Ethernet - Why?

[Grum]

Recently I have noticed that a lot of people are upgrading their DMX networks to ethernet.

I work for a projects company and whilst we do install ethernet networks we always run in a DMX network as well. This is because at the present time there is no universal ethernet based lighting control protocol. True enough all manufacturers produce their own ethernet to DMX converters but....

 

What happens when one of these converters dies just before a show?

 

Where do you go to get a replacement quickly?

 

I am not trying to hold things back and I know that DMX does have its limitations but it seems to me that we are just getting ethernet because its there. My concern is this, before DMX each manufacturer had thier own control protocol so if you wanted a strand desk you would have to have strand movers or a converter. Then came DMX suddenly everything talked to everything else this was great. I could use ADB dimmers with a compulite desk and Martin moving lights.

 

This situation is not so good if you make the good desk that everyone wants but your intelligent fixtures are not exactly the best, it means that you are no longer going to sell so many of your intelligent fixtures. By opting for ethernet at this early stage are we not running the risk that in a couple of years we will only be able to use either ethernet devices made by our desk manufacturers or that antiquated DMX stuff that no one makes any more?

 

There is also the question of converting the signal from ethernet to DMX. If your plan is to have ethernet points around your venue that you plug your converter(s) into then you are actualy limiting your networks flexibility, as in most cases in this country venues don't have the money for too many of these "nodes". Plus, I am not to keen on the idea of putting something in the way of the signal from my desk to my dimmers (at least) that has the potential to go wrong, why take the risk?

 

If you are installing a new control network then yes get the cabling run in for ethernet, it is the way of the future and I know how hard it is to get the money to have it done a couple of years after you just had that expensive refit. But unless your using more that say 4 DMX universes in a regular size venue is it not better to spend your money on producing a better and more flexible DMX network?

 

In short do you want ethernet because its the latest toy, because you supplier / consultant is telling you you do, or because it actually offers you a cost saving / increase in control over your network?

 

Like I said I work for an installation company and we do install ethernet networks as an add-on to the DMX but I am interested in your opinions on this. Do you realy want ethernet or are you just being told you do?

 

G

[Guest lightnix]

It could all have been so different, if only Vari-lite Inc. had chosen to license the Series 200 system instead of being all precious and protectionist about it. That way we could all have had two way data comms back in the 80s and they could have become the Micro$oft of stage lighting, with a little circuit card that said "VARI*LITE®" in each and every moving light and desk.

 

OK, I know that (as an protocol created in the 80s) it's be a little creaky by todays standards, but it would still be streets ahead of DMX (or RS485 as it maybe should be rightly known). But then Series 400 was introduced a couple of years ago with the Virtuosoâ„¢. That was fantastic - the desk would give you a little graphic of your system, with exactly what kind of light was connected to which port on each Smart Repeaterâ„¢ and what address it was set to. Ah well...

 

Maybe Genlyte will take a more enlightened :D (sorry) view and realise that if you control the system (like M$) it doesn't matter who makes the kit, but somehow I doubt it.

 

Both DMX and Ethernet are classic examples (like Socapex & Camlok) of how this business borrows technology from elsewhere, because it is too cheap to develop its own. I suppose that the major manufacturers could get together and develop new core technologies, but why bother cooperating when you can fight each other instead ?

 

As a result DMX has spread like a nasty little rash all over the lighting business; an 8-bit technology introduced when every other industry on the planet was introducing 16 bit and then "upgraded" to 16 bit just as everybody else was moving up to 32. Originally intended as a dimmer only protocol, despite the fact that moving lights were very clearly on the horizon at the time. Thanks, USITT.

 

Now Ethernet is set to do the same. It's not just the converters that are a problem - those naff little connectors are never going to take any serious touring abuse, simply because they were never intended too. Not only that: don't forget that the Ethernet signal only gets as far as Dimmer City or (maybe) the truss, before it gets converted back to... DMX ;) The "industrys" love of cheap, second-hand, ageing technology means that Ethernet only systems are decades away and in the meantime the business will be stuck with this ropey, hybrid system - a situation that will probably still exist by the time that most of todays students are about to retire.

 

Even when Ethernet only systems become a reality, you can bet that every other industry on the planet will be tranferring data with something 64 times faster - maybe even direct from thought power alone.

 

Or am I just whinging ? Thought so... I'll get my coat...

[Ike]

Now Ethernet is set to do the same.  It's not just the converters that are a problem - those naff little connectors are never going to take any serious touring abuse, simply because they were never intended too.

Neutrik make an Ethernet connector (basically a RJ-45 in an XLR housing) which looks quite tough and roadworthy however the plug at least is not compatible with the standard RJ-45 socket.

 

As for the rest of your points I agree totally. Is Ethernet not seen as quite old fashioned now compared to WiFi etc?

 

Maybe the next step should be wireless control for dimmers and moving lights, or maybe sending of control signals over mains wiring? :D

 

 

Ike

[peter]

Isn't part of the reason for using Ethernet that you can link in remote consoles, network based PC emulators (WYSIWYG etc) and also something to do with the fact that an RJ45 Networking card for a PC (after all, thats what a lighting desk is, essentially) costs around £5, compared to a DMX output card which can set you back £150? :D

 

As regards Wireless data, I certainly wouldnt rely on it for controlling moving lights - we had a simple video send/reciever system on a show last week which worked fine in rehearsal. Introduce an audience with mobile phones and it went haywire - when the projector wasnt displaying "no signal", it had interference on the screen. A moving head is a much more sensitive piece of equipment and costs a lot more than most video projectors. Dirty DMX is something we all dread, so why add a weak link into the system (after all, radio, by nature isnt going to be 100% reliable) which could result in the problems of the older moving lights?

 

Not sure if this makes sense.

 

 

PEter

[Ben]

it depends on the type of wireless network you set up, there is now the ability to se up wireless networks that are faster adn more relaiable than most of the wire networks used especialy in the theatre where lets face it we are so far behind every other industry that we may as well be going backwards.

 

think about the fact that a company like UPS have a full wireless network set up and they deal with large amounts of info that is worth litrally millions every day and if we were to put the same amount down a system used in the theatre, none of us would be surprised when the whole thing crashes. the problem is not introducing week links, its the fact that there is not the funding to spend on strong links.

 

as for ethernet, if you dont run it in now then you will probably be asking for money in 2 years time to put it in as not all but a lot of new features will probably be using it. if this is not the case then I think the question has to be asked: why, when there is a better system out there and in use already, are we still making items with an out dated system. you dont expect to find a computor today that takes punch cards and is the size of a truck simply because the technoledgy is there to do better. the thing that should be asked is why are we not making use of that technoledgy?

 

sorry if this makes little (if any) sense but its been a long day and an intoxicating night.

 

Ben

[Grum]

Ben,

 

I agree, we should be making more use of technology but to use your example at the moment its like having a super fast computer that can ONLY talk to other super fast computers of the same make (think Apple 10 years ago) and having to buy several boxes made by that company to create punch cards so that you can still control your old machines.

The result? Just like personal computers in the early eighties we are setting ourselves with kit that won't work with anything in as little as 2 or 3 years when the manufacturers finaly get together and decide on one universal protocol (like artnet) and then everyone with a Strand (and I use Strand only as an example) desk is left just like an Amstrad owner. They own a perfectly good machine that can do exactly what it was built for but when it goes wrong........

 

Yes, when you do your install run in the cable, take advantage of such things as synchro link via ethernet, but don't spend money on a system that requires boxes at every outlet to convert your signal from one protocol to another. More importantly, DO NOT install a system that limits you to one manufacturer for example a desk that uses its own ethernet protocol to talk to dimmers and movers from the same company.

 

I only say this because I am in a position to see so many theatres being spec'd by different consultants who all spec differend manufacturers kit and I know that they are not going to care when in a couple of years your desk dies in the morning before your show and you cannot hire anything that will work with the "integrated system" they sold you. In fact the chances are they will just turn around and tell you that the kit is out of warranty but they can sell you a new one if you want. The chances are that when you go to your boss and tell him this you will be told you have got to make do because there just isn't the money for a new desk (this applies even more so if you work in a council venue).

 

And as for the idea of using wireless gizmos.......... Don't even get me started.

 

G

 

P.S.

I am not trying to hold things back but hoping that we can make things move forward in a more unified manner.

[Ben]

I agree that we need a uniform way of addressing this problem but if we go back to my example of the computor, we now have windows (dont get me started on that) which is not compatable with any other system unless yo spend alot of money, however you can get a printer for example that can run on both windows and macs and can be set up in linix. the way all these systems work is very different and even different versions of windows are not often compatable but the end item (eg printer or moving light) can understand what system it is being used with.

maqybe the answer is to have a system similar to the plug and play method on computors, where once connected you have to tell the desk to "install" the drivers, but the drivers are all in the desks memory already and all it does is patch the driver to the item in a similar way to a soft patch.

desks would be compleatly upgradeable by use of disk (or cd because we really should be making desks with cd writers to back up)

 

this way you could use whatever system you wanted and it would give more scope to increase in the future without making lanterns or desks obsolite.

 

well just one idea there, I have others and if others think its a god idea I have a friend in intel I could ask how easy it would be to have a prototype built and investigated to see if it really works.

 

any other ideas?

Ben

[Grum]

So, now that you have had time to think about it here is my final question.

 

As a project engineer my job is to fulfill a consultants spec (this usualy has nothing to do with what the people working in the venue want) but I always try to ensure that whatever they (the consultant) ask for the solution I offer is the best one for those working in the venue.

 

I asked the question because I have my own views on the subject but wanted to talk to as many end-users as I could. From the response I have recieved am I right in saying that....

 

Ethernet is (at the moment) the best technology the industry has to offer and is the way of the future like it or not. When designing a system you would prefer increased compatibility rather than a fully integrated system, but you would rather a full ethernet system with boxes to give you DMX where you want it than a DMX network with a decent patch bay and several splitters and ethernet cabling ready to use in the future.

 

Is this right?

 

As for the "plug and play" idea, when I next talk to the people in R&D I will mention the idea.

[robloxley]

Surely the solution when starting from scratch is to decide what equipment you want to install (desk, dimmers, outstations etc.) and put in the appropriate control cabling to make it all work? This is dictated by the equipment manufacturer, and the M&E engineer / consultant specs a suitable cabling system based on the specific equipment. If you are already putting in a structured cabling system for phones, computers etc., then putting in some more points at the same time for lighting may be more cost effective than a separate DMX system. However until you know the specific circumstances and equipment to be used, how can you guess what cabling will be needed? Especially years down the line...?

 

I don't see any reason to upgrade to 'ethernet' from existing DMX systems (or to DMX from D54 etc.) unless you have a specific need - e.g. to install a particular piece of equipment etc.

 

For example, the theatre I work in mainly, still runs its dimmers on D54 (from a 520 via D54 demux boxes for the installed dimmers and via the D54 input of our portable Act6s). A few years ago we added a pair of DMX tielines to the LX box and a splitter to points around the stage, to run moving lights and other DMX items - the dimmers were left on D54 as it works fine. More recently we have installed cat5 cabling ('ethernet') to the LX and sound boxes and stage area - this was to provide internet access in the boxes and at stage level for conference use. In the near future, refurbishment will extend the cat5 cabling to give telephone points in the stage area and boxes.

 

If in the future we want to run the lighting system down cat5, the cable to the LX box and stage will be there, but it was put in for other reasons. Also we haven't speculatively put cat5 into the dimmer room as, when our dimmers/demux boxes die at some unspecified point, who can tell what type of cabling will be needed?

 

So in summary - put in what you need today; don't put in what you think you may need in the future, as how can you tell?

[themadhippy]

wot another control protocol? weve olny just put DMX in,all this moden technolegys too much for the hippy,next thing youll be telling me theres such a thing as digatal audio

[Ben]

I don't see any reason to upgrade to 'ethernet' from existing DMX systems (or to DMX from D54 etc.) unless you have a specific need - e.g. to install a particular piece of equipment etc.

Rob

 

the point is not so much "well its there so lets upgrade" so much as its a new venue so what should I install?

 

are you saying that I should work out what exactly the incoming companies to the space will be using so I can spec the new building for it? thats mad, the chances are that in 6 months yime when a company comes into my new space with a new peice of kit I wont have the time or the budget to install a new system for them to be able to use it, thats just the sort of thing that should be in the original spec.

 

I dont think anyone is saying the technoledgy is there so we should all upgrade... because I agree thats not the right thing to do BUT if you had put in the cables for telephone points you would not have to spend as much having to call people in to do yet another referb.

 

this is just whiy I think that my idea about a plug and play system would be perfect, you could just plug whatever you wanted in to anywhere it was needed adn then tell the desk that its there and what method it is using to comunicate, this way upgrades would be needed less frequently, and new equipment would be usable with older systems.

 

Ben

[robloxley]

the point is not so much "well its there so lets upgrade" so much as its a new venue so what should I install?

 

are you saying that I should work out what exactly the incoming companies to the space will be using so I can spec the new building for it? thats mad, the chances are that in 6 months yime when a company comes into my new space with a new peice of kit I wont have the time or the budget to install a new system for them to be able to use it, thats just the sort of thing that should be in the original spec

OK - let me say I don't know the specifics of this or any particular venue, and on that basis:

I would start in the original spec by putting in what cabling you will need to run the house desk, dimmers and any other proposed equipment. If this is plain DMX then so be it, if you need to go for an ethernet-based solution (e.g. for remote video nodes or multiple desks/servers) then go for it. Obviously cost is a major factor.

 

As for incoming companies' equipment, yes I am saying that you need to put in a system that will accomodate it. However, nobody has a crystal ball and can tell what protocols are going to be in use in years to come or what cabling they will need. But, if people come in next week with a pile of moving lights, they will want some DMX feeds - whether this is plain DMX out of the back of the desk via splitters or DMX from nodes on an ethernet system frankly doesn't matter to them.

 

They are unlikely to come in next week with anything (except computers and I'll come back to that) that needs a cat5 feed specifically; maybe they will in 5 years time but maybe by then it'll be cat7 and your lighting cat5 network will be obsolete without ever being used! Given that the future is unpredictable, unless you have other reasons (or lots of money) to put in a cat5 solution on top, I'd stay with just a simple DMX system if it does everything you want it to do.

 

Having said all of this, you may well - on the non-theatre-specific side of the services - be putting in structured cabling to the LX box and stage anyway (for phones, computers, printers etc.) so extending this a bit more around the stage at the original installation time may be cost effective for potential future theatrical use.

 

At the end of the day, if you can't predict exactly what cabling will be needed in your new building in 5 or 10 years time, the more important thing is to spec the rest of the services to allow for easy upgrade in the future. This means spec'ing your trunking and conduits to have 30-50% spare space, spare blank ways on all facility panels, ease of accessibility to trunking and conduit etc. etc. When you look at it this way, pulling a few new cables is so cheap compared to laying new trunking and conduits around the building as all the original ones are full.

[Ian Ferguson]

Whilst there are several companies using the ethernet protocol in different ways the main manufacturers are all participating to reach a common standard for the use of ethernet. Strand and Artisitic licence have both said to me that once this standard is reached their products will comply with it. Artistic licence has also got software so that its products will work with Strand control desks now.

 

Also there has been published a standard for "RDM" the "plug and play" that you anticipate for DMX.

 

Artistic Licence Notes on RDM

 

Artistic licence Notes on Ethernet and DMX

 

Perhaps we are nearer to a common standard than you thought.

[Ben]

Rob,

I agree that we dont know what will be used in 5 years time btu we do know what will be used in 1 years time so if you put in the "top spec" now, you are less likly to need to upgrade for a few years at least rather than just in 6-12 months

 

but this is exactly why I am looking into a plug and play method where you could have a simple wire from the box and all the conversions are done desk side, its mainly just a software thng, see my poll on it.

Ben

[robloxley]

so if you put in the "top spec" now, you are less likly to need to upgrade for a few years at least rather than just in 6-12 months

Sorry, you forget to mention money was no object. In that case, yes, certainly, install ethernet everywhere, DMX too - and maybe some fibre links around the place for when they might be needed as well! Do point me in the direction of projects where someone will agree to put in 'top spec' now in the hope that it may be used in the future.

 

As for me, until a new protocol is standardised and widely adopted, I'll stick by my original suggestion to put in what you need today (be that plain DMX or one of the existing ethernet-based solutions if you need it), but put in the backbone infrastructure (trunking, conduit etc.) to allow for easy upgrade as and when you need it.

 

I'd argue with any suggestion that in a year's time we'll all be wanting to run everything on ethernet and throw out our DMX systems.