Re: [ecasound] writing multiple files

[e d. <ede...@gm...>]

Aaron,

Still not right .... each of the resultant 4 mono files is still
larger than the original 4 channel file.  so this is what I've got
now:

ecasound \
  -a:input \
  -f:16,4,44100 \
  -i:resample-hq,auto,WXYZ.wav \
  -el:Ambisonics-square-decoder,0,1,1.44,1.44,500,2  \
  -f:16,4,44100 \
  -o:loop,1 \
  -a:LF,RF,LB,RB -f:16,1,44100 -i:loop,1 \
  -a:LF -o LFront.wav \
  -a:RF -o RFront.wav -erc:2,1 \
  -a:LB -o LBack.wav -erc:3,1 \
  -a:RB -o RBack.wav -erc:4,1

... It kind of smells like a bug. I could see a logic if the resultant
files were all the same size as the original 4 channel file, but since
they are all larger ...

Etienne


On 5/12/07, Aaron Heller <he...@ai...> wrote:
> That last -erc:4,2 should be -erc:4,1
> But I doubt that's the problem
>
> I looked up some of my old ecasound scripts and see that I put -f s on loop
> inputs and outputs
>
> ecasound \
>   -a:input \
>   -f:16,4,44100 \
>   -i:resample-hq,auto,WXYZ.wav \
>   -el:Ambisonics-square-decoder,0,1,1.44,1.44,500,2  \
>   -f:f32_le,4,44100 \
>   -o:loop,1 \
>   -a:LF,RF,LB,RB -f:f32_le,4,44100 -i:loop,1 \
>   -a:LF -f:16,1,44100 -o LFront.wav \
>   -a:RF -f:16,1,44100 -o RFront.wav -erc:2,1 \
>   -a:LB -f:16,1,44100 -o LBack.wav -erc:3,1 \
>   -a:RB -f:16,1,44100 -o RBack.wav -erc:4,2
>
>
> On 5/11/07 10:47 PM, "e deleflie" <ede...@gm...> wrote:
>
> > Still not right. The below code results in 4 files that grow ever
> > bigger (I killed ecasound when each of the mono files were 10 times
> > bigger than the original 4 channel file).... I really cant see what is
> > wrong here:
> >
> > ecasound \
> >   -a:input \
> >   -f:16,4,44100 \
> >   -i:resample-hq,auto,WXYZ.wav \
> >   -el:Ambisonics-square-decoder,0,1,1.44,1.44,500,2  \
> >   -o:loop,1 \
> >   -a:LF,RF,LB,RB -i:loop,1 \
> >   -a:LF -f:16,1,44100 -o LFront.wav \
> >   -a:RF -f:16,1,44100 -o RFront.wav -erc:2,1 \
> >   -a:LB -f:16,1,44100 -o LBack.wav -erc:3,1 \
> >   -a:RB -f:16,1,44100 -o RBack.wav -erc:4,2
> >
> > -----------------
> > the below code, however, produces the correct results, except I have
> > my 4 tracks in 1 file (which is no good to me).
> >
> > ecasound \
> >   -a:input \
> >   -f:16,4,44100 \
> >   -i:resample,auto,WXYZ.wav \
> >   -el:Ambisonics-square-decoder,0,1,1.44,1.44,500,2  \
> >   -o:out.wav \
> >
> > can anyone see what is wrong? ... I suspect it is the loop back that
> > is doing something strange.
> >
> > Etienne
> >
> > On 5/12/07, e deleflie <ede...@gm...> wrote:
> >> Aaron, actually you are right ... That previous code was accepted by
> >> ecasound,  but didn't give the right results.
> >>
> >> funny you should answer, because the test file that I am converting is
> >> actually your Stravinsky recording! .. I am decoding that one so that
> >> I can compare it to your decodes you sent me ages ago (which I still
> >> have not listened to but will do and send you feedback!).
> >>
> >> I concur on your choice of ecasound for processing multi-channel content...
> >>
> >> Etienne
> >>
> >> On 5/12/07, Aaron Heller <he...@ai...> wrote:
> >>> Hi Etienne...
> >>>
> >>> Actually, I think there is a remaining error.  Chains 2,3,4, and 5 are four
> >>> channels each and the -f:16,1,44100 simply says to write the first channel
> >>> of the chain to the file.  So I think you will get four identical files,
> >>> each containing LFront.wav.
> >>>
> >>> The fix is to do some channel copying in the output chains.
> >>>
> >>> ...
> >>>>>   -a:2 -f:16,1,44100 -o LFront.wav   \
> >>>>>   -a:3 -f:16,1,44100 -o RFront.wav -erc:2,1 \
> >>>>>   -a:4 -f:16,1,44100 -o LBack.wav  -erc:3,1 \
> >>>>>   -a:5 -f:16,1,44100 -o RBack.wav  -erc:4,1
> >>>
> >>> -erc:2,1 says replace channel 1 in the chain with channel 2, and so forth.
> >>> I think -erc has been superceded by -chcopy in newer versions of ecasound.
> >>> I'll warn you that I'm writing this from memory and have not tried it.
> >>>
> >>> Best regards,
> >>>
> >>> Aaron
> >>>
> >>> --
> >>> Aaron Heller <he...@ai...>
> >>> Menlo Park, CA  US
> >>>
> >>>
> >>> On 5/11/07 3:58 AM, "e deleflie" <ede...@gm...> wrote:
> >>>
> >>>> jeepers, would you believe my only error was the missing slash at the
> >>>> end of the line!
> >>>>
> >>>> BTW ... I was at first frustrated learning ecasound's syntax .... but
> >>>> now , as I get to know it better, I like it more and more ... in
> >>>> combination with ladspa plugins, its a really powerful tool ...
> >>>>
> >>>> please keep up the work!
> >>>>
> >>>> Etienne
> >>>>
> >>>>
> >>>> On 5/11/07, e deleflie <ede...@gm...> wrote:
> >>>>> Hi,
> >>>>>
> >>>>> I am trying to send a 4 channel file to a ladspa effect, and then
> >>>>> write the 4 resultant tracks as independent files .... I'm having
> >>>>> problems routing the output of the LADSPA effect into 4 separate
> >>>>> tracks ... this is what I have (which seems to break at the loop
> >>>>> business):
> >>>>>
> >>>>> ecasound \
> >>>>>   -a:1 \
> >>>>>   -f:16,4,44100 \
> >>>>>   -i:resample,auto,WXYZ.wav \
> >>>>>   -el:Ambisonics-square-decoder,0,1,1.44,1.44,500,2  \
> >>>>>   -o:loop,1 \
> >>>>>   -a:2,3,4,5 -i:loop,1 \
> >>>>>   -a:2 -f:16,1,44100 -o LFront.wav
> >>>>>   -a:3 -f:16,1,44100 -o RFront.wav
> >>>>>   -a:4 -f:16,1,44100 -o LBack.wav
> >>>>>   -a:5 -f:16,1,44100 -o RBack.wav
> >>>>>
> >>>>> (BTW: that ladspa plugin takes 4 channels in and spits out 4 channels)
> >>>>>
> >>>>> can anyone see what I am doing wrong?
> >>>>>
> >>>>> Etienne
> >>>>>
> >>>>
> >>>> -------------------------------------------------------------------------
> >>>> This SF.net email is sponsored by DB2 Express
> >>>> Download DB2 Express C - the FREE version of DB2 express and take
> >>>> control of your XML. No limits. Just data. Click to get it now.
> >>>> http://sourceforge.net/powerbar/db2/
> >>>> _______________________________________________
> >>>> Ecasound-list mailing list
> >>>> Eca...@li...
> >>>> https://lists.sourceforge.net/lists/listinfo/ecasound-list
> >>>>
> >>>
> >>>
> >>>
> >>
>
>
>