EMUSIC-L Digest Volume 15, Number 2 Today's Topics: Finale for the IBM Brother sequencers answer The way forward? Or just nostalgia? (18 messages) Roland Juno 60 (2 messages) Moog Prodigy (2 messages) MIDI expansion problem (7 messages) MIDI switchers (3 messages) T3 upgrade for the M1? Who's in San Diego? Amiga librarian for the DW-8000? (2 messages) Good (cheap) mail-order? Your EMUSIC-L Digest moderator is Joe McMahon . You may subscribe to EMUSIC-L by sending mail to listserv@american.edu with the line "SUB EMUSIC-L your name" as the text. The EMUSIC-L archive is a service of SunSite (sunsite.unc.edu) at the University of North Carolina. ---------------------------------------------------------------------- Date: Sun, 8 Apr 90 14:20:52 EDT From: MICRO@MOREKYPR.BITNET Subject: Finale for the IBM Is anyone currently running or has run the IBM version of Finale I am trying to decide whether to buy it and I have never seen it? Brian micro@morekypr ------------------------------ Date: Mon, 9 Apr 90 09:35:00 EDT From: "John Young, Southern CT State Univ" Subject: Brother sequencers >Has anyone had any experience with Brother's (the typewriter company) MIDI >sequencers? How much are they selling for on the market? Keyboard Magazine did a review of them in the April issue. They seem to have some serious limitations. Check the review. john young ------------------------------ Date: Mon, 9 Apr 90 11:26:00 EST From: METLAY@PITTVMS.BITNET Subject: Re: Re : The way forward? Or just nostalgia? Some people WILL be prepared to pay for the flexibility. But how much will it cost? The Synclavier is a moneymaker because it does the moneymaking stuff in a moneymaking fashion, not because it's an innovator's dream. Does anyone believe a monster modular setup will sell in the same way? I sure don't. Until it becomes cheap and easy to do without wrecking a company's market share, we won't see it. metlay ------------------------------ Date: Mon, 9 Apr 90 11:43:00 EST From: METLAY@PITTVMS.BITNET Subject: Re: Re : The way forward? Or just nostalgia? Cecilia: If "MIDI 2.0" were ever announced, the backlash in the end user community would make tthe copy code fracas look like a lunch in the park. What has happened is that literally millions of boxes out there right now rely crucially on MIDI 1.0, and any attempt to update MIDI itself would obsolete them all in one swoop. Even adding a use for those extra two pins would be a disaster-- ask any Atari user how he'd feel about a 5-pin MIDI upgrade! It's my belief that the best way to "upgrade" MIDI is to leave it intact and to encapsulate it within a LAN system like the LoneWolf MIDITap, so one can have rapid bidirectional communication between many boxes, all of which are individually just ordinary MIDI devices. The limiting factor THERE would be the capability of each box itself to handle dense data streams, which has intteresting implications for the current generation of Black Boxes(tm) that gag on even moderately complicated instruction sets....modular MIDI, anyone? |-> metlay ------------------------------ Date: Mon, 9 Apr 90 14:16:48 BST From: "Benjamin M. J. Ellis :->" Subject: Analogue filters and other such thingies... Going back to all the chatter about analogue filters: First off, since I live in England (arg...) can any one give me the address of Digisound, so that I can get in touch with them, with a view to buying some PCB's and plans etc... Also, I seem to remember Electronics and Music maker magazine being into that sort of thing at some point, and for anyone who is feeling really suicidal with a soldering iron, there is (or rather are) a whole series of books, they are in the Bernard Bibani series of publications, I think thats how you spell it, I can give details for anyone who is interrested... its a series of little books which are about 2 quid each, or at least were last time a went out and bought one, but that was erm, some time ago, even by my standards, but still, they are cheap, even compared to the cost of a piece of sheet music (not, I didn't really mention that horrible stuff, honnest!) Ok, next, suprise, suprise, its ask for advice time, and guess what, this is a REALLY new and orgional question, 8-; (that's a sarcastic smilely by the way): I play classical guitar, bass, and piano, I have an amiga, and MIDI interface, and a little sampler for the Amiga, but what I really want is, yes you guessed it, a KEYBOARD, really I'd like some thing that can split its polyphony over a few channels, and have different sounds on each channel, but then we can all dream, the unusual opposing restraint is that of cost, ie I don't want it to... I'm perfectly happy to buy something second hand, velocity sensitivity would be nice, but again, cost is really more important I guess... but one MUST is full size keys, I can't hack those tinky tiny little bits of plastic the size of a finger nail, I'm a pianist, not a microsurgeon! any help/suggestions greatfully recieved: thanks in advance, Benjamin M. J. Ellis -------------------------------------------------------------------------------- What goes up, must come down, unless some one pulls out the MIDI lead, in which case it drones on until some one pulls the plug, or kicks it hard... - The voice of experience -------------------------------------------------------------------------------- ------------------------------ Date: Mon, 9 Apr 90 13:31:57 EDT From: "Kurt A. Geisel" Subject: Re: Analogue filters and other such thingies... I'll give you the address when I get home this evening. I am interested in hearing about those books or Electronics and Music Maker. In fact, that publication is where I think the D-80 started. - Kurt Kurt Geisel SNAIL : Carnegie Mellon University 65 Lambeth Dr. ARPA : kg19+@andrew.cmu.edu Pittsburgh, PA 15241 UUCP : uunet!nfsun!kgeisel "We just need to short-circuit the continuum on a BIX : kgeisel 5 or 6 parsec level." - Forbidden Planet ------------------------------ Date: Tue, 10 Apr 90 09:50:40 BST From: "Patrick C.K.Tsang" Subject: The way forward ... >Some people WILL be prepared to pay for the flexibility. But how much will >it cost? The Synclavier is a moneymaker because it does the moneymaking >stuff in a moneymaking fashion, not because it's an innovator's dream. >Does anyone believe a monster modular setup will sell in the same way? >I sure don't. Until it becomes cheap and easy to do without wrecking a >company's market share, we won't see it. >metlay How about the NeXT computer ? How is that coming along ? Even moneymakers nowadays will gradually get away from the moneymaking beast and do it at home with the new wave of digital multitracks and soundtools editors. Is there a future for SynClavier ? Patrick. ------------------------------ Date: Tue, 10 Apr 90 11:52:00 EDT From: METLAY@PITTVMS.BITNET Subject: Re: The way forward ... YES there's a future for Synclavier! As fast as the next wave of home studio abilities is moving, there will always be an envelope oof higher-cost stuff just ahead of it that pros will be willing to pay for, and NED KNOWS tthat. metlay ------------------------------ Date: Tue, 10 Apr 90 09:48:54 -0700 From: PMY@VIRGINIA.BITNET Subject: Re: The way forward ... >How about the NeXT computer ? How is that coming along ? Barry Vercoe's "csound" has been ported to NeXT and given a nice, friendly graphic user interface. Csound, for those who don't know it, is a software synthesis system based on the modular unit-generator model. It is descended from a long line of similar programs such as Music360, Music4bf and Music5. I have also adapted or written tools to accompany NeXT-CSound: - a graphic sound file editor - an adaptation of Paul Lansky's LPC program for linear predictive analysis. - a sound file limiter/compressor (actually, the editor does this as well) All of this is ftp-able from uvaarpa.acc.virginia.edu:/pub/next, and I'm willing to e-mail to those who don't have ftp access. - Pete ------------------------------ Date: Wed, 11 Apr 90 01:56:53 EDT From: "S. Krusee" Subject: Re: The way forward ... Does anyone have any information on the whereabouts of an A.R.P.2600,huh? IIm quite serious about this and I'd liketo purchase one. Please let me know. While my music lays festering I remain .....Stephanie I'd appreicate any responses kids. If this is the the wrong place to ask please foeard me to the correct list. Tanks! ------------------------------ Date: Wed, 11 Apr 90 11:27:09 +0100 From: White Rabbit Subject: Analogue filters and other such thingies... 9 Apr 90 14:16:48 BST The Digisound modules first saw the light of day in Electronics Today International magazine, as well as Electronics and Muisc Maker (which was the original title of Music Technology). The company were a family firm, and were based in Brighton, I believe. However the designer of the modules died a couple fo years ago, and the designs etc were sold off. I don't have the address of the guy who bought up the pcb designs, but the folk who distribute the Curtis chips the range is based on is called Cicada Engineering, I think. You can now buy the PCB's, the plans, kits (Maybe) and fully built modules from the new company (wish I coud remember what its called..!), and the guy that runs it was intending to expand the range last time I heard. He still sends me a catalogue every so often, but my D-80 is 300 miles away and I haven't lived in the same house as it for three years, due to college etc... THIS year I'll do something about it (famous last words!) As far as I know the range includes Power Supply Unit (of course) VCO (with 4 wave outs simultaneously, FM, AM and PWM) VCLFO (as above, but slower) Dual VCA with log and lin controls, plus another control input VCF in lo-pass, hi-pass, notch and band versions, with resonance to oscillation VCF with selectable type (12db and 24db hi and lo, as well as notch and bandpass) with resonance but without oscillation Dual resonant VCF (band pass I think), good for formants Dual EG : ADSR type with several triggering options VCEG : Separatly Voltage controlled Attack, delay, sustain and release Dual Ring Modulator VCDO : Voltage controlled digital oscillator : 32 Waveforms stored in EPROM, and selectable one at a time, or stepping through them. Noise / Sample and Hold Module VC Mixer : Voltage controlled 4:2 mixer Dual LFO Quad LFO Reverb Module : Analogue multi-tapped delay : No Voltage control of time , though Waveform Multiplier : Gives multiple wavesfrom outputs from one input, like a chorus, but a slightly different method Preset Module : For selecting 'patches' or as a sequencer or whatever, selectable groups of preset pots... A control signal adder/subtractor/invertor module Mono keyboard interface Joystick/Octave/ switch module Alphadac 16-note polyphonic keyboard contoller, with sequencer EPROM's etc Input Module : trigger/Waveform follower unit That's all I can remember at the moment, but the price per module varies between about 25-40 pounds, depending on complexity etc. ( may have gone up slightly since last i saw, but not much) but if you buy the components yourself and make up your own PCB's that drops. I'll post the address when I can find it unless anyone else already has. Oh yeah, its ALL CV/gate so will (and has been) interfaced to roland's 100 series modules. And anyone with a soldering iron should be able to make up bits and pieces like an analogue sequencer, manual triggers, 'drone' controllers with pots, and even non-VC filters etc. Check out the Babani books for other ideas if you can get them... Maplin still sell them I think. Hope this inspires more interest in my favourite beastie, and I just wish I could afford a few hundred modules to play with... (Modular-head) White Rabbit. ------------------------------ Date: Wed, 11 Apr 90 10:59:02 BST From: Nick Rothwell Subject: The way forward? Or just nostalgia? 6 Apr 90 17:31:53 EDT <18294.9004071619@lfcs.ed.ac.uk >Speaking from experience, I can tell you that my latest incarnation of >MIDI studio gives me far more productivity than before MIDI was >around! ...when it works, of course. I spent a good 6 hours yesterday chasing down a serial port driver problem between the MOTU MTP and Apple's MIDI drivers. A working MIDI studio is a dream, but it's a tough battle against all the crappy software, nonexistent quality control, and rushed beta-release products out there. Nick. ------------------------------ Date: Wed, 11 Apr 90 11:14:13 BST From: Nick Rothwell Subject: Re : The way forward? Or just nostalgia? 7 Apr 90 11:34:21 EDT <2116.9004100913@lfcs.ed.ac.uk >Have any of you better-informed people heard about work on MIDI 2? Nope, and all the major manufacturers say there's no such thing. Nor should there be. MIDI is fine (well, passable) for short-distance performance link-up between devices. For anything else, use a gateway to SCSI, MIDILAN, RS422 (a la MTP), or something better. Why do we need a MIDI-2? Nick. ------------------------------ Date: Wed, 11 Apr 90 10:54:57 BST From: Nick Rothwell Subject: The way forward? Or just nostalgia? 6 Apr 90 10:52:56 EDT <17054.9004070948@lfcs.ed.ac.uk >Well, here's a thought on the subject of "Black Boxes" (by the way, the >original "Black Box", the DX-7, was really Brown. "Brown Boxes" anybody? >:-) ). Was it my imagination, or were the very first DX7's bright red on the top panel? I saw Steve Jolliffe using one in 84, and I get the feeling it was red. So, my brain is going... sue me... I now have a full 12-unit rack of gear, and luckily, I've managed to do it all in black (well, the M-6R is dark grey...). Looks nice. I've lost count of how many LEDs there are in there, though. >Considering the time factor which Kurt mentioned, which is very real, where >is all the music that must be produced so rapidly? Look at the charts. >Proportional to the >number of Black Boxes available, shouldn't there be more music to show for >it? No, because music production (*real* music, that is) requires some talent and a lot of perseverance. Nick. ------------------------------ Date: Thu, 12 Apr 90 14:30:43 EDT From: Dean Swan Subject: Re: The way forward ... There's been a lot of talk here, on r.m.s. and on comp.dsp about using the NeXT for synthesis. The problem is that the 56001 is only a 12.5 MIPS chip (yes,yes I know it can do a multiply-accumulate in one cycle, but so can every other DSP in the universe), and that just isn't enough bandwidth to implement a complex, polyphonic synthesis algorithm in real time. I told Metlay about my "pet synth project", so I might as well tell you all too. I feel that synthesizers currently available in black box form are only distantly related to the "Acoustic Model" of sound production, so I'm designing a synth more closely tied to that model. The Acoustic Model, for those unfamiliar with it, is a two part model. The sound begins with some form of excitation (the Excitation Function, I'll call it EF for short) which is modified by some system, usually of either variable or fixed complex resonators (the System Function, or SF for short). Analog subtractive systems are somewhat similar to this model. The VCF would correspond to the EF, and the VCF and VCA would correspond to the SF. The problem with this is that it's obviously not a very complex system (yes, Mike, even in the form presented in the Xpander, it's still pretty simple relative to ....oh, say a violin). Real sounds can be grouped according to how they are produced, and modeled with a fixed or variable EF, and a fixed or variable SF. For example, you might model an acoustic piano as a fixed excitation (the impulse of the hammer striking the strings) and a complex set of resonators that model the oscillation and decay of a steel string, the sound board, case, etc. The resonators would be varied based on pitch (at least the ones modelling the strings. The resonators modelling the soundboard and case would be fixed.) The trouble with this model of an acoustic piano is that it becomes very complex, very quickly, and hence impractical to implement. Another problem with it is that one of the goals of synthesis is to give the performer a large degree of control over the production of sound with a fairly small set of parameters. Somebody (i forget who. I could look it up if anybody really *needs* to know) was doing research on the tone of violins, and built a fixed filter bank that was excited by bowing real strings over a magnetic pickup. The filter bank modeled many of the natural resonances and anti-resonances of a real violin, and the results were surprisingly good. To extend this concept, you could drive such a filter bank with a sawtooth oscillator through a VCA with an EG, and also get very good results. Anyway, I percieve most conventional synths as only half the model because a signle variable cutoff and resonance LPF does not allow you to model the resonant characteristics of a finely crafted wooden resonator. Also, they put the VCA and EG in the wrong place in the audio path. They should go at the input to the filter, not the output. This simple modification would allow some interesting sounds to be made, but more resonators are needed to complete the sound. A violin can be modelled nicely with three resonances and one anti-resonance (again, I'm not citing my references, but I could hunt them up if pressured), so you'd need four filters, in addition to the oscillator and VCA+EG. Anyway, I'm toying with two current designs. They differ, largely in their implementation of the System Function. They both have an excitation source consisting of a variable pulse width (from 0% to 100%, 0% being a pulse train - i.e. a single 1.0 sample surrounded by 0.0 samples) oscillator with a pulse width control and a frequency control, a white noise source (it's not a pe perfectly Gaussian distribution. Just a shift register type generator.) and a mix control that determines the proportion of noise to pitched source, followed by a DCA which then feeds the System Function. All of the controllable elements of the excitation function are controlled by a multi-staged EG that can be looped for LFO type effects. My EG design is pretty complex. Maybe I'll go into it later. The two possible implementations of the System Function that I'm looking at are: 1) a Finite Impulse Response filter based on Linear Predictive Coding and 2) a network of probably 8 recursive IIR state variable (i.e. multi- mode) filters each of which is capable of high, low, band pass, or notch functions with variable cutoff and resonance. I'm currently partial to option number 2 because it's more intuitive to program, and you could create patches without the aid of computer analysis software. The LPC method is easier to implement, and easier for resynthesis software to generate parameters for, but less intuitive, and requires a lot of parameter data to run. LPC would have nearly no advantages over multi-sampling instruments currently available. My intention is to use this synth engine as the basis for a resynthesis system which would allow the user to mix and match various portions of different real sounds. For example, you could combine a pitch envelope generated by a violin passage with the System function of a female singing voice, and the excitaion function of a grand piano, or other such things. Any of you who read about Mark Medovich's work in this direction on r.m.s. might think all of this sounds familiar, and it should. I think Mark is heading in the right direction with his project, but I think he's getting caught up in a lot of things at work (aren't we all), and can't spend as much time on it as he'd like to. One thing I disagree with him about is using a transversal filter. It makes sense if you don't want people to program the thing without using analisys software, but I consider that limiting. Anyway, a good synth engine is only half of it. A good user interface is just as important, if not more important. I'm also working on those aspects, but I have a "real" job too, so time is limited to weekends more or less. Well I've babbeled on long enough. I hope this is interesting to someone. By the way, I haven't built any hardware yet, like Mark has. My background is really in software, so this just exists as ideas in my head, and on paper, tied together with bits of C code, and the obligatory cardboard, bailing wire, and duct tape. :-) -Dean Swan dean@sun.soe.clarkson.edu ------------------------------ Date: Thu, 12 Apr 90 14:41:39 EDT From: Dean Swan Subject: Re: The way forward? Or just nostalgia? True, Nick, *real* music does require talent and perseverance, but that's not what I'm talking about. There are studio synthesists in L.A. (and New York, i imagine) that are great players, and sound programmers, but they are music *machines*. They crank out stuff very methodically, usually written by someone else, for someone else's record. I guess I just wonder why people at the top of the charts put out an album, ride on it for two years with lot's of dance mixes and single released, the they do something new. I suppose the considerations are largely economic. -Dean ------------------------------ Date: Fri, 13 Apr 90 11:04:36 MEZ From: Christian Teuscher Subject: Re: The way forward ... 12 Apr 90 14:30:43 EDT from Dean, I am really excited to learn that there are people (a few, tho) concerned with alternate sound synthesis based on the models of REAL instruments. Currently I'm working out some algorithms by G. Rabl, which model a simple string - a digital monochord, so to speak. And there are indeed pounds of parameters and no idea of real-time computation. There is a 1:100 relation run on a 326 with math companion. I am trying to implement it for a MacII with DigiDesign SoundAccelerator board. It seems that you have done a lot of research, lately, and I would be very pleased if you could provide some information. I am a composer/musician and math isn't just the thing I'm particularly trained in. Moreover, does anyone know the Sound Designer file format. I decided not to implement a 'play' function so far, rather write the files to disk and let the professional software do the playing. What is comp.dsp? Can I reach it from here (Vienna, Austria)? --Christian Teuscher, Vienna ------------------------------ Date: Fri, 13 Apr 90 11:14:03 EDT From: Dean Swan Subject: Re: The way forward ... I have specs detailing the Sound Designer file format, as well as Turtle Beach's SampleVision file format, and Syntonyx's Overtone Sample file format. I personally use a file format of my own. My file format uses IEEE single precision floats, mostly because it's more flexible for research purposes, and all my code is for an AT&T UnixPC, so it makes it more portable. I'd be interested in reference to the string modeling algorithm that you're using, Christian. -Dean dean@sun.soe.clarkson.edu ------------------------------ Date: Sat, 14 Apr 90 03:39:03 MEZ From: Christian Teuscher Subject: Re: The way forward ... 13 Apr 90 11:14:03 EDT from Dean, I just learned that X. Rodet and the group in Grenoble use physical modelling for sound synthesis. But I think their approach differs slightly, but fundamentally. The oscillator algorithm is best described as being