Message written at 16 Oct 1997 11:21:28 +0100
--- Copy of mail to mogambo@real.net.au ---
In-reply-to:  (message from Arne Hanna
	on Tue, 14 Oct 1997 08:50:54 +1000)
References:  

What score are you using?  I am particularly interested in p9, as if
it were 0 it would loop 2^31 times, which is rather a long time.
Without the score there is insufficient information.

==John ff

PS Looks like a Mikelson orchestra to me.

>The instrument below is from an old post, (apologies to the author, whose
>name has been lost in time)and refers to the Lorenz Equation.  When I try
>to process with a ten second score file, everything seems to set up just
>fine and then waiting... waiting... waiting...  My PPC doesn't hang, but
>two hours later still nothing, and still no crunching.  Anyone?
>
>sr=44100
>kr=4410
>ksmps=10
>nchnls=1
>
>
>instr 1
>
>; Lorenz Equations System at a-rate with all parameters Time-Variable











Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa11788;
          16 Oct 97 14:19 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa20455;
          16 Oct 97 14:19 BST
Received: (qmail 3028 invoked from network); 16 Oct 1997 13:16:31 -0000
Received: from hermes.ex.ac.uk (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 16 Oct 1997 13:16:31 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (NAA03235); Thu, 16 Oct 1997 13:29:54 +0100
Received: from hermes.ex.ac.uk by maths.exeter.ac.uk; Thu, 16 Oct 97 13:29:37 +0100
Received: from sparticus.bright.net [205.212.123.14] by hermes via ESMTP (NAA21877); Thu, 16 Oct 1997 13:29:13 +0100
Received: from brutus (root@find3-cs-6.dial.bright.net [205.212.145.169]) by sparticus.bright.net (8.8.7/8.8.7/FNG) with SMTP id IAA17498; Thu, 16 Oct 1997 08:30:47 -0400 (EDT)
Message-Id: <34460A96.77968CCC@bright.net>
Date: Thu, 16 Oct 1997 08:37:42 -0400
From: Dave Phillips 
X-Mailer: Mozilla 3.0 (X11; I; Linux 2.0.29 i486)
Mime-Version: 1.0
To: csound@maths.ex.ac.uk
Subject: Linux Csound: Help needed !
References: 
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

I'm trying to prepare the Linux version of Csound 3.47, but I'm having
trouble with the new MIDI output opcodes system from Gabriel Maldonado.
His code is fine: where I need the help is in setting up the MIDI output
code for a "typical" Linux machine. I have reached this point:

  1. I can initialize, open, and close the interface (using either
/dev/sequencer or /dev/midi);

  2. My test programs indicate the interface functions properly;

  3. I've activated code in main.c to provide the -Q flag (MIDI output
opened);

  4. But GM's opcodes aren't working yet.

Gabriel checked out my test orc/sco on his Windows system and said it
worked fine, so I know the problem is with coding the connection to the
Linux MIDI interface. I've already received much advice, written a lot
of tests, and puzzled over the soundcard API. At this point I know I
need external assistance.

Are there any Linux Csound users who can help me here ? Everything else
in the upcoming 3.47 seems to be working well, but I'd like to the Linux
version to be as full-featured as any other. I can provide sources as
well as what code I've cooked up already.

Please reply via email, thank you.

== Dave Phillips

       http://www.bright.net/~dlphilp/index.html
   http://www.bright.net/~dlphilp/linux_soundapps.html











Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa12846;
          16 Oct 97 18:15 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa29276;
          16 Oct 97 18:15 BST
Received: (qmail 7598 invoked from network); 16 Oct 1997 17:15:32 -0000
Received: from hermes.ex.ac.uk (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 16 Oct 1997 17:15:32 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (RAA20726); Thu, 16 Oct 1997 17:50:37 +0100
Received: from hermes.ex.ac.uk by maths.exeter.ac.uk; Thu, 16 Oct 97 17:50:15 +0100
Received: from ella.mills.edu [144.91.3.20] by hermes via SMTP (RAA00366); Thu, 16 Oct 1997 17:50:02 +0100
Received: (qmail 14960 invoked by uid 3049); 16 Oct 1997 09:51:31 -0700
Message-Id: <19971016165131.14959.qmail@ella.mills.edu>
From: David Madole 
Subject: PPC Perf Hangs - PPC Fixes and 3.47 beta
To: csound-list 
Date: Thu, 16 Oct 1997 09:51:31 -0700 (PDT)
Reply-To: madole@mills.edu
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk


Hi PPC Csound Users,

This is probably the "+" bug.

There are two perf binaries on the Mills ftp site
(mills.edu:/pub/ccm/csound.ppc).  One, perf.ppc.fix.hqx, fixes a bug
that crept in when I modified the score reader to read scientific
notation, which is probably the reason your perf hangs ("+" in the
score).  The other is a beta of the 3.47 release (perf3.47(beta).hqx).
You probably need the first (which has been there for a long time), if
you don't need the second you should wait until jpff gets done with
3.47 and I get done incorporating his changes, which should be soon.

Dave

Dave Madole
Technical Director, Center for Contemporary Music
Listserv Administrator

Mills College
Oakland, CA 94613
510-430-2336

madole@mills.edu











Received: from stork.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa14048;
          17 Oct 97 3:44 BST
Received: from pat.bath.ac.uk by stork.maths.Bath.AC.UK id aa14741;
          17 Oct 97 3:44 BST
Received: (qmail 8598 invoked from network); 17 Oct 1997 02:44:34 -0000
Received: from hermes.ex.ac.uk (144.173.6.14)
  by pat.bath.ac.uk with SMTP; 17 Oct 1997 02:44:34 -0000
Received: from noether [144.173.8.10] by hermes via SMTP (DAA03827); Fri, 17 Oct 1997 03:29:43 +0100
Received: from hermes.ex.ac.uk by maths.exeter.ac.uk; Fri, 17 Oct 97 03:29:26 +0100
Received: from root@discover-net.net [208.134.196.20] by hermes via ESMTP (DAA17293); Fri, 17 Oct 1997 03:29:17 +0100
Received: from lizard (max-ec2-186.discover-net.net [208.134.205.186]) by discover.discover-net.net (8.8.6/8.7.3) with SMTP id VAA07091 for ; Thu, 16 Oct 1997 21:28:16 -0500 (CDT)
Message-Id: <3.0.3.16.19971009160912.30c7cb2e@mail.discover-net.net>
X-Sender: hljmm@mail.discover-net.net
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.3 (16)
Date: Thu, 09 Oct 1997 16:09:12
To: csound@maths.ex.ac.uk
From: Hans Mikelson 
Subject: Re: Lorenz Orchestra
In-Reply-To: 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk

Hello,

At 08:50 AM 10/14/97 +1000, Arne Hanna wrote:
>The instrument below is from an old post, (apologies to the author, whose
>name has been lost in time)and refers to the Lorenz Equation.  When I try
>to process with a ten second score file, everything seems to set up just
>fine and then waiting... waiting... waiting...  My PPC doesn't hang, but
>two hours later still nothing, and still no crunching.  Anyone?

>if kcontzoom==kzoom kgoto sortida
>if kcontzoom!=kzoom kgoto loop1

I suspect it is stuck in a loop for some reason.  Change == to >=.
Actually this is not mine although I think I had it up on my web site for a
while.  I think it may be by either Josep M* Comajuncosas or a friend of
his originally.  The one that was posted had a bug in but I don't recall
exactly where it was.  The following one should work:

sr=44100
kr=44100
ksmps=1
nchnls=1

instr 1

; Lorenz Equations System at a-rate with all parameters Time-Variable


ipi init 6.283184
amix init 0

;p4 p5 i p6 = initial values for control parameters
;p7 = time differential (not less than 0.02 - may be improved using
Runge-Kutta
;       integration methode but it is very time consuming )
;p8 amplitude
;p9 zoom factor (a kind of temporal window applied to trajectories, values of
;               aprox. 5 are equivalent-almost- to downing p7 at the price of
;               a bit slowing down computing time...)
;p10 p11 p12 = deviation value for p6 p7 p8
;               (It is implemented with a simple line statement, but
;                improving it is straightforward)


kcontzoom init 0
kzoom init p9

iprof init p8
idt init p7

; NEWTON integration methode

adx init 0     ;diferentials
ady init 0
adz init 0
ax init .6       ;values
ay init .6
az init .6

;aa, ab and ac holds for the values of control parameters

       aa line p4,p3,p4+p10
       ab line p5,p3,p5+p11
       ac line p6,p3,p6+p12

aa init p4
ab init p5
ac init p6

loop1: adx=aa*(ay-ax)
       ady=ax*(ab-az)-ay
       adz=ax*ay-ac*az
       ax=ax+(idt*adx)
       ay=ay+(idt*ady)
       az=az+(idt*adz)

kcontzoom=kcontzoom+1

if kcontzoom>=kzoom kgoto sortida
if kcontzoom!=kzoom kgoto loop1

sortida:amix = ax*iprof

out amix

kcontzoom=0

endin

; score******************************************************************
;  P2     P3   P4   P5    P6     P7         P8         P9     P10 P11 P12
;  istart idur A1   B1    C1     Time (dT)  Amplitude  KZoom  A2  B2  C2
i1  0     .5    22   28   2.667      .01       600     3      5   0    0
;i1  0     3    26   28   2.667      .01       600     4      0   10    0

;e

Or this simplified version:

sr=44100
kr=44100
ksmps=1
nchnls=2
;Lorenz Attractor
        instr  1
;--------------------------------------------------------------------
ax      init   p5
ay      init   p6
az      init   p7
as      init   p8
ar      init   p9
ab      init   p10
ah      init   p11
;--------------------------------------------------------------------
kampenv linseg 0, .01, p4, p3-.02, p4, .01, 0
axnew   =      ax+ah*as*(ay-ax)
aynew   =      ay+ah*(-ax*az+ar*ax-ay)
aznew   =      az+ah*(ax*ay-ab*az)
;--------------------------------------------------------------------
ax      =      axnew
ay      =      aynew
az      =      aznew
;--------------------------------------------------------------------
        outs   ax*kampenv,ay*kampenv
        endin

f1 0 8192 10 1
t 0 400
;  Start  Dur Amp  X   Y   Z    S    R   B      h
;i1  0     8   600  .6  .6  .6  10   28  2.667  .01
;i1  +     .   600  .6  .6  .6  22   28  2.667  .01
;i1  .     .   600  .6  .6  .6  32   28  2.667  .01

The loop will allow you to get higher frequencies out of the system without
going to very high sample rates.

Bye,
Hans Mikelson