[CELT-dev] celtclient 64kbps more than expected

Paul Sykes paul at paulsykes.co.uk
Thu Jun 4 05:08:48 PDT 2009


Hi Jean-Marc,

Mmm indeed, I should have thought of that.  It's amazing how much the IP
overhead starts to add up.

thank you again,
Paul.



On Thu, 2009-06-04 at 07:26 -0400, Jean-Marc Valin wrote:
> Hi Paul,
> 
> I think the issue you're seeing here is simply the overhead of the IP
> (20 bytes) and UDP (8 bytes) headers (plus our 8-byte timestamp/header).
> With a frame size of 256 at 48 kHz, you send 188 packets/sec, which
> means about 45 kbit/s of header overhead. Depending on how you're
> measuring, you may be seeing the Ethernet overhead of something like
> that. All I can assure you is that when you ask CELT for 43 bytes, this
> is *exactly* what you receive. Does that explain the numbers you have?
> 
> Cheers,
> 
> 	Jean-Marc
> 
> Paul Sykes a écrit :
> > Hi,
> > 
> > The default settings for celtclient are:-
> > 
> > Sampling rate 48000
> > Frame size 256
> > Packet size 43
> > 
> > According to my calculations this should give a bitrate of 64kbit/sec +
> > a little extra for the 32bit time stamp and 4 null byte pseudo header.
> > 
> > Using vnstat I was very surprised to see that it was reporting a bitrate
> > usage of around 16kbyte/sec  which is 128kbit/sec, double what I was
> > expecting.
> > 
> > I then changed packet size to 86 which I expected should give a bitrate
> > of 128kbit/sec, vnstat said I was using 24kbyte/sec which is 192kbit/sec
> > 
> > I then changed packet size to 128, I expected to see a bitrate of
> > 192kbit/sec, vnstat gave me a bitrate of 32kbyte/sec which is
> > 256kbit/sec
> > 
> > Therefore all measurements were 64kbit/sec greater than I anticipated,
> > can somebody tell me why, or I have got something wrong here?
> > 
> > thank you,
> > Paul.
> > 
> > 
> > _______________________________________________
> > celt-dev mailing list
> > celt-dev at xiph.org
> > http://lists.xiph.org/mailman/listinfo/celt-dev
> > 
> > 
> _______________________________________________
> celt-dev mailing list
> celt-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/celt-dev




More information about the celt-dev mailing list