Ethernet lighting system quesitons...

SteveB

Well-Known Member
Joined
Mar 20, 2004
Location
Brooklyn, NY
Remember that just because you specify Strand, doesn't mean that a dealer/contractor won't attempt a substitution with another "approved equal" product - ETC, as example. Unless you have existing equipment that must remain compatible and/or have, and can prove a need for Strand/ET, if it's a public bid process, you sometimes have to accept substitutions.

One trick is to list "approved local dealers" who only represent Strand. If someone wants to bid with ETC, they have to find out about it thru public records in a newspapers, etc.. - note that this is all assuming a non-private theater. Trouble is, in most cases, Strand dealers may also be ETC dealers, and while they generally will price out the listed product, they may also recommend, and it's perfectly legal to do so, that an ETC product be accepted as substitute.- especially if they can make a better deal. Most dealers don't bother, especially if they have to do a complete system re-design (read as lots of paperwork) for the substituted gear.

That said, in general the manufacturers don't really want to be sending you old/obsolete gear. They will make the case that's the latest stuff is going to be provided if only that they may well no longer make some of the stuff specified. The longer the time frame between initial proposals and the shipment date, the more likely there will be substitutions.

This was how I ended up with a Net2 system with complete Ethernet, including CEM+ rack CEM's. We knew that ETC no longer made key products - Net 1 nodes as example. By re-designing ( I did the draftings in Vectorworks and sent them to the Theater Consultant and ETC) around an Ethernet wiring distribution, the cost was actually LOWER then the plain old DMX system.

So, I wouldn't worry too much as in the long run, reputable companies and dealers want to make the client happy. An example of this is common public sector contracts for computer equipment. Companies such as Dell will ship the latest stuff they make, especially when an order is spec'd as stuff they no longer make. It doesn't pay them to do it any other way.

If I were you though and if it's not too late, I'd do a re-read of the spec's and add in (if it's possible) wording that all products much be of the current manufacturers product line, not be old stuff sitting on shelves, compatible with all existing manufacturers equipment, yada, yada.

SB
 

SteveB

Well-Known Member
Joined
Mar 20, 2004
Location
Brooklyn, NY
This whole package went to bid over a year ago. The foundation is poured and the walls begin going up soon. At this point any minor change involves tons of paperwork, and expensive change order fees.

Sorry. Missed this post.

My comments stand about the dealers and manufacturers doing the right thing. At this point they will most likely be providing the console specified, as it's a new product. Just for giggles, have you thought of contacting Strand and asking where ACN fits in with your system ?.

SB
 

SHARYNF

Well-Known Member
Joined
Sep 3, 2006
I'm with Steve
One of the things I usually recommend in this type of bidding process is to split out the theater systems from the construction. You can spec the cableing etc, but if you look at the bid process and then the construction process, you wind up bidding the systems almost 3 years before you install. This tends to cause issues with old equipment spec'd, outof date solutions etc. With Strand changing ownership, there is a certain amount of risk in exactly where they will be in a few years.

The change order process is complex, but if the contract were worded to take into consideration these issues, having a review period that would allow with the schools approval to re look at the equipment spec'd makes sense. This is were the construction bid process needs to allow for just this sort of thing on the equipment side of things. I have seen many instances where people get trapped into this sort of thing, and wind up with out dated poorly supported systems, because the industry has moved on during the bid to actual acquisition process. Construction and facilities systems tend to have long stable lives, theater/av systems are on much shorter development cycles.

If you were looking at the safest bet at this point in time, ETC probably would be the lowest risk. Chances are that when you finally are at the point of actually getting this equipment purchased you will have to make changes.

Again, this is a major reason why typically there are different terms and conditions, and time frame for the theater/av systems bids. Construction and facilities
Sharyn
 
Last edited:

SteveB

Well-Known Member
Joined
Mar 20, 2004
Location
Brooklyn, NY
We tried getting a 520 as a temporary replacement for our 300 series and apparently there are none available at all, even in stock storage, it has just been dumped entirely. Strand said they will provide support for 500 series for five years, and I hope Genlyte will help to extend that. Also I would wonder if it is more likely to see ET take over the Strand side of things with all the shownet capabilities and

The Strand website now lists 3 new consoles - the Light Palette series - Classic, Live and VL. These are one level above the Palette series, from the way I read it, but the same basic OS, just more monitors and channels.

Some of the description, including Abstract Control Module terminology, seems to come from the ET Marquee OS, so maybe there's been a bit of blending between the 2 companies in terms of consoles.

SB
 

lightbyfire

Member
Joined
Jul 28, 2005
Location
Northern Virginia
SteveB

Strand did a significant amount of blending with ET after the merger, including incorporating the windows XP system as apposed to Linux, which was the direction before the merger. marquee has also benefited from Shownet compatability. I just wonder if both systems will continue to sell together or whether there is a plan to phase out one or the other.


Also the light pallete series all together are basically a replacement for the whole 500 series. the pallete series is a slight step up but also replacement for the 300 series. Both are running Pallete OS, since they all basically contain the same computer module, which as I understand is easily replaceable (which I see more as an admission that it will have to be replaced, possibly often. Also the floor model of a classic pallete i saw a few months ago was not aging well, when I saw it it had already ripped a few internal connectors just from opening the cover.) Although it looks like all the models have shownet and thus are extremely compatable with the new C21 dimmers, and the website seems to indicate a possible future ACN compatability, although it is not clear.
 

bdesmond

Active Member
Joined
Mar 8, 2004
Location
Chicago, IL USA
Aside from everything else here, you want to coordinate this project with your IT department so that the networks work well for both of you. Ideally you want your lighting stuff on its own VLAN and subnet. If you're going to be spanning multiple switches you'll want Quality of Service (QoS) deployed across all of those so that your DMX traffic is getting higher priority and won't be delayed.

As far as PoE goes, this is easier in general, it's 48V DC on an extra pair in the CAT5/6 cable when you're doing 10/100. Gig is a little different but thats irrelevant here. Don't buy cheap switches and certainly not hubs, they'll just come back to haunt you.
 

SteveB

Well-Known Member
Joined
Mar 20, 2004
Location
Brooklyn, NY
Aside from everything else here, you want to coordinate this project with your IT department so that the networks work well for both of you. Ideally you want your lighting stuff on its own VLAN and subnet. If you're going to be spanning multiple switches you'll want Quality of Service (QoS) deployed across all of those so that your DMX traffic is getting higher priority and won't be delayed.
As far as PoE goes, this is easier in general, it's 48V DC on an extra pair in the CAT5/6 cable when you're doing 10/100. Gig is a little different but thats irrelevant here. Don't buy cheap switches and certainly not hubs, they'll just come back to haunt you.
Personally, I think your suggestion of keeping them separate is good advice, while utilizing someone elses LAN a bad idea:
The lighting manufacturers are going to recommend complete isolation between the lighting network and all other networks. For very good reason. There is no reason to add complexity. There are usually no reasons to want to use portions of an existing - or new, backbone for your lighting system. About the only exception I can think of is the ET Marquee console system that can utilize an internet connection to ET for support and downloads. I'm going to guess that ET then recommends pulling the internet plug.
My experience is that unless you're building a network that requires you to link multiple and separate spaces in differing locations, the lighting network will most likely be wholly contained inside one space, thus doesn't require access to the facility LAN. Exceptions to this are facilities like major film or television facilities where it's desirable for multiple servers to interact between spaces (CNN NYC comes to mind) or huge events like the Olympics or the Super Bowl where running a separate and big lighting network is expensive and impractical when a generic LAN pre-exists.
In general, even big theaters don't operate this way, thus there is usually no requirement for the theatrical user to utilize a larger network. I've also discovered that the IT folks would rather NOT have the lighting network to deal with, as often times the users (us theater folks) know just enough to be a PITA. IT folks also (again - in general) don't have an understanding of the concept of "there are 2500 paying patrons in the theater tonight, thus the network shall not fail".
One thing that might be desirable, would be the ability to have the facility LAN provide tap points adjacent to the lighting network tap points. Our system almost provides for this as we have a 19" rack unit that houses out Unison system as well as the Net2 nodes, E-Net patch bays, switchers, PoE switcher and UPS for the lighting network. I could not get the IT folks to actually wire their LAN into this patch bay, but they provided taps adjacent, thus I can utilize the unused lighting network as needed to route facility LAN if desired.
Bottom line though, and my recommendation is to keep it separate and keep it simple.
SB
 

soundlight

Well-Known Member
Joined
Oct 27, 2005
Location
NJ & NYC
IMO, if you get the extra money for ML's, go with the new Strand Light Pallette consoles. They are really just different versions of the Marquee ILC, if I read the rumors correctly. And the Marquee ILC is a nice console.
 

bdesmond

Active Member
Joined
Mar 8, 2004
Location
Chicago, IL USA
Well, as someone whose primary background is in large scale IT, the last thing I want is a network I don't control parallel to one I do. This is a disaster waiting to happen (and I've seen it happen spectacuarly more than once).

I and any IT guy worth 2 cents ought to understand the concept of stability aka "there are 2500 paying patrons in the theater tonight, thus the network shall not fail". If you have some guy who doesn't get that and know how to deliver, send him to dice.com for a new job and find someone new. Think of corporate networks where 5 minutes of downtime is hundreds of thousands if not millions of dollars in revenue. Your 2500 seat show is nothing.

The basic design I would follow is one or more seperate VLANs for the show networks. This makes the networks seperate but running on the same hardware. I can stick a router or a firewall in and move traffic between VLANs as necessary. I also would have end to end QoS deployed already or would certainly add it in this scenario.

One of the other things that happens when someone who doesn't know how to buy network gear buys it is that they buy cheap crap. It might look good, might claim to be business class, or some other nonsense, but its still cheap. Cheap crap breaks far more often than the good stuff, and it tends to have little to no management capability. Part of the basic services any half ass IT department will provide is end to end monitoring. They will have a system which is constantly polling numerous status indicators and counters on each device, and in the event a threshold is crossed, pages are automatically sent out. If your DMX nodes or whatever support snmp or some other form of remote management, I can easily add them into monitoring system and easily give you and me a rollup view of the health of the whole system.

Bottom line is if you have IT folks who are competent and give a hoot, they will be willing to support and integrate your needs and they will provide value add rather than a hassle.


Personally, I think your suggestion of keeping them separate is good advice, while utilizing someone elses LAN a bad idea:
The lighting manufacturers are going to recommend complete isolation between the lighting network and all other networks. For very good reason. There is no reason to add complexity. There are usually no reasons to want to use portions of an existing - or new, backbone for your lighting system. About the only exception I can think of is the ET Marquee console system that can utilize an internet connection to ET for support and downloads. I'm going to guess that ET then recommends pulling the internet plug.
My experience is that unless you're building a network that requires you to link multiple and separate spaces in differing locations, the lighting network will most likely be wholly contained inside one space, thus doesn't require access to the facility LAN. Exceptions to this are facilities like major film or television facilities where it's desirable for multiple servers to interact between spaces (CNN NYC comes to mind) or huge events like the Olympics or the Super Bowl where running a separate and big lighting network is expensive and impractical when a generic LAN pre-exists.
In general, even big theaters don't operate this way, thus there is usually no requirement for the theatrical user to utilize a larger network. I've also discovered that the IT folks would rather NOT have the lighting network to deal with, as often times the users (us theater folks) know just enough to be a PITA. IT folks also (again - in general) don't have an understanding of the concept of "there are 2500 paying patrons in the theater tonight, thus the network shall not fail".
One thing that might be desirable, would be the ability to have the facility LAN provide tap points adjacent to the lighting network tap points. Our system almost provides for this as we have a 19" rack unit that houses out Unison system as well as the Net2 nodes, E-Net patch bays, switchers, PoE switcher and UPS for the lighting network. I could not get the IT folks to actually wire their LAN into this patch bay, but they provided taps adjacent, thus I can utilize the unused lighting network as needed to route facility LAN if desired.
Bottom line though, and my recommendation is to keep it separate and keep it simple.
SB
 
Last edited:

koncept

Active Member
Joined
Mar 6, 2005
Location
.
i'll second bdesmond on that. it is what I am studying. i do know our high school theatre that has that is on a separate switch but i think that has to do with there being no existing infrastructire in the venue.
 

SteveB

Well-Known Member
Joined
Mar 20, 2004
Location
Brooklyn, NY
BD,
Were you to call Strand or ETC on Monday morning and have them read and then comment on our posts and I can almost certainly guarantee you that they are going to disagree with you.
I have had this very discussion with both David Fox and Turk Wilson at ETC and that's where I learned that they do not recommend utilizing non-lighting system LANs, unless it's a system as I have described previously - I.E. a really big infrastructure, is needed.
This is not to say that it cannot be done, and done well. We've all read the trade journals and seen articles about the Olympics or CNN NYC where the lighting system is part of the LAN infrastructure. Your comment about "having a separate system you don't control is a disaster waiting to happen" is totally false. This kind of scenario is very common, the Santa Fe Opera as example, which is a facility I was very familiar with. I sincerely doubt that the IT folks managing that facility lose a nights sleep over the potential failure of the lighting network, which in any case (and like most similar systems) is managed by the lighting department. That's how my system functions and is one of the reasons my General Manager was more then willing to pay for my attendance at 4 days of ETC End User Training.
But, in any case, the manufacturers are going to tell you that if you do not have to do this, don't.
Note that I did not mean to disparage IT folks, though I can tell you that in many of the applications we are talking about on this site - namely schools, are not high volume corporate environments and in most schools. the IT folks tend to work 9-5, M-F and are not around on a weekends. This is certainly true of my facility, thus there is no way I am going to utilize the college LAN for any purpose, as if it goes down with my system on it, there's no one to call.
My point was that for 90 percent of applications, there's no need to. It costs little to install a separate backbone for the lighting system. Such systems are generally not that big and rarely requires more then one or two switch points
In a way, I'm surprised you would even WANT a lighting network integrated into a general LAN. This is a perfect example where keeping it simple works.
SB
 
  • Like
Reactions: v75vette

bdesmond

Active Member
Joined
Mar 8, 2004
Location
Chicago, IL USA
Hi Steve,

I honestly haven't read the trade journals about the Olympics or CNN or what have you, but I'll certainly take your word.

The IT folks don't lose sleep over the lighting network if they have nothing to do with it. They'll lose sleep when someone goofs up and bridges the two networks and builds a loop and both networks become useless until someone finds and breaks the loop. This is the very scenario that I've seen happen time and time again, and I've seen it take a very long time to find and break that loop with alot of guys that know what they're doing looking for it.

Commercial gear has something called spanning tree protocol (STP) and various derivatives which will for the most part take care of the loops and block one port automatically, but sometimes it doesn't if the switchport is configured in a manner not to do STP or something else is hosed. The stuff that tends to show up in the networks the lighting guys, the video guys, etc build is cheaper and doesn't have this feature. I've fixed a couple of these networks and everytime its been someone didn't realize what they were connecting and bam, got a loop and it's useless, or a hub in the wrong place and now you have half duplex communications and/or a speed and duplex mismatch.

Schools I endorse my plan even more. I spent two years working for one of the largest districts in the country as a senior engineer and escalation point - I was the end of the line for my team in terms of problem resolution. Said district had published standards which every facility (>700) were required to adhere to in terms of equipment, design, and configuration. We provided all of this equipment complete with a rolling refresh schedule to every school. In some cases we could even get additional equipment into places that weren't originally needing it (e.g. a theater). All of this was centrally managed, and it worked great. The network was extremely stable. The problems were almost always the school's fault. Schools buy stuff that's cheap, they hook it up not really knowing what they're doing. The next guy comes along and hooks some more cheap stuff in and the pattern continues.

At this point, we now have helpdesk calls rolling in to the tune of network down, performance complaints, etc. These rarely can be solved remotely which means field resources have to be dispatched, and this costs alot of money. Far more than following the rules would have. Field support is often backed up with issues like this and it might take more than a day to get someone out there and the school isn't working as it should thus interrupting the education side of the picture.

I worked on a few of these very vertical market systems - both video, audio, and lighting while I was at this district and every time I had them fully integrated into the network. This added complexity if you look at it that way cost *way* less than when someone at the school did something stupid and we had to send field support out.

So, that's where I'm coming from. My stance is that I would much rather work with you to meet all of your needs (connectivity, availability, etc) than have you either duplicate or addon to my infrastructure with your own stuff. IT shops are there to meet the needs of the business and if the business needs to run a specialized application on the network (which they virtually all do), that's the job of the IT department to make that application run as it should.
 

cutlunch

Active Member
Joined
Jan 12, 2005
Location
Auckland, New Zealand
From talking to the Strand Rep last week they prefer to keep the desk off a main network. This is because of the loading that could possibly cause latency problems. Of course you can connect a computer so you can use Powerpoint linked to cues. Also use it to print out stuff. But one computer shouldn't really cause to much of a load.

BD when you link up the lighting and data networks do you have to plan it so that the lighting is not being connected to a heavy usage switch /router?

Also what is the necessity of the lighting being on a data network?

With a lighting network once it is setup there should not be many changes. Once the ethernet / dmx convertors have been setup correctly you shouldn't really have to do much with it. Theorectically once setup you shouldn't have to touch it except for the ocaasional bit of preventive maintenance maybe once a year.

I would bring in a data network connection into the theatre on a patch panel. Then you could send it down Cat 5 to the stage if a presenter needed access to the network. You could also use it in the booth to feed images to data projector from the network etc.

I believe in the old adage KISS "Keep It Simple Stupid". If you don't have a burning need to have the console directly connected to the data network why risk trouble.

I would probably talk to your IT support guy just as a matter of courtesy. He may have some tips for you in such things as good design and implementation. You will need at least a hub for the separate legs to the ethernet / dmx decoder. He might tell you which is good and which to avoid. He might suggest you go with the type the school uses. So if it packs up at a critical time you could borrow one from the school.

The main thing I believe is to minimise the number of things that go wrong during the show.
 
Last edited:

bdesmond

Active Member
Joined
Mar 8, 2004
Location
Chicago, IL USA
Hi cutlunch-

To deal with latency issues you deploy what's called Quality of Service aka QoS. Basically you classify and setup queuing on the switches & routers. Traffic is matched according to specified criteria and prioritized. Aytime you have VoIP (Voice over IP) this is deployed to minimize jitter, blips, etc on the call.

Hubs are bad by the way - you want a switch. Hubs are half duplex and one big collission domain - only one thing can talk at any given moment. If two devices try at the same time they both have to stop and execute a backoff algorithm and try again. Switches don't have this issue.
 

RonaldBeal

Well-Known Member
Joined
Jan 24, 2004
Location
TN
EVERY major manufacturer (and I do mean EVERY!!!) recommends keeping the lighting network isolated from other networks. It is so easy to do, doing it any other way doesn't make any sense. You shouldn't need WANs, or other LANs in the dimmer room, catwalks, or any lighting location other than the FOH position. Interconnecting them should be difficult, because access points should be in separate locations.
A combined network violates the most basic tenant of the K.I.S.S. (Keep It Super Simple) You add more points of failure, you add more points of potential network interference, etc...
Once a lighting based Ethernet network is up, it shouldn't need any additional management, (remote or otherwise.)
I've been running major tours with 4 or 5 parallel networks, and NEVER had a problem. Interconnecting them and having to manage switches etc... would add significantly to my workload, and troubleshooting with no measurable gain.
Ronald Beal
Lighting Crewchief
TV, Rock and Roll:


P.S.
RDM does NOT use pins 4 and 5 on DMX, it sends return data down the same data pair as the normal DMX signal. It injects the return data during the DMX break times.
More info can be found from ESTA
RB
 

Users who are viewing this thread