Single fixture type per DMX run

[twiceonsundays]

I've recently been introduced to the idea that one should run a speperate data line for each fixture type in a rig to avoid data issues. Meaning that if you hung a pipe as follows

Mac 2k .... Luster .... Mac 2k .... TW1 .... Luster .... TW1 .... Mac 2k

You would have three lines of data running down the pipe, 1 for Mac 2ks, 1 for lusters, and 1 for TW1s.

I've never heard this before, and I'm extremly skeptical. Is there anything to this idea? It seems to me like it might be a hold over from back in the early days of the DMX protocol when some manufacturers were using the second twisted pair to send non-DMX data between their controllers and equipment?

 

[Erik456]

Never heard that before, nor done it. It would sure make for some stupid cable hods. I have never had ANY issues mixing families of lights on one line.

 

[theatricalmatt]

Occasionally, on lower-end or older equipment, you run into issues where a unit takes in DMX, but spits out its own flavor of control signal that other lights misunderstand. If that's the case, it makes sense to make a dedicated run to each type of equipment.

But that being said, on many larger shows, I split universes based on position, so universe 1 might be (for example) everything FOH, universe 2 the 1E and 2E, universe 3 the 3E and 4E, and so on, depending on the capabilities of the board. Even when I'm not filling an entire universe, it makes sense, and makes it easier to track down some issues. Most modern equipment can handle this with no issues.

 

[Jeff Lelko]

I've never heard of doing this either. I'll occasionally split by position as well, but never by fixture type. Once in a great while I'll come across something that doesn't play well with other lights (even of the same type), but I find that quite rare and just deal with it on occurrence. One asterisk to this is if you're using older lights that may need to be phase reversed. In that case, yes, it might make more sense to split by fixture type so that you don't need dozens of phase reversers, but otherwise no.

 

[techieman33]

I've done it with Chinese fixtures to keep them separate from the rest of the fixtures. They sometimes have weird things going on with them that can give other fixtures issues. I've seen plenty of data runs split by fixture type, but it's usually because they are each on separate universes.

 

[MNicolai]

The "belt and suspenders" approach.

May be of benefit for someone with low-end equipment or low-grade cables/infrastructure. For an already sensitive system, compartmentalizing the DMX distribution by fixture could make for easier troubleshooting, particularly if there's a potential for a few lower-grade fixtures to disrupt the entire rig. Not really an issue I would expect with Martin.

Sounds a little like someone who started their career on lower tier systems, developed some superstitions that were well-founded at that level, and hasn't adapted to working on higher tier systems.

 

[JD]

Not a bad idea, just a pain. Some low-end fixtures don't play well with others, but may play well with each other. Just a guess.

 

[danTt]

DMX passthru is entirely passive in every fixture I've encountered, so the idea that a lower end fixture might modify the data on it's way through in such a way that it would effect other fixtures is strange. If it "damaged" the signal, I certainly wouldn't expect chaining multiple of the same type together to be forgiving to damaged DMX, I'd expect it to compound the problem.

The one possibility i can think of where this might be useful is if you have a group of fixtures that are RDM capable, and a set of fixtures that are not. If you cable them separately you'll be able to use the RDM features on the fixtures that support it without causing flicking and issues on the older fixtures.

 

[Colin Bishop]

DMX passthru is entirely passive in every fixture I've encountered, so the idea that a lower end fixture might modify the data on it's way through in such a way that it would effect other fixtures is strange. If it "damaged" the signal, I certainly wouldn't expect chaining multiple of the same type together to be forgiving to damaged DMX, I'd expect it to compound the problem.

The one possibility i can think of where this might be useful is if you have a group of fixtures that are RDM capable, and a set of fixtures that are not. If you cable them separately you'll be able to use the RDM features on the fixtures that support it without causing flicking and issues on the older fixtures.

I think on cheap fixtures it can, something on my DMX line caused my Mac 250 entours to lamp off during service, pretty sure it was my cheap Chinese led pars.

 

[TNasty]

There's only two instances where this would make sense.

One being what other people here have said about lower end fixtures.

The other deal being on High End (company, not quality) fixtures, whenever you're attempting to flash their firmware.

 

[RonaldBeal]

In the infancy of DMX controlled moving lights (early 1990s) this was a fairly common practice, because some combinations of movers would not play nice together. Martin mac500s did not like sharing a line with high end dataflashes; Intellabeams and Goldenscans didn't like to share a line, etc, so was common practice back then. As technology has improved, the need for separate lines has diminished.

I have had to do it recently. I found that SGM XC-5s and Martin Mac Vipers don't play well together. When the XC-5 strobes, the crappy powersupply puts noise on the DMX line (as viewed with an oscilloscope) and causes the Vipers to twitch. The more the XC-5s strobe, the more the Vipers twitch. Putting them on separate optically isolated data lines by fixture type solved the problem. (It also works well on PRGs Series 400 system, because the DMX nodes monitor the line and errors cause the DMX output to stop, and start a new packet, thus eliminating the twitch)
RB

 

[MikeJ]

I've had older LEDS that have standalone master/slave modes, that were supposed to be overridden by DMX cause problems with still sending Slave data and causing things to twitch or blink down the line. It's not very common these days though.

 

[JD]

I've had older LEDS that have standalone master/slave modes, that were supposed to be overridden by DMX cause problems with still sending Slave data and causing things to twitch or blink down the line. It's not very common these days though.

More common than you would think. The fixture is only suppose to listen NOT speak, but there are a number of "non-DMX" features in some of these Chinese fixtures, usually for their own purpose and generally their own fixtures are immune to the nonsense. DMX opto-splitter also works like a one way valve, information can only come in the input and go out each of the outputs. Any weird stuff is kept from branching into another leg of the system.
Should it be? NO. Is it? YES.

 

[FMEng]

I can think of one reason to do that: If you have mixture of instruments, some with 3 pin DMX, and some with 5 pin DMX, a run for each might keep things a lot simpler without the need for a bunch of adapters.

 

[Jeff Lelko]

The fixture is only suppose to listen NOT speak, but there are a number of "non-DMX" features in some of these Chinese fixtures, usually for their own purpose and generally their own fixtures are immune to the nonsense.

True, and I've actually had a Chinese fixture shock me through a DMX cable. Things would certainly be much easier if the fixtures actually just listened and didn't broadcast. Even still, I just deal with these situations on occurrence, otherwise even a basic system would require a tremendous amount of cable and other hardware that in most likelihood isn't necessary.

 

[SteveB]

I've a mixed MAC 700 (6) and MAC Aura (9) electric, all daisy chained off 1 DMX. Done a lot of rental rigs with mixed manufacturers with daisy chained off same DMX run, no issues.

 

[cbrandt]

I keep this one as a back pocket problem solution. When all else fails, isolate things down to find the problem. I generally only plan on doing it when I'm doing installs with architectural grade fixtures, or low end stuff. You see the bad dmx/modify dmx problem a lot more often on fixtures intended for permanent installations. I've seen it on a couple brands of chalice or house style lights that really need to be on their own. Most often it comes down to bad internal wiring choices on routing the dmx inside of the fixture.

 

[Lightguy5]

One other instance of it being needed:
We have some older Martin gear with the data polarity reversed. Separate line with only one custom polarity reverser at the head.