Csound Csound-dev Csound-tekno Search About

Re: Subject of research ??

Date1998-07-04 12:53
FromCarlton Wilkinson
SubjectRe: Subject of research ??

Ben Jefferys wrote:

> > like because of the "sounds".
>
> Absolutely - the endless stream of guitar/drum/vocal based music is bound to
> leave us wanting a greater variety of aural stimulation, at the waveform
> level, if not the composition.

At every level. Familiar popular styles seem to overwhelm everyone who has ever
put drums bass and guitar into the same room. (It's not so much the
instruments--although there's that--but the way they're used. That's all I'm
sayin')

> > with alternate forms of notation recently?)
>
> Would there be any real use to this? Shifting, say, a Csound composition to a
> human-readable form (as I guess is the point of sheet music) would be
> reasonably pointless since no human could ever hope to reproduce the exact
> sound with any accuracy. An analogy might be the idea that the most complete
> specification of what a program does is the progam itself. (That's not to
> say the most "best" specification of a piece of music is its waveform...)
>

This is a question I've been struggling with for a while, since I've moved in
the last few years into writing a lot for performer and tape. What is necessary
to include in the score? How is it done? My solution has been to compromise into
a system of notated cues for the performer (as opposed to simply numbering
anonymous "events" in the performer's score) as you would for an orchestral
part. The cues are labeled "foreground events." The notation (since it is not
trying to accurately represent the entire Csound score) can take huge liberties
with its representation of the sound. This allows for a more concise score as
well. Techniques vary from traditional notation to the timed "pretty pictures"
variety for sounds outside the ability of traditional notation. I'ld be
intereested to hear what others are thinking along these lines.

> >  I would love
> > to hear what Beethoven would do with modern instruments
> > and recording techniques.
>
> I wonder if Beethoven would even be a composer if given modern equipment.
> Perhaps his skill was limited to the orchestral and standard musical notation
> domain. I'm no expert of course... ;)

L.v.B. was always pushing the technology/physical envelope according to whim. I
like to picture him in the 20th Century as a reclusive, Eno-type producer who's
genius lies in making everybody else sound like him. The big thing with the
mature B. became expression; expression at all costs. I can't believe he
wouldn't be deeply involved with gear.Of course, this type of speculation always
annoys me when others do it. So I'm sure I'm annoying the hell out of somebody
on the list right now so I'll just shut up.

> --

Carlton Joseph Wilkinson
http://excaliber.net/alex/wilkwrks.htm





Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa23643;
          4 Jul 98 20:03 BST
Received: from mercury.bath.ac.uk by stork.maths.Bath.AC.UK id aa22837;
          4 Jul 98 20:03 BST
Received: (qmail 1778 invoked from network); 4 Jul 1998 19:03:44 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by mercury.bath.ac.uk with SMTP; 4 Jul 1998 19:03:44 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (UAA17493); Sat, 4 Jul 1998 20:00:41 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Sat, 4 Jul 98 20:00:23 +0100
Received: from first1.lnk.telstra.net [139.130.48.118] by hermes via ESMTP (UAA02208); Sat, 4 Jul 1998 20:00:14 +0100 (BST)
Received: from wira (wira.firstpr.com.au [203.36.57.201])
	by gair.firstpr.com.au (8.8.7/8.8.7) with SMTP id GAA01335
	for ; Sun, 5 Jul 1998 06:01:40 +1000
Message-Id: <199807042001.GAA01335@gair.firstpr.com.au>
Comments: Authenticated sender is 
From: Robin Whittle 
Organization: First Principles
To: Csound list 
Date: Sun, 5 Jul 1998 04:51:49 +1000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: rnd, birnd doco problems
Reply-To: rw@firstpr.com.au
Priority: normal
In-Reply-To: <359E17A9.213ECA1C@prodigy.net>
X-Mailer: Pegasus Mail for Win32 (v2.53/R1)
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

I was testing out the random opcodes that Barry Vercoe added last 
year.  

Eg:

kfoo =    rnd ( kbar )

kfoo =  birnd ( kbar )

All instances of these ugens use the same algorithm with a single 
state, so they produce a fixed sequence of pseudo random numbers.
There is no way of setting the seed - the seed is 0.5 at the start 
of the Csound run.

The algorithm is simple.  The new "random" number is the fractional 
part of the old one multipled by 105.947.  This is calcuated and 
stored as a double precision float (64 bits).

I don't know anything about the quality of this Pseudo Random 
Number Generator (PRNG).  I did an Altavista Advanced search for 
this number and came up with nothing related to random number 
generation.  (However searching for "16807 and random" turned up a 
lot because it is a magic number of some note in pseudo random 
algorithms.)  I haven't heard of a PRNG using floating point states 
- all of them seem to be based on long integers.

The manual page is misleading, because it gives the opcodes:

  irnd
  krnd
 
  ibirnd
  kbirnd

when in fact these words with i and the k at the start cannot be 
used.

The orchestra parser does some tricky things with certain forms of 
opcode, sticking an "i", a "k" or an "a" in front of them before 
looking into the list in entry.c.

Therefore the opcodes to use are "rnd" and "birnd".

This causes some tricky situations however:

   kfoo = rnd ( 1 )

causes the "i" rate opcode irnd to be executed, which means that 
kfoo only gets one serve from the random number generator!

Other similarly organised EVAL opcodes do the same, but there is no 
problem, since they are supposed to do the same thing every time - 
for instance:

   kfoo = abs ( -5 ) 

will fire up the iabs opcode and write 5 into kfoo once only - at 
the init time of the instrument.

Therefore, to use these opcodes for k rate, do something like:

   krange = 1
   kfoo = rnd ( krange )



- Robin




 






===============================================================

Robin Whittle     rw@firstpr.com.au  http://www.firstpr.com.au
                  Heidelberg Heights, Melbourne, Australia 

First Principles  Research and expression: music, Internet 
                  music marketing, telecommunications, human 
                  factors in technology adoption. Consumer 
                  advocacy in telecommunications, especially 
                  privacy. Consulting and technical writing. 

Real World        Electronics and software for music: eg.
Interfaces        the Devil Fish mods for the TB-303. 

===============================================================



Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa23767;
          4 Jul 98 22:48 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa29056;
          4 Jul 98 22:48 BST
Received: (qmail 12131 invoked from network); 4 Jul 1998 21:48:46 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 4 Jul 1998 21:48:46 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (WAA25220); Sat, 4 Jul 1998 22:46:23 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Sat, 4 Jul 98 22:46:06 +0100
Received: from ella.mills.edu [144.91.3.20] by hermes via SMTP (WAA23698); Sat, 4 Jul 1998 22:45:59 +0100 (BST)
Received: (qmail 5964 invoked by uid 1964); 4 Jul 1998 14:45:31 -0700
Date: Sat, 4 Jul 1998 14:45:31 -0700 (PDT)
From: "Matt J. Ingalls" 
Cc: Csound List 
Subject: Re: Notation &= ~Subject of research ??
In-Reply-To: <006501bda743$3bf066a0$34015e9e@default>
Message-Id: 
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

> Maryanne Amacher very well suited to listen to.
        i thought she hated recordings..  how much of her
work is available on CD?
>
> > I have been noticing that there is a difference between
> > music that I like because of the "notes" and music that I
> > like because of the "sounds".

        i have been noticing that i *dont* like any music meant
to be liked because of the "notes"
        i have been noticing that i *dont* like any music meant
to be liked because of the "sounds"

> abstract 'high level' notation (e.g. gestural notation of their perceived
> There have been *many* uses of graphical and other un-conventional notations

	is "sono" still around/supported/werking?  its an (X-Windows??)
app that converts a soundfile into a postscript sonigram score. has it
been ported to any other platforms? (or anyone know of similiar apps?)


-matt
Date1998-07-04 19:51
FromRobin Whittle
Subjectrnd, birnd doco problems
I was testing out the random opcodes that Barry Vercoe added last 
year.  

Eg:

kfoo =    rnd ( kbar )

kfoo =  birnd ( kbar )

All instances of these ugens use the same algorithm with a single 
state, so they produce a fixed sequence of pseudo random numbers.
There is no way of setting the seed - the seed is 0.5 at the start 
of the Csound run.

The algorithm is simple.  The new "random" number is the fractional 
part of the old one multipled by 105.947.  This is calcuated and 
stored as a double precision float (64 bits).

I don't know anything about the quality of this Pseudo Random 
Number Generator (PRNG).  I did an Altavista Advanced search for 
this number and came up with nothing related to random number 
generation.  (However searching for "16807 and random" turned up a 
lot because it is a magic number of some note in pseudo random 
algorithms.)  I haven't heard of a PRNG using floating point states 
- all of them seem to be based on long integers.

The manual page is misleading, because it gives the opcodes:

  irnd
  krnd
 
  ibirnd
  kbirnd

when in fact these words with i and the k at the start cannot be 
used.

The orchestra parser does some tricky things with certain forms of 
opcode, sticking an "i", a "k" or an "a" in front of them before 
looking into the list in entry.c.

Therefore the opcodes to use are "rnd" and "birnd".

This causes some tricky situations however:

   kfoo = rnd ( 1 )

causes the "i" rate opcode irnd to be executed, which means that 
kfoo only gets one serve from the random number generator!

Other similarly organised EVAL opcodes do the same, but there is no 
problem, since they are supposed to do the same thing every time - 
for instance:

   kfoo = abs ( -5 ) 

will fire up the iabs opcode and write 5 into kfoo once only - at 
the init time of the instrument.

Therefore, to use these opcodes for k rate, do something like:

   krange = 1
   kfoo = rnd ( krange )



- Robin




 






===============================================================

Robin Whittle     rw@firstpr.com.au  http://www.firstpr.com.au
                  Heidelberg Heights, Melbourne, Australia 

First Principles  Research and expression: music, Internet 
                  music marketing, telecommunications, human 
                  factors in technology adoption. Consumer 
                  advocacy in telecommunications, especially 
                  privacy. Consulting and technical writing. 

Real World        Electronics and software for music: eg.
Interfaces        the Devil Fish mods for the TB-303. 

===============================================================
Date1998-07-05 17:06
FromRobin Whittle
SubjectRandom number generation - 31 bit inside Csound
It has been a rainy Sunday in Melbourne - but Csounders would 
understand why someone would spend a day searching the world for 
software which could be integrated into a computer program which 
would make a high powered computer (with quietened fans and 
sound-proofed hard disk) generate a satisfying "Hisssssssssssssss" 
sound.


I have found a good 31 bit Pseudo Random Number Generator - a 
Park-Miller linear-congruental generator with a good pedigree.  I 
have integrated it into csmath.h, where it drives all the "x-class 
noise" ungens.   It sounds fine - but I need to test the seed-setting 
department.  I will also add a bipolar uniform set of opcodes:

    ibunirand, kbunirand and abunirand

which will do the -1 to +1 range (when given 1 as an input), to 
compliment the 0 to 1 set of opcodes:

    iunirand,  kunirand  and aunirand

which, by the way, are incorrectly documented in the manual as 
"xuniform" and therefore by implication as:

    iuniform,  kuniform  and auniform 


The new self-contained PRNG to serve all these removes Csound's 
dependency on an external library function.  This external function 
"rand()" which used in the current version of Csound could lead to a 
crappy PRNG with a 65535 long cycle, which is audible if used at an 
audio rate.

In RedHat 5.1 Linux, it leads to the glibc library - which does seem 
to provide a 31 bit PRNG which is probably quite respectable.

I timed the old and the new code and the new code is 28% faster.  It 
seems that a Pentium Pro can execute an entire audio cycle of this 
PRNG, including writing it to a word in the Csound a rate variable, 
in about 60 clock cycles!  (With ksmps = 10.) This means it can 
generate a random number and write it to the variable in 0.25 usec on 
a 233 MHz CPU.

A 31 bit PRNG has a cycle period of 2,147,483,645 - so if you are 
calling it 100 times per audio cycle, at 44.1 kHz, it takes 
13.5 hours before it repeats itself.

The details to date are at the bottom of:

    http://www.firstpr.com.au/csound/

and I should be able to finish this in the next few days and provide 
the source code and patches to cmath.c cmath.h and entry.c.

- Robin








===============================================================

Robin Whittle     rw@firstpr.com.au  http://www.firstpr.com.au
                  Heidelberg Heights, Melbourne, Australia 

First Principles  Research and expression: music, Internet 
                  music marketing, telecommunications, human 
                  factors in technology adoption. Consumer 
                  advocacy in telecommunications, especially 
                  privacy. Consulting and technical writing. 

Real World        Electronics and software for music: eg.
Interfaces        the Devil Fish mods for the TB-303. 

===============================================================



Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa25524;
          5 Jul 98 18:54 BST
Received: from mercury.bath.ac.uk by stork.maths.Bath.AC.UK id aa19237;
          5 Jul 98 18:54 BST
Received: (qmail 27904 invoked from network); 5 Jul 1998 17:54:32 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by mercury.bath.ac.uk with SMTP; 5 Jul 1998 17:54:32 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (SAA14397); Sun, 5 Jul 1998 18:51:59 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Sun, 5 Jul 98 18:51:43 +0100
Received: from [193.121.99.70] by hermes via ESMTP (SAA09794); Sun, 5 Jul 1998 18:51:36 +0100 (BST)
Received: from turing.hogent.be ([193.190.88.183]) by hurricane.netgate.be
          (post.office MTA v2.0 0813 ID# 0-32575U60) with ESMTP id AAA286
          for ; Sun, 5 Jul 1998 19:56:55 +0200
From: David Schuyeteneer 
To: Csound List 
Subject: [?] How to read Postscript philez
Date: Sun, 5 Jul 1998 19:47:32 +0200
X-Msmail-Priority: Normal
X-Priority: 3
X-Mailer: Microsoft Internet Mail 4.70.1155
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Message-Id: <19980705175653828.AAA286@turing.hogent.be>
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

Yo people !  I downloaded some great (i think so, at least) technical
documents
about MP3 stuff.....But the files are in PS format....How can I read those
?? 
Which program is used to read them ?? What is PS anyway ??

later,
David




Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa25610;
          5 Jul 98 20:12 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa22333;
          5 Jul 98 20:12 BST
Received: (qmail 2682 invoked from network); 5 Jul 1998 19:12:13 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 5 Jul 1998 19:12:13 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (UAA06367); Sun, 5 Jul 1998 20:09:01 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Sun, 5 Jul 98 20:08:43 +0100
Received: from clavin.interaccess.com [207.70.126.132] by hermes via ESMTP (UAA22989); Sun, 5 Jul 1998 20:08:36 +0100 (BST)
Received: from Default (d21.focal5.interaccess.com [207.208.188.21])
	by clavin.interaccess.com (8.9.0/8.9.0) with SMTP id OAA29217;
	Sun, 5 Jul 1998 14:10:15 -0500 (CDT)
From: "Bob Hays, Computer Geek" 
To: David Schuyeteneer , 
    Csound List 
Subject: RE: [?] How to read Postscript philez
Date: Sun, 5 Jul 1998 14:08:02 -0500
Message-Id: <000501bda848$3b524140$15bcd0cf@Default.interaccess.com>
Mime-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-Msmail-Priority: Normal
X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0
X-Mimeole: Produced By Microsoft MimeOLE V4.72.2106.4
In-Reply-To: <19980705175653828.AAA286@turing.hogent.be>
Importance: Normal
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

PostScript(TM) is a programming language for displayable documents from
Adobe Systems (http://www.adobe.com).  The language is postfix, like Forth
or an HP calculator (people sometimes call it reverse-polish).  To add 1 and
2 you do:

1 2 add

which pushes 1 and then 2 on the stack and then executes add, which pops the
top two values from the stack and adds them together.

Normally, PostScript files are ASCII text - there are binary forms, but
those are not portable.  The text versions, if they adhere to a proper
version of PostScript, will work on any adhering printer; this is what made
PostScript a defacto standard - you could output a PostScript document to a
file and print it on any PostScript printer.  Probably more than you wanted
to know about PostScript, the language, but I used to teach a beginning
course in PostScript programming so....

Tools....  The most popular and common tool is GhostScript, which provides
rendering of PostScript for viewing.  This is ported to the PC/Windows
platform.  Don't know where to find it right now, but a good Yahoo or Lycos
search will find it somewhere.

A very popular extension to PostScript is Adobe Acrobat format, with the
file extension .pdh.  Another popular extension is Display PostScript, which
provides extensions for writing GUI programs in PostScript (Sun used this
for a time for their GUI of choice on SunOS for their UNIX variant).

Hope all that helps.  Have fun! - Bob

--
"I did not ask for the anal probe" -- Passion Fish
"Discipline is never an end in itself, only a means to an end." - King
Crimson

Bob Hays                               | Aviva Kramer
---------------------------------------+--------------------------------
rhays@interaccess.com (home)           | a1kramer@interaccess.com (home)
Bob.Hays@abnamro.com (work)            |
http://homepage.interaccess.com/~rhays |

> -----Original Message-----
> From: owner-csound-outgoing@noether.ex.ac.uk
> [mailto:owner-csound-outgoing@noether.ex.ac.uk]On Behalf Of David
> Schuyeteneer
> Sent: Sunday, July 05, 1998 12:48 PM
> To: Csound List
> Subject: [?] How to read Postscript philez
>
>
> Yo people !  I downloaded some great (i think so, at least) technical
> documents
> about MP3 stuff.....But the files are in PS format....How can I read those
> ??
> Which program is used to read them ?? What is PS anyway ??
>
> later,
> David
>
>
>




Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa25626;
          5 Jul 98 20:17 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa22551;
          5 Jul 98 20:17 BST
Received: (qmail 2799 invoked from network); 5 Jul 1998 19:17:03 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 5 Jul 1998 19:17:03 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (UAA08616); Sun, 5 Jul 1998 20:13:17 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Sun, 5 Jul 98 20:12:18 +0100
Received: from f12.hotmail.com [207.82.250.23] by hermes via SMTP (UAA15496); Sun, 5 Jul 1998 20:11:51 +0100 (BST)
Received: (qmail 14808 invoked by uid 0); 5 Jul 1998 19:11:25 -0000
Message-Id: <19980705191125.14807.qmail@hotmail.com>
Received: from 207.172.114.43 by www.hotmail.com with HTTP;
	Sun, 05 Jul 1998 12:11:24 PDT
X-Originating-Ip: [207.172.114.43]
From: Paul Winkler 
To: csound@maths.ex.ac.uk
Subject: timout, reinit & rireturn
Content-Type: text/plain
Date: Sun, 05 Jul 1998 12:11:24 PDT
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

Just experimenting with basic stochastic stuff for the first time and I
have two questions:

1. I'm trying to use timout, reinit & rireturn to generate a certain
number of random notes over the duration of one score "note". I've never
used these before and I don't seem to correctly understand how to get 
the
behavior I want.

I expected the following orc/sco to produce 5 random notes in the first 
10
seconds and 10 random notes in the second 10 seconds, but instead I only
get one random note that sustains for the dur of each score note.

2. Is there a way to seed the x-class noise generators? Or is there
another i-rate random generator that can be seeded? I don't want to hear
the same stuff every time...  I found an old mailing-list mention of
rnd(x) which is presumably in the value converters, but it's not in the
3.48 HTML manual.

orc & sco follow...

regards,

PW

; orc
sr = 22050
kr = 2205
ksmps = 10
nchnls = 1

	instr 1 ; very basic random notes, only upper & lower bound set

ilbound = 6.04    ; these bounds are set for guitar.
iubound = 9.09   ; for bass, use 5.04 to 8.00 and change the offset in 
ifreq

reset: timout 0, p3 / p4, notepick
	reinit reset
notepick: ioct ilinrand 3 ; octave range
ipch ilinrand 11 ; 12 semitones
ifreq = ioct + 6 + (ipch * 0.01)
if ifreq > iubound igoto notepick  ; test limits
if ifreq < ilbound igoto notepick
ivol ilinrand p6
rireturn
a1 oscil ampdb(ivol + p5), cpspch(ifreq), 1
out a1
	endin






;sco

f1 0 8192 8 0 4080 1 32 -1 4080 0

; 
;i	at	dur 	notes	amp-av	amp-dev
i1  	0 	10	5	50      20
i1  	10      10   	10	70      20
e


______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com



Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa25835;
          5 Jul 98 23:29 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa00113;
          5 Jul 98 23:29 BST
Received: (qmail 6431 invoked from network); 5 Jul 1998 22:29:09 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 5 Jul 1998 22:29:09 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (XAA11173); Sun, 5 Jul 1998 23:16:54 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Sun, 5 Jul 98 23:16:31 +0100
Received: from howl.werewolf.net [206.103.224.20] by hermes via ESMTP (XAA12874); Sun, 5 Jul 1998 23:16:21 +0100 (BST)
Received: from hans7.cray.com (dial30.werewolf.net [206.103.225.40])
	by howl.werewolf.net (8.9.0/8.9.0) with SMTP id RAA09767
	for ; Sun, 5 Jul 1998 17:15:35 -0500 (CDT)
From: Hans Mikelson 
To: csound@maths.ex.ac.uk
MMDF-Warning:  Parse error in original version of preceding line at UK.AC.Bath.maths.stork
Subject: Re: timout, reinit & rireturn
Date: Sun, 5 Jul 1998 17:18:07 -0500
Message-Id: <01bda862$c92a5ce0$28e167ce@hans7.cray.com>
Mime-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-Msmail-Priority: Normal
X-Mailer: Microsoft Outlook Express 4.71.1712.3
X-Mimeole: Produced By Microsoft MimeOLE V4.71.1712.3
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

Hi,

Paul Winkler wrote:

>1. I'm trying to use timout, reinit & rireturn to generate a certain
>number of random notes over the duration of one score "note". I've never
>used these before and I don't seem to correctly understand how to get
>the
>behavior I want.


Try making the following changes:

 instr 1         ; very basic random notes, only upper & lower bound set
ilbound =       6.04    ; these bounds are set for guitar.
iubound =       9.09    ; for bass, use 5.04 to 8.00 and change the offset
in ifreq
reset:
       timout   0, p3 / p4, next1
notepick:
ioct   ilinrand 3       ; octave range
ipch   ilinrand 11      ; 12 semitones
ifreq  =        ioct + 6 + (ipch * 0.01)
if ifreq > iubound igoto notepick  ; test limits
if ifreq < ilbound igoto notepick
ivol ilinrand p6
       reinit   reset
next1:
a1 oscil ampdb(ivol + p5), cpspch(ifreq), 1
out a1
 endin

Bye,
Hans Mikelson




Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa26280;
          6 Jul 98 4:41 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa20724;
          6 Jul 98 4:41 BST
Received: (qmail 12491 invoked from network); 6 Jul 1998 03:41:02 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 6 Jul 1998 03:41:02 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (EAA26239); Mon, 6 Jul 1998 04:37:10 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Mon, 6 Jul 98 04:36:53 +0100
Received: from send1e.yahoomail.com [205.180.60.64] by hermes via SMTP (EAA03362); Mon, 6 Jul 1998 04:36:46 +0100 (BST)
Message-Id: <19980706033555.6521.rocketmail@send1e.yahoomail.com>
Received: from [192.9.25.21] by send1e; Sun, 05 Jul 1998 20:35:55 PDT
Date: Sun, 5 Jul 1998 20:35:55 -0700 (PDT)
From: Qian Chen 
Subject: csound archive available!
To: Csound 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

Dear all,

Good News!

The csound archive at 
http://www.maths.ex.ac.uk/~james/csound_list/
is available now.  I just found it available several hours ago.

Regards



==
Qian Chen

_________________________________________________________
DO YOU YAHOO!?
Get your free @yahoo.com address at http://mail.yahoo.com




Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id ab26721;
          6 Jul 98 10:31 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa00924;
          6 Jul 98 8:30 BST
Received: (qmail 16487 invoked from network); 6 Jul 1998 07:30:14 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 6 Jul 1998 07:30:14 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (IAA14032); Mon, 6 Jul 1998 08:26:55 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Mon, 6 Jul 98 08:26:37 +0100
Received: from varis.cs.tut.fi [130.230.4.2] by hermes via ESMTP (IAA08526); Mon, 6 Jul 1998 08:26:31 +0100 (BST)
Received: from kalalokki.cs.tut.fi (kalalokki.cs.tut.fi [130.230.14.118])
	by cs.tut.fi (8.8.8/8.8.8) with ESMTP id KAA03177;
	Mon, 6 Jul 1998 10:26:34 +0300 (EET DST)
Received: (from jams@localhost)
          by kalalokki.cs.tut.fi (8.8.5/8.8.4)
	  id KAA14933; Mon, 6 Jul 1998 10:26:33 +0300 (EET DST)
To: Jim Stevenson 
Cc: ben@qsure.demon.co.uk, csound@maths.ex.ac.uk
Subject: Re: MPEG compression?
References: <199807031715.KAA18609@eos.arc.nasa.gov>
From: Jarno Seppanen 
Reply-To: Jarno Seppanen 
Comment: Redistribution via Microsoft Network prohibited
X-Url: http://www.modeemi.cs.tut.fi/~jams/
Date: 06 Jul 1998 10:26:33 +0300
In-Reply-To: Jim Stevenson's message of "Fri, 3 Jul 1998 10:15:42 -0700"
Message-Id: 
Lines: 19
X-Mailer: Gnus v5.4.66/Emacs 19.31
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

	The MPEG Audio Page contains information on MPEG-1, MPEG-2 and MPEG-4
audio formats: http://www.tnt.uni-hannover.de/project/mpeg/audio/.  The MPEG-4
Final Committee Draft is available online.

Jim Stevenson  writes:

> Where can I get a technical description, and the source code for the
> algorithms that do the compression and decompression?
> 
> What compression ratios can be achieved?
> 
> Is stereo treated as two mono chanels, or can we do better by encoding
> sum and difference?
> 
> Thanks.
> 

-- 
-Jarno



Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa26750;
          6 Jul 98 10:59 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa08591;
          6 Jul 98 10:59 BST
Received: (qmail 26283 invoked from network); 6 Jul 1998 09:59:30 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 6 Jul 1998 09:59:30 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (KAA01201); Mon, 6 Jul 1998 10:56:23 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Mon, 6 Jul 98 10:55:56 +0100
Received: from first1.lnk.telstra.net [139.130.48.118] by hermes via ESMTP (KAA20407); Mon, 6 Jul 1998 10:55:44 +0100 (BST)
Received: from wira (wira.firstpr.com.au [203.36.57.201])
	by gair.firstpr.com.au (8.8.7/8.8.7) with SMTP id UAA10981
	for ; Mon, 6 Jul 1998 20:57:43 +1000
Message-Id: <199807061057.UAA10981@gair.firstpr.com.au>
Comments: Authenticated sender is 
From: Robin Whittle 
Organization: First Principles
To: Csound list 
Date: Mon, 6 Jul 1998 19:52:54 +1000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: Random number generation - 31 bit inside Csound
Reply-To: rw@firstpr.com.au
Priority: normal
In-Reply-To: <199807051711.DAA06680@gair.firstpr.com.au>
References: <199807042001.GAA01335@gair.firstpr.com.au>
X-Mailer: Pegasus Mail for Win32 (v2.53/R1)
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

I have finished the 31 bit Park Miller PRNG system for the x-class 
noise opcodes.  There is a seed-setting function there too - actually 
there always was one, but it wasn't in the manual.

There are three new opcodes for bipolar uniform random numbers:
 
    ibunirand, kbunirand and abunirand

For instance

kxyz kbunirand 6

produces numbers evenly distributed in the range -6 to +6.  This is 
also known as white noise.

Changes are to:

   cmath.c
   cmath.h
   entry.c

The changes are not platform dependent, so they should work with any 
operating system and compiler.

The new files, a single patch file to document or make the 
changes, and a new Csound executable for Intel Linux are all in a 
subdirectory linked to from the "Gutsy stuff" section of:

     http://www.firstpr.com.au/csound/

There you will also find a suitably updated HTML manual page for the 
x-class noise opcodes, and all other relevant details.

The random functions in other opcodes and which are embedded in 
"grain" are not affected.

By the way, I found that the xexprand and xbexprnd opcodes did 
something different from what the manual says, and I can't figure out 
what they might be useful for.  I don't have the reference books, and 
Paris Smaragdis hasn't been heard of in these parts for a while.  
Does anyone have any clues.  I have not tested the 

        xcauchy         
        xpcauchy        

        xpoisson        
        xgauss          
        xweibull        
        xbeta        

opdodes since I either haven't had time to look up the references on 
what they should do.

- Robin



===============================================================

Robin Whittle     rw@firstpr.com.au  http://www.firstpr.com.au
                  Heidelberg Heights, Melbourne, Australia 

First Principles  Research and expression: music, Internet 
                  music marketing, telecommunications, human 
                  factors in technology adoption. Consumer 
                  advocacy in telecommunications, especially 
                  privacy. Consulting and technical writing. 

Real World        Electronics and software for music: eg.
Interfaces        the Devil Fish mods for the TB-303. 

===============================================================



Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa26788;
          6 Jul 98 11:10 BST
Received: from mercury.bath.ac.uk by stork.maths.Bath.AC.UK id aa09386;
          6 Jul 98 11:10 BST
Received: (qmail 1330 invoked from network); 6 Jul 1998 10:10:54 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by mercury.bath.ac.uk with SMTP; 6 Jul 1998 10:10:54 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (LAA06445); Mon, 6 Jul 1998 11:06:45 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Mon, 6 Jul 98 11:06:19 +0100
Received: from first1.lnk.telstra.net [139.130.48.118] by hermes via ESMTP (LAA00026); Mon, 6 Jul 1998 11:06:03 +0100 (BST)
Received: from wira (wira.firstpr.com.au [203.36.57.201])
	by gair.firstpr.com.au (8.8.7/8.8.7) with SMTP id VAA11062
	for ; Mon, 6 Jul 1998 21:08:13 +1000
Message-Id: <199807061108.VAA11062@gair.firstpr.com.au>
Comments: Authenticated sender is 
From: Robin Whittle 
Organization: First Principles
To: Csound list 
Date: Mon, 6 Jul 1998 20:03:36 +1000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Manual correction: printk and printks
Reply-To: rw@firstpr.com.au
Priority: normal
X-Mailer: Pegasus Mail for Win32 (v2.53/R1)
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

The syntax in the manual for printk is wrong.  The correct syntax is:

    printk itime, kvar [,ispace]


I need to rework the printks ugen.  Currently, to get a newline, you 
have to use "\n" in the quoted string.  However, since I wrote this 
something else has been introduced into Csound which catches the 
backslashes - I think for line extension purposes.

I haven't decided, and am open to suggestion, but my thoughts are to
abandon \n, \N, \t and \T as and use ~n, ~N, ~t and ~T instead. 
This means I will take out the use of tilde to produce the ANSI
escape sequence.

By then it will be time for me to rewrite the whole manual page.

- Robin



===============================================================

Robin Whittle     rw@firstpr.com.au  http://www.firstpr.com.au
                  Heidelberg Heights, Melbourne, Australia 

First Principles  Research and expression: music, Internet 
                  music marketing, telecommunications, human 
                  factors in technology adoption. Consumer 
                  advocacy in telecommunications, especially 
                  privacy. Consulting and technical writing. 

Real World        Electronics and software for music: eg.
Interfaces        the Devil Fish mods for the TB-303. 

===============================================================



Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa26918;
          6 Jul 98 11:41 BST
Received: from mercury.bath.ac.uk by stork.maths.Bath.AC.UK id aa11509;
          6 Jul 98 11:41 BST
Received: (qmail 2354 invoked from network); 6 Jul 1998 10:41:53 -0000
Received: from hermes.ex.ac.uk (HELO exeter.ac.uk) (144.173.6.14)
  by mercury.bath.ac.uk with SMTP; 6 Jul 1998 10:41:53 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (LAA12054); Mon, 6 Jul 1998 11:38:50 +0100 (BST)
Received: from hermes.ex.ac.uk by maths.ex.ac.uk; Mon, 6 Jul 98 11:38:26 +0100
Received: from nicb@ax-nicb.axnet.it [194.184.60.149] by hermes via ESMTP (LAA21148); Mon, 6 Jul 1998 11:38:17 +0100 (BST)
Received: (from nicb@localhost)
	by ax-nicb.axnet.it (8.8.8/8.8.8) id MAA13859;
	Mon, 6 Jul 1998 12:39:18 +0200
Date: Mon, 6 Jul 1998 12:39:16 +0200 (MET DST)
From: Nicola Bernardini 
To: Csound mailing list 
Subject: Re: Manual correction: printk and printks
In-Reply-To: <199807061108.VAA11062@gair.firstpr.com.au>
Message-Id: 
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

On Mon, 6 Jul 1998, Robin Whittle wrote:

[snip]
> I haven't decided, and am open to suggestion, but my thoughts are to
> abandon \n, \N, \t and \T as and use ~n, ~N, ~t and ~T instead. 
> This means I will take out the use of tilde to produce the ANSI
> escape sequence.

...mmm... I would rather say that it would be the score parser's problem
to avoid to escape \[A-z] inside strings, is'nt it? Otherwise we'll
end up having the first tilde escape system after the good old
mail system (which could mean that some code would'nt fare well over
the mail - just joking, tilde must be at BOF). Seriously, escape characters
are a very important issue and they should be thoroughly thought out
(otherwise the printks case is a typical case of what happens).
My proposal is to make the score parser more intelligent.

Nicola
------------------------------------------------------------------------
Nicola Bernardini
E-mail: nicb@axnet.it
 
Re graphics: A picture is worth 10K words -- but only those to describe
the picture.  Hardly any sets of 10K words can be adequately described
with pictures.
Date1998-07-06 10:52
FromRobin Whittle
SubjectRe: Random number generation - 31 bit inside Csound
I have finished the 31 bit Park Miller PRNG system for the x-class 
noise opcodes.  There is a seed-setting function there too - actually 
there always was one, but it wasn't in the manual.

There are three new opcodes for bipolar uniform random numbers:
 
    ibunirand, kbunirand and abunirand

For instance

kxyz kbunirand 6

produces numbers evenly distributed in the range -6 to +6.  This is 
also known as white noise.

Changes are to:

   cmath.c
   cmath.h
   entry.c

The changes are not platform dependent, so they should work with any 
operating system and compiler.

The new files, a single patch file to document or make the 
changes, and a new Csound executable for Intel Linux are all in a 
subdirectory linked to from the "Gutsy stuff" section of:

     http://www.firstpr.com.au/csound/

There you will also find a suitably updated HTML manual page for the 
x-class noise opcodes, and all other relevant details.

The random functions in other opcodes and which are embedded in 
"grain" are not affected.

By the way, I found that the xexprand and xbexprnd opcodes did 
something different from what the manual says, and I can't figure out 
what they might be useful for.  I don't have the reference books, and 
Paris Smaragdis hasn't been heard of in these parts for a while.  
Does anyone have any clues.  I have not tested the 

        xcauchy         
        xpcauchy        

        xpoisson        
        xgauss          
        xweibull        
        xbeta        

opdodes since I either haven't had time to look up the references on 
what they should do.

- Robin



===============================================================

Robin Whittle     rw@firstpr.com.au  http://www.firstpr.com.au
                  Heidelberg Heights, Melbourne, Australia 

First Principles  Research and expression: music, Internet 
                  music marketing, telecommunications, human 
                  factors in technology adoption. Consumer 
                  advocacy in telecommunications, especially 
                  privacy. Consulting and technical writing. 

Real World        Electronics and software for music: eg.
Interfaces        the Devil Fish mods for the TB-303. 

===============================================================