Hello, All Csounders.

I'd like to put csound audio into some filters in realtime on Linux 2.0.30.
I did try as follows according to "CsManual3.48b1.HTML/Command/CSCOMM.html".

---from Manual------------

csound -o osfname 
  The name stdout will cause audio to be written to standard output.
------------------------

fmtk%> csound -o stdout ???.orc ???.sco | some filter(like a mp3enc)

But, csound tell me like this.
--------------------------------------------------------

  Csound Version 3.51 (Feb 18 1999)
  Csound Command ERROR:   stdout audio not supported
--------------------------------------------------------

And from "Release Notes for 3.48"
------------------------------------------------------------------------

  Language Changes
  ----------------

  The input or output file in -i and -o can start with a | to indicate a
  process which is started to create or process audio files.  This works
  on Windows and Unix, but not (yet) DOS.
------------------------------------------------------------------------

What is the meaning of thie sentence? 
Is there strange point on my way or is this a bug or something?

And if I do comment-out as follows at argdecode.c, version 3.51,
I can use "-o stdout". (Thanks Mr. Christopher Penrose for his help).
------------------------------------------------------------------------

	case 'o':
            if (strcmp(O.outfilename,"stdout") == 0) 
/*
#if defined macintosh || defined SYMANTEC || defined BCC || defined __WATCOMC__ || defined __unix || defined WIN32
*/
#if defined WhatIsThat              /*fmtk edited*/

              dieu("stdout audio not supported");

------------------------------------------------------------------------

I'm sorry if this question is one of FAQ, but I'd like to know why this 
functionthere do not use default version. Is there any problem?

thanks.
fmtk