E682 using multicast but ran out of universes. Would moving to indicate give me more channels as I can patch to More universes? What are the pro and cons of each. I can add another controller and may need to just to save cable lengths but wanting to understand the network side better if someone can explain in simple terms.
Thanks!
Sent from my iPhone using Tapatalk
This has been written about many times, so for other's experiences and opinions, just do a search of the forums and the wiki.
To put it in a nutshell:
1. Multicast is easier to setup, according to some.
2. Multicast takes more processing power at the controllers (searching the stream for the data intended for that controller, then acting upon it).
3. Unicast takes a bit more during setup in the form of telling the sending program/device where each of the data bits go.
4. Unicast takes less processing power at the controller, freeing up room to allow more channels to be used on that controller.
I have always used Unicast and it has been flawless. There are those that have used Multicast and say the same.
The number of channels available at any one controller is the selling point for me. That, and the lesser amount of traffic over the show network as a whole. (No "broadcasts" to all devices, but point-to-point communications only.)
algerdes wrote:That, and the lesser amount of traffic over the show network as a whole. (No "broadcasts" to all devices, but point-to-point communications only.)
I have always used unicast and have had no problems, but I'm not sure if you are correct on this point. I think multicast uses udp and unicast uses tcp, so unicast may have more traffic because the destination (controllers) are sending received packets back where in multicast they don't.
It would take some time to setup, but I'll try to setup (as in unpacking the controllers and the comm wire) a test show network and put some data online. Will try to measure the same sequences hooked up via a Unicast setup and a Multicast.
I thought there was already something up here on the forums with empirical data supporting network traffic of both methods, but I can't find it. Perhaps it is on one of the other sites. If anyone knows of this (or any like data), it would be nice to know.
Once again, it will take several days before I can put this together (that "day job" thingy getting in the way - again.)
Thanks all,
I will convert over to unicast tonight and go for more channels as it seems for most there is really no difference between the two formats to a pixel. Network guys will have preferred methods but I like more channels at the expense of a few extra minutes of setup and mapping
Sent from my iPhone using Tapatalk
