i have recently decided that recordings i make have to be little endian, as the architecture has that byte order.
so i have used [writesf~] sending it a message: open -bytes 4 -nextstep -little -rate 96000 /tmp/recording-XX
after the recording is done, i have a file /tmp/recording-XX.snd, which is not recognised by the file(1) command ..it just says that's a data file ;( i can open the sf in Snd editor, and set header to what i know it is supposed to be - everything works.
it seems like [writesf~] has a glitch with that ..it doesn't write header totaly incorrectly or may be somewhat doesn't write it at all..
it's in the latest version of pd, compiled localy on x86_64, gcc-4.1.2, gentoo linux.
Hi Ilya,
As far as I can tell, Pd write little-endian .snd files OK but nobody else but Pd seems to support them any more. ".wav" files are little-endian and widely supported; perhaps that's a better option.
cheers Miller
On Fri, Feb 15, 2008 at 04:05:26AM +0000, errordeveloper@gmail.com wrote:
i have recently decided that recordings i make have to be little endian, as the architecture has that byte order.
so i have used [writesf~] sending it a message: open -bytes 4 -nextstep -little -rate 96000 /tmp/recording-XX
after the recording is done, i have a file /tmp/recording-XX.snd, which is not recognised by the file(1) command ..it just says that's a data file ;( i can open the sf in Snd editor, and set header to what i know it is supposed to be - everything works.
it seems like [writesf~] has a glitch with that ..it doesn't write header totaly incorrectly or may be somewhat doesn't write it at all..
it's in the latest version of pd, compiled localy on x86_64, gcc-4.1.2, gentoo linux.
-- cheers, ilya .d
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list