rtpsend: "Line ???? is invalid" about rtptools

[njh]

Hi,

I have been using rtpdump with an AES67 compatible Dante device:
https://www.aelius.com/njh/posts/0002

The error Line {RIFF???WAVEfmt } indicates that you are trying to send a WAV file. The rtpsend command uses its own special text based file format - as output by rtpdump -F ascii. And rtpplay uses the rtpdump -F dump file format. The dump file formats contain timing and RTP packet information, not just the raw audio samples / payload.

Unfortunately the official rtptools website is offline at the moment (see #135) but there is a copy of the page documenting the file format here:
http://www.cs.cmu.edu/~./libra-demo/rtptools-1.17/rtptools.html

You would need some kind of tool to convert the WAV file into RTP packets before it could be sent by rtpplay / rtpsend.

Maybe your Omnia One will be more tolerant than my Dante device, but I suspect that you may run into trouble with getting the AES67 device to accept RTP packets generated by rtpplay because of clocking problems. You will have to make sure that your system clock is synced to the Omnia's clock (using PTPv2) and that the RTP media clock matches.

I really wish there was more Open Source tooling available for AES67. I keep meaning to spend some time on it... what is you end-goal?

nick.

[Tratosca]

Thanks for this detailed answer ! As a student in audio & video engineering I really like to try to achieve some pointless goals. I call this pointless because a Livewire driver does exist for Windows, linking the entire Livewire network and up to 32 virtual Windows WASAPI interfaces, however this driver is proprietary. And as a macOS/Linux user I’d love to create a way to « talk » to the Omnia both ways without having to boot a Windows VM (but as I said the main idea is to elevate my knowledge). I talked to one of my teacher about this challenge some weeks ago and he also told me about clock issues (I achieved to send some audio using SoX and VLC ‘sout’ but audio was really stuttered, hardly audible. Vu-meters were all in red). I will continue to experiments and will have a look into clock synchronization. Don’t know where to start, but I will start. I don’t know much how RTP does work, I guess this is where I have to come to first.