Hi everyone,
The little endian/big endian theory is definitely the problem, since I
converted the dat file on my linux to big endian and it then played as
expected on my ibook. It is a rather simple c swap float code that
compiles good on my linux. Unfortunately, when I tried to compile my
linux c code on my ibook I get an error or with minor adjustments, I
get a bus error after starting the conversion. I will have to look
into that.
As far as I know, one would need to make an endian sniffer which would
either read a header in the file or decide on its own how to
read/convert the file.
I will look into this as time permits. Seems to be a good introduction
to gnuradio. This hasn't dampened my interest in gnuradio; on the
contrary, it has only excited it. Thanks for the timely feedback and
ideas.
Josh
On 8/17/06, Eric Blossom <[EMAIL PROTECTED]> wrote:
On Thu, Aug 17, 2006 at 03:36:41PM -0400, Michael Dickens wrote:
> Jan - Thanks for adding weight to the "endianness" theory. Back
> again to the previous question: Does the .dat file format somehow
> internally specify the data endianness? If so, then it would seem
> simple to write a block to handle this. If not, not so simple. - MLD
>
> On Aug 17, 2006, at 12:42 PM, Jan Schiefer wrote:
> >I just tried this with rev 3340 on an Intel Mac, and I hear the guy
> >talking. This would confirm the endianness theory.
There are no headers on the files read or written by
gr.file_{sink,source}. You are free of course to create another set
of sinks and sources that perhaps encode the data type and
representation in a header.
Eric
_______________________________________________
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
