EMUSIC-L Digest Volume 64, Issue 11 This issue's topics: I saw and heard VL1 played properly (14 messages) PM references Speaking of waveguides (2 messages) 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, 1 May 1994 18:46:45 +0100 From: Nick Rothwell Subject: Re: I saw and heard VL1 played properly >In my teaching, I have always put forth the >notion that synthesis instruments must be "acoustically viable"; that is, >they obey certain fundamental laws of acoustics (i.e. a loud sound is >not just more amplitude, but brighter spectrum). If synthesized sounds >are acoustically viable, then a composer or performer will be better >able to express musical ideas because their instrument is more expressive. Maybe. But I don't think I'd want to have my synthesisers restricted to such paradigms. If we had Acoustic Viability Police we'd never have got the PPG. >If you have ever wondered why sampling synthesizers never sound quite >right, it is because the technology is inherently "acoustically non- >viable." Transposing sound samples by sample-rate conversion completely >ignores the changes in spectra due to frequency and resonance. BZZT! Straw man argument, on two counts. Firstly, I agree with your feeling about sampling synthesisers (sic.) but I thought we were talking about synthesisers in general, including proper ones. Secondly, some of us might actually like the artificial nature of brute-force transposition; I can think of pieces of music which exploit that. >The >only way to make a sampler work is to have enough memory for three >samples per note (one for low velocities, one for medium, one for high). Nonsense. A sampler works if you think it works. Oh, and please show me where the number "three" is engraved in my ears or auditory system. >The bottom line here is that the VL1 creates a paradigm shift away from >older synthesis techniques, and that naturally has many people confused, >dismissive and nonplused. Oh dear, marketing hype alert. Actually, I agree with you to an extent, but we've seen revolutions before and we'll see them again. The VL1 is reasonably significant IMHO. So, a shrug of the shoulders here in Scotland. My music, and my current instruments, are no worse today than they were pre-VL1. To believe that is to be taken in by the marketing hype. >The VL1 (and the yet to be released VP1) >changes the way we think about making sounds and (more importantly) >making MUSIC. "Unleash your creativity." >It is an instrument that must be learned :-(, and >practiced :-(). I think you'll find that they've been around for a few centuries at least. I'm curious: who's the "we" that you refer to? Sequencer-using sample-synth keyboardists (who also learn and practice, if they wish to be good)? Avant-garde modular synthesiser sound designers? I think you're being a little close-minded here. >You can alter the way your performance affects the instrument, >but I doubt there are many people who understand the technology well >enough to change the instrument model without destroying it (the model, >not the VL1). Very few people can program DX7's either. I fail to see the significance of this. Perhaps some of us want to hear what a destroyed VL1 model sounds like, anyway. >I don't care that the "bassoon" sound is better >on a SY99 or a GSU2200 turbo or a KiwiABCD ultra chrome with remote >control. I want a sound that is expressive on a human level, and this >does that. SO, the VL1 is expressive. Great. I'm actually impressed with what I've heard of it. So? Expressiveness is not the only criterion. My synths can run rings round it in some respects, and some of them are quite old and obsolete. Nick Rothwell CASSIEL Contemporary Music/Dance nick@cassiel.com ------------------------------ Date: Mon, 2 May 1994 09:04:04 -0500 From: Stephen David Beck Subject: Re: I saw and heard VL1 played properly Nick Rothwell responds: > > >In my teaching, I have always put forth the >notion that > synthesis instruments must be "acoustically viable"; that > is, >they obey certain fundamental laws of acoustics > (i.e. a loud sound is >not just more amplitude, but > brighter spectrum). If synthesized sounds >are > acoustically viable, then a composer or performer will > be better >able to express musical ideas because their > instrument is more expressive. > > Maybe. But I don't think I'd want to have my synthesisers > restricted to such paradigms. If we had Acoustic Viability > Police we'd never have got the PPG. My ideas about "acoustic viability" are strictly an aesthetic approach to synthesis design, not synthesizer design. A good synthesizer IMHO will provide synthesis controls that help approach this aesthetic, with the ultimate goal of making interesting sounds. Besides, the AV Police are busy ticketing people for failure to maintain control of their sequencers ;-). > > BZZT! Straw man argument, on two counts. Firstly, I agree > with your feeling about sampling synthesisers (sic.) but > I thought we were talking about synthesisers in general, > including proper ones. Secondly, some of us might actually > like the artificial nature of brute-force transposition; > I can think of pieces of music which exploit that. BZZZZZT! Tin man argument. Since when has endless rambling off the topic been a crime on the internet? > >The bottom line here is that the VL1 creates a paradigm > shift away from >older synthesis techniques, and that > naturally has many people confused, >dismissive and > nonplused. > > Oh dear, marketing hype alert. Actually, I agree with > you to an extent, but we've seen revolutions before and > we'll see them again. The VL1 is reasonably significant > IMHO. So, a shrug of the shoulders here in Scotland. My > music, and my current instruments, are no worse today > than they were pre-VL1. To believe that is to be taken > in by the marketing hype. > Nick, I think you'll find no greater critic of the music instrument industry and it's marketing hype than me. It pisses me off no end the way instrument manufacturers claim new and revolutionary synthesis techniques with every year's newest model synth. When in fact, these guys simply repackage or reinvent basic techniques that are 30+ years old. Everything out there is either a form of additive, subtractive, or distortion synthesis, or sample playback (or combinations thereof). The technology behind the VL1 is based on physical modelling using waveguides. There are a few at Stanford experimenting with real-time synthesis using waveguides (NeXT Music Kit), but the VL1 is the first commercial synth to use this technology. I'm not promoting Yamaha's hype. Do they tell you about waveguides? I don't think so... > >You can alter the way your performance affects the > instrument, >but I doubt there are many people who > understand the technology well >enough to change the > instrument model without destroying it (the model, >not > the VL1). > > Very few people can program DX7's either. I fail to see > the significance of this. Perhaps some of us want to hear > what a destroyed VL1 model sounds like, anyway. > Actually, I'm a little disappointed that users can't get into the model and manipulate some of the parameters. Their just trying to make it easier for everyone. Gee, thanks... > SO, the VL1 is expressive. Great. I'm actually impressed > with what I've heard of it. So? Expressiveness is not > the only criterion. My synths can run rings round it in > some respects, and some of them are quite old and > obsolete. > I must admit that the following question did keep running through my mind while playing the VL1: What is the point of this synth if you can't transform the instrument out of the imitative mode? They're just thoughts... ========================== Stephen David Beck School of Music Louisiana State University Baton Rouge, LA 70803 ------------------------------ Date: Mon, 2 May 1994 21:23:25 +0100 From: Nick Rothwell Subject: Re: I saw and heard VL1 played properly >My ideas about "acoustic viability" are strictly an aesthetic >approach to synthesis design, not synthesizer design. So, you're saying that this aesthetic applies in the abstract but not in the actual design and construction? A good >synthesizer IMHO will provide synthesis controls that help >approach this aesthetic, with the ultimate goal of making interesting >sounds. Besides, the AV Police are busy ticketing people for >failure to maintain control of their sequencers ;-). > >> >> BZZT! Straw man argument, on two counts. Firstly, I agree >> with your feeling about sampling synthesisers (sic.) but >> I thought we were talking about synthesisers in general, >> including proper ones. Secondly, some of us might actually >> like the artificial nature of brute-force transposition; >> I can think of pieces of music which exploit that. > >BZZZZZT! Tin man argument. Since when has endless rambling >off the topic been a crime on the internet? You gotta have a heart... But my point was that your arguments don't hold water here; it's the Douglas Adams dolphin argument again. Your criticisms are not criticisms to some people. >Everything out there is either a form of additive, subtractive, >or distortion synthesis, or sample playback (or combinations thereof). Er, OK; so what makes physical modelling as a new technique any different to any other technique when *it* was new? >I must admit that the following question did keep running through >my mind while playing the VL1: What is the point of this synth if >you can't transform the instrument out of the imitative mode? Well, there's something to be said of it if it's robustly MIDI-controllable, for some folks; but the the fun starts when you start imitating imaginary instruments (IYSWIM). Nick Rothwell CASSIEL Contemporary Music/Dance nick@cassiel.com ------------------------------ Date: Tue, 3 May 1994 08:32:02 EDT From: Keith Maynard - CCC Subject: Re: I saw and heard VL1 played properly >>Everything out there is either a form of additive, subtractive, >>or distortion synthesis, or sample playback (or combinations thereof). Nick Rothwell responded to Stephen Beck thusly: > Er, OK; so what makes physical modelling as a new technique any different > to any other technique when *it* was new? I don't think in this case it is a matter of the novelty as much as the fact they may have made a breakthrough in the approach to synthesis. Matthews and Hiller(sp) at MIT proposed that the initial portion of an instrument's sound was the most important portion in creating the timbre. In the additive, subtractive and distortion techniques, a steady state wave is analyzed then synthesized; thus the ramp up portion of the wave is never recreated. This is why, to date, synthetic sounds have sounded markedly different from the real thing. From what I have heard about the VL-1 they crunch the physical model very quickly to produce the 'actual' response that the real instrument would have undergone; this would make the sound match that of the real instrument more closely from t=0 until the sound decays. I was impressed by the kinds of behaviours that the VL-1 captured like vibrato in a clarinet tone, which I have heard is very difficult to create even in the real thing. I certainly was not impressed by the number of parameters apparently neccessary to produce these sounds. My initial reaction was: ' damn, sounds like I'm better off learning to play the real thing'. Keith Maynard ------------------------------ Date: Tue, 3 May 1994 09:04:40 -0400 From: Mark G Simon Subject: Re: I saw and heard VL1 played properly > > I was impressed by the kinds of behaviours that the VL-1 captured like vibrato > in a clarinet tone, which I have heard is very difficult to create even in the > real thing. Most clarinetists trained in classical music don't use vibrato at all, and are even less pleased with the cheesy LFO vibrato usually applied to "clarinet" patches. I would hope that when the VL-1's "clarinet" plays vibrato it does so in response to embouchure pressure (assuming there are any wind controllers out there that can transmit such data, and if there aren't there should be) or breath pressure. Mark Simon mgs2@cornell.edu ------------------------------ Date: Tue, 3 May 1994 10:38:20 -0500 From: Joe McMahon Subject: Re: I saw and heard VL1 played properly >I certainly was not impressed by the number of parameters apparently neccessary >to produce these sounds. My initial reaction was: ' damn, sounds like I'm >better off learning to play the real thing'. Key point! Key point! As far as I personally am concerned, this is very true for the VL-1. Why am I learning to play a clarinet through a non-clarinet interface, when I'm going to have to put in the same amount of effort to learn to play it? Okay, I can't plug headphones into a clarinet, but ... I want a physical modeling instrument that can tell me what a piano would sound like if it were strung with 20-foot rubber bands. I want to know what a plastic trombone sounds like. I want a quartz guitar with glass strings. I want to hear things that are impossible to contruct in the real world. I mean, jeez, I can always find a violinist or an oboist if I want one. A contra-bassoon made of cast-iron pipe! Fun stuff! Yeah, that's more like it. --- Joe M. -- "When Brian Eno ruled Chicago..." (MST3K) ------------------------------ Date: Tue, 3 May 1994 13:02:00 -0500 From: Stephen David Beck Subject: Re: I saw and heard VL1 played properly Nick Rothwell responded: >>(sbeck) My ideas about "acoustic viability" are strictly >>an aesthetic approach to synthesis design, not synthesizer >>design. > >So, you're saying that this aesthetic applies in the >abstract but not in the actual design and construction? No. The aesthetic applies to the programming of a synthesizer, not the manufacturing of the synth hardware. Of course I would like all manufacturers to provide programming options that allow me to program my synths in this manner. I guess I'm the only one out here who actually DOES program his/her synth ;-) (I can't help but paraphrase Bones McCoy, "Dammit, Jim. I'm a composer, not an electrical engineer!") >>(sbeck)Everything out there is either a form of additive, >>subtractive, or distortion synthesis, or sample playback >>(or combinations thereof). > > Er, OK; so what makes physical modelling as a new > technique any different to any other technique when *it* > was new? As a new technique, physical modelling is no different from any other NEW technique when *it* was new. At some point it will become an **OLD** technique. But as a new SYNTHESIS technique, physical modelling changes the paradigm by which sound is created. In all (I repeat ALL) of the aforementioned techniques (additive, filtering, FM, sampling), the basis for synthesis is a modelling of acoustic spectra. On the other hand, physical modelling is concerned with creating a computer model of a resonant body, not with creating spectra. This is not to say that people working in this field are not concerned with spectra. Only that they are focused on the physical process of making sounds, not just the end product. The VL1 is certainly an exciting product. But I, too, am looking for (and working towards) models of more imaginary instruments. ========================== Stephen David Beck School of Music Louisiana State University Baton Rouge, LA 70803 ------------------------------ Date: Tue, 3 May 1994 23:12:54 -0500 From: "J. Michael Roach" Subject: Re: I saw and heard VL1 played properly > I want a physical modeling instrument that can tell me what a piano would > sound like if it were strung with 20-foot rubber bands. I want to know what > a plastic trombone sounds like. I want a quartz guitar with glass strings. > I want to hear things that are impossible to contruct in the real world. I > mean, jeez, I can always find a violinist or an oboist if I want one. A > contra-bassoon made of cast-iron pipe! Fun stuff! Yeah, that's more like > it. this guy is my hero! I thought I was the only one who felt this way! J. Michael ------------------------------ Date: Wed, 4 May 1994 10:52:41 EDT From: Keith Maynard - CCC Subject: Re: I saw and heard VL1 played properly Joe M. wrote: > I want a physical modeling instrument that can tell me what a piano would > sound like if it were strung with 20-foot rubber bands. I want to know what > a plastic trombone sounds like. I want a quartz guitar with glass strings. > I want to hear things that are impossible to contruct in the real world. I > mean, jeez, I can always find a violinist or an oboist if I want one. A > contra-bassoon made of cast-iron pipe! Fun stuff! Yeah, that's more like > it. Joe, I believe from the interview I heard, they claimed that it was possible to construct 'impossible' instruments. The example they gave was putting a flute's mouth-piece on an oboe's body etc. You introduce a material parameter now which is rather interesting, I wonder if they have a means for specifying material for the various parts of the instrument. In my opinion it would not alter the algorithm they use, they may just have to store something like a Young's modulus for the material and use that in the crunch. I think you could also run into designing unplayable instruments since one could conceivably model a heavy rock to be pushed up a hill :) ( no can do ). By the way with the imagination you displayed in your post, you should definitely insure your speakers before playing with a VL-1 and take care bending those glass strings :) Bye Keith ------------------------------ Date: Wed, 4 May 1994 16:30:37 +0100 From: Nick Rothwell Subject: Re: I saw and heard VL1 played properly >I don't think in this case it is a matter of the novelty as much as the fact >they may have made a breakthrough in the approach to synthesis. FM was a breakthrough as well. I still don't see what makes modelling a "revolutionary" breakthrough. Nick Rothwell CASSIEL Contemporary Music/Dance nick@cassiel.com ------------------------------ Date: Wed, 4 May 1994 11:26:28 -0500 From: Arne Claassen ISE Subject: Re: I saw and heard VL1 played properly > >I don't think in this case it is a matter of the novelty as much as the fact > >they may have made a breakthrough in the approach to synthesis. > > FM was a breakthrough as well. I still don't see what makes modelling a > "revolutionary" breakthrough. I think the point was that previous methods of synthesis that imitated real life (ok, i mean acoustic instruments, nothing fake about synths) performed modulation on some set of oscillators and these were tuned so that they sounded similar to the instrument they tried to copy. This is sort of an empirical approach to sysnthesis (it doesn't work the same, but it sounds the "same"), while the VL-1 and the like try to simulate the way the acoustical instrument creates its sounds and imitates the sound in that way. I don't know if that qualifies as revolutionary, but it is a new class of synthesis. -- Arne F. Claassen ----------------------------------------------------------------------------- <<< From "Netwhore Incantations": Blessed be thy MIDI implementation >>> ------------------------------ Date: Thu, 5 May 1994 09:46:41 -0400 From: Mark G Simon Subject: Re: I saw and heard VL1 played properly > > I want a physical modeling instrument that can tell me what a piano would > > sound like if it were strung with 20-foot rubber bands. I want to know what > > a plastic trombone sounds like. I want a quartz guitar with glass strings. > > I want to hear things that are impossible to contruct in the real world. I > > mean, jeez, I can always find a violinist or an oboist if I want one. A > > contra-bassoon made of cast-iron pipe! Fun stuff! Yeah, that's more like > > it. This is certainly a very admirable goal, and we ought to have a synth that can do this sort of thing. Perhaps when they come out with a VL2 they'll let you hack around with the sounds more than they do now. For all we know there might be a secret "hacker mode" such as supposedly exists on certain Roland samplers, something they don't mention in the manuals, but leave it to the user to discover. There has been much talk about this on the Roland sampler list. I think, though, that the point of an instrument like the VL1 is not so much creating new sounds, or even emulating the sound of acoustical instruments, but rather to emulate the control over sounds that one has with acoustical instruments, the subtle inflections one can bring about with breath pressure, tonguing, embouchure, or bow pressure, bow speed, etc. Music becomes expressive when we apply these things to a sound, and electronic instruments up to now have been inferior to acoustic instruments in applying these subtle inflection in real time (and I think in order to really sound convincing you have to do it in real time. Tweaking controller data after the fact can do wonders for sequences, but it still lacks the instinctive immediacy of an accomplished instrumentalist playing straight off). So *potentially* the VL1 could be a great advancement for electronic instruments. Whether this is a revolutionary advancement or just your ordinary great advancement I'll leave for others to debate. I was impressed by the sentence in the review of the VL1 in Keyboard where it implies that unless you get proficient at the breath controller you may not be able to play the thing at all. If this is so, why the heck did they build it with a piano keyboard? Why didn't they give it a wind instrument interface instead? They could have reduced the cost of the thing had they put it in a module and let you plug in the controller of your choice. They could then market VL1 compatible wind controllers and violin controllers. Why should I have to buy the keyboard when I'm going to use it like a wind instrument? Mark Simon mgs2@cornell.edu ------------------------------ Date: Thu, 5 May 1994 16:14:46 +0200 From: Tommy Anderberg Subject: Re: I saw and heard VL1 played properly Most probably a wind-instrument interface without conventional keyboard would put off 99.9% of all potential buyers before they even got to reading the reviews, even less the manuals. So I guess the answer to Mark's question is "marketing"... Tommy Anderberg ------------------------------ Date: Fri, 6 May 1994 21:13:08 EDT From: Jeff Rona Subject: Re: I saw and heard VL1 played properly Actually, Yamaha has a "VL1 Development System" that runs on Macintosh. It allows full access to the algorithms in the VL1 not available from the machine itself! However, they say that they will *never* release it to the public. I think of the VL1 as sort of Yamha's second coming of the GS1, the first, non-programmable/woodgrain/terribly expensive FM machine of years back. Next, maybe Yamaha might do the "DX1" of the virtual acoustic synth finally leading up to the "DX7" of virtual acoustics - a programmable and inexpensive machine :-). Jeff Rona ------------------------------ Date: Fri, 6 May 1994 17:55:10 BST From: Andy Farnell Subject: Re: PM references On the recent VL1/modelling thread.... Does anyone know of any really good references on PM, (apart from the ones in DePoli and the last couple of CMJ's) particularly those concerned with what I heard called 'waveguides'. Not constrained even to synthesis and emusic but anything relevant to fundamental theories/algorithms connected to vibration simulation. Is there a physics or math list/usenet group I could plug into for ideas. If anyone is connected to such a source I would be grateful if you can forward this. Cheers loads. Andy F. inet: afarnell@bournemouth.ac.uk janet: afarnell@uk.ac.bmth ------------------------------ Date: Mon, 2 May 1994 10:15:45 +1871500 From: "Dr. Kirk Corey" Subject: Speaking of waveguides Can anyone give me a pithy explanation (or appropriate reference to CMJ) of what exactly waveguide synthesis is? I mean, other than the latest from Yamaha. :) A pointer to some do-it-yourself algorithms would be great. Kirk -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Dr. Kirk Corey [insert Electronic Systems Administrator kirk-corey@uiowa.edu disclaimer School of Music here] The University of Iowa "Msdos? MSDOS? I've seen CALCULATORS with a better operating system!" ------------------------------ Date: Mon, 2 May 1994 15:59:06 -0500 From: Stephen David Beck Subject: Re: Speaking of waveguides You can find articles by Perry Cook and Julius Smith discussing physical modelling synthesis through the use of waveguides in: Computer Music Journal; vol 16, #4 and vol 17, #1. There are some Mathmatica files which also help explain them on the ccrma ftp site (ccrma-ftp.stanford.edu). Dig in..... ========================== Stephen David Beck School of Music Louisiana State University Baton Rouge, LA 70803 ------------------------------ End of the EMUSIC-L Digest ******************************