issue08

EMUSIC-L Digest                                      Volume 56, Issue 08

This issue's topics:
	
	(fwd) FM feedback (2 messages)
	FM Synth Software with Oberon Source for Amiga (5 messages)
	the fm simulator (8 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:         Mon, 27 Sep 1993 08:37:00 EDT
From:         Jeff Harrington 
Subject:      (fwd) FM feedback

Found this on comp.music - since we've been discussing his efforts I've
invited him to join in our discussion.  Does anyone know the answer to his
query?

>From: stiens@math.uni-muenster.de (Christian Stiens)
>Newsgroups: comp.music
>Subject: FM feedback
>Date: 27 Sep 1993 10:43:35 GMT
Hi!

I wrote a program for the amiga which calculates samples with the
FM synthesis. But I don't know how to implement the feedback correctly.
If I have op6 -> op1 with feedback lets say 4 , then the prog calcs
it as op6 -> op6 -> op6 -> op6 -> op6 -> op1. BTW '->' means 'modulates'.
Is that how the DXs and SYnths do it?

Bye, Chris



------------------------------
Date:         Mon, 27 Sep 1993 10:29:12 EDT
From:         ronin 
Subject:      Re: (fwd) FM feedback

retain the last output of the feeding-back operator in a register and
use it as the modulation source for the next sample iteration.
this is op6(t-1) -> op6(t) -> op1(t), for the simplest configuration.
with sufficiently high sample rates (say, 1024 points per cycle) the
phase offset introduced by the delay in feedback is negligible.
i have no idea if that's how yamaha did it, but it's the only way i could
think of to do it in a serially structured program.

-----------< Cognitive Dissonance is a 20th Century Art Form >-----------
Eric Harnden (Ronin)
 or 
The American University Physics Dept.
4400 Mass. Ave. NW, Washington, DC, 20016-8058
(202) 885-2748  (with Voice Mail)
---------------------< Join the Cognitive Dissidents >-------------------

------------------------------
Date:         Fri, 10 Sep 1993 12:13:43 -0400
From:         idealord 
Subject:      FM Synth Software with Oberon Source for Amiga

Fish Disk (if you have an Amiga you're probably familiar with the Fred Fish
series of PD software disks) number 895 has an FM synthesis program closely
modelled after the Yamaha 6 op synthesizers.  I just ftp'ed it from
wuarchive.wustl.edu in the /pub/aminet/fish/ff895 directory if you're
interested.

Complete source in Oberon included.  There's a demo Oberon compiler available
for the Amiga, too.

Jeff Harrington
idealord@dorsai.dorsai.org

------------------------------
Date:         Fri, 10 Sep 1993 16:15:47 -0400
From:         Joe McMahon 
Subject:      Re: FM Synth Software with Oberon Source for Amiga

>Complete source in Oberon included.  There's a demo Oberon compiler available
>for the Amiga, too.

There's a Mac implementation of Oberon, but who knows if the sound support
even vaguely resembles the Amiga's.

 --- Joe M.

------------------------------
Date:         Fri, 10 Sep 1993 19:43:45 -0400
From:         idealord 
Subject:      Re: FM Synth Software with Oberon Source for Amiga

>
> >Complete source in Oberon included.  There's a demo Oberon compiler available
> >for the Amiga, too.
>
> There's a Mac implementation of Oberon, but who knows if the sound support
> even vaguely resembles the Amiga's.
>
>  --- Joe M.
>

The Amiga Sound chip is totally different.  I unpacked the thing and got it
running and it's pretty nice.  It'll even do an FFT on your FM patch.  Saves
samples as either software specific or as 8 bit samples in the Amiga IFF 8SVX
format.    You could easily add an ARexx port to it and do that genetic
algorithm thing we've been talking about for reproducing samples by running
the population generation, etc. as a front end to it.

One question came to mind, though, it only has 10 algorithms - doesn't the DX7
have a few more?  I've never even played a DX7 - just my SY77 :)

Not sure if _anybody_ is interested in this - but since Eric is probably not
going to risk a suit and make his system PD :( - here's a full blown DX7
emulator with source for anyone interested.  I just don't have the time for
this or many other emusic things at the moment.

Jeff Harrington
idealord@dorsai.dorsai.org

------------------------------
Date:         Fri, 24 Sep 1993 16:02:02 EDT
From:         Simon.Weatherill@COAT.COM
Subject:      Re: FM Synth Software with Oberon Source for Amiga

Is Oberon an environment?  Is there a version for a PC?  If it
generates "samples" (raw audio data), maybe it would be fairly easy to
convert it for PC sound cards.

Yes, I'm interested (if it'll work for me :-).

Simon Weatherill

------------------------------
Date:         Sat, 25 Sep 1993 10:14:54 -0400
From:         idealord 
Subject:      Re: FM Synth Software with Oberon Source for Amiga

>
> Is Oberon an environment?  Is there a version for a PC?  If it
> generates "samples" (raw audio data), maybe it would be fairly easy to
> convert it for PC sound cards.
>
> Yes, I'm interested (if it'll work for me :-).
>
> Simon Weatherill
>

Oberon is a language conceieved of as being best used in its own enviroment.
 There
are PD versions of the language (slightly crippled) for all platforms.  The
Amiga code will have to be digested so that you can strip the GUI stuff out
and you'll want to re-write the "save" routines as it does save in samples but
with Amiga headers.  It's going to be a task.  Plus you would probably want to
get rid of the 8 bit routines and keep it 16 bit - do an integer mulitply up -
I think it uses floats at the moment.

The nicest thing about it was that it was concise.

It's in wuarchive.wustl.edu /pub/aminet/fish/ff895 (I think).  It's on a Fred
Fish disk number 895 as FMSynth.lha.  If you can't find the precise directory
try archie or I'll look for it again.  People have had problems getting it
de-archived - you'll probably need a friend with an Amiga to do it... small
lharc format differences.

Jeff Harrington
idealord@dorsai.dorsai.org

------------------------------
Date:         Thu, 2 Sep 1993 14:20:58 EDT
From:         ronin 
Subject:      the fm simulator

the dxtutor is a pascal program i wrote instead of completing my contracted
graduate work, using calibration measurements that we made of our dx7.
ultimately, the goal was to have it interface with psychoacoustically
determined timbre-space specifications, to achieve 'best-fit' approximations
of arbitrary complex spectra. that researcher disappeared, so it never
went any farther. so what i've got is a program that presents you with a
screen to edit the algorithm, envelope, frequency, and level parameters
for a dx7 patch, then calculates the resultant waveform and executes
an fft on that. the waveform and the fft are displayed with a few viewing
options, and the whole shebang can be saved to disk. the idea was to do
a bunch of them offline, then call them up from files in a classroom. now that
we've got 386's and up, it's hardly necessary... it's reasonably quick.
the editing screen has a toggle that allows you to enter level parameters
either in standard dx7 format, or as indices of modulation, which allows you
to examine non-dx indices (the maximum index on the dx, at a level of 99,
is just a little less than 13). ratio mode is the only mode modeled, but
feedback can be entered as an empirical fm parameter 'beta'. envelopes,
while represented onscreen, are not implemented. the thing does not generate
dynamic timbres. and the command parser is the ugliest damn piece of code
i've ever written. but it works.
it has occurred to me that if i let it out i'll get my ass sued off.
i understand that yamaha actually holds a patent on the fm process itself,
and that you can't build an fm instrument compatible with their parameter
set without infringing on it. this came up when a friend of mine who's
building a dsp board told me that, even though i could provide him with
code which would emulate the yamaha chips that are used on other commercial
boards, he couldn't use it because of the patent. if he wanted that sound
on his board, he had to use the yamaha chip. which makes me wonder about
the consequences of my having effectively reverse-engineered the process
for my program. at any rate, i haven't released it anywhere yet, even in the
public domain, so i assume i'm safe for now. hmmm. now i'm unsure again.
any thoughts or prior knowledge out there on this topic?

-----------< Cognitive Dissonance is a 20th Century Art Form >-----------
Eric Harnden (Ronin)
 or 
The American University Physics Dept.
4400 Mass. Ave. NW, Washington, DC, 20016-8058
(202) 885-2748  (with Voice Mail)
---------------------< Join the Cognitive Dissidents >-------------------

------------------------------
Date:         Fri, 3 Sep 1993 09:42:22 -0400
From:         idealord 
Subject:      Re: the fm simulator

Mouth-watering descrip of Eric's dxtutor program deleted =->
> i understand that yamaha actually holds a patent on the fm process itself,
> and that you can't build an fm instrument compatible with their parameter
> set without infringing on it. this came up when a friend of mine who's
> building a dsp board told me that, even though i could provide him with
> code which would emulate the yamaha chips that are used on other commercial
> boards, he couldn't use it because of the patent. if he wanted that sound
> on his board, he had to use the yamaha chip. which makes me wonder about
> the consequences of my having effectively reverse-engineered the process
> for my program. at any rate, i haven't released it anywhere yet, even in the
> public domain, so i assume i'm safe for now. hmmm. now i'm unsure again.
> any thoughts or prior knowledge out there on this topic?

1. Isn't Yamaha's patent going to expire this year?
2. If you're skepical about Yamaha's litigatinous potential you should talk to
Don Buchla :( - He used FM in most of his early instruments and after that
asshole Chowning "patented" (what a nice word) FM (what a laugh as if Bessel
himself never existed) he was forced to never use FM again.  This is why Mort
and Don aren't on speaking terms :) (free gossip!).

Jeff Harrington
idealord@dorsai.dorsai.org

------------------------------
Date:         Sat, 4 Sep 1993 08:32:22 -0400
From:         Andy Farnell 
Subject:      Re: the fm simulator

> it has occurred to me that if i let it out i'll get my ass sued off.
> i understand that yamaha actually holds a patent on the fm process itself,
> and that you can't build an fm instrument compatible with their parameter
> set without infringing on it.
>

This sounds like a load of scrote to me. I am not so sure about American law but
I always had the idea that it was not possible to patent any conceptual,
abstract or genearally defined idea. If anyone has a claim to a patent on
FM synthesis (as a concept) you might think it was Chowning, but then again
it might just as easily be Shannon or any number of scientists. FM synthesis
as an audio production mechanism is a conceptual process, it exists by
virtue of the nauture of mathematics, it was not invented it was discovered
and cannot as such be patented. Yamaha may however have a patent on the VLSI
designs for their chip, or a more general software copyright on some of the
production algorithms (their intellectual property). If you have written your
own code then no one can stop you from distributing it provided that it does
not import any copyright material.
As to whether you may be infringing by producing some output compatible with
a parameter set - this sounds the most frightening and unlikely thing I have
heard so far. If it is true then I think we should all take a good look at
ourselves and ask just what the hell is going on. Of course no one can
really stop you publishing just what the hell you like anyway, it doen't have
to be tracable to you. Anyone with a good idea can share it, unless you want
to make money from the idea of course.

Andy Farnell
Computing and Cognition
Bournemouth University
England

------------------------------
Date:         Sat, 4 Sep 1993 09:28:49 -0400
From:         idealord 
Subject:      Re: the fm simulator

A couple of comments on Andy Farnell's understanding of Yamaha's patent on FM:

1.  Yamaha owns FM thanks to John Chowning.  Stanford University reaps
millions because of Chowning's manipulation of U.S. Patent Law.  Mr. Chowning
receives nothing.

2.  Yamaha has sued or threatened to sue many companies who employ FM-like
capabilities.  Don Buchla's instruments and Robert Moog's instruments can no
longer employ FM or phase distortion algorithms.  If you wish to take Yamaha
on, you'd better get your head examined.  Mr. Chowning has practically
destroyed America's homegrown E-instrument development.  THANKS!  and your
music sucks too, Mr. Chowning!

3. Yamaha's patent is supposed to expire very soon - (this year?) this is why
they're putting most of their efforts into sampling tech.

4. Patented "ideas" or "concepts" is destroying America's inventiveness.
Check out the story of RSA encryption algorithms.  What about patents on
neural net configurations.  Weird... patent lawyers will try anything and then
depend on American jury or judicial technological incompetence to win in
court.

Sigh.... I hate Yamaha but I love my SY77!

Jeff Harrington
idealord@dorsai.dorsai.org

------------------------------
Date:         Mon, 6 Sep 1993 14:42:00 EDT
From:         John Rossi III 
Subject:      Re: the fm simulator

Chowning had nothing to do with the failure of the American homegrown
E-instrument development.  The companies that failed (from Moog trhough ARP
through Oberheim through New England Digital) did so either because of
idiotic management or failure to come to grips with changing technology.
As for the whole picture, Kurzweil, Ensoniq, Alesis, E-mu and Peavey have
been doing a hell of a job in the development side.  As a matter of fact,
each of the mentioned companies has done a lot more productive post-FM
R&D than ANY Japaneese company I can mention.

John

------------------------------
Date:         Mon, 6 Sep 1993 19:17:18 +0000
From:         Nick Rothwell 
Subject:      Re: the fm simulator

>Mr. Chowning has practically
>destroyed America's homegrown E-instrument development.  THANKS!

Didn't he approach a number of American instrument designers back around
1973 when FM looked like being a good long-term investment (when hardware
got cheap enough to do it)? They all showed him the door. He went to
Yamaha. Looks to me as if the American companies destroyed themselves at
least partially through short-sightedness and short-termism.


                        Nick Rothwell   |   cassiel@cassiel.demon.co.uk
     CASSIEL Contemporary Music/Dance   |   cassiel@cix.compulink.co.uk

------------------------------
Date:         Tue, 7 Sep 1993 08:49:13 -0400
From:         idealord 
Subject:      Re: the fm simulator

>
> >Mr. Chowning has practically
> >destroyed America's homegrown E-instrument development.  THANKS!
>
> Didn't he approach a number of American instrument designers back around
> 1973 when FM looked like being a good long-term investment (when hardware
> got cheap enough to do it)? They all showed him the door. He went to
> Yamaha. Looks to me as if the American companies destroyed themselves at
> least partially through short-sightedness and short-termism.
>

Well I got this second hand from a good friend of Buchla's - so - whatever -
both Buchla's boxes and Moogs synths used FM - I used to use a big set of moog
modules at LSU - dense amazing analog sounds - NO MORE!

As far as Mr. Rossi's comparison of Buchla's and Moogs efforts to Ensoniq's
and Kurzweil's - no comment - but this - there is a small limited market for
experimental instruments - analog - and that is what Chowning/Yamaha killed.
Yamaha harassed Buchla out of the instrument business and into that wretched
thunder stick (whatever - I've used them and they're nightmarishly
unreliable).  These were not necessarily real big Business enterprises - but
more like custom imagineering.  Why did Yamaha care about them?

Jeff Harrington
idealord@dorsai.dorsai.org

------------------------------
Date:         Tue, 7 Sep 1993 10:27:00 EDT
From:         John Rossi III 
Subject:      Re: the fm simulator

What was it that you got second hand?  What is - NO MORE!?  Actually, from
the composition of that heavily dashed (i.e., '-') string of words, it is
impossible to determine what you are saying.  Is it that there are no more
Moog modules at LSU?  Perhaps you can no longer make dense amazing analog
sounds (maybe because you are no longer at LSU)?  If you are trying to
imply that there is a conection between Moog's demise and the fact that
you could do analog FM on a Moog you are mistaken.  As far as Buchla goes,
the company was never serious about marketing their products (which is
a term I use very loosely because of the extremely limited production
runs... i.e., there used to be a joke going around some 20 years ago about
the 'three' of everything Buchla made) so it is not surprising that any
of the old analog stuff remains in limited supply.

How any of this relates to Yamaha remains a mystery.  As far as I can tell
the only relationship between the demise of analog production and Yamaha
is that the DX7 demonstrated the 'bang for the buck' potential of digital
synths.  Now, Yamaha has been rumored to have taken legal action against
several manufacturers over the FM patent, but those were digital synths
which borrowed heavily from Chowning's work (which is what Yamaha had a
patent on).  Oberheim's Xpander and Matrix-12 synths fully supported
analog FM, as a matter of fact, they made a big deal about it in the
manuals for those products.  Clearly, Oberheim's failure last year had very
little to do with Yamaha, per se, and asbsolutely nothing to do with FM.

The fact is that analog synth production was killed by market forces.  People
wanted polyphonic synthesizers with more than 16 oscillators.  Analog
manufacturing requires expensive components and the cost of an analog
machine which could compete feature wise with a digital synth would be
absurd.  Hell, most people looked at the price of an Xpander as being absurd,
let alone the cost of a Matrix-12.

Now, if you are really interested in dense timbres, you might want to listen
to a K2000 in 96 oscillator mode.  Pete Towshend summed it up in his retro-
synth player interview.  The K2000 is where it's at, currently.

John

P.S. If you must use a title in front of my name you can choose between Dr.
or LCDR.  Please don't call me Mr.

------------------------------
End of the EMUSIC-L Digest
******************************