General DMX Question

[eviljohn2]

I'm still fairly new to this lampy business so please bear with me. I think I understand how DMX works and how the Pearl works but this bit confuses me...

 

Say I'm patching a Mac500 on the Pearl. There's a few personalities but why do only the correct ones work? I would expect that just increasing the value of a certain DMX string would affect that light attribute appropriately.

 

How does the Mac/desk know that the wrong personality has been used when the attributes would work fine if I just patched them as dimmers?

 

TIA. ;)

[Tomo]

When you control a Mac500 (or other moving light) you are only sending a set of DMX values.

 

Say the Mac500 is patched into DMX channel 1

Looking up the manual of the Mac500 tells me that the channels do the following:

1 - Shutter, Lamp control & reset

2 - Dimmer

3 - Colour wheel 1

4 - Colour wheel 2

...

11 - Pan

At this point things become complicated, as the Mac500 has several different personalities with different numbers of channels.

12 can either be Pan Fine (least sig byte (LSB) of 16-bit number for pan) or Tilt

13 either doesn't exist, is Tilt or is the pan & tilt speed (for vector programming, not used much)

14 either doesn't exist, is the Tilt LSB or the effects speed.

15 and 16 only exist in Mode 4

 

Personalities are used to make the trackball do movements and to group the various features of the moving light in logical ways - they are purely to make it easier to program.

 

Therefore, if you use the wrong personality in the desk (eg Pearl), what happens is that you program all the things you want, but the light doesn't do what you expected.

 

Eg - the Mac500 is set to Mode 1, but the Pearl is set to use the Mode 4 personality.

When you pan the light, you find that the tilt will move as well because you are giving the LSB of Pan to the Tilt channel.

 

In answer to your last question - the desk and Mac have no way to figure out if you are using the wrong personality. The desk does exactly what you tell it to, and the Mac does exactly what the desk tells it to.

Except this won't be what you want if the personality is wrong.

 

Clear as mud?

[eviljohn2]

Thanks for that Tomo.

 

What you describe is exactly what I would expect to happen, only it doesn't seem to work that way when I'm trying it.

 

Say you had a Mac 500 starting at 1 through to 11 but used some other fixture personality from 1 through to some other value. By increasing the DMX value of each string on the desk I would expect something to happen at the Mac end (obviously not necessarily the right thing) but how come the light just doesn't respond at all? ;)

 

I understand that there's no real reason to try this in practice but I'm just trying to get a handle on how the whole thing works if you follow. :)

[niclights]

If you have a Mac500 addressed to DMX 001 and you patch, say, 11 dimmers ranging from 001 to 011 they will control the Mac. The same would be true of patching a fixture using similar number of channels. Your assumptions are correct, just your observations or method of experiment must have gone wrong! The desk has no idea what it is controlling and the fixture has no idea what is controlling it. The only restrictions would perhaps be in the personality file. ie. for each attribute (read channel in this case) you can specify it's min & max values, whether it is inverted etc. etc.

 

You might find it easier to understand by editing a personality file on a PC. They aren't too complicated and have notes explaining the syntax. Essentially a personality file defines the number of channels to be used, to which attribute buttons each channel (offset of patch start address) is assigned and things like max/min value and whether the attribute is HTP or LTP etc.

 

HTH

[Tomo]

Assuming it's definately got power (there are LEDs to check with), and definately plugged in with a good cable (check with a dimmer you know works):

 

The most probable reason is that you've set it to Martin protocol.

Go through the menus and check that it's set to DMX-512

Hit <MENU>, then find PSET, and choose one of the DMX options, instead of Martin, hit <Enter> to confirm

Then find dAdr, hit <Enter>, and set the DMX address.

If it says mAdr instead of dAdr, it's still in Martin protocol.

 

Alternatively:

Most Martin fixtures do nothing until they have reset and the lamp has struck. (I'm not certain if the Mac500 needs the lamp struck as it's years since I used one)

Usually they are set to reset and Lamp On when powered, but not always.

You may have to send the Lamp On and/or reset command on Channel 1 before anything else responds.

 

Or: (Least likely as I think the Pearl nevers sends that few channels)

The DMX standard doesn't require that you send all 512 channels, and many desks only transmit the patched channels to save on processing power.

So it could be that you aren't sending enough channels to the Mac (it needs 16, you send only 10)

Various fixtures respond in various ways to a lack of control channels - most deal with it by doing nothing as it's not part of any standard!

[paulears]

I think you may be coming at this with the wrong head on, so to speak.

 

You are quite correct that if you patched the desk as ordinary dimmer channels, then you could indeed work every function on the mac. This is how many people with basic controllers manage. The real problems are in very basic control difference. To work dimmer channels, the faders tend to work on what's called HTP - highest takes priority (or precedence) Two faders can work the same dimmer, but the result is whichever fader is highest is the one that is being used. If you use movers like this, then if you pull down the master fader, not only does it get dimmer, but it turns towards the left, lowers down, the colours change and the gobos alter - they all go down. So like tomo says, on a proper moving light desk, it abandons this method of working apart from brightness and moves to LTP, as in the 'L' = latest. these get treated individually, and all the desk has to do is change one value to another and things happen. Only this change is recorded, everything else remains the same. Now part of the 'personality' or 'fixture library' maps which DMX channel does what, and sticks control onto an appropriate button, fader, knob, trackball or whatever. Tomo mentioned mode - this is set on the macs and uses up some extra DMX channels to help make the movement smooother. If you use the wrong personality, two things happen. The obvious one is that the wrong amount of DMX channels get eaten up. As in, a basic scanner with pan, tilt, col, gobo and shutter - it takes just 5 channels, so the DMX output would be putting out pan messages on channels 1,6,11 etc - swap this unit for a mac and you are using up a chunk of 12 or more channels for the one fixture.

 

If, like me, you end up writing your own fixture files and you get it wrong, then like one I did last week, pan wasn't next to tilt - tilt came up on the colour menu. I could sort of work it, but no way was hitting the button labelled red going to work. I had to put it right.

 

So I guess the answer to why you need different personalities is simply so the desk can do all the features it's design for. you could use dimmer channels but how ould you grab say a dozen or so movers and change every colour to green. You'd have to do the maths and then select channels 3,7,34,56,89 (I made those up - I couldn't work it out in my head) What a pain. Thank god this isn't required.

 

as for why it doesn't appear to work - well getting pan and tilt to match are the obvious ones you'll spot - but many attributes amy well start at zero, whcih will be no gobo, no colour, no intensity etc - so although it is getting data, it just won't appear to do anything. My experience of getting it wrong is that something happens, usually a faint whirr inside.

 

Does this help?

[eviljohn2]

Many thanks for all of the detailed and helpful answers guys; from what you've said I don't think my brain's barking up the wrong tree but that:

 

Your assumptions are correct, just your observations or method of experiment must have gone wrong!

 

Thanks also Paulears for the point about lowering the master fader affecting every attribute if the fixture was patched as dimmers (not something I'd ever really want to try but useful in an emergency I suppose) as I hadn't thought about that and it's not something to find out at the last minute! ;)

[Tomo]

Thanks also Paulears for the point about lowering the master fader affecting every attribute if the fixture was patched as dimmers (not something I'd ever really want to try but useful in an emergency I suppose) as I hadn't thought about that and it's not something to find out at the last minute! ;)

It's something that I did find out at the last minute a few years back. (Into The Woods - sorry Colin!)

Running a rig including Mac300s on a generics desk (Celco Explorer - don't do this! It triples, or maybe quadruples your plotting time)

I pulled down the GM on the curtain call on the last night of the run because it was obvious that there would be an encore.

Brought it back up for the encore, only to have the Mac300s sweep across the audience into position.

Oops.