Conference napalm::commusic_v1

    Well, some days have passed, and no one has responded.. I tell myself
    it's because of the plethora of confusion your topic contains.  Let's
    tackle things one at a time..
    
>                 ...  "Wait  a  sec, I have a 386,  wonder  if
>                there  is  any  software/hardware  to  do the A/D
>                conversion and  upload  it to the PC and edit it,
>                run it back  through  a  D/A  to  tape".  
    
    Well, yes, technically you CAN sample stereo direct to hard disk using
    A/D, D/A technology.  It's REAL expensive.  BIG disk required.  Dyaxis 
    and other companies offer PC kits to allow you to do just this.. HOWEVER :
    
    >            recently, I've been  trying  to  create dance mix
>                tapes,  primarily by taking  two  and  with  a  4
>                track,  recording  one on channels  1&33333333333
>                and  the other on 2&4.   I'm  finding  it  rather
>                difficult  to  match  the two up, 
    
    okay.. your motivation to try the 'sample to hard disk' thing is to
    allow you to synchronize two separate stereo dance pieces.  I'm unaware
    of any pure hard disk digital *4-track* machines that would allow you to 
    slide entire stereo mixes in time.  If they do exist, they're probably
    of the Synclavier variety, and can cost up to $500K.
    
>                On another note, if  I  were  going  to be
>                writing my own stuff (no live performances,  just
>                sampling  and    some    effects  and  multitrack
>                recording), what would  I  need?   I have a DX-7,...
    
    Hang out in this conference.  You'll need to decide if you want to be
    totally MIDI, with no acoustic music, or a mix of MIDI and acoustic.
    If you're totally MIDI it's possible to get along without a multitrack
    tape recorder - rather, you 'multitrack' MIDI events into either a
    hardware-based or software-based (PC) sequencer.  You'll need a small
    mixer.  A playback system - many times your home stereo will do.  A
    stereo mixdown medium -      "     "     "    "  cassette recorder will
    do.
    
    There's a lot more.. subscribe to the various music comix - Electronic
    Musician, Keyboard, Music Technology, etc.  Get familiar with MIDI.
    
    best of luck, karl

                
                WOW!   $500K!??    I'll  pass.   Let's start with
                this,  for  starters,    what   could  I  use  to
                synchronize two songs with  the  same tempo (like
                radio  stations  do on what  they  would  call  a
                'powermix')  so  it  sounds like one  long  song.
                (There  is  a  station  in  Florida  that'll  run
                something  like  this  for  hours at a time  with
                never a break and never will you hear what sounds
                like   two  songs  at  once  when  the  join  two
                different songs).
                
                Thanks!
                
                

    Two turntables with variable speed control, and lots of practice.

    When I worked in college radio we did this kind of stuff all the time.
    In fact, it was an informal kind of contest to see who could come up
    with the smoothest seque.  If you use variable speed turntables, you
    have to make sure that they have accurate speed stops as well, so that
    the neat transition that you auditioned in the production studio sounds
    just as good in the air studio.  Being able to disengage the drive
    motor is fun, too...  we used to have really long versions of Edgar
    Winter's Frankenstein, which we created by spinning the table by hand
    (forwards and backwards) during the synth break.
    
    If you really want to do this at home, you're going to need a couple of
    good, accurate turntables (Technics maybe...), a mixer, and an
    open-reel tape deck.  Practice your transitions in advance, and leave
    the tape rolling.  When you botch a transition, let the tape continue,
    don't touch any record levels, re-cue the second table, and move the
    stylus on the first table back a few grooves.  Redo the transition
    until you get it right.  After you get all your cuts on tape, go back
    and splice out the glitches.  Or...
    
    send your playlist and $100k (what a deal!) and we'll make your tape
    for you!
    
    P.S. lest anybody get excited, it's just me (John Stewart) hanging out
    in AZ, again...
    
    a few grooves

    No need to disconnect the motor, if you use felt pads on the turntable.
    You hold the edge of the record with one finger and start the motor.
    (Broadcast turntables have really hefty motors that can spin a
    weighted turntable up to speed in one-quarter revolution.)  The
    turntable will spin but the record won't.  Then at the proper instant
    you let go and the record instantly starts playing at the cued point.
    
    You may not be able to do this on home-type turntables, as the motors
    are too wimpy.

    any of you guys out there can do this for well under $5k. Here's
    how: Technics markets a CD player with pitch (speed) control and
    balanced audio outs. Combine 2 of these thru a decent quality mixer
    to the DAT recorder of your choice. No editing, but digital sound
    quality nonetheless. Also check out the new 386- based digital audio
    editor in June issue of Electronic Musician.

                                                -elf
    

    I have been looking for a CD player with pitch control for some time now.
    Always been told that there was no such beast. Do you have a model #? And
    even more pertinent a price on such a player?
    
    Just the tool for diggin' out some licks.
    
    Errol

>    I have been looking for a CD player with pitch control for some time now.
>    Always been told that there was no such beast. Do you have a model #? And
>    even more pertinent a price on such a player?

There certainly is such a beast, as I own one! I bought it used for $100 about 
a year ago. The pitch control is certainly nice, as I play along with my CDs to 
learn songs. When bands tune to Eb (Smithereens, Guns and Roses), I just speed 
up the CD rather than retune my guitar.

I'll look at the model number when I get home tonight and post it (unless 
someone else knows and posts it first...)

Alan S.

    
    Technics SL1200CDP, about a grand. +/- 8% pitch control, fine/coarse
    search speeds, jog wheel.
    
    Numark/PPD just announced a two player two box system that supposedly
    will beat match at the push of a button. Rumour is it's about $2K.
    The transport is in one box and the electronics in another.
    
    The $100 used one must be an older Technics export model that is no
    longer available in the US.
    
    CdH
    
    

    What I want is one that is 1/2 speed to cop licks from records.
    
    As it stands now, I sometimes use my sampler for that.
    
    They have stuff that "compresses time" - that is plays it back
    faster, but without an increase in pitch, I'd like to them to
    start looking at the opposite: play it back slower without
    decrease in pitch.
    
    I know, it's a lot harder...  but I don't think it's impossible,
    at least on the theoretical (unpractical) level.
    
    	db

    Anyone who wants my lecture cassette deck that plays back
    variable to twice as fast and shift pitch variable to an octave lower
    roughly let me know for say $20.  It sounds noisy and cruddy but works.
    I modified it to lower pitch of a line input without recording first.
    Sure you get a pitch shifter.  Tell it to shift pitch up an octave.
    Get an old open reel deck with 1 7/8 and 3 3/4 and use it to 
    play stuff back twice as slow.  Done all the time.
    I don't remember if I have used that technique for transcription.  I
    have done the following:
    play a record faster to hear the bass better.
    Play a record slower to hear licks more clearly.  Or less clearly
    depending on their chops ;-)
    Tom
    

    Twice the duration at the same pitch would be real easy to do digitally;
    just repeat each waveform cycle once.  E.g., watch for the zero crossings,
    and you'd need a sample buffer that was as long as one cycle of the lowest
    pitch you want to reproduce.  To get down to 20 Hz, you'd need 1/20th
    second's worth for two channels, or 2 * 2 bytes/sample * 44100
    samples/sec * 1/20 sec = 8820 bytes.
    
    Spin the disk at half speed to keep the buffer filling.
    
    I'm surprised somebody hasn't already done this as a gimmick feature.
    
    len.
    

re: variable speed CD players

>    The $100 used one must be an older Technics export model that is no
>    longer available in the US.
>    CdH

I don't know if this was an export-only model, but it does appear to be an 
older model. The model number is SL-P8, for those on the lookout for one. It 
has the basic features as most players (remote control, search, memory), as 
well as the pitch control.....

Alan S.

Hi there !

There is also a little toy from AKAI that would give you the desired
functionality. I think it's called U5 and/or Trackman Phase Trainer. If you
want i can look it up for you. There was a review in a German magazine. It got
very good ratings. It can record "phrases" up to so-and-so-many seconds and
replay it at any speed with no no pitch change and also loop  of course.
I think the price was about 400,- DM.

	best regards

		Richard

    re.12
    
    Uh, no.  Watching for zero crossings (even, say only positive-going
    zero crossings) and then repeating the sample since the previous zero
    crossing will NOT do time-doubling while keeping freqency constant.
    
    In the case of a pure sine wave, yes.  But imagine a 100 Hz sine wave
    at 1V-peak-to-peak and a 10 KHz sine wave at 0.1 volt pp.  For much
    of the cycle ( > +- 5.73 degrees away from 0 and 180), the 100 Hz wave 
    has enough amplitude to prevent a zero crossing.  But at around +-5
    degrees, the signal will start zero-crossing rapidly.  
    
    So, for much of the signal, you'll get a repeat of the last 1/100'th
    of a second of sound, then a burst of very short (1/10000th of a sec)
    repeats of very short sound segments.  Because these repeats are NOT
    quite equally long (remember, they are triggered by 
    
    	sin (100 t) + sin (10000 t ) = 0
    
    and the sin (100 t) is slowly changing.  So, the high frequency tone
    will see a sort of FM freqency smearing.  For example, the cycle
    right after the 100Hz positive-going crossing will take approximately 
    99.043 microseconds instead of the 100 microseconds it was supposed 
    to (corresponds to a frequency of 10.096 KHz- about 20 cents sharp.)
    Likewise, the cycle just before the 100Hz negative-going zero crossing 
    will take too long to complete, and be pitch-shifted downwards about
    20 cents as well.
    
    Sorry, it did sound nice and easy at first- but I bet it'd sound
    like Yamahabells if you actually used it.
    
    	-Bill
    

    Ok, it'd have to be a little smarter than just looking for zero
    crossings, which for most *musical* signals does a pretty reasonable
    job of determining the fundamental frequency.  Probably the easiest way
    to fix this is to take advantage of the fact that musical signals
    don't change their fundamental much from one cycle to the next.
    
    But I've still got to wonder why it wouldn't sound right.  I mean, that
    100Hz 10000Hz combo is going to sound a bit strange anyway.  What's
    its "frequency" when the 100Hz component is essentially zero amplitude?
    What does it sound like to the ear?  What would a frequency meter read?
    
    It would also probably not work terribly well on noise.
    
    len.
    

Music is almost always more than one frequency at a time; also, many waveforms
are non-repetitive in nature (cymbals, percussion, noise...)

Using a multispeed tape recorder is really the simplest, easiest, and cheapest
solution, if all you want is to play �X or 2X speed. I used to use one to
transcribe Coltrane solos, and even had to use �X speed for parts, although at
thaqt point the pitch gets too low to make out.

If you're not into open-reel recorders, you could get a cassette deck with two
speeds.

    Hi there !
    
    Due to popular demand some details to the little box mentioned in .14
    First of all, it's called AKAI U4 Phrase Trainer. It's basicly a
    8-bit sampler which can sample up to 7.3 sec of any audio source (at a
    rate of 8.4kHz). The playback allows accelerate up to 250% or slow down
    to 45% without any change in the pitch. It has 2 recording modes, REC
    samples the next 7.3 sec, and PRE((view)REC keeps the last 7.3 sec in
    memory (i.e. permanently updates the memory). You can then define a
    startand end point of a loop within this 7.3 sec. It also contains a
    relatively simple parametric EQ that allows boosting or filtering
    specific frequency ranges (e.g. an instrument).
    The size is about of a portable CD player, it's 300 grams (whatever
    that is in pounds...) and the price has been lowered to 280,- DM at the
    dealer I asked last saturday.
    
    	best regards
    
    		Richard

    With a 4.2K bandwidth, just what does this thing sound like??
    
    (The first person who says "a Mirage" dies! " :^) )
    
    Edd

    Hi Edd !
    
    I haven't tried it, but in the magazine they say "Well, not HiFi
    quality, but reasonable for the kind of usage, especially when used
    with EQ." 
    
    	Richard
    	

    re .11
    
    My father has one of those cassette players that you can vary the speed
    and the pitch independantly. It is intended for speeded up listening to
    voice recordings (meetings, etc). The instruction manual specifically
    states it is NOT for music... so I tried it! I popped in a cassette and
    hit play. On 1:1 it sounded tinny but comprehendable. Anything else
    sounded chopped up badly - totally useless mess.
    
    By the looks of it, they used a CCD bucket delay line to do the
    frequency shifting. The "chunking" rate (sampling has other
    connotations) sounded like it was about 100 milliseconds or so - like a
    10 Hz LFO.
    
    Amazingly, it worked pretty well for voices, but horribly for music.