Ron Hardin <rhhardin@mindspring.com> wrote in message news:<416A5504.AA6@mindspring.com>...
> Take a collection of good real audio files and put their first N bytes
> ahead of what you've got, and increment N until it plays for each one.
> 
> You can look for strings in the good files to see where it's a good
> place to start N other than 0.
> 
> You don't need to use the whole file that you've got since you just
> want to see at what N it starts to play.

that's a good suggestion, but it will take potentially N*M trials
(where M is some chunk of the headless file - i think you are right,
the whole file is not necessary to find N)

i good this programatically, then and maybe use rmfix every iteration
to see if it complains.  the only problem is that i believe the header
info contains a bunch of specific decoding instructions for that
particular file - number of 'packets'/'frames' or whatever, sizes,
etc. and it seems like just taking a random head from a good file and
plopping it in the bad file won't work out.

does anyone know the RM format well enough to comment on this?
also, what if it turns out its another format?  any way to figure this
out?

Take a collection of good real audio files and put their first N bytes
ahead of what you've got, and increment N until it plays for each one.

You can look for strings in the good files to see where it's a good
place to start N other than 0.

You don't need to use the whole file that you've got since you just
want to see at what N it starts to play.
-- 
Ron Hardin
rhhardin@mindspring.com

On the internet, nobody knows you're a jerk.

hi all,

i've got a ~10MB file which i think is a realmedia audio file (because
the filename ended with .rm and i found it with other realmedia audio
files) but the index header is missing.  it all seems to be just
"data".

could anyone give me a clue as to how to:

1-identify what format this raw data is in?
2-plop a header onto it so i can use it?

any experts out there?