Conference napalm::commusic_v1

    I just recently got a Mirage rack-mount. I have no complaints about
    the 8-bit fidelity. The reason it sounds so good is that it's not
    a linear 8-bit -- they have "scaled" the 256 numbers so as to optimize
    it for this application. This is one of the functions of the famous
    Q-chip.
    
    There's a lot to be said for:
    The price.
    The large user base.
    The time it's been on the market.
    Myself and others will swap samples with you.
    
    I would rule out the Sequential box. As you say, they could go belly-up
    at any moment.
    
    Peter
    

    BTW, if you have the cash, buy TWO Mirages and I think you will
    be able to meet ALL of your ideal requirements, right? I would think
    that a dealer would cut you a good deal on a PAIR of them, and you
    would potentially have more flexibility than with any of the other
    units.
    

    The sound quality on the original Mirages was a bit shaky, but I
    think they've spiffed up the filtering enough to make it quite usable
    at this point. A friend of mine had one of the first ones in Boston
    and some patches, especially strings, had noticeable noise when
    we were in the studio. But he just recently had the filter upgrade
    installed (this comes standard on new models) and it made a big
    difference. There's still just a bit of hiss in the upper reaches
    of the string patch, but overall I think it sounds really good.
    
    As mentioned in previous notes the sound library for the Mirage
    is very large and generally inexpensive. The choice of wave editors
    , which is definately necessary if you're planning on doing any
    of your own sampling, is also probably more extensive and cheaper
    than any of the other samplers.
    
    Rob

	Just curious.   How much does the Mirage rack mount cost these days?

    $1300 - $1400
    
    Edd

    huh?
        
    I thought Mirage *with* a kbd was $1395.
    

    rr
    

    Just got off the phone with Ensoniq (I drove home to call). In no
    particular order,
    
    Yes, 5 octaves only, wrapped top&bottom octaves, but both the mks-20
    and kx88 have octave transpose features so if I wanted a solid bottom
    I could 'transpose' the other 2 units up an octave. No 'solid bottom'
    jokes solicited here.
    
    I've spoken to 3 Ensoniq factory reps so far and gee, none of them
    have a S/N ratio for me. Actual quotes: "I don't know, I'm not an
    engineer" "I think it's +12db" "I have that here somewhere but can't
    find it now". Think there's corporate amnesia here?
    
    Claim was the rack Mirage is quite a bit quieter than the keyboard
    because of lack of 'keyboard tracking noise'... huh?
    
    They Ensoniq market a MAC-resident visual waveform editor called
    'Soundlab' for $399.
    
    Tried to sell me an ESQ-1 to drive it even after told I have a KX88.
    I told him to call when it came out as a rack unit.
    
    Sending me a complimentary copy of the Transoniq Hacker. Thought
    there was no official link with Ensoniq. Suspicion is a terrible
    thing.
        
    Stacked voices cause no loss in polyphony. good.
    
    As to affording two.. would be great, na? Then I could also get
    the KAWAI K3M digital wave rackmount for ~$600...
    Gonna get a private demo using headphones, not a low-fi P.A. system
    as playback... I just may join the ranks of Mirage owners !
    
    karl in arizona not nashua.. DEC stock funds another project!!        

     Keyed Mirages must have some sort of scanning circuitry that looks for
     note events. Rack-o units don't need it.
    
     ... no idea why such a thing would be noisy, though.
    
    Edd groovin'_to_his_new_Arizonian_paradiddles_this_mornin'_also!

    Linearly, 8 bits is around 40 to 60 db sig/ noise.
    (too lazy to look it up). 
    
    Who cares about the numbers. Proofs in the HEARING.
    
    Soft samples will not cut it. otherwise: you decide.
    rr
    

    The rule of thumb I use is 6 db of s/n per bit.  So 8 bits would
    be about 48 db.   Not great, but usable.  For comparison a quiet
    phono input to a preamp would run about 65 db and a quiet high level
    input about 80 db.  Most people consider 60 db or better to be
    essentially inaudible.
    
    I don't know where I picked up the 6 db per bit r.o.t..  I always
    get power and voltage confused, but 3 db is a factor of 2, as is
    1 bit, and voltage and power dbs are related by a factor of 2, so
    it sort of works out if you have blind faith.
    
    Ok, Tom, tell us how it really is...
    
    len.
    

    
    Thats the right number. Equation in Chamberlains book.
    
    rr
    

    Let's see, if I remember my log tables correctly, the log base 10
    of 2 is .301, so 20 times that would be... 6 !  (Not to be confused
    with 6!.)
    
    Gee, did I actually get something right for a change?
    
    len.
    

    Noise isn't the only problem with 8-bits per sample.  I have found
    that simple waveforms (pure sine waves) don't sound as good as
    complex waveforms (orchestra patch).  There is a lot of audible
    sound off the pure feequency.  I suspect this is not objectionable
    in a complex waveform because you *expect* a lot of energy spread
    around the spectrum.  Using a 1/3 octave (i.e., cheap) analyzer
    I found that I could whistle a purer sine wave than I could get
    out of an 8-bits-per-sample-linear synthesizer.
    
    I guess if you are a purist what I have been describing counts as
    "noise", but it doesn't sound like the usual kind of noise from
    an audio system.
        John Sauter

    Strictly speaking, it is a form of noise, but there's a special
    form of noise called "distortion" which is usually singled out.
    Specifically, "distortion" seems to mean noise that's related to signal
    content, and we tend to reserve "noise" to mean noise that's
    independent of signal content.
    
    The "6 dB/bit" "signal to noise ratio" is really a dynamic range
    assessment, which can be misread as a S/N ratio if you assume that
    the noise is all due to quantization error, a gross oversimplification.
    Thus, it's more correct to use the 6 dB rule to bound the dynamic
    range of the signal.  The actual signal to noise ratio can be no
    better than this bound, and will in practice be quite a bit less.
    
    len.
    

    Update: yesterday was my personal day, use it or lose it... so I
    used it.
    
    The latest issue of Electronic Musician has a Mirage mod which makes
    it STEREO (one of my pre-demo peeves) and claims to better the hotly
    debated S/N ratio by 6db.
    
    So I went and was able to futz with a keyboard Mirage for 1 hour
    using MY headphones. There is an audible thump as the key reaches
    bottom. P'raps this is the famed 'keyboard tracking noise'. Overall
    I wasn't too displeased. Quite an extensive library.. the thought
    was 'I could live with this', especially with stereo voice assignments.
    
    HOWEVER !!! On the way home I stopped in at another music store
    which carries both the E-mu EMAX and the AKAI S900. Well, this was
    still early in the day (no heavy metal guitarists, they don't awaken
    until 1pm) and I got a real good overview (again) of these two
    samplers. They sound so much alike.. I played the S900 from a Roland
    88note RD1000 keyboard, and found that the S900 has true 88note
    range ! There was a cello patch SO clear I swear I could hear the
    rosin flaking off the bow.. or was that just hiss ?
    
    So the local store has a decent S900 library, and continues to port
    Emulator samples over, AND has beat the taxless price out of Los
    Angeles.
    
    The quality difference was clear. If I hadn't heard the two higher-
    end samplers only 10 minutes after the Mirage, I'd never have known..
    
    Let you know... k the m

    
    Yo, Karl, how much they want for the 900?
    
    ron_who_sees_a_sampler_on_the_Xmas_list
    

Went over to play with Stevie K's Ensoniq ESQ-1 the other night, and wasn't 
impressed with the synth part but was EXTREMELY impressed by the quality, 
flexibility and ease of use of its sequencer.  Or is MIDI recorder a more 
appropriate term?

Is there a sequencer out there that allows recording of different tracks 
without permanent merger, where you can determine how much of the final mix is 
each track, play back with any combinations of the recorded tracks, more than 
two buttons to operate?  After a 45-minute office interruption, I've 
forogotten my other questions.


				Mark

    The old (now discontinued) Roland MSQ-700 allowed you to keep up
    to 8 tracks separate, but the aggregate capacity was limited to
    about 6500 notes.
    
    The MC500 allows you to have up to 4 tracks, which (like the MSQ)
    you can combine anyway you want.  You can extract a single channel
    from a merged track, so if you're willing to do some shuffling,
    you can extract/combine in various ways to get the effect of the
    MSQ's more simple (8 tracks, but no channel extract capability,
    except via some legerdemain - you connect the output to the input,
    and tell the receiving track to only listen to the channel you want,
    then play the source track to the receiving track) scheme.  The
    MC500 holds about 25000 notes onboard, up to 100000 on disk.  However,
    you can't "page" from disk, so you're effectively limited to 25000
    note compositions, which with an 8 track recorder really means 175000
    note compositions before you have to go second generation on tape
    (you have to save one track for sync).
    
    The MSQ requires two buttons to use this way, the MC500 two, maybe
    three.  If you need more, you can make a nonfunctional button box.
    
    len.
    

Oh yeah, I remember the other basic question:  Are all the sequencers like the 
Yamaha QX-7, which loses all its memory upon power down?  The ESQ has static 
memory, so if you have to rewire after doing something you don't lose what you 
did.  This was why I sold my digital delay pedal; unplugging the input shut 
the pedal off, which lost the data.

Don't believe the Rolands are what I'm looking for, and I require human 
interface and feedback since I'm stupid and have a short memory.  I'm 
wondering if a microcomputer is the way to go, especially with prices these 
days.  Or maybe the Yamaha music computer's successor.

Playing with Steve's ESQ showed me how much faster it would be to do 
songwriting with a top notch sequencer rather than always doing the drum 
machine on 4-track for everything (which locks you in, takes longer to 
program, takes longer to reprogram, etc.).


				Mark

    QX-7 looses memory at powerdown? Mine doesn't. At least it doesn't
    immediately. It takes about 3 days to loose a sequence.
    
    It also remembers the parameters.
    
    Edd

    Mark, before you write off the MC500, you really outght to look
    at one.  The interface is pretty straightforward once you understand
    the principles behind it.  And it's *very* powerful.  It does forget
    everything when you power it down, but you just pop a disk in it
    and there you are!
    
    len (who is *not* paid by Roland).

    This is a SAMPLER topic. Sequencers are/have been discussed at length
    elsewhere.
    
    Ron; re AKAI S900.. I've been quoted $2400 from West L.A. and $2355 
    at a local Tucson store... including whatever library..
    
    However ! The latest issue of KEYBOARD has a E-Mu EMAX product review..
    keyboard model only. I'm still hiliting various points, trting to
    figure what features I really need vs. wish for.. plus, how long
    can I wait for the EMAX rack to come out?
    
    Stacking sounds..The S900 CAN stack samples, with loss of polyphony.
    The EMAX can, but in 'dual' mode (analog filters common both samples)
    it will NOT cause voices to drop out.
    
    MIDI implementation using external sequencer: S900 can set up different
    presets to respond to separate MIDI channels... spanning different
    or the SAME portion of the 'keyboard'. The EMAX review states that
    each preset can be assigned to a specific MIDI channel. It does
    NOT state whether it will respond to multiple MIDI channels in 'mono'
    mode or not.. still unclear. Call to E-Mu coming up.
    
    Programmable stereo panning: The EMAX review stated how this is
    accomplished, the S900 review didn't even mention it, nor did the
    dismally noninformative glossy sheet.. tho the S900 does have stereo
    outs as well as eight individual outputs (same w/EMAX). Haveta go
    back to the store for this one.
    
    Both machines are superb, CD quality.. not that the rest of my
    recording gear is quite as good. The $300 difference between the
    AKAI S900 and the EMAX is really not the deciding factor here. It's
    probably youthful impatience to finally DO IT .
    
    karl

    
    Karl
    
    3 things to help you on your way.
    
    1
    I spoke to a guy this week in the U.K. re the purchase of a Mirage
    rackmount and he said that they are O.K. providing you don't want
    to make your own samples. Apparently the factory pre-sets are recorded
    on a larger machine and the Bits down-line loaded. You could never
    get the same quality "at home". Any comments Mirage owners?
    
    2 
    The E-MAX 8 outs are not implemented in the first (prototype?)
    software revision. Will they be implemented in the production versions.
    (according to a local review).
    
    3
    The Roland S50 will be out in rack-mount soon. This machine will
    do just about everythin I seem to remember you wanting including
    built in video wave-form editing.
                  
    		Best of luck.
    
    			Paul
    

    Got a chance to hear the Roland S50 this weekend.  This machine
    sounds *very* nice, it is the cleanest sounding sampler I have heard,
    so clean it doesn't sound like a sampler, it sounds like an incredibly
    versatile synth.  The video interface is neat - all you need add
    is a b&w monitor, or color if you feel like it.  Just plugs right
    in and you get visual editing capability.  Like the Mirage and the
    MC500, it is dedicated hardware controlled by software loaded from
    disk (builtin 3.5" drive) so it offers the option of increasing
    functionality as they update the software.
    
    I was told a rack mount version was coming, but not to hold my breath.
    Maybe it will be shown at the forthcoming NAMM (February?).
    
    Still a little pricey at $2600 or more, but it is *nice*.
    
    Most of the samples I heard were homebrew, not factory supplied,
    including an 8 second heavy metal power chord complete with feedback,
    virtually indistinguishable from the real thing.
    
    I think I've found my sampler.  Now all I need to do is find some
    money.
    
    No samplers from Yamaha yet?
    
    len.
    

    Battle of the 12bit 8 voice samplers continues... round 45..
    
               AKAI S900 vs. E-Mu EMAX
    
    Back to the store Saturday, spent time poking thru the manuals.
    The EMAX manual was written by Craig Anderton.. quite thorough,
    with the GLARING lack of a MIDI implementation chart. The AKAI 
    manual, while short, covered the material adequately.
    
    Voice mix outputs : The EMAX' programmable voice panning DOES work, 
    into the stereo outs, and in conjunction with with the programmable 
    (relative) volume cut/boost of 45db, has the equivalent of an 8channel 
    mixer implemented in software. The AKAI has 8 outs (needs 8chan
    mixer) and stereo outs, BUT the voices can be programmed to 'go
    left' and 'go right' only.
    
    Note range : Here's where it gets confusing. EMAX: MIDI response
    range from the spec sheet, 'full 88 key'... I couldn't find any of 
    the factory samples/programs which would respond beyond 6 octaves, 
    using an external 88note controller. The AKAI, on the other hand,
    which is advertised as having only a 6 octave range, had no problem
    responding to a full 88 notes. Weird.
    
    Sidebar (yes I know this is a SAMPLER note) I found a nice inexpensive
    8chan stereo mixer from Yamaha, the KM802, ~$295 list, 3 effects sends
    per channel, 2 chans allow attenuated mic in. NO EQ, parametric
    or otherwise. This is why it's cheap. Sliders, LED level indicators.
    Pan pot per channel. If I get the AKAI I'll also get one of these...
    which brings the price right up to the EMAX... AND the EMAX has
    its own sequencer, downloadable from the MAC, AND dual samples per
    voice with no polyphony loss, AND programmable panning... I do believe
    that putting this stuff into the conference has helped me quantify
    my requirements and to see the gear a bit more clearly !
    
    karl moeller
    
    
    
    
    T

    
    re. Yamaha mixer...
    
    have one of these. Nice unit. VERY quiet.
    There IS separate EQ on the 2 'acoustic?' channels,
    but for this you lose send on all but bus 3. THESE
    channels can also be used for synths!
        
    Note separate output level faders for main stereo outs
    and headphone outs. Which is nice live cause you can
    go main out to the PA and use phone out (I use a stereo to mono
    1/4 phone plug ) to a monitor amp. You have separate main and monitor
    LEVEL controls at your fingertips.
    
    I thought it was a bargain (do mail order) for a key mix.
    
    Sure you can get a 'real' board. Not for ~$250.
    
    2 cents.
    
    ron
    

    The votes are just about in, and here's how it stacks up. The Mirage
    is out due to my ears. So it's the EMAX and the S900. I'll do this
    by category, and attempt a 'score' per category, with total. 0=low,
    5= high.
Score/Feature    
    Voices:
[5]	EMAX: 8. 
[5]    	S900: 8.
[5]    	EMAX: stack no polyphony loss. 
[3]    	S900: stack, polyphony loss.
    Availability:
[1]	EMAX: 3wks-2months. Real Soon Now. 
[5]     S900: NOW. Like, IN STOCK.
    Library:
[4]	EMAX: factory disks already in stores. Emulator ports likely.
    	      E-Mu has good track record, sampling parties, etc. A matter
    	      of time, esp. as various MAC-resident visual waveform
    	      editors support EMAX, allowing MAC-MAC conversions.
[1]]    S900: Crappy, highly dependent on individuals. Dismal after
    		full year on the market.
    MIDI Implementation:
[5]    	EMAX: each preset can be assigned to a specific MIDI channel
    		for sequencing.
[5]	S900: each preset can be assigned to a specific MIDI channel
    		for sequencing.
    Audio Outs:
[5]    	EMAX: 8 outs, 2 stereo w/programmable panning, mono
[3]    	S900: 8 outs, 2 stereo, programmable L/R only, mono
    MIDI Note range:
[5]    	EMAX: 128 notes logical, 88 notes physical
[5]    	S900: 88 notes logically/physically
    Price:
[3]	EMAX: $2700 no discounts incl library
[5]    	S900: $2400 discounted   incl library    
    User Groups/Support:
[4]    	EMAX: anticipate major factory support/ user group
[0]    	S900: after one year, nothing anywhere.
    Users Manual:
[4]    	EMAX: ~200 pages, by Craig Anderton. Very thorough. NO MIDI
    	      spec.
[2]    	S900: ~45 pages. Adequate. 
     Memory:
[3]    	EMAX: 512k	
[5]	S900: 750k    
     Unique Features:
[2]    	EMAX: Full feature arpeggiator. 
[3]    	      Internal sequencer can take entire sequence from external 
    	      unit in one pass. Good performance feature.
[2]    	S900: More memory.. more presets, longer samples.
    
    The envelope please :  EMAX: [49], S900: [42] ! ~18% gap... what
    would YOU do ?
    

    Go for the E-max.
    
    Edd

    
    A. Go for the Akai		(just had to say it Edd)
    
    B. Get both. Use em. Choose. Sell loser.
    
    C. Get a Kurzweil

    D. wait till namm in Jan.
    
    rjr
    

    I've been thinking about the difference in memory between the AKAI
    at 750K and the EMAX at 512K.
    
    The EMAX uses E-Mu's hotly debated 8bit companding scheme. Whatever
    it is, it works, and seems to give results commensurate with 12bit
    linear.
    
    As the AKAI uses 12bit bytes, it would seem that the extra 250K
    would be used just holding these extra bits.
    
    And, indeed, the maximum sampling time available on these two units
    is virtually identical.
    
    k the m

    
    Its surprising no one mentioned the Korg DSS-1 in this note.  I know 
    it has a keyboard, but so does the S-50 discussed here.   The Korg
    seems like a nice unit at under $2k, yes? 

    I'm first in line for an Emax Rack at Synthony Music in Scottsdale
    Az. I was gonna buy out of West L.A. Music but I was 10th in line
    there... 1st in line at the 3rd biggest dealer in the US is better
    than 10th at the 1st biggest.
    
    Price has stablilized at $2750 + tax. aaaarrg. I've never spent
    so much on anything, including autos, in one place, in my life.
    
    kero miller

>    Price has stablilized at $2750 + tax. aaaarrg. I've never spent
>    so much on anything, including autos, in one place, in my life.

    Gee, as a S/W Engineer with a $10K studio, this should be a drop
    in the bucket for me. I would like to point out that I've been 
    evaluating samplers and collecting the money for this big step for
    over a year. Hardly an 'impulse purchase'. Which makes it nonetheless 
    startling to actually face spending the bux. And, incidentally, shows 
    where some of my priorities lie. And my wife's. 
    
    Frankly, Dave Blickstein is doing a bit too much of our thinking
    to suit me. I feel cheated of an opportunity to respond to Tom's 
    latest scattergun cheap shot, which I feel was aimed in part at me.

    As I know noone on the net or in my hometown that is looking for
    or using a top-of-the-line sampler, 'keeping up' with anyone else
    doesn't enter into it. As for this conference 'egging on' unneccessary
    purchases... I've found this a wonderful forum to lose some of my
    fear of MIDI and computer-based MUSIC systems. And since I hear
    far beyond solo piano timbres in my imagination, a good sampler
    and a computer-based sequencer which allows editing of my
    improvisations seem to me to be very valid tools for MUSICmaking.

    If one is fiscally at survival level, some purchases beyond survival
    needs might be seen as superfluous. I am very fortunate in being
    able to 'indulge myself' in what has become not only a very satisfying
    hobby, but MAY evolve into a profitable tape-sale operation. So
    I would ask, is it only moral and ethical if it makes money?
              

    After reading thru all these, I thought it might be good to get this
    topic started again, with the advent of all the new toys hitting the
    market.

    What exactly are the major issues as far as a sampler is concerned?
    Here are the ones I know about:

      - amount of memory (usually maps to sample time)
      - sampling rate(s)
      - no. of bits used for A/D/A
      - no. of bits used for internal data manipulation

    And of course there are things like:

      - number of outs
      - number of voices (simultaneous notes)
      - auto-looping
      - poly-timbrality
      - dynamic voice allocation (a biggie in my mind)

    But are there other things to consider?  Number of filters?  Envelopes?
    Loop points?  I've heard some things about the Emax sounding better
    than the S-50 because of the "way it plays back samples".  What in the
    heck does that mean?

    Any comments from sampler owners?

-b

    Digidesign support...
    
    Nice clean *input* filters...
    
    Available Library...
    
    Load time...
    
    Edd

    
    
    re: Brad .-2
    
     Yeah, those are all major issues.
    
     Amt of memory & sample rate together relate to sample time.
     
    I think the two more are:
    
    - Ease of Use.
    - Understandability.
    
    The human factor should not be ignored. I get really sick of
    programming both by day and by night and not getting to play music
    which is what I wanted to do in the first place. One of the reasons
    I bought the EPS is that it is an extremely expressive instrument,
    and it is quite simple to program, therefore it is simple to make
    programs expressive. And it's laid out well, IMHO.
    
    The subjective side of the discussion ("it plays samples better")
    is kinda like every argument you ever heard about what the best
    stereo speakers are. Everybody's box adds some kind of color to
    your sound, if only to smooth those nasty non-natural harmonics
    that they all add in. That's purely subjective once you get past
    the THD, IMD, etc. analyses.
    
    How much bang for the buck? What features? All have subsets of one
    another, most have proprietary features (i.e. S-50 is mouse driven,
    and has cute little keyboards that play along on the screen, the
    EPS has 'poly-key' and some somewhat intelligent looping algorithms,
    the Korg DSS-1 has on-board delays, etc.)...
    
    Also, what do they cost? Are you willing to shell out 1,2,3,4,9,or
    30K for a sampler?
    
    /pjh

    Right now, music stores are dumping S-10's for about $800-900. 
    I picked up one.  The reason for the blowout is that nobody wants the 
    QD's because they cost $3-5 each and can only store � to 1 sample on 
    each disk.  Also, there's not as much sample time (up to 8 seconds)
    and there's no aftertouch.  On the plus side, it's an easy, simple, 
    stock sampler.  The keyboard is nice (IMO) and pretty standard (not
    a toy).  Word has it that the guts are S-50.  For me, it's perfect.
    I'll buy a few disks.  Maybe down the road I'll look into some other
    form of storage.  But, it sounds great and is pretty easy to use, as
    has been discussed elsewhere.
    
    Steve_who_is_days_away_from_submitting_stuff_for_Commusic_V

    The MC-500/MRB-500 combo supports the S-10, so if you've got an
    MC-500, you could "upgrade" to 3.5" floppies.
    
    len.
    

    > Digidesign Support....
    
      Pardone, pardone moi, es "SoundDesign"...
    
    Edd