EMUSIC-L Digest Volume 14, Number 1 Today's Topics: Atari Software (2 messages) Environments (3 messages) rec.music.synth? (2 messages) CMU MIDI Toolkit (2 messages) MIDI Manager and the Mac Plus (7 messages) Cassette deck query Intro from a new member Ensoniq VFX - the good, the bad, and the ugly (5 messages) FTP help (2 messages) Editor/librarian for PC? Spectral interpolation synthesis HMSL Music, OOP, and request for help Job opening at Oberlin 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, 1 Mar 90 16:02:09 EET From: Riitala Subject: Re: Atari SysEx Software ?? 27 Feb 90 09:40:50 GMT from On Tue, 27 Feb 90 09:40:50 GMT Tony Jewell said: >Does anyone out there have a SysEx filer for the Atari ??? I do. Want it? (Kinda dumb question, I think?). Email me (directly). >Thanks in advance. My pleasure. I've been helped too. --------------------------------------------------------------------------- Jari Riitala email: mainly to JRIITALA@KONTU.UTU.FI still for a while also T6M-RIIT@MAMMUTTI.UTU.FI snail: Kuusiharjunkatu 20660 Littoinen Finland --------------------------------------------------------------------------- 'We've got nothing to fear but the fear itself' - Neil Peart of Rush (The Weapon) --------------------------------------------------------------------------- ------------------------------ Date: Sat, 3 Mar 90 00:04:00 EST From: AD0834@ALBNYVMS.BITNET Subject: Atari-ST Sequencers Hello all! I am an aspiring electronic musician who would like some advice on purchasing a sequencer. I have a 520STFM and a Roland D-10 and I'd like a sequencer which has notating abilities... Anyone got an ideas? - Andy Dickinson (AD0834@ALBNYVMS.BITNET) ------------------------------ Date: Thu, 1 Mar 90 15:57:15 EST From: Christian Teuscher Subject: Very alternate controllers From a new digest member: ...some questions are burning underneath my nails: I am working on an interactive environment and want to use MIDI to get things going, but I can't use keyboards to control the stuff, what I am looking for are gear like MIDI DANCER and MIDI BEAM, isn't Bob Moog now involved with things like that?? Greetings --Christian Teuscher ------------------------------ Date: Tue, 6 Mar 90 12:26:12 EST From: Christian Teuscher Subject: Remote MIDI controllers Does anybody know something about remote MIDI Controllers? I am experimenting with interactive environments. Christian Teuscher ------------------------------ Date: Tue, 6 Mar 90 16:10:00 EST From: METLAY@PITTVMS.BITNET Subject: Remote MIDI controllers What sort? Keyboards, non-keyboards? Or are you referring to wireless? metlay ------------------------------ Date: Thu, 1 Mar 90 14:37:14 CST From: Jason Keeler Subject: rec.music.synth? 27 Feb 90 13:46:39 EST from Just wondering, what is rec.music.synth and how can I get access to it? Thanx, Jason. ------------------------------ Date: Thu, 1 Mar 90 18:05:00 EST From: METLAY@PITTVMS.BITNET Subject: Re: rec.music.synth? rec.music.synth is a USEnet newsgroup dealing with electronic music and synthesis. It's accessible from any UNIX machine on USEnet with news reading programs like rn or vnews. There are about 600 active newsgroups on the Net as of now, on various subjects. Because .synth isn't moderated, it has a much wider audience and no control over what's posted; as a result, between 60 and 90 percent of postings are "What's a good price for a....?" or "I'm selling my...." or "You useless dweeb, get a life, get a clue, everyone knows that you're a total loser if you don't own a...." The restriction of equipment sales and pricing from EMUSIC-L, and the higher proportion of educated readers, make EMUSIC-L a far superior medium of discussion, and if I had to choose one or the other I'd undoubtedly drop .synth. But there are a few people there who are worth listening to (Like Nick, and Dan, and me....|-> ), and if you want to sell a used synth for far more than it's worth, I highly recommend it. There's also a lot more volume: three dozen or more posts per day. Which is okay, if you use a utility that lets you skip the boring stuff easily. metlay ------------------------------ Date: Thu, 1 Mar 90 18:37:45 EST From: RRH101@PSUVM.PSU.EDU Subject: Re: CMU Midi Toolkit Could someone please send me a copy of the CMU Midi Toolkit also. I FTP'd it from ucsd.edu, but it is compressed with 16 bits and I only have a 12 bit version of COMPRESS. Thanks Much. Randy Hammond RRH101@PSUVM ------------------------------ Date: Fri, 2 Mar 90 08:18:00 EST From: FNELSON@OBERLIN.BITNET Subject: Re: CMU Midi Toolkit CMU toolkit is in the public domain and can be purchased from the Center for Art and Technology, Carnegie-Mellon University, Pittsburgh, PA 15213. My last information was that they were charging $20 for a disk with all source code and a nice manual. Both IBM and Mac versions are available. You must purchase the appropriate C compiler separately. I have used MOXC extensively. I think its great. Gary ------------------------------ Date: Fri, 2 Mar 90 15:57:10 CST From: Eric Huffman Subject: MIDI Manager and the Mac+ I have the found the MIDI Manager to operate correctly on a Mac IIcx but not on a Mac+. Have others observed this? To clarify: On a Mac IIcx the MIDI Manager sample applications do their thing. Squid- Cakes monitors input accurately. HourGlass monitors and generates time correctly. SimpleKey sends notes perfectly. And, PatchBay connects these applications and the Apple MIDI Driver together. On a Mac+ I try to monitor MIDI in by patching the output of the Apple MIDI driver to the input of SquidCakes. Nothing appears in the message list except when I send the same note in succession. And then I occasionally get a time-stamp followed a 0xFF value. Very strange. Then PatchBay sometimes crashes when I switch to it from SquidCakes. I am using MIDI Manager 1.2. Have others had these problems? - Eric Huffman ------------------------------ Date: Fri, 2 Mar 90 18:10:00 EST From: Brad Miller Subject: Re: MIDI Manager and the Mac+ > Date: Fri, 2 Mar 90 15:57:10 CST > From: Eric Huffman > I have the found the MIDI Manager to operate correctly on a Mac IIcx but > not on a Mac+. Have others observed this? I've heard 1.2 is broken on the plus and the IIci; I use version 1.1 successfully on my plus without problems. ------------------------------ Date: Fri, 2 Mar 90 21:14:00 EST From: FNELSON@OBERLIN.BITNET Subject: RE: MIDI Manager and the Mac+ It is a toss up if you hope to test the MIDI Mgr witht the sample programs they include. I rebuilt the C examples as projects in Think C. Even after the obvious translation problems (mostly the comments) the applications didn't work the first time. It is hard to believe that they made the sample\ applications with the source code they sent. I can't comment on your problems (Eric Huffman) since everything worked fine for me after I corrected the code. I'd like to hear more detail about your problems. Gary ------------------------------ Date: Sat, 3 Mar 90 23:43:21 MST From: schabtac@SPOT.COLORADO.EDU Subject: RE: MIDI Manager and the Mac+ Funny, I had no trouble at all recompiling and running the Arpeggiator demo MIDI Manager program, under Think C 4.0. Worked right off, as I recall. This was the first thing I did with it, just to try to get something up and running with it and TC4. --Adam ------------------------------ Date: Sun, 4 Mar 90 10:38:00 EST From: FNELSON@OBERLIN.BITNET Subject: RE: MIDI Manager and the Mac+ After reading Adam's recent message I loaded my Midi Mgr disk and tried to compile the Aepeggiator .c file in Think C 4.0. #include files are missing and the // style comments are objectionable to TC4. If by "worked right off" you mean that you changed all of // comments and located the equivalent #include files then my version also worked right off too. There were some other problem that I can't recall at the moment however, one of the programs reported the MidiMgr version when it started up. The compiled version and the one that I made from source report different versions (or at least the messages are different.) More later, Gary ------------------------------ Date: Sun, 4 Mar 90 11:27:43 MST From: schabtac@SPOT.COLORADO.EDU Subject: RE: MIDI Manager and the Mac+ Well, yeah, Gary, I guess I should've been more explicit about getting MIDI Mgr. going with TC4. I did indeed have to change the // comments and do the other little fixes necessary to get an MPW C program to compile under TC4. What I meant by my message was that there seem to be no obstacles to using the MIDI Mgr. with TC4 -- I thought that you were having trouble getting MIDI Mgr. to do *anything*. Sorry about the confusion. I think I saw similar weirdness with version numbers in the Simple Keyboard demo program. It doesn't seem to handle opening its connection to the MIDI Mgr. properly -- charming that Apple's demo programs don't work, isn't it? Do you have the example MIDI Mgr. code that Nick so kindly handed out a few months ago? I found it to be a big help; it proved that someone else out here, down in the trenches, had managed to get the thing working. I could send you a copy if you'd like (I assume you don't mind, Nick?). --Adam ------------------------------ Date: Sun, 4 Mar 90 15:28:00 EST From: FNELSON@OBERLIN.BITNET Subject: RE: MIDI Manager and the Mac+ Please do send Nick's examples. I joined the list only a month or so ago so I didn't see the posting. I retraced my steps with Arpeggiator and SquidCakes. In Arpeggiator the version reported is 1.2 for the precompiled application and 1.2?0 for the one that results from compiling the sources under TC4. With Squidcakes I had to use ResEdit to move the resources from the precompiled application to a .rsrc file. This is not difficult but it is not a skill that the teach you at Conservatory. In fact it is something I learned between my first failure with SquidCakes and my success today. There are no problems with MidiMgr and TC4 that I can see. It's mostly the code translation form MPW to Think. Gary Reading the Pascal examples is also quite useful. If you don't already know how to make a keyboard appear on the Mac screen and scan it for key numbers, SimpleKey is a very clear example. ------------------------------ Date: Sun, 4 Mar 90 12:19:00 MST From: DEYOUNG@ARIZMIS.BITNET Subject: cassette decks I am looking for information on cassette decks, with an eye towards purchasing a new one... what would people out there reccommend? If anyone has used them, how are the Nak CR2A? The CR3A? Yamahas? Denons? Are there others that you would particularly reccommend? -Thanks, Chris ------------------------------ Date: Fri, 2 Mar 90 07:45:25 -0800 From: PMY@VIRGINIA.BITNET Subject: Howja do, I'm... ...Pete Yadlowsky, as you can probably tell from the mail header, and a newcomer to emusic-l. I'm a computer systems engineer with the Academic Computing Center at the University of Virginia, and my specialty machines are Suns, NeXTs and unix boxes in general. On the musical side of things, I serve UVa's music dept as programmer/engineer to the computer&electronic music studio. It's my job to install new stuff, maintain old stuff, do sound engineering and...(most fun of all)...work with composers and students to help them realize compositions and performances using the many neat toys we have lying around here, and innovative and experimental software to drive them. We make fairly heavy use of the Hierarchical Music Specification Language (HMSL), which is a programming language for developing interactive human-machine environments for computer-aided composition and performance. Also, I've been writing alot of NeXT code lately, for software synthesis and audio processing...taking advantage of that machine's wonderful audio capabilities. By the way, the NeXT stuff is free for the taking via ftp or email, for you NeXTers out there. Finally, I'm something of a budding composer m'self, though I haven't produced anything yet. I feel I'm presently in a development stage; contemplating music and the nature of sound, and building software tools that help me work out ideas that I have. I've had some classical training in piano, flute, cello, and a little theory, but now find myself strongly drawn to the ideas and philosophy of John Cage. Well, enough about me... My thanks to those who operate this mailing list. See y'all. - Pete ------------------------------ Date: Thu, 1 Mar 90 16:58:36 GMT From: Nick Rothwell Subject: May Ensoniq rot in Hell I spent a good length of time last night playing with a VFX. I'm getting to like the machine. A lot. I was rather concerned that the filtering is digital and doesn't have resonance, but this doesn't seem to be too bad in places; the resonance wavetables do a pretty good job of getting the effect of true resonance, and the wave layering and modulation facilities are good enough to get plenty of oomph and movement into the sounds. If anything, the bare wavetables in the VFX have more body in them than those in the PPG/Waldorf, although I'll be checking up on that at the weekend. The VFX's wavetable modulation is what makes it special; without that, it would just be another M1/K1 clone. For your interest, that chipmunk (thanks, Metlay...) voice sound from the Tangerine Dream Kiew Mission track is trivial to do on a VFX, if you use a bit of common sense. It took me about 20 minutes to come up with something practically identical. The user interface is not bad at all; plenty of buttons and a nice cut-and-paste facility between edit pages and voices. I'm extremely impressed with performance architecture. The voice-stealing is very well done, and you can (for example) sustain notes and then change entire performance setups underneath them without interfering with the notes which are sounding. The performance controls are very good; you can program the left-hand patch buttons to bring in different sets of voices, and this gives a very nice touch indeed. So: I like the machine. It can do all sorts of things which I can make use of. It can make the kinds of sounds I like. I can get hold of one, without any problem. I can just about afford it. So, I'm going to buy one, yes? Nope. Well, not just like that. Let me run you through a Keyboard magazine review, at least as far as I can remember some of the details. Last August, Keyboard took delivery of four VFX machines. At this time they were on the market (I know someone who had one two months previously). For a start, they reported problems with the onboard signal processor; clicks and pops coming out of the machine at random intervals. These were early models which engineers from Ensoniq were able to come in and fix. Bully for you if you're Keyboard magazine, not so hot if you're a punter in Edinburgh. During a close analysis of the voice architecture, two of the machines didn't do LFO wavetable modulation properly; one would modulate rather randomly, and another would lose its place in the wave memory, and start playing back random cycles from other samples (this is my interpretation of their interpretation). There were also rather nasty performance problems; playing a glissando of a brass patch along the keyboard would confuse the VFX totally, and notes would follow on after arbitrary delays. Even the "fixed" versions of the OS had problems with playing large chords without flamming. Keyboard's bottom line was that it was a very nice machine indeed, except for these problems. Last night's play with the VFX was rather plagued, as well. My colleague has an Atari ST, so he has no decent software to analyse MIDI SysEx messages and so on. Me, I have Performer and a number of software (and one hardware) MIDI analyser on a Mac, for precisely this kind of thing. He has managed to lock the VFX a number of times in the past; it works fine for a while, and then crashes, perhaps rebooting and recalibrating itself. This problem had gone away; however, we managed to get the VFX into a state where it was listening to four MIDI channels at once, except that it ignored part #1 listening on channel 2. We set part #4 to listen to channel 2 - no problem, that worked. Send program changes, and part #2 responded to them. Terr-IF-ic. Cycle the power - nope, still in a funny state. Eventually we found that a MIDI system reset command would fix it. Perhaps it was caused by my messing about with SysEx transmissions earlier that evening. Back to the Keyboard magazine; there was a quote from Ensoniq that they would be fixing these bugs for the finalised software version, 2.0. Now, anybody involved in software development knows that the word "finalised" doesn't make any sense. If Ensoniq really have "finalised" their software for the VFX, then what about remaining bugs? If they aren't working on the software any more, then the VFX is, to all intents and purposes, obselete as a current product. I have some mail here from Donald Maghrak (who has saved my bacon, or at least a large number of $$$'s). He claims that a friend of his who owns a VFX now has a version 2.0 machine. This is after three trips back to Ensoniq and 11 weeks of downtime. Wanna guess what that will be from the UK? He says that v2.0 still doesn't do bulk MIDI SysEx transfers: these will crash the machine and require a total reset. This makes the VFX about as useful to me as a chocolate firegrate. He suggests that the VFXsd is perhaps the way to go, since then the machine can do its own bulk storage. No thanks; if they can make such a mess of the OS for the bare machine, I'm sure they've managed to make a pretty impressive balls-up of the sequencer and floppy file system routines as well. I was veering towards the VFXsd as a good roadworthy sequencer; however, I was at the Ensoniq roadshow last week, and during a live set, one out of the three VFX's had to have its power cycled, and the other failed to load a song sequence first time. Compared with this, gigging with the Mac is childs' play. I've gone on long enough. I am going to arrange a detailed long demo of a *recent* VFX machine, and go in with my Mac, all the software tools and the MIDI analyser. If I don't have any problems with it, and it seems solid under everything I can throw at it, I might, just might, part with my money. If the SysEx problem is still there, I'll call Ensoniq UK and present them with an ultimatum: fix this problem and I'll consider buying your (otherwise extremely nice) machine. Don't fix it and I'll buy something else. There is nothing more frustrating than to be faced with a very nice synthesiser indeed, programmed by engineers who have as much feel for software quality and reliability as a hedgehog. If Ensoniq go down the drain over this one, I won't be in the least surprised. Nick. ------------------------------ Date: Mon, 5 Mar 90 12:16:41 GMT From: Nick Rothwell Subject: can you love two synthesisers at once? Yup, silly subject line, I know. Still trying to decide on whether I buy a Waldorf or a VFX. The Waldorf can certainly do all those wonderful rich filter-swept iiiieeeaaaaoooouuuuw sounds that the PPG was legendary for; I'm quite willing to believe that it has all the functionality of the PPG (sans WaveTerm of course). One minor flaw: Each of the wavetables contains 64 cycles; 61 of these are the wavetable proper, and the top 3 are things like square/saw/sin, so that you always have access to these regardless of wavetable. I guess this means that the Waldorf can function as a conventional (and very nice) analogue synthesiser if you want, or you can tie one DCO per voice to a conventional wave. Beware: it's possible to sweep through the wavetable and off the edge, and find yourself playing a sawtooth instead. This is probably no great problem for predictable modulation routings, but if you're using key velocity to gate a random LFO shape or something, you might have to watch out. I don't know whether the Waldorf can do a great deal more than the legendary PPG sounds. It should, in theory, be pretty much a superset of (say) my MKS-70 and hence capable of everything that it is. Of course, such rationale doesn't hold for synthesis. Now, the VFX. Digital filtering, no resonance, hence no wonderful filter sweeps (although you can sweep the resonance wavetables to get similar effects). But, a much more flexible voice and performance architecture (for example, upto six different wavetables per note, switched by the patch buttons), and much nicer user interface. The problem is Ensoniq, of course, who have a dreadful reputation for reliability and software quality. (there's a discussion raging at the moment over in rec.music.synth about this. I'm saving the articles; give me a shout if you want to see them.) I'm getting a chance to spend some time in the comfort of my own home with a v2.0 VFX connected to the Mac, so I'll get a chance to see how it fares. I'd probably choose a reliable VFX over the Waldorf, even though the latter can do things which the former can't, for reasons of overall versatility. But, there's no such thing as a reliable VFX, as far as I can tell. Now, here's the Scottish entry for rumour-of-the-month. (i) The Roland D-70 has an analogue filter section. Nope, I don't believe it either. OK, try this one: (ii) Yamaha's SY-22 synth is a vector synthesiser with a centre-weighted trackball for wave mixing, price 800 quid. Nick. ------------------------------ Date: Tue, 6 Mar 90 20:20:00 EST From: METLAY@PITTVMS.BITNET Subject: Re: can you love two synthesisers at once? I do; they're both Xpanders. |-> metlay PS. The SY-22 was only a prototype as of the NAMM show, and Yam was uncertain as to whether a production model would be made. ------------------------------ Date: Wed, 7 Mar 90 10:21:16 GMT From: Nick Rothwell Subject: Well, maybe Ensoniq aren't all bad... As a follow-up to my long tirade about Ensoniq software quality and so on, here's roughly what happened when a friend of mine, a VFX owner, contacted Ensoniq UK getting his machine upgraded (i.e. working properly: he was considering threatening Ensoniq under the consumer protection act, for selling goods of unmerchantable quality...): Ensoniq:Howdy. Punter: I want to upgrade my machine to v2 software. E: What version do you have now? P: 1.52. E: 1.52? The first machines shipped to the UK were 1.6. Where did you get a 1.52 machine? [FX: sounds of police sirens, SWAT teams breaking down the door, helicopters etc.] P: Sound Control in Edinburgh. E: How do you know it's 1.52? P: I sent it a version ID request over SysEx, and it told me. E: Ah. Press Preset+Master and it'll tell you in the display. VFX: "I am a 1.7 machine." [Aside: Tiddly-pom... Popping the lid of the VFX revealed a ROM marked 1.7 as well. If a VFX tells you its version number via SysEx, don't believe it. Btw, there are other diagnostic routines available through similar ALT-meta-cokebottle key combinations.] P: So, can I have an upgrade, please mister? E: If you've got a v1 machine, I'm not surprised you want it upgraded, to be honest... We can ship the upgrade kit to your local dealer. P: Why not send me the ROM? E: It's more than that: a v2 machine has a new chip-set, with a new DSP chip and better D/A conversion, so it has to be installed by a qualified electrical engineer. P: I am a qualified electrical engineer... E: In that case we'll get the upgrade kit off to you in the mail, no problem. [End credits] So, there you have it. I have a v2 machine on order, and will be hitting it hard over a brief trial period before laying out the clams. Ensoniq are apparently beta-testing v2.4 of the software (not on customers, I hope...). I'm sort-of willing to give Ensoniq the benefit of the doubt over this one; v1 of the VFX should never have got out of the door, and any benefits Ensoniq might have gotten by releasing it early are almost certainly overshadowed by the bad reputation of all the faulty v1 machines floating around out there. The v2 machines appear to be a lot more solid, although I shall of course let you know in more detail later. Nick. ------------------------------ Date: Wed, 7 Mar 90 09:14:00 EST From: MCINTYRE@MSUPA.BITNET Subject: Time to stock up The State News (our student-run paper) has been running the following "Misc. for Sale" ad this week: Inventory Clearance Sale NOW IN PROGRESS Ensoniq VFX synthesizers Reg. $19.95 now $15.95 at Marshall Music in Frandor. Call 337-9700 If anyone wants to help them clear their inventory, the area code is 517 8-) John McIntyre Physics - Astronomy Dept Michigan State University ------------------------------ Date: Sun, 4 Mar 90 15:44:41 EST From: FNELSON@OBERLIN Subject: ftp I have been using ftp to download public domain software and synth patches from various sites around the US. Several people at Oberlin have asked for help on this and it occurred to me that people on emusic-l might be interested too. If this has already been discussed, I apologize. I been a member only for a month or so. I note that "Nick" is one of the contributors to the library that I probe in this example. Here is the transcript of my ftp session with ucsd.edu. You learn to logon to a remote ftp site as an anonymous (guest) user, to serach the public directories, and to "get" a file. The goal of this sample session is to get a DX7 bank from ucsd's public library. NOTE: Universities and research sites provide ftp access as a service to the community. Rules vary but most sites prefer that you connect during off peak hours. Stay away from 9-6 (their time, not yours!) /* <- I have enclosed my annotations with C language style delimiters -> */ ======================== begin transcript ========================= $ftp ucsd.edu /* identify the site */ ocvaxa.cc.oberlin.edu Wollongong FTP User Process (Version 5.0) Connection Opened Using 8-bit bytes. 220 ucsd.edu FTP server (SunOS 4.0) ready. Name (ucsd.edu:fnelson): anonymous /* your guest ID */ 331 Guest login ok, send ident as password. Password: /* anything can be typed here */ 230 Guest login ok, access restrictions apply. /* only certain actions permitted */ *dir /* list the main directory */ 200 PORT command successful. 150 ASCII data connection for /bin/ls (132.162.32.243,1487) (0 bytes). total 23 drwxrwxrwx 5 ftp ftp 512 Feb 24 07:47 amiga drwxr-xr-x 2 0 1 512 Jan 5 23:28 anime dr-xr-xr-x 2 3 500 512 Sep 12 17:05 bin dr-xr-xr-x 2 3 500 512 Jan 29 1988 etc drwxrwxrwt 3 102 500 2560 Feb 19 12:34 graphics drwxrwxrwt 8 102 500 1024 Feb 23 07:54 hamradio drwxr-xr-x 2 3 500 512 May 22 1989 hosts drwxr-xr-x 5 102 80 512 Dec 5 20:37 images drwxrwxrwt 2 3 500 2560 Feb 7 13:50 mailarchives drwxrwxrwt 6 102 500 512 Jan 13 21:34 midi drwxrwxrwt 7 3 500 1536 Feb 23 09:47 pub drwxrwxrwt 2 0 1 2560 Jan 10 21:03 sparcaudio dr-xr-xr-x 2 3 500 1024 Jan 5 21:17 sun-fixes drwxrwxrwt 2 0 1 512 Feb 20 00:52 tmp drwxrwxrwt 2 0 1 512 Dec 13 20:22 u3b2 drwxr-xr-x 7 6 6 512 Feb 25 10:17 usenet 226 ASCII Transfer complete. 983 bytes transfered in 8.38 seconds (0.11 Kbytes/second) /* The DX7 voices are probably in the midi directory */ *cd midi /* change directory to midi */ 250 CWD command successful. *dir /* list the midi directory */ 200 PORT command successful. 150 ASCII data connection for /bin/ls (132.162.32.243,1488) (0 bytes). 226 ASCII Transfer complete. total 5 -r--r--r-- 1 102 500 732 Jan 16 1989 README drwxrwxrwt 3 102 500 1024 Jan 17 17:19 doc drwxrwxrwt 8 102 500 512 Oct 2 23:31 patches drwxrwxrwt 5 102 500 512 Jan 17 19:05 scores drwxrwxrwt 8 3 500 512 Dec 3 16:52 software 313 bytes transfered in 2.67 seconds (0.11 Kbytes/second) *cd patches /* change directory to patches */ 250 CWD command successful. *dir /* list patch directory */ 200 PORT command successful. 150 ASCII data connection for /bin/ls (132.162.32.243,1489) (0 bytes). 226 ASCII Transfer complete. total 11 drwxrwxrwt 2 102 500 5632 Aug 9 1989 DX7 drwxrwxrwt 2 102 500 512 May 12 1989 K1 drwxrwxr-x 2 102 500 512 May 12 1989 K5 drwxrwxrwt 2 102 500 512 May 12 1989 M1 drwxrwxrwx 2 ftp ftp 512 Oct 2 23:31 V50 drwxrwxrwt 2 102 500 512 May 13 1989 casio 356 bytes transfered in 3.03 seconds (0.11 Kbytes/second) *cd DX7 /* change to DX7 directory */ 250 CWD command successful. *dir READ* /* list all the filenames that begin with READ */ 200 PORT command successful. 150 ASCII data connection for /bin/ls (132.162.32.243,1493) (0 bytes). 226 ASCII Transfer complete. -r--r--r-- 1 102 500 2146 Feb 19 1989 README.1ST -r--r--r-- 1 102 500 966 Feb 14 1989 README.32 -r--r--r-- 1 102 500 1413 Feb 14 1989 README.AGray -r--r--r-- 1 102 500 1819 May 9 1989 README.bank -r--r--r-- 1 102 500 162 Feb 14 1989 README.dx7 327 bytes transfered in 2.78 seconds (0.11 Kbytes/second) /* These README files describe the contents of several collections of DX7 voice banks */ *get README.dx7 /* copy file from ucsd to local mainframe */ 200 PORT command successful. 150 ASCII data connection for README.dx7 (132.162.32.243,1496) (162 bytes). 226 ASCII Transfer complete. 162 bytes transfered in 0.13 seconds (1.20 Kbytes/second) /* At this point you may want to return to your local mainframe to look at README.dx7 to get more information before you go thru the effort of downloading files that may not be what you want */ *dir b*.dx7 /* list all the files that begin with b and end with .dx7. */ 200 PORT command successful. 150 ASCII data connection for /bin/ls (132.162.32.243,1495) (0 bytes). 226 ASCII Transfer complete. -r--r--r-- 1 102 500 4097 Feb 14 1989 bank01.dx7 -r--r--r-- 1 102 500 4097 Feb 14 1989 bank05.dx7 -r--r--r-- 1 102 500 4097 Feb 14 1989 bank08.dx7 -r--r--r-- 1 102 500 4097 Feb 14 1989 bank11.dx7 -r--r--r-- 1 102 500 4097 Feb 14 1989 bank14.dx7 -r--r--r-- 1 102 500 4097 Feb 14 1989 bank18.dx7 -r--r--r-- 1 102 500 4096 Feb 14 1989 brass2.dx7 -r--r--r-- 1 102 500 4096 Feb 14 1989 brass3.dx7 520 bytes transfered in 4.42 seconds (0.11 Kbytes/second) *get bank18.dx7 /* copy a file from ucsd to local mainframe */ 200 PORT command successful. 150 ASCII data connection for bank18.dx7 (132.162.32.243,1496) (4097 bytes). 226 ASCII Transfer complete. 4048 bytes transfered in 3.29 seconds (1.20 Kbytes/second) *quit /* exit ftp and log off ucsd.edu */ 221 Goodbye. ======================== end transcript ========================= Gary Lee Nelson TIMARA Program Conservatory of Music Oberlin, OH 44074 ------------------------------ Date: Mon, 5 Mar 90 13:42:02 EST From: "ISA-HRDD Robert N. Muth" Subject: Re: ftp >I have been using ftp to download public domain software and synth patches >from various sites around the US. Where else?????!!!! (Besides ucsd.edu) Thanks. \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ | | | | | Robert N. Muth MILNET: rmuth@cor1.pica.mil | | Student WhateveritsnameisNET: u91_rmuth@VAXC.STEVENS-TECH.EDU | | | | "If money is the root of all evil, then I should be the pope" | | | | | \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ ------------------------------ Date: Wed, 7 Mar 90 06:06:00 EST From: FNELSON@OBERLIN.BITNET Subject: Re: ftp FTP SUMEX-AIM.STANFORD.EDU for PD mac software. Not much music however. Gary ------------------------------ Date: Sun, 4 Mar 90 21:28:06 EST From: RRH101@PSUVM.PSU.EDU Subject: Editor/Librarian for PC Could someone please tell me where I can find a PD version of an editor/librarian for the PC? I have a Yamaha DX7IID and would really like to dump patches to it. Thanks Much. Randy Hammond RRH101@PSUVM.bitnet ------------------------------ Date: Mon, 5 Mar 90 10:42:16 EST From: Roger.Dannenberg@SPICE.CS.CMU.EDU Subject: Spectral Interpolation Synthesis Another very interesting approach, courtesy of one of our digest readers... ------------------------------------------------------------------------------- Several posts about new synthesis methods and raves about "vector synthesis" have encouraged me to contribute some information about research at Carnegie Mellon and elsewhere. The idea of interpolating between waveforms is an old one. Matsushita (Panasonic & Technics) holds a patent on a technique whereby interpolation between a succession of waveforms is used to create spectral evolution. I have an unpublished report from 1979 describing a very similar approach and this approach was reportedly tried at IRCAM and well-known there around the same time if not earlier. Switching between waveforms (0th-order interpolation) has also been described and reinvented a few times. The first mention I know of is Covitz and Ashcraft in 1981. For the sake of discussion, I'll divide the approaches into two camps: (1) those who say "let's take a few fixed waveforms and see how we can use interpolation to generate interesting, evolving new ones", and (2) those who say "let's assume an arbitrary sequence of waveforms and interpolate from one to the next". In either case, selection of waveforms is the interesting problem. Case (1) is "vector synthesis". We have found, in studies based on analysis and synthesis of orchestral instruments, that a small set of waveforms is not really adequate to model what goes on in acoustic instruments, but if you are willing to pay for a large set, then you might as well choose a set of sinusoids as the basis and use additive synthesis. We have therefore focussed on Case (2), where we are always interpolating between only 2 waveforms, but both are arbitrary. We have implemented software that automatically analyzes digitized tones and produces a sequence of waveforms to be interpolated. We call this "Spectral Interpolation Synthesis" rather than "Waveform Interpolation Synthesis" to emphasize that we [can] store the waveforms as spectra and that we keep everything in phase so that interpolation between waveforms is equivalent to interpolation in the spectral domain. The results are very promising. Our synthesized sounds are almost indistinguishable from the originals (most people learn to tell the difference by listening for noise and other details but these are not apparent unless you are very critical). These are synthsized, so it is easy to perform manipulations such as stretching and transposition. Storage is several hundred bytes per second, or 5 to 20 spectra per second. We are working on techniques for characterizing instruments as functions from control parameters (pitch, loudness, "bending notes", bow position, etc.) into spectra so that we can generate spectra from control information rather than by analyzing individual tones. A report on this work should appear in JAES soon. I'd like to make one other point. I've seen the term "resynthesis" coming up quite often as if it will save the world from the stagnating sounds of samplers. The analysis of sounds into sinusoidal partials with slowly changing parameters and the use of additive synthesis to reconstruct sounds has been going on in the computer music community since the 60's. It has not been very apparent from the literature, but this approach does not work well when signal parameters change very rapidly as in attack portions of musical tones. We have successfully combined sampled attacks with spectral interpolation synthesis to avoid these limitations, but that's another story and I've gone on long enough. -Roger ------------------------------ Date: Mon, 5 Mar 90 12:31:59 -0800 From: PMY@VIRGINIA.BITNET Subject: HMSL In a previous post I wrote: >>We make fairly heavy use of >>the Hierarchical Music Specification Language (HMSL), which >>is a programming language for developing interactive human-machine >>environments for computer-aided composition and performance. And an interested reader wrote privately to ask: >How do I find out more about this (HMSL)? I think this is probably something that will be of general interest to the group, so I'll reply here. HMSL, as I said, is a programming language (rather, programming environment) that was originally written for the Commodore Amiga and also runs on Macs of most flavors. Actually, going back a bit further, HMSL had its very beginnings on a home-brew S100 (remember S100?) box. Anyway, HMSL is itself built on another popular language, Forth...not just any Forth, but object-oriented Forth. In the Amiga's case, this means JForth (the World's Greatest Forth); in the Mac's case, a pseudo-JForth layer applied to MachII Forth. So, before one can begin using HMSL, one must first have a good working knowledge of Forth. Then, one must become acquainted with object-oriented programming. Yes, there's a fair amount of learning to be done before diving into HMSL, but power and flexibility always have their price. Briefly, object-oriented programming is a programming philosophy which has the programmer constructing intelligent data structures called "objects". That is, these structures contain not only space for holding data, but "methods" which operate on these data as well. The exact nature of the data and methods are private to the object, and can only be accessed or called by sending "messages" to the object. A given message, sent to two or more objects of different classes, will have different effects because each object class will interpret that message in a different way, appropriate to itself. This greatly facilitates practices such as procedure abstraction and data hiding, which are Very Good Things. But, back to HMSL... HMSL provides a number of different object classes to work with during the composition and/or performance of a piece of music. Actually, HMSL is general enough that it need not be limited to music at all...I've heard of its being used to drive kinetic sculpture and to control lighting. Anway, two of the object classes provided are the "instrument" and its sub-class, the "midi-instrument". The instrument object is what actually talks to whatever hardware you may wish to control from your computer. Moving up a level, we come to the "player" object class. The player has the task of reading "shape" objects (analogous to musical scores), and sending events coded therein to the player's assigned instrument object, at the right times. Moving up again, we come to the "collection" object class and its subclasses. This is where the hierarchy comes in. Collections may contain players, jobs (lists of Forth routines executed at regular time intervals), productions (lists of Forth routines executed once, immediately), or...other collections. A collection's constituents may be executed sequentially, in parallel (as an ensemble) or according to some random-access algorithm. Furthermore, collections may be nested to virtually any depth, any combination of sequential/parallel behaviour. For real-time input and stimulus, HMSL provides an object class called the "action". An action simply sits and waits for a particular stimulus. When/if received, the action collects stimulus data (if any) and effects its programmed response, which may be most anything, including execution of another HMSL object. Finally, and very importantly, HMSL, as a language, is extensible. This is a trait it inherits from its mother language, Forth, and from its object-oriented nature. If there's a language feature you'd like to have, make it. If you like, install it as a permanent feature of the system. If there's an object class not in the system that would be useful to your work, create it. Likewise, install it permanently. Customize. All the source code is written in Forth (except perhaps for some machine- specific stuff) and is provided for your hacking pleasure. HMSL is not a slick-but-closed application, like a sequencer, nor is it used to produce such applications, like a C compiler (though under Amiga's JForth it can be made to do so). HMSL is an interactive, open-ended programming environment for those who wish to involve their computers actively and intelligently in the processes of composition and performance. HMSL was developed at the Center for Contemporary Music at Mills College in Oakland, CA. Last I checked, HMSL was going for $150 plus a Forth for your Mac or Amiga. For more info write: Frog Peak Music Box 9911 Oakland, CA 94613 or you might try e-mailing to Jeanne Parson: jeanne@mills.berkeley.edu Also, there's an HMSL BBS: 415-928-8246 Disclaimer: I'm an HMSL/JForth developer and an Amiga fan, but I receive no inco or other material benefits from sales of HMSL, Amigas or JForth. - Pete ------------------------------ Date: Tue, 6 Mar 90 12:27:00 N From: MUSIC01@UNIPAD.INFNET Subject: Music, OOP, and Help Request. Hello, you all. My name is Andrea Provaglio, and I'm talking to you from Venezia, Italy. I have a communication for you and a help request, but first let me tell you some historical fact.... I am involved in the field of musical application of computer technology, and, as primary activity, I cooperate since 1985 with an university research center in Padova (Centro di Sonologia Computazionale), which is the most important in Italy. We work on mainframe (IBM 3090, VAX 8650) systems as well as on mini and micro systems (PDP 11, AT, Atari, Apple) and a large part of our software has been developed here. We perform researches in the acoustic and musical field, and support musicians for the realization of their works. In the past years I've been involved in speech synthesis on the IBM mainframe, developing a software synthesizer based on the D.J. Klatt cascade/parallel formant model; I developed software tools for the digital sound processing such as transposers, mixers, tools for spatial simulation and so on; I cooperated with many composers (James Dashow, Richard Karpen from USA) aiding then in obtaining from the computers what they want; I have been the coordinator of research and software developement for our "Real Time" project using the G. Di Giugno 4I and 4X processors and I did some other minor activity. Actually I am responsible of the Object Oriented Musical Software developement on PC stations at our centre, I cooperate with the Italian CNR (Consiglio Nazionale delle Ricerche), departement of Phonetics, in a ESPRIT project (a European research project with EEC funds) for the comparate analysis of the main Westren Europe languges; I work with a commercial company developing part of the support sofware for a digital audio board that they produce for IBM PC. And, for the past's all. Now the future.... Another important computer music centre in Italy ("Tempo Reale" in Firenze) has the project to build a set of software libraries of "musical objects" joining the effort of the main italian computer music centres as well as of individual contributions. The aim is to have efficient, modular, easy- maintenible and easy-transportable musical software. Since the Object Oriented Programming seems to be the best way, the language adopted is C++. The SW developed should run on the following systems: - MsDos - MacIntosh - Atari TOS - UNIX System V - NeXT and a initial division of the libraries is: - General purpose music oriented objects. - Graphic objects for musical applications. - Drivers for MIDI and AES/EBU protocols. - Drivers for DSP and AD/DA add-on boards. This is the communication: For this project a user group will be created. The group is open. Any contribution will be welcome. ALL the software developed by the users will be shareware (probably with source). The project is about to start. If someone is interested or would like to have more specific information, don't hesitate to contact me. This is the help request: I am planning to develop objects for the handling of sound files in the following formats: - Raw - IRCAM format - NeXT sound file format - MIDI Sample Dump format - Composer's Desktop Project format. and to design drivers for AD/DA converters add-on cards for IBM PC and other personal computers. PLEASE, any direct or indirect information (such as references), suggestions, experiences, ideas the could help me will be greatly appreciated, especially about the AD/DA boards (tell me what you use, if it works, what there's on the market AND HOW TO CONTACT THE COMPANIES THAT PRODUCE THE BOARDS to obtain technical reference of the product) Remember that, if you'll join this group, all the software developed will be available. Thanks a lot for your time and excuse my uncertain English. Regards, Andrea Provaglio How to contact me... E-Mail Address : MUSIC01@UNIPAD.CINECA.IT CompuServe : 73520,3205 BIX : 'aprovaglio' FAX : +39 - 49 - 875-4094 Please contact me via private mail whenever possible. ------------------------------ Date: Wed, 7 Mar 90 21:01:00 EST From: FNELSON@OBERLIN.BITNET Subject: job posting Faculty Vacancy in Electronic and Computer Music The Conservatory of Music at Oberlin College invites applications for a non-continuing full-time position for the first semester 90-91. Teach courses in analog and digital sound synthesis. Doctorate in composition or computer music preferred but consideration will be given to individuals with doctorates in other fields if they can demonstrate substantial achievement in composition or publication. For more information write to: David Boe, Dean Conservatory of Music Oberlin, OH 44074 (216) 775-8200 Application deadline: April 17 Please re-post to any BB's that you think may reach qualified applicants. ---------------------- End of EMUSIC-L Digest **********************