Profile photo of pngaudioguy


Originally posted by mervaka

Do these controllers that you guys are proposing all support NRPN messages? I don’t see any other way in which the MIDI protocol could cope with the extensive addressing requirements. Metering feedback will undoubtedly not be possible via MIDI, as it lacks the bandwidth required.

Quite the contrary, sir. MIDI system exclusive commands (or sysex for short) are fully capable of addressing every parameter on the console, including metering. Bandwidth is adequate for metering 24 channels, dependent on the hardware capabilities. I personally haven’t tried to push it beyond that.

For an example of the level of control you can achieve to include metering, check out the Yamaha LS9 midi spec sheet (http://download.yamaha.com/file/47461 ~295Kb .zip file) – definitely not an A&H link! My Lemur template makes use of this, and the 01V MIDI and DM1000/DM2000 MIDI for control of those consoles. The 01V doesn’t quite handle metering smoothly, the others do fine. Actually, the MIDI spec for 01V/DM1000/DM2000 is nearly identical. They changed metering data to a single byte in their newer series of consoles (older ones were 10 bit resolution, which requires 2 bytes of data to convey hence more processing power which I believe the 01V lacks.)

I’ve done some limited reverse engineering via wireshark of the commands that A&H live editor is sending to the rack. There is a 13 byte message starting with F0 and ending with F7 (standard sysex start and end of exchange messages), but they don’t correspond to the specs they have posted. The rack then responds with a series of 2 or 3 messages depending on whether you have channel window open in the editor, which the editor sends an “ACK” packet to. The first response message from the rack also activates the MIDI “in” LED on my midiman when I adjust a level, and the data packet of that message terminates with the standard NRPN message as described in their spec page (which does not match the level range of 0000 to 8a00 that the editor sends, btw – 35329 point resolution – as NRPN is limited to a single byte with range 00 to FF – 256 point resolution.)

Attempts at sending the same messages via MIDI result in a “received end of sysex message, but no start of message was found” error in the editor. I haven’t been able to determine if some sort of handshake takes place between the editor and the rack to “authorize” these modified messages, but it appears as though everything the editor sends to control the rack is just a MIDI sysex message. This is consistent with how A&H describe their TCP/IP protocol in that spec sheet.

There’s one 16 byte and one 27 byte message every second as well, the 16 byte appears to be a timecode of sorts, and my hunch is that the 27 byte is metering that gets averaged in the app somehow, but I’ll have to do some more testing to see if I can figure that out and make use of it.

I’ll be sure to keep you all posted if I am able to make any more headway, but it will be a good month before I have the rig off the road to bench test some more.