DiyLightAnimation
Software => Light Show Pro => Topic started by: IndianaChristmas on July 16, 2013,
-
This is my first attempt at using Pixelnet in LSP so I want to be correct when I set up my controllers.
I will be using the EtherDongle.
When using SSC's I need to drag a Pixelnet controller over to the sequencer. There will be as many PixelNet controllers added as I have SSC's (assuming each controller does not mirror another).
This seems obvious to me but just wanted to make sure. Didn't want to get going and learn I was to bring something less intuitive over to the sequencer.
Thanks. I am kicking myself already for not going to the Academy...maybe next year!
-
No you need to use the e1.31 controller. There is a setting in there for the etherdongle. The pixelnet controller is for the usb dongle with pixelnet firmware.
-
One etherdongle can supply 16,384 pixelnet channels. It does this by passing 4 universes of 4096 channels each. One active hub is used to receive one universe and pass that on to the SSCs (smart string controllers). The active hub gives you up to 16 strings and a total of 1365 pixels.
while these numbers represent the "max", you can reduce the number of hubs... and/or the number of strings.
if 16,384 is still not enough for your mega display -- you can have multiple etherdongles receiving data from your show computer. you could even go computer-less with the Conductor/Slave units.
-
My suggestion is that in LSP you do not think of them as controllers, but as elements in your display. If you are going E1.31 simply drag a controller and set up the channels that you will be using for that particular element. In your options that is where you set up your 16384 channel E1.31 universe.
Rethink what you are doing in LSP. no controllers just elements.
-
I just purchased LSP, I am following the route that rdebolt has suggested. It seemed to be the most logical way of doing it at the time. Time will tell as I am just in the beginning of things. I have to build more hardware and set some things up before I know just how long each of my flex strips will be.
-
My suggestion is that in LSP you do not think of them as controllers, but as elements in your display. If you are going E1.31 simply drag a controller and set up the channels that you will be using for that particular element. In your options that is where you set up your 16384 channel E1.31 universe.
Rethink what you are doing in LSP. no controllers just elements.
If you are saying that for a megatree use one huge controller, I would think it would be harder to manually have to calculate the start channel for each SSC. That is why I have created separate controllers for each SSC. Especially if I need to change start channels at a later date or if needing to change out SSC's if one becomes defective.
-
If a ssc or string fails, you just replace it and program the replacement ssc with the same settings that the failed one had.
As for manually calculating the addresses, you need to do that anyway to ensure that the strings don't overlap and to avoid wasting channels between strings.
The only good reasons for separating the strings that I see is if you are crossing the boundary between two pixelnet universes or you are using something like a sandevices controller that cannot handle pixels with addresses in two e1.31 universes. In both of these cases, there can be a couple of unusable addresses that you have to skip.
-
Actually in LSP when you tell it how many pixels are in the controller you are creating and what the 1st channel number is, it knows how many channels to use and numbers them automatically so it is easy to tell what the start channel needs to be for the 2nd controller.
-
If you are saying that for a megatree use one huge controller, I would think it would be harder to manually have to calculate the start channel for each SSC. That is why I have created separate controllers for each SSC. Especially if I need to change start channels at a later date or if needing to change out SSC's if one becomes defective.
That is exactly what I am saying. You don't have to manually set up a start channel for each controller in LSP. I have a spread sheet that I use for my SSC channel settings. What you do have to do is set up the Megatree (in the image editor) for each string channel start and end.
-
I use basically one controller per element with some exceptions. Mini trees, live trees, and luminaries, which are programmed 3 channel string but I group these items left and right since they are on separate universes. Very intuitive.
LSP is really flexible. This year for my mega I went as far as 1 controller per strand and then created a mega tree layer and assigned the strands to it. I have a house outline layer and am planning an arch layer where each arch has it's own controller.
Manual calculation of the channels is the hardest aspect of setting LSP up but it really isn't bad at all after you get the hang of it. Once and done. If you don't have your elements built I suggest you leave plenty of room between channels for tweaking. Also, I find it very helpful to put the channels in the controller description. Screen shot of last years attached. I printed it and used it for programming my SSCs.
-
Good point Sam. I do have 1 exception for myself as well. My spinners are set to 10 controllers so that I can get the effects that I want. Each arm is a controller.
-
I should point out that I arranged the controllers in channel order for the screen shot. Normally controllers are grouped together (same color) in a matter that is conducive to my sequencing.
Maybe it should also be said the beauty of "controller per element" is being able to add effects with the controller minimized which applies effect to all channels which very powerful. Trust me, this may all appear to be Greek to new users but it will eventually click.
-
I hope I don't confuse the issue. I'm just wondering for myself and good practice. I have dmx controllers with open channels left with in the elements, and I've set this in LSP as an open controller with the left open channels starting the next controller or element what ever works in your situation to the next start channel. This lets me add if needed to the open controller and keeps the channel counts in order. Follow?. Now if I could and I didn't need the extra channels I could set the next controller for the next channel count even if it mirrors the open controller since I'm not going to use it anyway and eliminate the open controller from my template, but I think having it there keeps things in order.
Guys I'm just thinking out loud here and maybe making to much of it. <fp. Just wondering if it would be good practice just to leave the open controller?
animal
-
That's the great thing about LSP, you can customize it to your needs. Like Animal I have my first 512 channels (DMX) and use those for my LEs and Aethers. I don't touch those channels with my RGB, but I could if I wanted. Youi can set it up for what ever you use to keep things straight for you. I don't necessarily do it the way that it was designed... I just do what works for me.
-
I don't think the first idea is bad especially with dmx controllers which have a rigid channel count. However, your second idea would end up confusing to me when it comes time to modify the setup in the future to add stuff. The problem becomes a domino effect and you would have to reconfig all controllers to account for the change. Something I try to minimize. That is if I understand you correctly. IE - Last year I left plenty of room between channels but this year I have to reconfig to stay under my license's channel count of 8,192. :o
-
Agreed
. my head hurts
animal,
-
In response to Keith's initial input.
The E1.31 controller has 512 channels (or pixels). If I had more than 512 I would just add a second E1.31 controller and start channel count at 513 (or where ever that next 512 might start).
Would it be advisable to create virtual controllers to "Break" out the individual elements or simply create layers for that? I suppose the layer method is better as I would have to create those anyway.
Thanks.
-
The Ether Dongle flashed for pixelnet is actually 16384 channels. 4 universes of 4096 channels. Or if flashed for DMX is 2048 DMX channels.
-
For RGB pixels once you create the controller you change the controller type from "Light Controller" to "RGB Light Controller" and you will then set the start channel and you will see the Logical channels (number of pixels) will be smaller than the physical RGB channels(actual channels used) 3 channels per pixel.
-
In LSP an E1.31 controller can have as many channels as you want. The default number might be at 512, but you can put what ever number channel count that you need.
Sorry about my last post was not thinking about what you were saying.
-
When setting the "controller protocol" in the E1.31 controller settings, it was noted there was an etherdongle selection but I don't see that as an option.
When I was reviewing MickPat's PDF on setting up controllers in LSP he selected the "E1.31 DMX over Ethernet". So I assume the "E1.31 DMX over Ethernet" setting is the correct one.
Sorry, I'm a bit slow on the uptake here.
And on setting the Universe settings on the Active hubs...
I was planning on a matrix with 1568 pixels or (4704 channels). If the Active Hub universe is only 4096 channels I suppose I will need two hubs to make this matrix work?
This poses another question I have...If a hub can theoretically support 16, 128 node strings (physically - meaning power consumption and safety), this far exceeds the 4096 channels per universe so technically you could never really "max" out an active hub. The limit is in deed 4096 channels (or 10.67, 128 node strings).
-
Additionally...
If my current matrix needs to use two hubs with two universes, I see two options...
1. Max out hub one and use all 4096 channels and then use 608 channels from the second hub (set to a second universe of course).
2. or simply split them evenly between two hubs.
The problem I see with two hubs is no matter what there will be some "wasted" channels which isn't good because I only have the Advanced Edition of LSP and limited to 8K channels. Wasting channels isn't a good thing. So here is my last question. To save "channels" and using my first option as an example...Can I use a third hub set to the second universe and have SSC's attached to it that have channel Number set above 608? Will there be a conflict between the 2nd and 3rd hubs using the same universe settings?
-
"options" "configure outputs" "add output" "select e1.31 dmx over ethernet" and zone. Then when you go to cofigure that output you can "load defaults" and choose etherdongle from the dropdown.
-
Thanks Keith, I did that for the output configuration already.
I wasn't very clear (something I do a lot). I meant for the E1.31 controller settings on the sequencer. The place where you define your physical channels, controller color etc.
-
When setting the "controller protocol" in the E1.31 controller settings, it was noted there was an etherdongle selection but I don't see that as an option.
When I was reviewing MickPat's PDF on setting up controllers in LSP he selected the "E1.31 DMX over Ethernet". So I assume the "E1.31 DMX over Ethernet" setting is the correct one. In creating the controller itself you are correct.
Sorry, I'm a bit slow on the uptake here.
And on setting the Universe settings on the Active hubs...
I was planning on a matrix with 1568 pixels or (4704 channels). If the Active Hub universe is only 4096 channels I suppose I will need two hubs to make this matrix work? Yes
This poses another question I have...If a hub can theoretically support 16, 128 node strings (physically - meaning power consumption and safety), this far exceeds the 4096 channels per universe so technically you could never really "max" out an active hub. The limit is in deed 4096 channels (or 10.67, 128 node strings). Yes
-
There is a pdf that mickpat created a while back at the top of this thread
http://diylightanimation.com/index.php?topic=7875.msg120511#msg120511 that may help. I would only use 2 smarthubs.
-
Thanks. That is indeed the PDF I mentioned earlier.
If I can't have two hubs using the same universe settings then it would be best to trim the number of nodes in my matrix to 1365 and use only one hub.
I would hate to waste all those other channels on the second hub. My matrix (and hence the hub) will really be too far away from other SSC's to use it for other elements.
-
Additionally...
If my current matrix needs to use two hubs with two universes, I see two options...
1. Max out hub one and use all 4096 channels and then use 608 channels from the second hub (set to a second universe of course).
2. or simply split them evenly between two hubs.
The problem I see with two hubs is no matter what there will be some "wasted" channels which isn't good because I only have the Advanced Edition of LSP and limited to 8K channels. Wasting channels isn't a good thing. So here is my last question. To save "channels" and using my first option as an example...Can I use a third hub set to the second universe and have SSC's attached to it that have channel Number set above 608? Will there be a conflict between the 2nd and 3rd hubs using the same universe settings?
I don't see a problem with a duplicate hub, in fact I plan to do the same thing with a Zeus for my mega. My Zeus will be set to same universe as my 2nd hub (No2). I can really see a benefit to 2 hubs set to same universe to minimize cable runs, easy of install, etc... For some reason knowing the pixelnet data stream consists of all 4 universes helped this stuff click for me. The hub universe is selected by using the jumpers. The SS system is nothing like DMX and I think that's good thing.
Also, you are ridiculously close to being able to use a 4 port (512ch) to cover those 608 channels.
Personally I think with a matrix that big, no one will notice 608 missing channels... It would be cheaper, easier to install, program, etc. but that's me. Gotta give you credit, dedicating half of your channels to a matrix!
-
By 4 port you mean the passive 4 port hub correct?
If so how does that fit into the Universe as there are no jumpers on it. Would it just "extend" what ever the universe was from the active hub that fed it? Meaning the active hub could be 100' away for instance?
-
<fp. I thought there were jumpers on that thing...
But, in my research on the matter I stumbled upon Zeus being a 16 STRING controller... 6144 channels? Not sure of the details of Zeus, but seems ideal for your situation...