony. 
   When the synthesizer's full polyphony limit is reached, the algorithm
   for "stealing" voices is not based on static limitation on the polyphony 
   of any given timbre. 

If the above interpretation is correct, then the notes over the past
couple of weeks have been helpful to me.  I think I can now make a better
informed decision on future equipment.  

As to layering, here's a try (I'm sure I'll learn from the responses :-))

   Layering is a measure of the degree "stacking" of patches or instruments 
   (within a split region or in a MIDI channel) so that for each note 
   (voice) requested, all of the layered instruments are given the same
   note/velocity/aftertouch information simultaneously.  

If the above layering definition is true, it explains why it is so hard to 
compare multitimbrality with layering.  It also allows one to describe a
Casio toy that has the ability to sound two instruments at once as a
keyboard with two-level layering.

Alex

    I fully agree with -.1 definition of multi-timbrality whether the
    syntho-marketeers do or not...
    
    Edd

Sounds good to me too on first read.

To me, layering is part of multi-timbrality.  You just simultaneously play
two (or more) timbres together.

Chad

Definition of terms describing the audio signal outputs.


Monophonic output:  This sort of output allows a single *note* at a time to
                    be output.  Playing a second note while the first is still
	            sounding will cut the first off.  Useful for that reason --
                    muting cymbals, opening a high-hat, etc.

                    The individual outs on the TX16W and S900 amongst others
	            are monophonic.

Polyphonic output:  This sort allows multiple notes to sound concurrently 
                    through it.  Previously played notes are not cut off when 
                    a new note sounds.

                    The R8, W30, and many other devices have poly-outs.

Stereo pair:        Usually a stereo pair is a pair of polyphonic outputs
                    used together to output a stereo field.

Individual outs:    These can be either monophonic or polyphonic outputs and are
	            outputs that allow voices/timbres/patches to be assigned
                    to individually to separate outputs.  Useful when a mixer
                    is used for panning and also for applying FX to individual
                    sounds and not the mix.

The definitions of monophonic and polyphonic outs seems to be an industry 
standard.  Though I have seen it many times in different literature, most
recently I noticed it while reading the KEYBOARD sampler issue feature 
comparison chart, in a review of the S1000 in KEYBOARD, as well as the W30
ads now in the different magazines as well as in the Ensoniq literature/brochure
for the EPSm sampler module.  All of the above (and many other places)
use the terms to differentiate between output capabilities.

Chad

    I'm new to this game & still trying to learn. I can't even understand
    some of you'r definitions !
    
    Re .16 Whats the SGU ?
    Re .18 Whats SYSEX ?
    Re .32 Whats DCA ?
    
    & while we're about it whats SPMTE (or whatever combination of letters
    it really is ?)
    
    	Stuart P.

>   <<< Note 923.46 by ODIHAM::POORE "Stuart Poore, SRAC, Basingstoke, UK" >>>
>
>    I'm new to this game & still trying to learn. I can't even understand
>    some of you'r definitions !
 >   
 >   Re .16 Whats the SGU ?
    sound generation unit, a synthesizer
>    Re .18 Whats SYSEX ?
    system exclusive.  System exclusive messages are special setup
    messages.
    The format is different for evevery path of every synth and it's all
    undocumented, so it's a b***h.
>    Re .32 Whats DCA ?
    delta current in amps.
>    & while we're about it whats SPMTE (or whatever combination of letters
>    it really is ?)
    society of motion picture technicians and engineers .
    SMPTE code is just a time code.  I think it may be twenty-fourths of a
    second because standard 35mm sound film runs at 24fpm.
    
    TOm
    >	Stuart P.
    

>>    Re .32 Whats DCA ?
>    delta current in amps.

	or Digitally Controlled Amplifier(Attenuator), the digital equivalent
	of a VCA (Voltage)... Probably there are more TLAs like that.

	/pjh

    SYSEX formats are not "all undocumented".  Perhaps this varies from one
    manufacturer to another.  I have never owned a piece of MIDI gear that
    did not document its SYSEX messages in the manual.  (Of course, I have
    never owned an SGU that wasn't made by Yamaha...)

    The Yamaha SPX90 sysex documentation is next to useless.  Note quite
    useless.
    Tom

    What do the society of motion picture technicians and enginneers have
    to do with midi sequencers ?? 
    
    Also, I have a Casio MT-540, which, I admit to you guys is a toy, but
    is all I could afford. It is MIDI but has no mention of SYSEX. What can
    you do with SYSEX messages. You're clearly talking something different
    from the normal NOTE-ON sort of message yes ? 
    
    Many Thanks,
    		Stuart P.

>>    What do the society of motion picture technicians and enginneers have
>>    to do with midi sequencers ?? 


	as little as they possibly can, I'm sure ;^)

	Seriously, SMPTE time code is THE thing that synchronizes films scores
	to films, and video soundtracks to video frames. MIDI spec also allows
	for a MIDI time code, which was put there because MIDI in and of itself
	doesn't give a tick about what relative time things are happening in.

	So if your MIDI sequencer does MIDI time code, you can get a box to
	translate it to SMPTE for less than $1000. US. If your sequencer 
	understands song position pointer, then I think you need to tack
	on an extra $1K for the hardware to translate. Nobody ever said this
	habit was cheap.

>>    Also, I have a Casio MT-540, which, I admit to you guys is a toy, but
>>    is all I could afford. It is MIDI but has no mention of SYSEX. What can
>>    you do with SYSEX messages. You're clearly talking something different
>>    from the normal NOTE-ON sort of message yes ? 
   

	Yes. SYSEX commands are anything that is unique to that unit and isn't
	covered by something else in the MIDI spec. Example: A command to 
	induce your CASIO to blow all of it's patch information out the MIDI
	port, if such a command exists, would surely be SYSEX. Your MIDI
	implementation chart (the one with the Xs and Os) should make mention 
	of SYSEX - if it says X there then you're outa luck. If you don't have
	an implementation chart then check with a music store and see if they
	have one for your board.

	I realize this is a vague answer but SYSEX is a vague concept, as it
	is designed to allow for a huge amount of extra functionality in the
	MIDI spec - depending on the manufacturer, however.

	/pjh

    re: SMPTE
    
    	I think it actually stands for Society of Motion Picture and
    Television Engineers.  This is a technical society along the lines of
    IEEE (Institute of Electrical and Electronics Engineers).  This group
    set the so-called SMPTE standards for making sure events such as
    pictures and sounds can be synchornized.  The previous notes already
    explained what it does.  I just felt a need to be picky.  Sorry.
    
    re: SMPTE <--> MIDI
    
>>>	So if your MIDI sequencer does MIDI time code, you can get a box to
>>>	translate it to SMPTE for less than $1000. US. If your sequencer 
>>>	understands song position pointer, then I think you need to tack
>>>	on an extra $1K for the hardware to translate. Nobody ever said this
>>>	habit was cheap.
    
    If you can get by with little bit of SMPTE, it can be substantially
    cheaper.  The JLCopper PPS-1 (a little grey box) can generate 1 form of
    SMPTE time code on your multi-track (30 frames per second, non-drop
    frame, I think).  Then you can feed that sync signal back into the box
    while overdubbing, the PPS-1 then has a MIDI out that can go to the
    MIDI in of your sequencer.  If your sequencer allows external sync
    based on MIDI time code, put it in that mode and you can have your
    audio tracks (vocals, acoustic guitar, etc.) play along in sync with
    your sequenced tracks while you play or record new tracks.
    
    I do most of my recording now using at least a little SMPTE-syncing
    along the way.  I've seen the newer PPS-1 version 3 advertised for
    ~149$US.  I got an old version 1 for $30 in a music store clearance
    sale!
    
    - John -

    RE: .51
    
    Here's an attempt at a clearer answer that is not clouded by hours
    of frustration :-)

    SYSEX simply means SYStem EXclusive.  MIDI has a number of things
    that might be called "public," e.g., numbers that correspond to
    keys (notes), velocity, after touch, pitch bend, etc.  These are
    parameters common to various brands of equipment and are therefore
    "easily" interchanged between units.  You play note 55 on a Roland
    and it is easily "understood" on a Yamaha unit.
    
    Many characteristics in units are specific to a particular brand--both
    in the syntax (how you tell it what to do) and the semantics (what
    it's able to do).  SYSEX are private codes set up for each
    manufacturer.  It doesn't mean that a different couldn't be programmed
    to emit such a code or respond to it if it saw one.  It's just that
    the manufacturer owning that SYSEX code gets to define what it means.
    And sometimes the definitions are anything but clear.
    
    Casio home units speak as much of the public MIDI code as might
    make sense (e.g., the note number and some fixed number corresponding
    to a fixed velocity).  They do not have SYSEX. The professional
    units do.
    
    With this introduction, you can go back to the dozens of notes
    about SYSEX throughout the notes file and convince yourself that
    you can get arbitrarily confused.
    
    Alex