E1.20 RDM (Remote Device Management) Protocol Forums

E1.20 RDM (Remote Device Management) Protocol Forums (http://www.rdmprotocol.org/forums/index.php)
-   RDM General Implementation Discussion (http://www.rdmprotocol.org/forums/forumdisplay.php?f=4)
-   -   Broadcasting to a device model (http://www.rdmprotocol.org/forums/showthread.php?t=194)

sjackman December 14th, 2007 09:15 PM

Broadcasting to a device model
 
I want to broadcast a SET command to all devices of a specific device model. Broadcasting to a specific manufacturer is easy of course, using the manufacturer-specific broadcast address of mmmm:ffffffff. I suspect other manufacturers have run into this question

A model-specific broadcast address of say mmmm:dddd:ffff, where dddd is the model number, could have been quite useful.

How then to select a specific device model? My two ideas at the moment are to abuse the SUBDEVICE field, which is 16 bits, to encode the model number, or to use a manufacturer-specific PID and then include the desired model number and actual PID in the parameter data. One downside of the latter solution is that it reduces the maximum length of the parameter data by four bytes. I'm leaning towards using the SUBDEVICE field. The RDM spec limits the range of the SUBDEVICE field to values 0-512. That leaves plenty of bits to encode a model number.

Any other suggestions?

Cheers,
Shaun

sblair January 20th, 2008 07:04 PM

Shaun,

Using the Sub-Device field to accomplish this would likely break a number of things as it was not intended to be used in this way, and there are a number of expectations of the specific behavior of Sub-Devices in the Standard.

The intent with the 32-bit Device ID is to allow manufacturer's to encode that Device ID any way that they choose and there is certainly the capability of encoding model ID's as part of your Device ID without breaking anything.

sjackman January 21st, 2008 12:10 PM

My preferred solution would be to encode the model number in the device ID as mmmm:ffff:dddd. However, since we're talking about a broadcast packet, I think it's important that the packet be recognised at the addressing level of the protocol as a broadcast packet, so that repeaters and other infrastructure behave correctly. mmmm:ffff:dddd would not be a broadcast address, as only xxxx:ffffffff are broadcast addresses.

For comparison, Ethernet gives each manufacturer 2^24 manufacturer-specific broadcast addresses, in the form mm:mm:mm:xx:xx:xx, where the least-significant bit in the first mm byte is set to indiciate the address is a broadcast address.

Cheers,
Shaun

sblair January 21st, 2008 01:30 PM

Shaun,

Yes, you are correct. While it would work with your own equipment, it could possibly confuse other in-line devices that are expecting a response and never get one (since it is a broadcast address).


All times are GMT -6. The time now is 12:43 PM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2020, vBulletin Solutions, Inc.