ETCnet resources?

I can see a network would be appropriate. You are close to needing a 3rd universe and in a pure DMX system

I have looked at taking the network approach, but I am just concerned that after I leave people will struggle to use it. If I output sACN on the element and run that cable and place a node at the end, will it convert the sACN to DMX for an 18 year old dimmer rack to understand. I would really like to add more universes and it seems that this is the most viable way without buying a new board. My plan involves cat-5e running from the board to a switch and patch panel, and then from that to 5 different places around the theater and terminating those ethernet lines with gateways. I guess I'm confused about if the gateways convert it to dmx or leave it as sACN.
 
I guess I'm confused about if the gateways convert it to dmx or leave it as sACN.
sACN only exists on the network and wouldn't be/can't be placed on 5-pin DMX cable. David O Smith is right - a gateway does require configuration, but generally the configuration involved is:
- Per DMX port, is it an input or output? (input = listen to DMX, send sACN on the Ethernet network; output = listen to sACN, send DMX on its DMX port)
- Which sACN universe should it convert? (for output, which sACN universe should it listen to and convert into a DMX universe on a port)
- What DMX port speed settings should it use? (just use default unless you have issues. You'll know if issues pop up pretty quickly - thing just won't work.)
- IP address - how should it live on the Ethernet network?

There are, of course, many more settings. This is just the core of it.
 
Just something I came across yesterday. I installed the Eos 2.6.0 update and have been reading up on the notes. As part of the update, "Any sACN universes within the 1-63,999 standard range can now be used. 256 universes total can be used."
(Page 11, ETC Supplement - Eos Family v2.6.0)

This leads me to believe that this is in addition to the 2 DMX universes on board. So really, the Element has the capacity to handle more addresses, as long as you you don't exceed the channel count. This is yet another reason to update use an ethernet infrastructure. Can someone verify or debunk this for me?
 
Just something I came across yesterday. I installed the Eos 2.6.0 update and have been reading up on the notes. As part of the update, "Any sACN universes within the 1-63,999 standard range can now be used. 256 universes total can be used."
(Page 11, ETC Supplement - Eos Family v2.6.0)

This leads me to believe that this is in addition to the 2 DMX universes on board. So really, the Element has the capacity to handle more addresses, as long as you you don't exceed the channel count. This is yet another reason to update use an ethernet infrastructure. Can someone verify or debunk this for me?
It has always been the case that the Eos Family counts the number of addresses used, and not the universes used. The previous limitation was that you could only patch between universes 1-128 (and newer versions allowed you to apply a universe offset so that instead you could patch to a higher range of universes (example - universe 129 thru 256).

With 2.6.0, you now can directly patch to any universe numbered 1 to 63999 without the use of any offsets. Channel 1 can be address 1/1 and channel 2 can be address 63999/512. All fine and acceptable.

As for output counts and how the desk figures it out - you could patch 256 channels, each in address 1 for 256 universes, and the desk will only count 256 outputs as patched, leaving 768 addresses available on Element.

This is a rather extreme example, but it shows the flexibility.
 
As for output counts and how the desk figures it out - you could patch 256 channels, each in address 1 for 256 universes, and the desk will only count 256 outputs as patched, leaving 768 addresses available on Element.
As an MA guy, I'm curious how ETC handles 16-bit parameters - if you patch 256 16-bit dimmers, is that eating 256 parameters (MA style) or is it eating 512 outputs (The Other Leading Brand style)?
 
Just fyi, i have on recent projects put in a small network with a few taps - for console, wifi, paradigm, dimmers if any, plus a few - and din rail 4 port nodes and switch in a bix - and distribute dmx from there to every lighting position on category cable. Centralizes nodes so easier install and service. Much less expensive than distributed nodes. Category wire can always be repurposed for network. Clean, simple, economical.
 
Just something I came across yesterday. I installed the Eos 2.6.0 update and have been reading up on the notes. As part of the update, "Any sACN universes within the 1-63,999 standard range can now be used. 256 universes total can be used."
(Page 11, ETC Supplement - Eos Family v2.6.0)

This leads me to believe that this is in addition to the 2 DMX universes on board. So really, the Element has the capacity to handle more addresses, as long as you don't exceed the channel count. This is yet another reason to update use an ethernet infrastructure. Can someone verify or debunk this for me?

It might be easier to think of it as more usable address numbers as opposed to the actual amount of addresses. The Elements max capacity remains at 1024 actual total outputs which is unfortunate as every other console on the EOS line besides the Element got an actual output expansion with 2.6. The update to the sACN protocol allows those to be "grouped" however the operator wants them to. For instance, fixture 1 on lux 1 starts at 101/1. lux 2 starts at 102/1, upstage cat starts at 201/1, house starts at 1001/1 etc... allowing you to organize however you want with 63,999 universes all on one network protocol. This is really useful if you do a lot of gigs in house with no outside op; or if you are running a tour all you have to do is see the physical location of a light and you know what its address is, but it's pretty much a nightmare for road techs who already have a programmed show and just want to tie in to your rig and update a couple fixtures.

For your average event this is meaningless, as you normally don't use enough fixtures to bottleneck the previous allowance, but in arenas and massive events like the Olympics, this is huge. You can have both organization and massive total outputs, allowing large scale hookups as in as much as 1000 EOS TI consoles hooked up to the same show running different stages or sections or moving lights or someone running moving spots while someone is running front wash. This is extreme overkill but the average live tv awards ceremony has somewhere around 16 board ops and if I recall correctly the London Olympics Avolites setup had over 100 board operators alone, never mind every single seat was pixel mapped... Its really just one more step in allowing you to be creative, mostly meaningless to a lot of us. But ETC is great at just getting out of the way so we as lampies can just be creative with our tools.

From my point of view, it appears that the Net3 Protocol is being broadened to a more open sACN instead of the ETCs proprietary NET3, my opinion is that this has to deal with the acquisition of Hog and allowing those consoles to tie directly into installs with ETCs sACN integrated networks already in place. Again that's just my opinion. But it would be awesome if we can just hook a Hog straight into our rig instead of having to go through another multiport node just to crossover dmx to tie in.

As an MA guy, I'm curious how ETC handles 16-bit parameters - if you patch 256 16-bit dimmers, is that eating 256 parameters (MA style) or is it eating 512 outputs (The Other Leading Brand style)?

EOS patch is the same as MA on this, if its a 3 channel 8 bit LED, its 3 addresses used, if its a 3 channel 16 bit LED it's still 3 addresses used, ETC works with bit depth the same way it does intensity and Hue Saturation, its all handled 'virtually'.


I have looked at taking the network approach, but I am just concerned that after I leave people will struggle to use it. If I output sACN on the element and run that cable and place a node at the end, will it convert the sACN to DMX for an 18 year old dimmer rack to understand. I would really like to add more universes and it seems that this is the most viable way without buying a new board. My plan involves cat-5e running from the board to a switch and patch panel, and then from that to 5 different places around the theater and terminating those ethernet lines with gateways. I guess I'm confused about if the gateways convert it to dmx or leave it as sACN.

Going network on an install is pretty much a set it and forget it setup, so there is no real reason to be concerned about others not being able to use it. In fact it would be best to have it set up and then password protect the whole network... through EOS you can password protect the nodes, plus they hold their settings if they lose power so still no reason for someone else to mess with them once its set up and everything is running correctly, especially at a school where you never know who might try to mess with it.

Simply put a node or gateway is specifically designed to take one protocol and convert it to another.
DMX is an old and extremely weak signal and is limited by distance, there is some debate as to how long you can go but essentially you don't want to go more than 50 feet in a single cable run, and patch more than 15 fixtures on one run. And in my opinion, that's pushing it depending on the type of fixtures you are using.

To solve this problem, as we often need more than 15 fixtures that each need dmx in the LED world, we use nodes. with a node, we can either go directly out of the console through Cat5 to a node 200 feet away and then amplify the signal while converting it to DMX at the same time. this is done automatically at the node, not something you need to setup as you would buy an sACN node or an ArtNet node which would directly convert by from sACN or ArtNet directly to DMX on its own, that's its sole purpose in life to take one signal and turn it into another.

It really is plug and play besides some complexities that could benefit from having an ETC Tech set it up, as far as password and static IP stuff, if everything is ETC brand than the signal running through the Cat5 is sACN (unless someone specifically changed it from factory settings) and if its 3 or 5 Pin DMX cable, than its DMX protocol running down that cable, and that DMX coming out of that brand new node is essentially the same as it was when DMX first came out, so it should absolutely work with that 18-year-old dimmer rack.

In ten years time, I'd like to see Cat5 in and out on all fixtures and eliminate the need for both nodes and DMX cable but, we just aren't there yet. Some fixtures in Robe, GLP, and Phillips offer this but it's just not mainstream yet. Changes to the lighting world happen slow...

Also it's important to understand that the network is all about what your bottleneck is. We've established with Net3 you won't have a bottleneck on the network because of protocol you can pretty much use as many fixtures as you can power. HOWEVER.... you will be limited by what your board can output. You are limited on an Element to 1024 outputs (addresses in use). if you need more than that, simply put you will need another board. You could rent another Element and put it on the same network and simply patch half the network to one Element and the rest to the other Element, or you can upgrade your Element to an ION for a maximum of 6,144 outputs. Simply put I'd say your school has outgrown an Element and should seriously consider an ION. The ION is still the standard in many colleges and performing arts theaters around the world so it would be extremely beneficial for a school to teach their students on one. ETC offers grants for free ION consoles for schools, you can always ask your instructor for permission to apply.
 
As an MA guy, I'm curious how ETC handles 16-bit parameters - if you patch 256 16-bit dimmers, is that eating 256 parameters (MA style) or is it eating 512 outputs (The Other Leading Brand style)?
On Eos, a 16-bit dimmer uses two DMX addresses, so it is counted as 2 parameters/addresses.
 
DMX is an old and extremely weak signal and is limited by distance, there is some debate as to how long you can go but essentially you don't want to go more than 50 feet in a single cable run, and patch more than 15 fixtures on one run. And in my opinion, that's pushing it depending on the type of fixtures you are using.

Protocols don't get weaker as they age, so I'm not sure where you're getting this idea. DMX is still very good at what it does, and it is actually extremely tolerant of all kinds of problems. That's why there are so many marginal installations that still work most of the time (using microphone cable instead of proper cable is probably the biggest example of this).

As long as you're using the right cable and devices that comply with the standard, then you can reasonably have DMX runs up to 500m/1640' (as opposed to 100m with ethernet) and 32 "device loads". Many fixtures use transceivers that are less than one "device load" so you can potentially have more than 32 fixtures on a run and still be safely under the limit. Opinion shouldn't really come into play when talking about well-established standards--especially when the actual document is freely available through ESTA's partnership with ProSight: http://tsp.esta.org/tsp/documents/published_docs.php
 
I wonder if folks who would like fixtures directly on the network using caregory witing stsndards accept dmx using rj45 plugs and receptacles or insist on 5 pin.
 
Protocols don't get weaker as they age, so I'm not sure where you're getting this idea. DMX is still very good at what it does, and it is actually extremely tolerant of all kinds of problems. That's why there are so many marginal installations that still work most of the time (using microphone cable instead of proper cable is probably the biggest example of this).

As long as you're using the right cable and devices that comply with the standard, then you can reasonably have DMX runs up to 500m/1640' (as opposed to 100m with ethernet) and 32 "device loads". Many fixtures use transceivers that are less than one "device load" so you can potentially have more than 32 fixtures on a run and still be safely under the limit. Opinion shouldn't really come into play when talking about well-established standards--especially when the actual document is freely available through ESTA's partnership with ProSight: http://tsp.esta.org/tsp/documents/published_docs.php

I stated that the signal was weak and old. not weak because it is old... as fixtures get updated and the transmission adds along the line, such as adding RDM and revisions to that RDM the signal on the cable gets weaker. Lights are not required to abide by said standards and many of them don't (unfortunately), and thus experience shows that signal drop varies between fixtures in a line, sometimes even between the same type of fixtures. You have to remember that you have a starting signal strength and everything else in the line causes a signal drop... drop it too far before the next amplification than you just amplify noise...

"500 metres should be regarded as an absolute maximum. 500m cables should work with conventional DMX systems but experience with RDM-capable DMX systems has shown this to be not always true. RDM circuits present double the termination load to a transmitter. Some types of transmitters operating in certain conditions may not be able to reliably function on a 500m run. The recommendation is now to limit lines to 300m length, particularly if they feed multiple devices." http://tsp.esta.org/tsp/documents/docs/DMX512-A_Guide_(8x10)_ESTA.PDF pg.17 That was back in 2008, since then there have been revisions to the RDM further limiting the strength/ noise of the signal.

In an install, I would have no real issues going 300m in DMX but outside of that... no way. Cables don't always get cared for properly (especially when you don't always get to control who handles it), they get ran over by subs, snagged here and there, and then you have that rare occurrence of actual frequency interference. No thanks... I need my gear to work, and there is no way I'm going back to 100m runs for DMX line. But hey if you just want to listen to the standards and assume all is well more power to you. The standards are important, however, they are more of a guideline for upper limits, SAFETY (which should always be heeded) and installs. Outside of safety, nothing beats good ole experience.

Hopefully that explains what i meant.
Cheers.
 
In a house with a rep or stock plot where fixtures seldom move the RJ45 connector would be fine, particularly an ethercon connector. I'd be concerned about longevity in most other instances. That little plastic locking tab is pretty unreliable.

proxy.php


That was my hesitation at the beginning too, but when I reluctantly switched to all Cat5 for all my speakers, and snakes for outdoor events... (not my preference) I was happy to know that the female end was shallow enough that it didn't use that plastic lock, instead it locked like a mic cable on the metal shield itself. That made me happy :)

So far two years of outdoor gigs with no issues.

I did, however, have one instance on a mover using ethercon where it locked into both the plastic lock and the metal lock... that was fun to get out. needless to say, the plastic clip cable side lost.



I wonder if folks who would like fixtures directly on the network using caregory witing stsndards accept dmx using rj45 plugs and receptacles or insist on 5 pin.

Personally, I would like simplicity, If dmx is running down the line, I'd want it to be 5 pin, and network (non dmx protocol) to be cat. if nothing else then for ease of knowing what signal is where in case i need to trace an issue.

Plus I don't have a tester to verify dmx through an rj45, only through 5 pin, but hey whats one more tool to carry... then again, I think I could use ETCs Concert at the fixture end to verify over the network....

For now, I'll stick with 5pin in the extremely high case I need to slap a fixture that doesn't have rj45. if it was more common than it'd be easier just to order bulk and use it for everything... audio, general network, and lighting; removing the expense of nodes and opto iso's. But I am looking forward to the eventual day when I only need to carry two kinds of cable to a gig, (cat and power).

Now if only i could figure out how to plug that sm58 into a cat5 then I'll be happy ;-)

So I guess the 3 pin will have to stay
 
If youre going to be plugging and unplugging more than once or twice a year, ethercon for network and 5 pin for dmx, both with a ruggedized cable like proplex. If like many high schools fixtures are installed and never moved, I think im fine with RJ45s.
 
I stated that the signal was weak and old. not weak because it is old...

Some of what you're saying has merit, but your basic premise is still wrong: Ethernet is more limited by distance than DMX, and DMX is actually a "stronger" signal both in terms of the actual voltages on the wires and in how tolerant it is of any problems along the way. In part this is precisely because it is older and slower (since higher speed signals are more picky about how they're being carried). It's true that RDM has some additional requirements over standard DMX, but you only need to worry about those requirements if you actually plan on using RDM. Devices that don't follow the standard are going to be a problem no matter what protocol you use to talk to them, and likewise Ethernet cables that are abused are just as likely (if not more likely) to fail as DMX cables. Neither of those issues are points in favor of one protocol over the other.

Obviously it would be silly to blindly assume that everything will magically work in all cases where the standard says it theoretically could, but it's just as silly to make equally blind assumptions that something won't work just because it's old. DMX is better at some things than Ethernet. Ethernet is better at some things than DMX. The standards for both act as a guide on how to make the best use of whichever one fits a particular application.
 
Or a typical early generation DMX system and its equipment installed in a theater is weaker than a typical new network system for lighting in a theatre because.....its 20-30 years older. I have both in a lot of projects and I'm certain with both installed at same time and of of similar quality, the DMX will be working longer with less maintenance than the network.
 
I'm surprised that no one has mentioned the one limitation of networked ethernet fixtures that would cause cascade failure...

DMX has In and Thru. If your fixture dies, even it it losses power, the signal still passes trough to the next fixture. This is not possible with ethernet. Even if you did get a device with a built in two-port switch, if it loses power, the switch is done. It won't pass data. Now your whole rig has been compromised.
 
I'm surprised that no one has mentioned the one limitation of networked ethernet fixtures that would cause cascade failure...

DMX has In and Thru. If your fixture dies, even it it losses power, the signal still passes trough to the next fixture. This is not possible with ethernet. Even if you did get a device with a built in two-port switch, if it loses power, the switch is done. It won't pass data. Now your whole rig has been compromised.
If your opto fails you're in the same boat. It is one drawback to switches built in to lights, but you can buy a lot of external ethernet switches for the price of an opto, and if it fails it's much easier to source a network switch locally.
 

Users who are viewing this thread

Back