DMX vs ACN

[dave e]

Hello,

Im a student at the RSAMD studying Technical and Production arts. As part of my course I am required to do a research project. Because of my interest in lighting I have decided to research the comparisons, advantages and disadvantages between DMX and ACN.

 

I am looking for any information that may not be readily available on the subject and for personnal opinions from those who will come in to contact with ACN in the near future.

 

I remember their being a forum on this subject a few months back and I followed it with interest but in my idiocy I forgot to ask those who made interesting posts, whether I could use their views in my research - and so if anyone has any thoughts about ACN or the problems in DMX 512, could they please email them to me at devans@rsamd.ac.uk.

 

Also if anyone knows of any ussefull websites that I may be able to use in my research then I would be much obliged if they could pass them on to me.

 

Cheers in advance.

[olistockman]

I haven't got this far in the world of protocols yet, but as a reminder to all...you can search previous threads, and, by tweaking the settings at the bottom of the forum pages, you can look back over the past few months, rather than past month....you should then be able to get in contact with the people who previously covered this topic

HTH

 

O

[dave e]

well I tried using the advanced search features up 365 days ago and couldnt find anything matching DMX or anc - so I thought I put up a post.

 

cheers for the suggestion

[martinw]

and so if anyone has any thoughts about ACN or the problems in DMX 512, could they please email them to me at devans@rsamd.ac.uk

It would be better if people just replied to the group, so we can all read the comments. Requesting replies by email to a question posted on a public group is considered rather poor form - certainly on usenet anyway.

 

Martin

[dave e]

im sorry - I didnt mean to cause any offence.

[martinw]

None taken :blink: - it just means that everyone can read and debate the replies presented, and the answers get stored in this board, so the next person who wants to ask the same question has somewhere to look.

 

Martin

[peter]

If you wish to recieve an email when a post is made in this topic, you can use the Track This Topic feature which is located under the notelet icon (between "Add Reply" and "New Topic"). From there you can also Download and Print topics, in friendly formats.

 

Now, back to the topic in hand...

[Steve B]

Well....

 

Chiming in from the left side of the pond....

 

Some differences between the protocols:

 

1) DMX is a one way communication sending a value for 512 separate DMX channels over one cable/wire. It has a limit of 30 listening devices without having to re-send via a splitter. It has no capability for error correction, nor is it "generally" used for bi-directional data - thought the original intention of the 5 pin connector/cable was to allow manufacturers to utilize the 2nd pair of wires for device talkback (not used in practical applications). It's a fairly robust signal though, with good fault tolerance - it generally either works or it doesn't. The exception would be poor device termination where the signal gets bounced back up the line as a ghost signal that gets re-read by devices/dimmers and causes flicker or other problems.

 

2) RDM - Remote Device Management is actually the next generation that will use the basic DMX signal for value transmission, while allowing talkback for smart devices so equipped and will utilize the existing DMX cabling as well as DMX over Ethernet systems.

 

3) ACN is the almost next communication amongst lighting devices. It is based on an Ethernet system, most likely using RJ45 connectors and using Cat5e or better cabling. Moving to Ethernet allows for significant savings in cable and installation costs, as well as utilizing inexpensive, yet robust devices for signal management that allow for greater fault detection and isolation. In theory, it will be bi-directional with the devices able to talk to the controller(s) in such a way as to allow automatic and fast remote device management and configuration, saving significant time in setup of medium to large systems.

 

A good brief description can be found at:

 

http://www.pathwayconnect.com/ahead.html

 

Steve B.

[martinw]

A few observations from an engineer and occasional user:

 

State and Delta systems

From an engineering level view, the two systems differ in that DMX is a 'State' based protocol, where as ACN is primarily a 'Delta' based protocol. By this, I mean that DMX continuously re-transmits the entire lighting state (up to 44 times per second), regardless of if any of the channels are changing in value. The 'Delta' nature of ACN means that a console should only send data when something on the rig is changing, and this means that if nothing is happening, then there is very little traffic on the network. This technique does however create wildly unpredicatable levels of network traffic, which can go from nothing to virtual saturation at the press of a 'Go' button.

 

Errors and Retransmission

DMX doesn't have any system of checking for and correcting errors in the data stream, but the continuous retransmission mostly overcomes this problem, because if corrupted data is received in one packet, it will get overwritten by the next packet very soon after.

In an ACN system, if a packet gets corrupted on the network and lost at some point, this needs to be detected so that the packet gets re-sent, otherwise a fixture will remain in its previous (now incorrect) state.

 

Tracking and Vector fades

When a DMX console fades a channel from one level to another, it has to generate many levels in between the start and finish points in order to create a smooth crossfade. In each DMX packet, that channel level will advance towards its target, and the fixture will 'track' the changing level, moving towards the desired postion.

ACN has the ability to generate the fade in the fixture, rather than the console, which will tell the fixture to go to a particular position in a specific time. The fixture can also be told to update the console with its current position periodically during the fade. This system works well for simple linear point-to-point fades, or fades where a dimmer is implementing a fade law, however when generating effects like circles, the console will still have to generate a tracking style output for the fixture to follow.

 

Martin

[Ellis]

ACN is more than a next generation control protocol. It is the starting point for a new suite of protocols whick may reach well beyond the lighting world.

 

As a software developer who has worked with real-time industrial controls I am very interested in the protocol and may consider ising it a a future non theatre related job if it is suitable.

 

Considering the amount of discussion about the correct cable to use for DMX, and the number of people who do not terminate! the physical layer limitations and pitfalls of ethernet are going to cause problems!

 

The maximum cable length for 100BASE-TX is 100m. Larger venues will need switches and routers (Max. 5 segments between routers!) or even fibre-optic backbones to cover the distance from control box to stage to dimmer room to rig...

 

Anyone who is exposed to a data network knows how unreliable that they can be in the wrong hands (like my employers IT department). Also bear in mind that any council run venue may come under the council IT department or, god forbid, and outsourced provider. If they get their hands on an ACN network, god help us.

 

I hope this (slightly jaded) comment is of use.

 

Regards

 

Ellis

[dave e]

Comments like this are very useful - it shows there is argument against the change as well as for the change to Ethernet. One thing im sure of though is how implementable such a system would be in a multi venue building.

 

If DMX equipment would be cheaper, then surely having two or three venues requiring ethernet would be very expensive, but if one venue was ethernet and the others DMX 512 - would this not cause problems with flexibility between venues?

[TeeJay]

Depending on how in-depth you want to go with this, it may be worth checking out The Show-Control Mailing List.

 

I'm sure there was a long thread on ACN there during spring but I'd recommend it in any case as a first port-of-call for show control info-seekers (after the Blue Room of course )

[martinw]

There is a discussion on this subject over at the Lightnetwork at present.

 

Martin

[dave e]

well just joined the show control mailing list and had a good trawl through the LightNetwork thread so cheers for all the comments so far.

Dave

[andystone]

Personally I don't think ACN (or any of the currently available lighting Ethernet protocols) will ever completely replace DMX - certainly not for a long time.

 

As already pointed out Ethernet cabling is in many ways more complex than DMX, although it has the advantage that hubs and routers are readily available at very low prices compared to DMX splitters/mergers.

 

The main problem is that there is so much DMX equipment around and there is likely to be for a long time to come.

 

ACN is more a way of distributing data around a building, not just for DMX which is only a very small part of it, but for remote controls, remote video, bi-directional communication with dimmers, tracking back-up systems, easy routing of DMX data, etc, etc - at the moment most of the major lighting manufacturers already have an Ethernet based protocol for communication between their equipment, but they still utilise DMX at some stage. Problem is (and the main reason ACN should happen) is that it is almost impossible to integrate different manufacturers equipment into one system, certainly without some loss of functionality.

 

Andy