EMUSIC-L Digest Volume 33, Number 5 This issue's topics: csound requirements (2 messages) The Birth of the Modern Synthesizer Re: Yamaha V-50 Re: Tuning "A Clockwork Orange" NeXT (Stop me before I reinvent the wheel!) synth user groups? (2 messages) Filter response measurement (7 messages) Casio VZ10M Starter synth (2 messages) Soundfile header conversion program 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: Thu, 24 Oct 1991 11:54:06 EST From: Kathleen Russell Subject: csound requirements Can anyone tell me what the suggested minimum hardware requirements are for doing CSound [amount of memory required, hard drive size, processor speed, etc.]. Thanks, Kathy ------------------------------ Date: Thu, 24 Oct 1991 09:54:46 -0700 From: Tom Erbe Subject: csound requirements The requirements for CSound are somewhat flexible. I actually compiled a Mac + (MPW) version of CSound 2-3 years ago, but it took about 8 hours to make 80 seconds worth of sound! I suggest you use a system with hardware floating point and a 32-bit processor (Any Mac II with FPU, a NeXT, Indigo). As far as memory requirements, that depends on how big your gen function tables will be. The same with hard disk size. If you want to make a 10 minute stereo piece at 44.1K sample rate, you will need 106 megabytes of disk space. (10.6 Megs per stereo minute at 44.1K, 11.6 at 48K). So the answer is... use as big of a system as you can buy or borrow. Tom Erbe * Technical Director * Center for Contemporary Music * Mills College tom@Mills.edu * Mills College, Oakland, CA 94613 * (510) 430-2191 ------------------------------ Date: Thu, 24 Oct 1991 15:11:25 EDT From: "Joseph D. McMahon" Subject: The Birth of the Modern Synthesizer (I reposted this message because LISTSERV didn't like it for some reason. -JM) Forwarded message: > From: Bryan Basham > Subject: The Birth of the Modern Synthesizer > Cc: EMUSIC-L@auvm.american.edu > >> Does anyone out there have a recording or original record of >> Switched on Bach? It was a important piece that proved electronics >> could make music and not just beeps and blips. > You're quite right! Walter (later Wendy) Carlos was a true pioneer in the > use of keyboard-based synthesizers. During graduate work at Columbia, she > met Robert Moog (an MIT electronics wizard) and a long standing colaboration > ensued. W. Carlos has created several albums reproducing classic works: > (a) "Switched-On Bach" > (b) "The Well-Tempered Synthesizer" > (c) "Clockwork Orange" soundtrack > (d) "Switched-On Bach II" > (e) "By Request" > (f) "Switched-On Brandenburgs" > and some of her own compositions: > (a) "Sonic Seasonings" > (b) "Tron" soundtrack > (c) "Digital Moonscapes" > Also of possible interest is a record she completed, in '87, that is a > tutorial on synth techniques that she has used, called "Secrets of > Synthesis". By the way, all of these albums can be found on the CBS label. > >> Also info on Bob Moog >> wanted.. > You might want to acquire an history book on synthesizers; he, along with > Donald Buchla (SP?), virtually created the world of analog synthesizers. > Buchla is a lesser known designer for the West camp of Electronic Music, but > he still has a market for synths. His style is opposed to the keyboard > setup; he uses multiple pressure sensor pads -- a potentially more > expressive technique, but with a larger learning curve. > A few comments: Wendy Carlos' most challenging (and many would say best) album, "Beauty in the Beast" is no longer in print. Look for it in used CD and record stores. "By Request" contains a number of Carlos' early works and some rather unusual compositions. --- Joe M. ------------------------------ Date: Tue, 22 Oct 1991 11:30:17 EST From: Jon Crystal Subject: Re: Yamaha V-50 The # for Yamaha Customer Service is 1-800-443-2232. Good Luck! ------------------------------ Date: Tue, 22 Oct 1991 11:37:23 EDT From: "Roger B. Dannenberg" Subject: Re: Tuning Originally submitted by rbd+@ANDREW.CMU.EDU EMUSIC-D: I don't want to start a big tuning debate, but much of what I read in Vol 39, Number 8 of EMUSIC-DIGEST is unsupported by actual measurement. I highly recommend reading the studies by Johann Sundberg who has made careful measurements of intonation in singing. The results are not nearly so simple as "The whole thing of ringing chords in barbershop is based on just intonation." Whether or not this and other statements made are true, there is more to the story, and there have been few careful studies. ------------------------------ Date: Fri, 25 Oct 1991 08:54:48 EDT From: ronin Subject: cw orange a brief note... there are actually two clockwork orange albums. the first is the one directly associated with the film, bearing its logo on the cover. the second is called 'music for clockwork orange', and includes a few very interesting things that never made it into the film. -------------< Extremism in the Pursuit of Good Noise is no Vice >------- Eric Harnden (Ronin) or The American University Physics Dept. Washington, D.C ------------------------------ Date: Fri, 25 Oct 1991 09:07:42 EDT From: ronin Subject: next(stop me before i reinvent the wheel!) i am embarking on a programming project on the next workstation. my intention is to develop a graphic tool for editing synthpatches, to be used as a module in a larger development environment. if this has been done, and particularly is there is any publicly available work of this nature, please let me know now, so that i don't generate redundant work. -------------< Extremism in the Pursuit of Good Noise is no Vice >------- Eric Harnden (Ronin) or The American University Physics Dept. Washington, D.C ------------------------------ Date: Fri, 25 Oct 1991 09:38:02 EDT From: Brian Adamson Subject: synth user groups? This is a call for info on synth-specific user groups/ lists on the net. Does anyone know of such for the Yamaha SY77/TG77 and/or Ensoniq EPS 16+? Also, whatever happened to the KORG users list that ran for a while? I haven't received mail in some time. Thank you, -- Brian Adamson NRL Code 5523 adamson@itd.nrl.navy.mil ------------------------------ Date: Mon, 28 Oct 1991 12:34:05 EST From: Jon Crystal Subject: Re: synth user groups? I don't know of existing lists, but as an SY77 owner I'm interested! ------------------------------ Date: Fri, 25 Oct 1991 09:11:04 EDT From: ronin Subject: filter project:1 this is the first in what will be a continuing series of postings on a research project underway here at au. i've brought up the question before of why certain filters sound the way they do. there are several aspects to this question, of course. 1) what is the nature of their electronics. 2) what effect does a given filter design decision have on their transfer characteristic, in both frequency and phase space. 3) what correlates exist between these transfer characteristics and a listener's perception of the filter's 'sound'. 4) to what degree is a filter's sound uniquely identifiable. we are going to do two sets of experiments to approach some kind of quantitative analysis if these questions. the first set will a set of filter analyses, and the second will be a round of psycho- acoustic tests. specifically, to begin with we will subject the lowpass filters of moog, oberheim, roland, arp, and sequential circuits instruments to standardized signal testing. we start on the moog next week. we will pass sawtooth tones of fixed amplitude at 250Hz, 500Hz, 1KHZ, and 2KHz through the lowpass filter, and measure the overall amplitude of the time domain signal and the frequency domain amplitudes of the fundamental, second harmonic, fourth harmonic, and eighth harmonic for each input frequency, for the filter set at a series of as-yet unspecified control positions. we will also pass a noise signal through at these positions, and make the same measurements. we're using a sawtooth because on several instruments we will be unable to inject a sine wave into the signal path, and the instrumements themselves do not generate sine waves. the sawtooth has, in a certain way of looking at it, the 'simplest' spectral structure available on a standard synth. it is well characterized, very general (since the whole harmonic series is represented), and should show pahse effects pretty readily. the redundancy in the test sequence is to satisfy ourselves about the generality of filter responses. we are, to a certain extent, calibrating ourselves as well as the equipment. for the moog test we will be simultaneously taking readings on a 10MHz storage scope, an older cathode display spectrum analyzer, and a Data Precision Data 6000 digital sampler/analyzer. we're calibrating our lab as well as our heads, and need to cross-correlate some information. if results are as expected, we will be using only the Data 6000 for all future analysis runs. the next posting on this will be delivered when the moog test run is complete. until then, if anyone has any thoughts or questions, please drop a line. -------------< Extremism in the Pursuit of Good Noise is no Vice >------- Eric Harnden (Ronin) or The American University Physics Dept. Washington, D.C ------------------------------ Date: Fri, 25 Oct 1991 09:57:19 EDT From: Brian Adamson Subject: Re: filter project:1 > of moog, oberheim, roland, arp, and sequential circuits instruments > to standardized signal testing. we start on the moog next week. > we will pass sawtooth tones of fixed amplitude at 250Hz, 500Hz, 1KHZ, > and 2KHz through the lowpass filter, and measure the overall amplitude > of the time domain signal and the frequency domain amplitudes of the > fundamental, second harmonic, fourth harmonic, and eighth harmonic > unspecified control positions. we will also pass a noise signal > through at these positions, and make the same measurements. > we're using a sawtooth because on several instruments we will be unable > to inject a sine wave into the signal path, and the instrumements themselves > do not generate sine waves. the sawtooth has, in a certain way of looking Are you using each instrument to generate its own noise and/or sawtooth for the experiment? If so, there can be significant differences in how each instrument generates its sawtooth and/or noise which could skew the results if you are interested in measuring the filter response. If you are injecting the signal externally, I'm not sure why you couldn't use a sine wave. You may want to look at the spectrum of sawtooth and/or noise prior to entering the filter. Also, this measurement is a static one, I assume. (i.e. filter is not changing dynamically via envelope, LFO, etc). A suggested extension to your measurement would be to see the dynamics (in the freq. domain) of the VCF's. This may require digitization of the output and offline signal processing, to look at spectral snapshots of the signal. LPC analysis of the filters would also be interesting. -- Brian Adamson NRL Code 5523 adamson@itd.nrl.navy.mil ------------------------------ Date: Fri, 25 Oct 1991 16:07:52 EDT From: ronin Subject: Re: filter project:1 On Fri, 25 Oct 1991 09:57:19 EDT Brian Adamson said: > > Are you using each instrument to generate its own noise and/or sawtooth >for the experiment? If so, there can be significant differences in how >each instrument generates its sawtooth and/or noise which could skew >the results if you are interested in measuring the filter response. > yes, in all cases except the moog i will be using the onboard oscillators. for some of the instruments, i could do some jury-rigging and inject the the signal, but that will be more difficult on the oberheim, for instance, than on the arp. i'm injecting on the moog because its own oscillators are pretty unsteady. as to the sawteeth, i'm running on the assumption that, since all of these are analog oscillators (even if digitally controlled), the sawtooth will be standard. please tell me... what difference could there be? symmetry, i suppose. i can imagine a sawtooth with a glitch at the turnover point. i'll have to check that. as for noise, you're absolutely right. different instruments use different methods to generate noise, which will probably have different probability distributions. i suppose it would be a good idea to characterize these sources independently. i will do so. > >Also, this measurement is a static one, I assume. (i.e. filter is not >changing dynamically via envelope, LFO, etc). A suggested extension >to your measurement would be to see the dynamics (in the freq. domain) >of the VCF's. This may require digitization of the output and >offline signal processing, to look at spectral snapshots of the signal. >LPC analysis of the filters would also be interesting. > at the moment, all measurements will indeed be static. the idea that a filter's dynamic and static responses are going to be insignificantly dissimilar is a tacit assumption in this test sequence, and one that i must admit i'm not willing to defend. you're right. these are useful extensions, that i might pursue after the initial analyses. -------------< Extremism in the Pursuit of Good Noise is no Vice >------- Eric Harnden (Ronin) or The American University Physics Dept. Washington, D.C ------------------------------ Date: Fri, 25 Oct 1991 16:51:53 EDT From: Brian Adamson Subject: Re: filter project:2 > than on the arp. i'm injecting on the moog because its own oscillators > that, since all of these are analog oscillators (even if digitally > controlled), the sawtooth will be standard. please tell me... what > difference could there be? symmetry, i suppose. i can imagine a sawtooth > with a glitch at the turnover point. i'll have to check that. as for noise, > you're absolutely right. different instruments use different methods to > generate noise, which will probably have different probability distributions. > I guess I was thinking in terms that the ramp of the sawtooth may exhibit different extremes of non-linearity depending on circuit design (assuming capacitor bridges are use to ramp up the voltage w/ some type of transistor switch at the turnover point). By the way, sawtooths /teeth? of the same frequency could vary somewhat. To illustrate: x x x x x x x x x x x x x x x x x x x x x ------A------ ------B------ ------C------ A ramps up exponentially, while B ramps up linearly, & C ramps up logarithmically. Give me while, and I could come with frequency spectrum for these different sawteeth (plural?). -- Brian Adamson NRL Code 5523 adamson@itd.nrl.navy.mil ------------------------------ Date: Sat, 26 Oct 1991 12:14:11 MDT From: Adam Schabtach Subject: Re: filter project:1 > yes, in all cases except the moog i will be using the onboard oscillators. > as to the sawteeth, i'm running on the assumption > that, since all of these are analog oscillators (even if digitally > controlled), the sawtooth will be standard. please tell me... what > difference could there be? symmetry, i suppose. i can imagine a sawtooth > with a glitch at the turnover point. Well, back when I had an ARP Odyssey, I spent some time looking at its signal at various places with a 'scope. The sawtooth wave was definitely not shaped like a nice, mathematically-perfect sawtooth function. Hence I would expect wave shapes to vary from manufacturer to manufacturer, and even model to model (e.g. the oscillators in a moog system 55 are probably different than those in a minimoog). In other words, I don't think your assumption that the sawtooth will be standard is valid. I'd either inject the signals from an external oscillator (preferrably a lab signal generator of some sort), or spend some time looking at the output of the oscillators before they enter the filter. --Adam ------------------------------ Date: Sun, 27 Oct 1991 11:55:03 EST From: ronin Subject: Re: filter project:1 i've changed the subject line back to filter project:1 so that i can keep straight which original posting this is a thread of. i'll update the numbers myself as the project progresses. as for the sawteeth... you know, it actually never occurred to me that the slope of that waveform that is labeled 'sawtooth' on a synth would ever be anything other than (essentially) completely linear. certainly, i have never noticed the slope differences described in any scope tracings. on the other hand, i may simply not have been looking close enough. given this concern, i will take some care to assure that the waveforms used *do* have standard sleep (plural of slope.. ha ha) before i attempt to characterize the action of the low pass on them. a quick guess on the drawings you provided... i'll bet that the spectra of the waveforms have respectively decreasing high-frequency content. -------------< Extremism in the Pursuit of Good Noise is no Vice >------- Eric Harnden (Ronin) or The American University Physics Dept. Washington, D.C ------------------------------ Date: Mon, 28 Oct 1991 08:26:11 EST From: Brian Adamson Subject: Re: filter project:1 > a quick guess on the drawings you provided... i'll bet that the spectra > of the waveforms have respectively decreasing high-frequency content. > My guess at spectral content concurs with yours... I'm going to try and do a quick spectral analysis .... I'll either use Mathmatica or my DSS-1 let's me draw single-cycle waveforms and then it will figure out the harmonic series for the waveform drawn. I guess Mathmatica will be alot quicker but it's fun to use musical instruments for lab experiments, right? :) -- Brian Adamson NRL Code 5523 adamson@itd.nrl.navy.mil ------------------------------ Date: Mon, 28 Oct 1991 12:41:45 EST From: Jon Crystal Subject: Casio VZ10M Does anyone have experience with the Casio VZ10M module? Seems pretty interesti ng from what I've read, and currently an inexpensive way to add some new sounds to a midi setup. Thanks for your thoughts ------------------------------ Date: Mon, 28 Oct 1991 20:50:00 EST From: I've ruined a life <00JJTUCKER@BSUVAX1.BITNET> Subject: question. I am very much on a budget, and am looking for a good (and cheap) synth. Anyone know anything about the ESQ-1? It's in my price range (unfortunately that is about all.) Any information would be useful. I know Ensoniq has the VFX, the SD and probably the SQ, but I haven't been able to find them used for the price that I can afford. I cannot stress enough how little money I have (a student... so sorry...) Jason Tucker. jjtucker@bsuvax1.bitnet ------------------------------ Date: Mon, 28 Oct 1991 22:37:29 LCL From: Norwich.School@DARTMOUTH.EDU Subject: Re: question. I think the ESQ-1 is a great machine. My school bought one for 1200 and it has done well. Depending of course on the price. Good sounds...decent little easy 8 track sequencer ------------------------------ Date: Sat, 26 Oct 1991 16:25:22 -0700 From: Tom Erbe Subject: Soundfile header conversion program I just wrote a little program for the Mac that does Soundfile header conversion. It translates between IRCAM, NeXT .snd, Sound Designer II, AIFF and DSP Designer files. It will open any file as a 16-bit linear soundfile, if you want. I am making no attempt (at least not yet) to support anything other than 16-bit linear. You can also change things in the header like sample rate and number of channels. This is my first program for the Mac, so it might have bugs. If you want a copy, you can get it through ftp to "mills.edu", it is "ccm/SoundHac.hqx". Or I could mail it to you if you don't have ftp. Please send bug reports! Tom Erbe * Technical Director * Center for Contemporary Music * Mills College tom@mills.edu * Mills College, Oakland, CA 94613 * (510) 430-2191 ------------------------------ End of the EMUSIC-L Digest ******************************