More and more clubs are going digital. When you go out to hear a band, they’re plugging into an ADC (analog-to-digital converter) box on stage, and the digitized audio data is transmitted to the mixing console over Ethernet. This saves the venue having to run many audio cables over long distances, but it’s a lot harder to hack on. So [Michael] trained popular network analysis tools on his ProCo Momentum gear to see just what the data looks like.
[Michael]’s writeup of the process is a little sparse, but he name-drops all the components you’d need to get the job done. First, he simply looks at the raw data using Wireshark. Once he figured out how the eight channels were split up, he used the command-line version (tshark) and a standard Unix command-line tool (cut) to pull the data apart. Now he’s got a text representation for eight channels of audio data.
Using xxd to convert the data from text to binary, he then played it using sox to see what it sounded like. No dice, yet. After a bit more trial and error, he realized that the data was unsigned, big-endian integers. He tried again, and everything sounded good. Success!
While this is not a complete reverse-engineering tutorial like this one, we think that it hits the high points: using a bunch of the right tools and some good hunches to figure out an obscure protocol.
I worked as a sound engineer a few years back and the big “snake” in the auditorium was about 100m long! The thing weighed a ton and was unwieldy to move around. Would have loved to have had something like this!
The cable cost quite a bit of money too, was something like $500/$1000 total.
90% of the work I do now is audio, and it’s interesting to see how things are changing. For the most part, live audio ten years ago was still operating with 30 year old technology. It makes sense, considering you’re dealing with analog signals. However, it seems all at once, live sound has made a shift to digital, along with stage lighting and projections. Digital soundboards can do the job of a room full of rackmount gear, and better yet, automate it all.
I worked the load in for Carrie Underwood a little while ago, and all the cable we used, bundled together, would have been about as thick as my wrist. A 16 channel snake alone is that thick, not to mention many times more expensive per foot. Lighting is doing the same, with LEDs replacing energy-hungry Tungsten. a 1kW PAR 64 has less punch than a 38 watt Slim Par Pro. Even better, there isn’t a need to lug around a ton (literally) of dimmer racks and stagepin or socapex cable.
It’s an exciting time in this field.
No mention of the protocol, I wonder if it’s their own proprietary blend, or they’re just being coy. There’s really no reason to come up with yet another standard, AVB & Dante are both pretty great. I’ve been deep in the ‘digital audio over ethernet’ weeds on a pretty complex project lately (mixed sample rates & protocols, lots of channels). BSS has a great overview of the various formats for anyone interested:
http://d18nzrj3czoaty.cloudfront.net/site_elements/resources/306_1371661657/Architectural_Media_Systems_Guide_To_Audio_Networking_-_white_original.pdf
Proprietary AFAIK. Entertainment industry loves proprietary protocols.
Not really true at the moment, there are a lot of companies using the standards (Cobranet, Dante, AVB etc), and various products to convert between them digitally. There are several standards, but hopefully now AVB is a stable standard it will meet enough needs to become ubiquitous.
Most companies realise that once you get past undemanding applications like 8-16 channels of digital audio, it’s better to let someone competent deal with the complexities (QoS, handling errors etc etc), and they can add their unique features.
It’s much like connectors (only way more complex!). Most microphone and balanced audio connections are now XLR, but that certainly didn’t happen overnight, it just became a ‘no-brainer’ even if it’s not perfect.
Are there any products in the market actually using AVB yet? (Not making a point, I genuinely don’t know =P)
I suppose I meant proprietary in the sense of owned by certain companies – Dante and Cobranet have had good success but must be licensed from Audinate and Cirrus, respectively – as opposed to IEEE, etc. standards like AVB or ACN, which would be nice to see but don’t seem to have been widely adopted. Hopefully they will, given time!
My money’s on Dante and ArtNet (both proprietary) to be the most popular for the foreseeable future, though.
AVB is cool and all but without support for IP routing it has its limitations for larger designs. I guess you could do VLAN trunking to get the media across a facility. You would need to translate to one of the other standards that is routable to go between facilities though.
Michael, do you by any chance have a recollection and/or the code / protocol / anything from back than? I am trying to do the same, with no luck so far, I have yet identified how channels are being transmitted. So far what I can see is that there are 144 bytes of data. which is probably 6 samples of 8 channels in 24 bit (3 byte) aka 3*8*6=144. I have tried few regular layouts such as “…….” and “<ch2…" or even byte encoding. with big and little endiness and signed/unsigned encoding. but non have been successful yet. Any help would be appreciated.
Anyone know how he got the data from tshark into excel? Unless he just did it by hand, it would be a neat trick to know.
I think he just used space/tab-separated text and imported that into Excel.
http://stackoverflow.com/questions/15112389/tshark-export-packet-info-from-pcap-to-cvs
Text to columns, fixed width, like a muppet =P
I remember my parents talking about the demise of the big-bands, now I seem to be seeing the demise what’s left.
A while back we had a short lived fest called Layflats. The grounds had several stages around including the Tippecanoe amphitheater. The loudest most aggressive sound at the whole fest was coming out of a digital front ended board. It sounded like a loud MP3. If it ain’t High Def or higher don’t even waste time. Engine at 32 bit with no delay!
Don’t get me wrong I play with [rakarrack] in surround.
Cobranet is a well-known digital snake protocol; it’s simple, and the details are available on the Web. I’ve written my own Cobranet source on an STM32F4.
Didn’t know details were readily available!
Re: STM32F4 code – that’s awesome! Kudos! Is it online anywhere? :)
I have never been a hackaday snob, but is “ADC (analog-to-digital converter)” really needed?
If someone isnt aware of what ADC stands for, should they be on HaD?
Or even this person exists, shouldn’t they google it?
If this is what this site is becomming, I am disappoint.
tl;dr
Set off my lawn, no good kids
They didn’t explain some similarly basic TLA a week or so ago and people complained. Can’t please everyone.
People who don’t know what ADC is still should to be on HaD – maybe they will learn something, maybe they just don’t care and are here only for mechanical hacks. And if they don’t care, they won’t google it . This way they might at least learn by osmosis.
SOrry, was knackered yesterday. Shifts moving at work ” should they be on HaD” wasnt what I was going for – “I wouldnt exect them on Had”
either way, Im sure I was just being cranky
As long as they don’t dumb down the information than absolutely. A few extra words is absolutely nothing to complain about. If they start leaving out words, or summarizing because of readership, well then fuck that
As long as they don’t dumb down the information than absolutely. A few extra words is absolutely nothing to complain about. If they start leaving out words, or summarizing because of readership, well then f that
(Didn’t mean to cuss, you can remove the last post admin)
Listen in a DANTE Audio Network is not really difficult: Lyrebird APP plays uDANTE:
http://www.tjaekel.com/udante/
Video Demonstration: https://youtu.be/1cR100Hs-Xw